mirror of
https://github.com/EyeTrackVR/EyeTrackVR.git
synced 2025-11-04 14:39:42 +08:00
nightly update
fall back to HSF in HSRAC is not correct. TODO
This commit is contained in:
parent
1ecbde4f7b
commit
db1883929a
@ -45,6 +45,7 @@ class EyeTrackSettingsConfig(BaseModel):
|
|||||||
gui_RANSAC3DP: int = 2
|
gui_RANSAC3DP: int = 2
|
||||||
gui_HSFP: int = 3
|
gui_HSFP: int = 3
|
||||||
gui_BLOBP: int = 4
|
gui_BLOBP: int = 4
|
||||||
|
gui_skip_autoradius: bool = True
|
||||||
|
|
||||||
class EyeTrackConfig(BaseModel):
|
class EyeTrackConfig(BaseModel):
|
||||||
version: int = 1
|
version: int = 1
|
||||||
|
|||||||
@ -153,25 +153,8 @@ class EyeProcessor:
|
|||||||
|
|
||||||
self.failed = 0
|
self.failed = 0
|
||||||
|
|
||||||
|
|
||||||
self.response_list = [] #This might not be correct.
|
|
||||||
#HSF
|
|
||||||
|
|
||||||
self.cv_mode = ["first_frame", "radius_adjust", "init", "normal"]
|
|
||||||
self.now_mode = self.cv_mode[0]
|
|
||||||
self.cvparam = CvParameters(default_radius, default_step)
|
|
||||||
self.skip_blink_detect = False
|
self.skip_blink_detect = False
|
||||||
|
|
||||||
self.default_step = (5, 5) # bigger the steps,lower the processing time! ofc acc also takes an impact
|
|
||||||
# self.default_step==(x,y)
|
|
||||||
self.radius_cand_list = []
|
|
||||||
self.blink_init_frames = 60 * 3
|
|
||||||
prev_max_size = 60 * 3 # 60fps*3sec
|
|
||||||
# response_min=0
|
|
||||||
self.response_max = None
|
|
||||||
|
|
||||||
self.auto_radius_range = (self.settings.gui_HSF_radius - 10, self.settings.gui_HSF_radius + 10)
|
|
||||||
|
|
||||||
#blink
|
#blink
|
||||||
self.max_ints = []
|
self.max_ints = []
|
||||||
self.max_int = 0
|
self.max_int = 0
|
||||||
@ -179,6 +162,8 @@ class EyeProcessor:
|
|||||||
self.frames = 0
|
self.frames = 0
|
||||||
self.blinkvalue = False
|
self.blinkvalue = False
|
||||||
|
|
||||||
|
self.prev_x = None
|
||||||
|
self.prev_y = None
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
@ -209,7 +194,7 @@ class EyeProcessor:
|
|||||||
self.previous_image = self.current_image
|
self.previous_image = self.current_image
|
||||||
self.previous_rotation = self.config.rotation_angle
|
self.previous_rotation = self.config.rotation_angle
|
||||||
except:
|
except:
|
||||||
print("E")
|
pass
|
||||||
def capture_crop_rotate_image(self):
|
def capture_crop_rotate_image(self):
|
||||||
# Get our current frame
|
# Get our current frame
|
||||||
|
|
||||||
@ -255,12 +240,22 @@ class EyeProcessor:
|
|||||||
|
|
||||||
def HSRACM(self):
|
def HSRACM(self):
|
||||||
cx, cy, thresh = HSRAC(self)
|
cx, cy, thresh = HSRAC(self)
|
||||||
|
if self.prev_x == None:
|
||||||
|
self.prev_x = cx
|
||||||
|
self.prev_y = cy
|
||||||
|
#print(self.prev_x, self.prev_y, cx, cy)
|
||||||
|
# if (cx - self.prev_x) <= 45 and (cy - self.prev_y) <= 45 :
|
||||||
|
# self.prev_x = cx
|
||||||
|
# self.prev_y = cy
|
||||||
out_x, out_y = cal_osc(self, cx, cy)
|
out_x, out_y = cal_osc(self, cx, cy)
|
||||||
if cx == 0:
|
if cx == 0:
|
||||||
self.output_images_and_update(thresh, EyeInformation(InformationOrigin.HSRAC, out_x, out_y, 0, True)) #update app
|
self.output_images_and_update(thresh, EyeInformation(InformationOrigin.HSRAC, out_x, out_y, 0, False)) #update app
|
||||||
else:
|
else:
|
||||||
self.output_images_and_update(thresh, EyeInformation(InformationOrigin.HSRAC, out_x, out_y, 0, self.blinkvalue))
|
self.blinkvalue = False
|
||||||
|
self.output_images_and_update(thresh, EyeInformation(InformationOrigin.HSRAC, out_x, out_y, 0, False))
|
||||||
|
# else:
|
||||||
|
# print("EYE MOVED TOO FAST")
|
||||||
|
# self.output_images_and_update(thresh, EyeInformation(InformationOrigin.HSRAC, 0, 0, 0, False))
|
||||||
def HSFM(self):
|
def HSFM(self):
|
||||||
cx, cy, frame = HSF(self)
|
cx, cy, frame = HSF(self)
|
||||||
out_x, out_y = cal_osc(self, cx, cy)
|
out_x, out_y = cal_osc(self, cx, cy)
|
||||||
@ -294,31 +289,24 @@ class EyeProcessor:
|
|||||||
|
|
||||||
|
|
||||||
def ALGOSELECT(self):
|
def ALGOSELECT(self):
|
||||||
print(self.failed, self.firstalgo)
|
|
||||||
if self.failed == 0 and self.firstalgo != None:
|
|
||||||
print('first')
|
|
||||||
self.firstalgo()
|
|
||||||
|
|
||||||
|
if self.failed == 0 and self.firstalgo != None:
|
||||||
|
self.firstalgo()
|
||||||
else:
|
else:
|
||||||
self.failed = self.failed + 1
|
self.failed = self.failed + 1
|
||||||
|
|
||||||
if self.failed == 1 and self.secondalgo != None:
|
if self.failed == 1 and self.secondalgo != None: #send the tracking algos previous fail number, in algo if we pass set to 0, if fail, + 1
|
||||||
print('2nd') #send the tracking algos previous fail number, in algo if we pass set to 0, if fail, + 1
|
|
||||||
self.secondalgo()
|
self.secondalgo()
|
||||||
else:
|
else:
|
||||||
self.failed = self.failed + 1
|
self.failed = self.failed + 1
|
||||||
|
|
||||||
if self.failed == 2 and self.thirdalgo != None:
|
if self.failed == 2 and self.thirdalgo != None:
|
||||||
print('3rd')
|
|
||||||
self.thirdalgo()
|
self.thirdalgo()
|
||||||
|
|
||||||
else:
|
else:
|
||||||
self.failed = self.failed + 1
|
self.failed = self.failed + 1
|
||||||
|
|
||||||
if self.failed == 3 and self.fourthalgo != None:
|
if self.failed == 3 and self.fourthalgo != None:
|
||||||
print('4th')
|
|
||||||
self.fourthalgo()
|
self.fourthalgo()
|
||||||
|
|
||||||
else:
|
else:
|
||||||
self.failed = 0 # we have reached last possible algo and it is disabled, move to first algo
|
self.failed = 0 # we have reached last possible algo and it is disabled, move to first algo
|
||||||
|
|
||||||
@ -352,7 +340,6 @@ class EyeProcessor:
|
|||||||
self.fourthalgo = self.RANSAC3DM
|
self.fourthalgo = self.RANSAC3DM
|
||||||
|
|
||||||
if self.settings.gui_HSRAC == True and self.settings.gui_HSRACP == 1:
|
if self.settings.gui_HSRAC == True and self.settings.gui_HSRACP == 1:
|
||||||
print("HERE")
|
|
||||||
self.firstalgo = self.HSRACM
|
self.firstalgo = self.HSRACM
|
||||||
elif self.settings.gui_HSRAC and self.settings.gui_HSRACP == 2:
|
elif self.settings.gui_HSRAC and self.settings.gui_HSRACP == 2:
|
||||||
self.secondalgo = self.HSRACM
|
self.secondalgo = self.HSRACM
|
||||||
@ -370,12 +357,6 @@ class EyeProcessor:
|
|||||||
elif self.settings.gui_BLOB and self.settings.gui_BLOBP == 4:
|
elif self.settings.gui_BLOB and self.settings.gui_BLOBP == 4:
|
||||||
self.fourthalgo = self.BLOBM
|
self.fourthalgo = self.BLOBM
|
||||||
|
|
||||||
|
|
||||||
if self.settings.gui_HSRACP == '1':
|
|
||||||
print("HERE")
|
|
||||||
|
|
||||||
print(self.settings.gui_HSRACP, self.settings.gui_HSRAC, self.firstalgo)
|
|
||||||
|
|
||||||
f = True
|
f = True
|
||||||
while True:
|
while True:
|
||||||
# f = True
|
# f = True
|
||||||
|
|||||||
@ -27,15 +27,22 @@ Copyright (c) 2022 EyeTrackVR <3
|
|||||||
'''
|
'''
|
||||||
|
|
||||||
|
|
||||||
import cv2
|
|
||||||
import numpy as np
|
|
||||||
import timeit
|
|
||||||
from functools import lru_cache
|
|
||||||
import os
|
|
||||||
import sys
|
|
||||||
import functools
|
import functools
|
||||||
import math
|
import math
|
||||||
#HSF \/
|
import os
|
||||||
|
import sys
|
||||||
|
import timeit
|
||||||
|
from functools import lru_cache
|
||||||
|
|
||||||
|
import cv2
|
||||||
|
import numpy as np
|
||||||
|
|
||||||
|
# from line_profiler_pycharm import profile
|
||||||
|
|
||||||
|
calc_print_enable = True
|
||||||
|
save_video = False
|
||||||
|
skip_autoradius = False
|
||||||
|
skip_blink_detect = False
|
||||||
|
|
||||||
# cache param
|
# cache param
|
||||||
lru_maxsize_vvs = 16
|
lru_maxsize_vvs = 16
|
||||||
@ -46,7 +53,6 @@ auto_radius_range = (default_radius - 10, default_radius + 10) # (10,30)
|
|||||||
blink_init_frames = 60 * 3 # 60fps*3sec,Number of blink statistical frames
|
blink_init_frames = 60 * 3 # 60fps*3sec,Number of blink statistical frames
|
||||||
# step==(x,y)
|
# step==(x,y)
|
||||||
default_step = (5, 5) # bigger the steps,lower the processing time! ofc acc also takes an impact
|
default_step = (5, 5) # bigger the steps,lower the processing time! ofc acc also takes an impact
|
||||||
response_list = []
|
|
||||||
|
|
||||||
"""
|
"""
|
||||||
Attention.
|
Attention.
|
||||||
@ -303,12 +309,6 @@ class HaarSurroundFeature:
|
|||||||
return kernel
|
return kernel
|
||||||
|
|
||||||
|
|
||||||
def to_gray(frame):
|
|
||||||
# Faster by quitting checking if the input image is already grayscale
|
|
||||||
# Perhaps it would be faster with less overhead to call cv2.cvtColor directly instead of using this function
|
|
||||||
return cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
|
|
||||||
|
|
||||||
|
|
||||||
@lru_cache(maxsize=lru_maxsize_vs)
|
@lru_cache(maxsize=lru_maxsize_vs)
|
||||||
def frameint_get_xy_step(imageshape, xysteps, pad, start_offset=None, end_offset=None):
|
def frameint_get_xy_step(imageshape, xysteps, pad, start_offset=None, end_offset=None):
|
||||||
"""
|
"""
|
||||||
@ -430,140 +430,171 @@ def conv_int(frame_int, kernel, xy_step, padding, xy_steps_list):
|
|||||||
np.multiply(kernel.val_in, inner_sum, dtype=np.float64, out=response_list)
|
np.multiply(kernel.val_in, inner_sum, dtype=np.float64, out=response_list)
|
||||||
response_list += kernel.val_out * outer_sum
|
response_list += kernel.val_out * outer_sum
|
||||||
|
|
||||||
# min_response, max_val, min_loc, max_loc = cv2.minMaxLoc(self.response_list)
|
# min_response, max_val, min_loc, max_loc = cv2.minMaxLoc(response_list)
|
||||||
min_response, _, min_loc, _ = cv2.minMaxLoc(response_list)
|
min_response, _, min_loc, _ = cv2.minMaxLoc(response_list)
|
||||||
|
|
||||||
center = ((xy_steps_list[0][min_loc[0]] - padding), (xy_steps_list[1][min_loc[1]] - padding))
|
center = ((xy_steps_list[0][min_loc[0]] - padding), (xy_steps_list[1][min_loc[1]] - padding))
|
||||||
|
|
||||||
frame_conv_stride[:, :] = response_list
|
frame_conv_stride[:, :] = response_list
|
||||||
# or
|
# or
|
||||||
# frame_conv_stride[:, :] = self.response_list.astype(np.uint8)
|
# frame_conv_stride[:, :] = response_list.astype(np.uint8)
|
||||||
|
|
||||||
return frame_conv, min_response, center
|
return frame_conv, min_response, center
|
||||||
|
|
||||||
|
|
||||||
|
# @profile
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
timedict = {"to_gray": [], "int_img": [], "conv_int": [], "crop": [], "total_cv": []}
|
||||||
|
# I'd like to take into account things like print, end_time - start_time processing time, etc., but it's too much trouble.
|
||||||
|
|
||||||
|
# For measuring total processing time
|
||||||
|
main_start_time = timeit.default_timer()
|
||||||
|
|
||||||
|
rng = np.random.default_rng()
|
||||||
|
cvparam = CvParameters(default_radius, default_step)
|
||||||
|
|
||||||
|
cv_mode = ["first_frame", "radius_adjust", "init", "normal"]
|
||||||
|
now_mode = cv_mode[0]
|
||||||
|
|
||||||
|
radius_cand_list = []
|
||||||
|
|
||||||
|
# response_min=0
|
||||||
|
response_max = None
|
||||||
|
response_list = []
|
||||||
|
|
||||||
def HSF(self):
|
def HSF(self):
|
||||||
|
|
||||||
|
global now_mode
|
||||||
|
global response_list
|
||||||
|
global radius_cand_list
|
||||||
|
global response_max
|
||||||
|
# default_radius = 15
|
||||||
frame = self.current_image_gray
|
frame = self.current_image_gray
|
||||||
if self.now_mode == self.cv_mode[1]:
|
if now_mode == cv_mode[1]:
|
||||||
|
prev_res_len = len(response_list)
|
||||||
prev_res_len = len(self.response_list)
|
|
||||||
# adjustment of radius
|
# adjustment of radius
|
||||||
if prev_res_len == 1:
|
if prev_res_len == 1:
|
||||||
# len==1==self.response_list==[self.settings.gui_HSF_radius]
|
# len==1==response_list==[default_radius]
|
||||||
self.cvparam.radius = self.auto_radius_range[0]
|
cvparam.radius = auto_radius_range[0]
|
||||||
elif prev_res_len == 2:
|
elif prev_res_len == 2:
|
||||||
# len==2==self.response_list==[self.settings.gui_HSF_radius, self.auto_radius_range[0]]
|
# len==2==response_list==[default_radius, auto_radius_range[0]]
|
||||||
self.cvparam.radius = self.auto_radius_range[1]
|
cvparam.radius = auto_radius_range[1]
|
||||||
elif prev_res_len == 3:
|
elif prev_res_len == 3:
|
||||||
# len==3==self.response_list==[self.settings.gui_HSF_radius,self.auto_radius_range[0],self.auto_radius_range[1]]
|
# len==3==response_list==[default_radius,auto_radius_range[0],auto_radius_range[1]]
|
||||||
sort_res = sorted(self.response_list, key=lambda x: x[1])[0]
|
sort_res = sorted(response_list, key=lambda x: x[1])[0]
|
||||||
# Extract the radius with the lowest response value
|
# Extract the radius with the lowest response value
|
||||||
if sort_res[0] == self.settings.gui_HSF_radius:
|
if sort_res[0] == default_radius:
|
||||||
# If the default value is best, change self.now_mode to init after setting radius to the default value.
|
# If the default value is best, change now_mode to init after setting radius to the default value.
|
||||||
self.cvparam.radius = self.settings.gui_HSF_radius
|
cvparam.radius = default_radius
|
||||||
self.now_mode = self.cv_mode[2] if not self.skip_blink_detect else self.cv_mode[3]
|
now_mode = cv_mode[2] if not skip_blink_detect else cv_mode[3]
|
||||||
self.response_list = []
|
response_list = []
|
||||||
elif sort_res[0] == self.auto_radius_range[0]:
|
elif sort_res[0] == auto_radius_range[0]:
|
||||||
self.radius_cand_list = [i for i in range(self.auto_radius_range[0], self.settings.gui_HSF_radius, self.default_step[0])][1:]
|
radius_cand_list = [i for i in range(auto_radius_range[0], default_radius, default_step[0])][1:]
|
||||||
# self.default_step is defined separately for xy, but radius is shared by xy, so it may be buggy
|
# default_step is defined separately for xy, but radius is shared by xy, so it may be buggy
|
||||||
# It should be no problem to set it to anything other than self.default_step
|
# It should be no problem to set it to anything other than default_step
|
||||||
self.cvparam.radius = self.radius_cand_list.pop()
|
cvparam.radius = radius_cand_list.pop()
|
||||||
else:
|
else:
|
||||||
self.radius_cand_list = [i for i in range(self.settings.gui_HSF_radius, self.auto_radius_range[1], self.default_step[0])][1:]
|
radius_cand_list = [i for i in range(default_radius, auto_radius_range[1], default_step[0])][1:]
|
||||||
# self.default_step is defined separately for xy, but radius is shared by xy, so it may be buggy
|
# default_step is defined separately for xy, but radius is shared by xy, so it may be buggy
|
||||||
# It should be no problem to set it to anything other than self.default_step
|
# It should be no problem to set it to anything other than default_step
|
||||||
self.cvparam.radius = self.radius_cand_list.pop()
|
cvparam.radius = radius_cand_list.pop()
|
||||||
else:
|
else:
|
||||||
# Try the contents of the self.radius_cand_list in order until the self.radius_cand_list runs out
|
# Try the contents of the radius_cand_list in order until the radius_cand_list runs out
|
||||||
# Better make it a binary search.
|
# Better make it a binary search.
|
||||||
if len(self.radius_cand_list) == 0:
|
if len(radius_cand_list) == 0:
|
||||||
sort_res = sorted(self.response_list, key=lambda x: x[1])[0]
|
sort_res = sorted(response_list, key=lambda x: x[1])[0]
|
||||||
self.cvparam.radius = sort_res[0]
|
cvparam.radius = sort_res[0]
|
||||||
self.now_mode = self.cv_mode[2] if not self.skip_blink_detect else self.cv_mode[3]
|
now_mode = cv_mode[2] if not skip_blink_detect else cv_mode[3]
|
||||||
self.response_list = []
|
response_list = []
|
||||||
else:
|
else:
|
||||||
self.cvparam.radius = self.radius_cand_list.pop()
|
cvparam.radius = radius_cand_list.pop()
|
||||||
|
|
||||||
radius, pad, step, hsf = self.cvparam.get_rpsh()
|
radius, pad, step, hsf = cvparam.get_rpsh()
|
||||||
|
|
||||||
# For measuring processing time of image processing
|
# For measuring processing time of image processing
|
||||||
cv_start_time = timeit.default_timer()
|
cv_start_time = timeit.default_timer()
|
||||||
|
|
||||||
gray_frame = frame
|
gray_frame = frame
|
||||||
|
timedict["to_gray"].append(timeit.default_timer() - cv_start_time)
|
||||||
|
|
||||||
# Calculate the integral image of the frame
|
# Calculate the integral image of the frame
|
||||||
int_start_time = timeit.default_timer()
|
int_start_time = timeit.default_timer()
|
||||||
# BORDER_CONSTANT is faster than BORDER_REPLICATE There seems to be almost no negative impact when BORDER_CONSTANT is used.
|
# BORDER_CONSTANT is faster than BORDER_REPLICATE There seems to be almost no negative impact when BORDER_CONSTANT is used.
|
||||||
frame_pad = cv2.copyMakeBorder(gray_frame, pad, pad, pad, pad, cv2.BORDER_CONSTANT)
|
frame_pad = cv2.copyMakeBorder(gray_frame, pad, pad, pad, pad, cv2.BORDER_CONSTANT)
|
||||||
frame_int = cv2.integral(frame_pad)
|
frame_int = cv2.integral(frame_pad)
|
||||||
|
timedict["int_img"].append(timeit.default_timer() - int_start_time)
|
||||||
|
|
||||||
# Convolve the feature with the integral image
|
# Convolve the feature with the integral image
|
||||||
conv_int_start_time = timeit.default_timer()
|
conv_int_start_time = timeit.default_timer()
|
||||||
xy_step = frameint_get_xy_step(frame_int.shape, step, pad, start_offset=None, end_offset=None)
|
xy_step = frameint_get_xy_step(frame_int.shape, step, pad, start_offset=None, end_offset=None)
|
||||||
frame_conv, response, center_xy = conv_int(frame_int, hsf, step, pad, xy_step)
|
frame_conv, response, center_xy = conv_int(frame_int, hsf, step, pad, xy_step)
|
||||||
|
timedict["conv_int"].append(timeit.default_timer() - conv_int_start_time)
|
||||||
|
|
||||||
crop_start_time = timeit.default_timer()
|
crop_start_time = timeit.default_timer()
|
||||||
# Define the center point and radius
|
# Define the center point and radius
|
||||||
center_x, center_y = center_xy
|
center_x, center_y = center_xy
|
||||||
upper_x = center_x + 25 #TODO make this a setting
|
upper_x = center_x + radius
|
||||||
lower_x = center_x - 25
|
lower_x = center_x - radius
|
||||||
upper_y = center_y + 25
|
upper_y = center_y + radius
|
||||||
lower_y = center_y - 25
|
lower_y = center_y - radius
|
||||||
|
|
||||||
# Crop the image using the calculated bounds
|
# Crop the image using the calculated bounds
|
||||||
cropped_image = gray_frame[lower_y:upper_y, lower_x:upper_x] # y is 50px, x is 45? why?
|
cropped_image = gray_frame[lower_y:upper_y, lower_x:upper_x]
|
||||||
|
|
||||||
if self.now_mode == self.cv_mode[0] or self.now_mode == self.cv_mode[1]:
|
if now_mode == cv_mode[0] or now_mode == cv_mode[1]:
|
||||||
# If mode is first_frame or radius_adjust, record current radius and response
|
# If mode is first_frame or radius_adjust, record current radius and response
|
||||||
self.response_list.append((radius, response))
|
response_list.append((radius, response))
|
||||||
elif self.now_mode == self.cv_mode[2]:
|
elif now_mode == cv_mode[2]:
|
||||||
# Statistics for blink detection
|
# Statistics for blink detection
|
||||||
if len(self.response_list) < self.blink_init_frames:
|
if len(response_list) < blink_init_frames:
|
||||||
# Record the average value of cropped_image
|
# Record the average value of cropped_image
|
||||||
self.response_list.append(cv2.mean(cropped_image)[0])
|
response_list.append(cv2.mean(cropped_image)[0])
|
||||||
else:
|
else:
|
||||||
# Calculate self.response_max by computing interquartile range, IQR
|
# Calculate response_max by computing interquartile range, IQR
|
||||||
# Change self.cv_mode to normal
|
# Change cv_mode to normal
|
||||||
self.response_list = np.array(self.response_list)
|
response_list = np.array(response_list)
|
||||||
# 25%,75%
|
# 25%,75%
|
||||||
# This value may need to be adjusted depending on the environment.
|
# This value may need to be adjusted depending on the environment.
|
||||||
quartile_1, quartile_3 = np.percentile(self.response_list, [25, 75])
|
quartile_1, quartile_3 = np.percentile(response_list, [25, 75])
|
||||||
iqr = quartile_3 - quartile_1
|
iqr = quartile_3 - quartile_1
|
||||||
# response_min = quartile_1 - (iqr * 1.5)
|
# response_min = quartile_1 - (iqr * 1.5)
|
||||||
self.response_max = quartile_3 + (iqr * 1.5)
|
response_max = quartile_3 + (iqr * 1.5)
|
||||||
self.now_mode = self.cv_mode[3]
|
now_mode = cv_mode[3]
|
||||||
else:
|
else:
|
||||||
if 0 in cropped_image.shape:
|
if 0 in cropped_image.shape:
|
||||||
# If shape contains 0, it is not detected well.
|
# If shape contains 0, it is not detected well.
|
||||||
print("[WARN] HSF: Something's wrong.")
|
print("Something's wrong.")
|
||||||
else:
|
else:
|
||||||
# If the average value of cropped_image is greater than self.response_max
|
# If the average value of cropped_image is greater than response_max
|
||||||
# (i.e., if the cropimage is whitish
|
# (i.e., if the cropimage is whitish
|
||||||
if self.response_max is not None and cv2.mean(cropped_image)[0] > self.response_max:
|
if response_max is not None and cv2.mean(cropped_image)[0] > response_max:
|
||||||
# blink
|
# blink
|
||||||
|
|
||||||
cv2.circle(frame, (center_x, center_y), 10, (0, 0, 255), -1)
|
cv2.circle(frame, (center_x, center_y), 10, (0, 0, 255), -1)
|
||||||
# If you want to update self.response_max. it may be more cost-effective to rewrite self.response_list in the following way
|
|
||||||
|
# If you want to update response_max. it may be more cost-effective to rewrite response_list in the following way
|
||||||
# https://stackoverflow.com/questions/42771110/fastest-way-to-left-cycle-a-numpy-array-like-pop-push-for-a-queue
|
# https://stackoverflow.com/questions/42771110/fastest-way-to-left-cycle-a-numpy-array-like-pop-push-for-a-queue
|
||||||
|
|
||||||
|
cv_end_time = timeit.default_timer()
|
||||||
|
timedict["crop"].append(cv_end_time - crop_start_time)
|
||||||
|
timedict["total_cv"].append(cv_end_time - cv_start_time)
|
||||||
|
|
||||||
|
# the lower the response the better the likelyhood of there being a pupil. you can adujst the radius and steps accordingly
|
||||||
|
print('Kernel response:', response)
|
||||||
|
print('Pixel position:', center_xy)
|
||||||
|
|
||||||
|
|
||||||
|
if now_mode == cv_mode[0]:
|
||||||
|
# Moving from first_frame to the next mode
|
||||||
|
if skip_autoradius and skip_blink_detect:
|
||||||
cv2.circle(frame, (center_x, center_y), 10, (0, 0, 255), -1)
|
now_mode = cv_mode[3]
|
||||||
# print(center_x, center_y)
|
response_list = []
|
||||||
|
elif skip_autoradius:
|
||||||
if self.now_mode != self.cv_mode[0] and self.now_mode != self.cv_mode[1]:
|
now_mode = cv_mode[2]
|
||||||
if cropped_image.size < 400:
|
response_list = []
|
||||||
pass
|
else:
|
||||||
|
now_mode = cv_mode[1]
|
||||||
if self.now_mode == self.cv_mode[0]:
|
|
||||||
self.now_mode = self.cv_mode[1]
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
try:
|
try:
|
||||||
self.failed = 0
|
self.failed = 0
|
||||||
@ -575,9 +606,3 @@ def HSF(self):
|
|||||||
|
|
||||||
|
|
||||||
|
|
||||||
#self.output_images_and_update(thresh, EyeInformation(InformationOrigin.FAILURE, 0, 0, 0, False))
|
|
||||||
# return
|
|
||||||
|
|
||||||
#self.output_images_and_update(larger_threshold,EyeInformation(InformationOrigin.HSF, out_x, out_y, 0, False),)
|
|
||||||
# return
|
|
||||||
#self.output_images_and_update(larger_threshold, EyeInformation(InformationOrigin.HSF, 0, 0, 0, True))
|
|
||||||
|
|||||||
@ -1,25 +1,298 @@
|
|||||||
import cv2
|
|
||||||
import numpy as np
|
|
||||||
import timeit
|
|
||||||
from functools import lru_cache
|
|
||||||
import os
|
|
||||||
import sys
|
|
||||||
import functools
|
import functools
|
||||||
import math
|
import math
|
||||||
|
import os
|
||||||
|
import sys
|
||||||
|
import timeit
|
||||||
|
from functools import lru_cache
|
||||||
|
|
||||||
#HSF \/
|
import cv2
|
||||||
|
import numpy as np
|
||||||
|
|
||||||
|
# from line_profiler_pycharm import profile
|
||||||
|
|
||||||
|
#RANSACAHA
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
thresh_add = 20
|
||||||
|
|
||||||
|
|
||||||
|
class TimeitResult(object):
|
||||||
|
"""
|
||||||
|
from https://github.com/ipython/ipython/blob/339c0d510a1f3cb2158dd8c6e7f4ac89aa4c89d8/IPython/core/magics/execution.py#L55
|
||||||
|
|
||||||
|
Object returned by the timeit magic with info about the run.
|
||||||
|
Contains the following attributes :
|
||||||
|
loops: (int) number of loops done per measurement
|
||||||
|
repeat: (int) number of times the measurement has been repeated
|
||||||
|
best: (float) best execution time / number
|
||||||
|
all_runs: (list of float) execution time of each run (in s)
|
||||||
|
"""
|
||||||
|
|
||||||
|
def __init__(self, loops, repeat, best, worst, all_runs, precision):
|
||||||
|
self.loops = loops
|
||||||
|
self.repeat = repeat
|
||||||
|
self.best = best
|
||||||
|
self.worst = worst
|
||||||
|
self.all_runs = all_runs
|
||||||
|
self._precision = precision
|
||||||
|
self.timings = [dt / self.loops for dt in all_runs]
|
||||||
|
|
||||||
|
@property
|
||||||
|
def average(self):
|
||||||
|
return math.fsum(self.timings) / len(self.timings)
|
||||||
|
|
||||||
|
@property
|
||||||
|
def stdev(self):
|
||||||
|
mean = self.average
|
||||||
|
return (math.fsum([(x - mean) ** 2 for x in self.timings]) / len(self.timings)) ** 0.5
|
||||||
|
|
||||||
|
def __str__(self):
|
||||||
|
pm = '+-'
|
||||||
|
if hasattr(sys.stdout, 'encoding') and sys.stdout.encoding:
|
||||||
|
try:
|
||||||
|
u'\xb1'.encode(sys.stdout.encoding)
|
||||||
|
pm = u'\xb1'
|
||||||
|
except:
|
||||||
|
pass
|
||||||
|
return "min:{best} max:{worst} mean:{mean} {pm} {std} per loop (mean {pm} std. dev. of {runs} run{run_plural}, {loops:,} loop{loop_plural} each)".format(
|
||||||
|
pm=pm,
|
||||||
|
runs=self.repeat,
|
||||||
|
loops=self.loops,
|
||||||
|
loop_plural="" if self.loops == 1 else "s",
|
||||||
|
run_plural="" if self.repeat == 1 else "s",
|
||||||
|
mean=format_time(self.average, self._precision),
|
||||||
|
std=format_time(self.stdev, self._precision),
|
||||||
|
best=format_time(self.best, self._precision),
|
||||||
|
worst=format_time(self.worst, self._precision),
|
||||||
|
)
|
||||||
|
|
||||||
|
def _repr_pretty_(self, p, cycle):
|
||||||
|
unic = self.__str__()
|
||||||
|
p.text(u'<TimeitResult : ' + unic + u'>')
|
||||||
|
|
||||||
|
|
||||||
|
class FPSResult(object):
|
||||||
|
"""
|
||||||
|
base https://github.com/ipython/ipython/blob/339c0d510a1f3cb2158dd8c6e7f4ac89aa4c89d8/IPython/core/magics/execution.py#L55
|
||||||
|
"""
|
||||||
|
|
||||||
|
def __init__(self, loops, repeat, best, worst, all_runs, precision):
|
||||||
|
self.loops = loops
|
||||||
|
self.repeat = repeat
|
||||||
|
self.best = 1 / best
|
||||||
|
self.worst = 1 / worst
|
||||||
|
self.all_runs = all_runs
|
||||||
|
self._precision = precision
|
||||||
|
self.fps = [1 / dt for dt in all_runs]
|
||||||
|
self.unit = "fps"
|
||||||
|
|
||||||
|
@property
|
||||||
|
def average(self):
|
||||||
|
return math.fsum(self.fps) / len(self.fps)
|
||||||
|
|
||||||
|
@property
|
||||||
|
def stdev(self):
|
||||||
|
mean = self.average
|
||||||
|
return (math.fsum([(x - mean) ** 2 for x in self.fps]) / len(self.fps)) ** 0.5
|
||||||
|
|
||||||
|
def __str__(self):
|
||||||
|
pm = '+-'
|
||||||
|
if hasattr(sys.stdout, 'encoding') and sys.stdout.encoding:
|
||||||
|
try:
|
||||||
|
u'\xb1'.encode(sys.stdout.encoding)
|
||||||
|
pm = u'\xb1'
|
||||||
|
except:
|
||||||
|
pass
|
||||||
|
return "min:{best} max:{worst} mean:{mean} {pm} {std} per loop (mean {pm} std. dev. of {runs} run{run_plural}, {loops:,} loop{loop_plural} each)".format(
|
||||||
|
pm=pm,
|
||||||
|
runs=self.repeat,
|
||||||
|
loops=self.loops,
|
||||||
|
loop_plural="" if self.loops == 1 else "s",
|
||||||
|
run_plural="" if self.repeat == 1 else "s",
|
||||||
|
mean="%.*g%s" % (self._precision, self.average, self.unit),
|
||||||
|
std="%.*g%s" % (self._precision, self.stdev, self.unit),
|
||||||
|
best="%.*g%s" % (self._precision, self.best, self.unit),
|
||||||
|
worst="%.*g%s" % (self._precision, self.worst, self.unit),
|
||||||
|
)
|
||||||
|
|
||||||
|
def _repr_pretty_(self, p, cycle):
|
||||||
|
unic = self.__str__()
|
||||||
|
p.text(u'<FPSResult : ' + unic + u'>')
|
||||||
|
|
||||||
|
|
||||||
|
def format_time(timespan, precision=3):
|
||||||
|
"""
|
||||||
|
https://github.com/ipython/ipython/blob/339c0d510a1f3cb2158dd8c6e7f4ac89aa4c89d8/IPython/core/magics/execution.py#L1473
|
||||||
|
Formats the timespan in a human readable form
|
||||||
|
"""
|
||||||
|
|
||||||
|
if timespan >= 60.0:
|
||||||
|
# we have more than a minute, format that in a human readable form
|
||||||
|
# Idea from http://snipplr.com/view/5713/
|
||||||
|
parts = [("d", 60 * 60 * 24), ("h", 60 * 60), ("min", 60), ("s", 1)]
|
||||||
|
time = []
|
||||||
|
leftover = timespan
|
||||||
|
for suffix, length in parts:
|
||||||
|
value = int(leftover / length)
|
||||||
|
if value > 0:
|
||||||
|
leftover = leftover % length
|
||||||
|
time.append(u'%s%s' % (str(value), suffix))
|
||||||
|
if leftover < 1:
|
||||||
|
break
|
||||||
|
return " ".join(time)
|
||||||
|
|
||||||
|
# Unfortunately the unicode 'micro' symbol can cause problems in
|
||||||
|
# certain terminals.
|
||||||
|
# See bug: https://bugs.launchpad.net/ipython/+bug/348466
|
||||||
|
# Try to prevent crashes by being more secure than it needs to
|
||||||
|
# E.g. eclipse is able to print a µ, but has no sys.stdout.encoding set.
|
||||||
|
units = [u"s", u"ms", u'us', "ns"] # the save value
|
||||||
|
if hasattr(sys.stdout, 'encoding') and sys.stdout.encoding:
|
||||||
|
try:
|
||||||
|
u'\xb5'.encode(sys.stdout.encoding)
|
||||||
|
units = [u"s", u"ms", u'\xb5s', "ns"]
|
||||||
|
except:
|
||||||
|
pass
|
||||||
|
scaling = [1, 1e3, 1e6, 1e9]
|
||||||
|
|
||||||
|
if timespan > 0.0:
|
||||||
|
order = min(-int(math.floor(math.log10(timespan)) // 3), 3)
|
||||||
|
else:
|
||||||
|
order = 3
|
||||||
|
return u"%.*g %s" % (precision, timespan * scaling[order], units[order])
|
||||||
|
|
||||||
|
|
||||||
|
def ellipse_model(data, y, f):
|
||||||
|
"""
|
||||||
|
There is no need to make this process a function, since making the process a function will slow it down a little by calling it.
|
||||||
|
The results may be slightly different from the lambda version due to calculation errors derived from float types, but the calculation results are virtually the same.
|
||||||
|
a = 1.0,b = P[0],c = P[1],d = P[2],e = P[3],f = P[4]
|
||||||
|
:param data:
|
||||||
|
:param y: np.c_[d, e, a, c, b]
|
||||||
|
:param f: f == P[4, 0]
|
||||||
|
:return: this_return == np.array([ellipse_model(x, y) for (x, y) in data ])
|
||||||
|
"""
|
||||||
|
return data.dot(y) + f
|
||||||
|
|
||||||
|
|
||||||
|
# @profile
|
||||||
|
def fit_rotated_ellipse_ransac(data: np.ndarray, rng: np.random.Generator, iter=100, sample_num=10, offset=80 # 80.0, 10, 80
|
||||||
|
): # before changing these values, please read up on the ransac algorithm
|
||||||
|
# However if you want to change any value just know that higher iterations will make processing frames slower
|
||||||
|
effective_sample = None
|
||||||
|
|
||||||
|
# The array contents do not change during the loop, so only one call is needed.
|
||||||
|
# They say len is faster than shape.
|
||||||
|
# Reference url: https://stackoverflow.com/questions/35547853/what-is-faster-python3s-len-or-numpys-shape
|
||||||
|
len_data = len(data)
|
||||||
|
|
||||||
|
if len_data < sample_num:
|
||||||
|
return None
|
||||||
|
|
||||||
|
# Type of calculation result
|
||||||
|
ret_dtype = np.float64
|
||||||
|
|
||||||
|
# Sorts a random number array of size (iter,len_data). After sorting, returns the index of sample_num random numbers before sorting.
|
||||||
|
# If the array size is less than about 100, this is faster than rng.choice.
|
||||||
|
rng_sample = rng.random((iter, len_data)).argsort()[:, :sample_num]
|
||||||
|
# or
|
||||||
|
# I don't see any advantage to doing this.
|
||||||
|
# rng_sample = np.asarray(rng.random((iter, len_data)).argsort()[:, :sample_num], dtype=np.int32)
|
||||||
|
|
||||||
|
# I don't think it looks beautiful.
|
||||||
|
# x,y,x**2,y**2,x*y,1,-1*x**2
|
||||||
|
datamod = np.concatenate(
|
||||||
|
[data, data ** 2, (data[:, 0] * data[:, 1])[:, np.newaxis], np.ones((len_data, 1), dtype=ret_dtype),
|
||||||
|
(-1 * data[:, 0] ** 2)[:, np.newaxis]], axis=1,
|
||||||
|
dtype=ret_dtype)
|
||||||
|
|
||||||
|
datamod_slim = np.array(datamod[:, :5], dtype=ret_dtype)
|
||||||
|
|
||||||
|
datamod_rng = datamod[rng_sample]
|
||||||
|
datamod_rng6 = datamod_rng[:, :, 6]
|
||||||
|
datamod_rng_swap = datamod_rng[:, :, [4, 3, 0, 1, 5]]
|
||||||
|
datamod_rng_swap_trans = datamod_rng_swap.transpose((0, 2, 1))
|
||||||
|
|
||||||
|
# These two lines are one of the bottlenecks
|
||||||
|
datamod_rng_5x5 = np.matmul(datamod_rng_swap_trans, datamod_rng_swap)
|
||||||
|
datamod_rng_p5smp = np.matmul(np.linalg.inv(datamod_rng_5x5), datamod_rng_swap_trans)
|
||||||
|
|
||||||
|
datamod_rng_p = np.matmul(datamod_rng_p5smp, datamod_rng6[:, :, np.newaxis]).reshape((-1, 5))
|
||||||
|
|
||||||
|
# I don't think it looks beautiful.
|
||||||
|
ellipse_y_arr = np.asarray(
|
||||||
|
[datamod_rng_p[:, 2], datamod_rng_p[:, 3], np.ones(len(datamod_rng_p)), datamod_rng_p[:, 1], datamod_rng_p[:, 0]], dtype=ret_dtype)
|
||||||
|
|
||||||
|
ellipse_data_arr = ellipse_model(datamod_slim, ellipse_y_arr, np.asarray(datamod_rng_p[:, 4])).transpose((1, 0))
|
||||||
|
ellipse_data_abs = np.abs(ellipse_data_arr)
|
||||||
|
ellipse_data_index = np.argmax(np.sum(ellipse_data_abs < offset, axis=1), axis=0)
|
||||||
|
effective_data_arr = ellipse_data_arr[ellipse_data_index]
|
||||||
|
effective_sample_p_arr = datamod_rng_p[ellipse_data_index]
|
||||||
|
|
||||||
|
return fit_rotated_ellipse(effective_data_arr, effective_sample_p_arr)
|
||||||
|
|
||||||
|
|
||||||
|
# @profile
|
||||||
|
def fit_rotated_ellipse(data, P):
|
||||||
|
a = 1.0
|
||||||
|
b = P[0]
|
||||||
|
c = P[1]
|
||||||
|
d = P[2]
|
||||||
|
e = P[3]
|
||||||
|
f = P[4]
|
||||||
|
# The cost of trigonometric functions is high.
|
||||||
|
theta = 0.5 * np.arctan(b / (a - c), dtype=np.float64)
|
||||||
|
theta_sin = np.sin(theta, dtype=np.float64)
|
||||||
|
theta_cos = np.cos(theta, dtype=np.float64)
|
||||||
|
tc2 = theta_cos ** 2
|
||||||
|
ts2 = theta_sin ** 2
|
||||||
|
b_tcs = b * theta_cos * theta_sin
|
||||||
|
|
||||||
|
# Do the calculation only once
|
||||||
|
cxy = b ** 2 - 4 * a * c
|
||||||
|
cx = (2 * c * d - b * e) / cxy
|
||||||
|
cy = (2 * a * e - b * d) / cxy
|
||||||
|
|
||||||
|
# I just want to clear things up around here.
|
||||||
|
cu = a * cx ** 2 + b * cx * cy + c * cy ** 2 - f
|
||||||
|
cu_r = np.array([(a * tc2 + b_tcs + c * ts2), (a * ts2 - b_tcs + c * tc2)])
|
||||||
|
wh = np.sqrt(cu / cu_r)
|
||||||
|
|
||||||
|
w, h = wh[0], wh[1]
|
||||||
|
|
||||||
|
error_sum = np.sum(data)
|
||||||
|
# print("fitting error = %.3f" % (error_sum))
|
||||||
|
|
||||||
|
return (cx, cy, w, h, theta)
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
# HSF
|
||||||
|
calc_print_enable = True
|
||||||
|
save_video = False
|
||||||
|
skip_autoradius = True
|
||||||
|
skip_blink_detect = False
|
||||||
|
|
||||||
# cache param
|
# cache param
|
||||||
lru_maxsize_vvs = 16
|
lru_maxsize_vvs = 16
|
||||||
lru_maxsize_vs = 64
|
lru_maxsize_vs = 64
|
||||||
# CV param
|
# CV param
|
||||||
|
default_radius = 10
|
||||||
#default_radius = 15
|
auto_radius_range = (default_radius - 10, default_radius + 10) # (10,30)
|
||||||
#auto_radius_range = (default_radius - 10, default_radius + 10) # (10,30)
|
|
||||||
blink_init_frames = 60 * 3 # 60fps*3sec,Number of blink statistical frames
|
blink_init_frames = 60 * 3 # 60fps*3sec,Number of blink statistical frames
|
||||||
# step==(x,y)
|
# step==(x,y)
|
||||||
default_step = (5, 5) # bigger the steps,lower the processing time! ofc acc also takes an impact
|
default_step = (5, 5) # bigger the steps,lower the processing time! ofc acc also takes an impact
|
||||||
response_list = []
|
|
||||||
|
|
||||||
"""
|
"""
|
||||||
Attention.
|
Attention.
|
||||||
@ -206,7 +479,6 @@ class CvParameters:
|
|||||||
self._step = step
|
self._step = step
|
||||||
self._hsf = HaarSurroundFeature(radius)
|
self._hsf = HaarSurroundFeature(radius)
|
||||||
|
|
||||||
|
|
||||||
def get_rpsh(self):
|
def get_rpsh(self):
|
||||||
return self._radius, self.pad, self._step, self._hsf
|
return self._radius, self.pad, self._step, self._hsf
|
||||||
# Essentially, the following would be preferable, but it would take twice as long to call.
|
# Essentially, the following would be preferable, but it would take twice as long to call.
|
||||||
@ -277,12 +549,6 @@ class HaarSurroundFeature:
|
|||||||
return kernel
|
return kernel
|
||||||
|
|
||||||
|
|
||||||
def to_gray(frame):
|
|
||||||
# Faster by quitting checking if the input image is already grayscale
|
|
||||||
# Perhaps it would be faster with less overhead to call cv2.cvtColor directly instead of using this function
|
|
||||||
return cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
|
|
||||||
|
|
||||||
|
|
||||||
@lru_cache(maxsize=lru_maxsize_vs)
|
@lru_cache(maxsize=lru_maxsize_vs)
|
||||||
def frameint_get_xy_step(imageshape, xysteps, pad, start_offset=None, end_offset=None):
|
def frameint_get_xy_step(imageshape, xysteps, pad, start_offset=None, end_offset=None):
|
||||||
"""
|
"""
|
||||||
@ -404,170 +670,388 @@ def conv_int(frame_int, kernel, xy_step, padding, xy_steps_list):
|
|||||||
np.multiply(kernel.val_in, inner_sum, dtype=np.float64, out=response_list)
|
np.multiply(kernel.val_in, inner_sum, dtype=np.float64, out=response_list)
|
||||||
response_list += kernel.val_out * outer_sum
|
response_list += kernel.val_out * outer_sum
|
||||||
|
|
||||||
# min_response, max_val, min_loc, max_loc = cv2.minMaxLoc(self.response_list)
|
# min_response, max_val, min_loc, max_loc = cv2.minMaxLoc(response_list)
|
||||||
min_response, _, min_loc, _ = cv2.minMaxLoc(response_list)
|
min_response, _, min_loc, _ = cv2.minMaxLoc(response_list)
|
||||||
|
|
||||||
center = ((xy_steps_list[0][min_loc[0]] - padding), (xy_steps_list[1][min_loc[1]] - padding))
|
center = ((xy_steps_list[0][min_loc[0]] - padding), (xy_steps_list[1][min_loc[1]] - padding))
|
||||||
|
|
||||||
frame_conv_stride[:, :] = response_list
|
frame_conv_stride[:, :] = response_list
|
||||||
# or
|
# or
|
||||||
# frame_conv_stride[:, :] = self.response_list.astype(np.uint8)
|
# frame_conv_stride[:, :] = response_list.astype(np.uint8)
|
||||||
|
|
||||||
return frame_conv, min_response, center
|
return frame_conv, min_response, center
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
#RANSAC \/
|
|
||||||
def ellipse_model(data, y, f):
|
|
||||||
"""
|
|
||||||
There is no need to make this process a function, since making the process a function will slow it down a little by calling it.
|
|
||||||
The results may be slightly different from the lambda version due to calculation errors derived from float types, but the calculation results are virtually the same.
|
|
||||||
a = 1.0,b = P[0],c = P[1],d = P[2],e = P[3],f = P[4]
|
|
||||||
:param data:
|
|
||||||
:param y: np.c_[d, e, a, c, b]
|
|
||||||
:param f: f == P[4, 0]
|
|
||||||
:return: this_return == np.array([ellipse_model(x, y) for (x, y) in data ])
|
|
||||||
"""
|
|
||||||
return data.dot(y) + f
|
|
||||||
|
|
||||||
# @profile
|
# @profile
|
||||||
def fit_rotated_ellipse_ransac(data: np.ndarray, rng: np.random.Generator, iter=100, sample_num=10, offset=80 # 80.0, 10, 80
|
|
||||||
): # before changing these values, please read up on the ransac algorithm
|
|
||||||
# However if you want to change any value just know that higher iterations will make processing frames slower
|
|
||||||
effective_sample = None
|
|
||||||
|
|
||||||
# The array contents do not change during the loop, so only one call is needed.
|
|
||||||
# They say len is faster than shape.
|
|
||||||
# Reference url: https://stackoverflow.com/questions/35547853/what-is-faster-python3s-len-or-numpys-shape
|
|
||||||
len_data = len(data)
|
|
||||||
|
|
||||||
if len_data < sample_num:
|
|
||||||
return None
|
|
||||||
|
|
||||||
# Type of calculation result
|
|
||||||
ret_dtype = np.float64
|
|
||||||
|
|
||||||
# Sorts a random number array of size (iter,len_data). After sorting, returns the index of sample_num random numbers before sorting.
|
|
||||||
# If the array size is less than about 100, this is faster than rng.choice.
|
|
||||||
rng_sample = rng.random((iter, len_data)).argsort()[:, :sample_num]
|
|
||||||
# or
|
|
||||||
# I don't see any advantage to doing this.
|
|
||||||
# rng_sample = np.asarray(rng.random((iter, len_data)).argsort()[:, :sample_num], dtype=np.int32)
|
|
||||||
|
|
||||||
# I don't think it looks beautiful.
|
|
||||||
# x,y,x**2,y**2,x*y,1,-1*x**2
|
|
||||||
datamod = np.concatenate(
|
|
||||||
[data, data ** 2, (data[:, 0] * data[:, 1])[:, np.newaxis], np.ones((len_data, 1), dtype=ret_dtype),
|
|
||||||
(-1 * data[:, 0] ** 2)[:, np.newaxis]], axis=1,
|
|
||||||
dtype=ret_dtype)
|
|
||||||
|
|
||||||
datamod_slim = np.array(datamod[:, :5], dtype=ret_dtype)
|
|
||||||
|
|
||||||
datamod_rng = datamod[rng_sample]
|
|
||||||
datamod_rng6 = datamod_rng[:, :, 6]
|
|
||||||
datamod_rng_swap = datamod_rng[:, :, [4, 3, 0, 1, 5]]
|
|
||||||
datamod_rng_swap_trans = datamod_rng_swap.transpose((0, 2, 1))
|
|
||||||
|
|
||||||
# These two lines are one of the bottlenecks
|
|
||||||
datamod_rng_5x5 = np.matmul(datamod_rng_swap_trans, datamod_rng_swap)
|
|
||||||
datamod_rng_p5smp = np.matmul(np.linalg.inv(datamod_rng_5x5), datamod_rng_swap_trans)
|
|
||||||
|
|
||||||
datamod_rng_p = np.matmul(datamod_rng_p5smp, datamod_rng6[:, :, np.newaxis]).reshape((-1, 5))
|
|
||||||
|
|
||||||
# I don't think it looks beautiful.
|
|
||||||
ellipse_y_arr = np.asarray(
|
|
||||||
[datamod_rng_p[:, 2], datamod_rng_p[:, 3], np.ones(len(datamod_rng_p)), datamod_rng_p[:, 1], datamod_rng_p[:, 0]], dtype=ret_dtype)
|
|
||||||
|
|
||||||
ellipse_data_arr = ellipse_model(datamod_slim, ellipse_y_arr, np.asarray(datamod_rng_p[:, 4])).transpose((1, 0))
|
|
||||||
ellipse_data_abs = np.abs(ellipse_data_arr)
|
|
||||||
ellipse_data_index = np.argmax(np.sum(ellipse_data_abs < offset, axis=1), axis=0)
|
|
||||||
effective_data_arr = ellipse_data_arr[ellipse_data_index]
|
|
||||||
effective_sample_p_arr = datamod_rng_p[ellipse_data_index]
|
|
||||||
|
|
||||||
return fit_rotated_ellipse(effective_data_arr, effective_sample_p_arr)
|
|
||||||
|
|
||||||
|
|
||||||
# @profile
|
|
||||||
def fit_rotated_ellipse(data, P):
|
|
||||||
a = 1.0
|
|
||||||
b = P[0]
|
|
||||||
c = P[1]
|
|
||||||
d = P[2]
|
|
||||||
e = P[3]
|
|
||||||
f = P[4]
|
|
||||||
# The cost of trigonometric functions is high.
|
|
||||||
theta = 0.5 * np.arctan(b / (a - c), dtype=np.float64)
|
|
||||||
theta_sin = np.sin(theta, dtype=np.float64)
|
|
||||||
theta_cos = np.cos(theta, dtype=np.float64)
|
|
||||||
tc2 = theta_cos ** 2
|
|
||||||
ts2 = theta_sin ** 2
|
|
||||||
b_tcs = b * theta_cos * theta_sin
|
|
||||||
|
|
||||||
# Do the calculation only once
|
|
||||||
cxy = b ** 2 - 4 * a * c
|
|
||||||
cx = (2 * c * d - b * e) / cxy
|
|
||||||
cy = (2 * a * e - b * d) / cxy
|
|
||||||
|
|
||||||
# I just want to clear things up around here.
|
|
||||||
cu = a * cx ** 2 + b * cx * cy + c * cy ** 2 - f
|
|
||||||
cu_r = np.array([(a * tc2 + b_tcs + c * ts2), (a * ts2 - b_tcs + c * tc2)])
|
|
||||||
wh = np.sqrt(cu / cu_r)
|
|
||||||
|
|
||||||
w, h = wh[0], wh[1]
|
|
||||||
|
|
||||||
error_sum = np.sum(data)
|
|
||||||
# print("fitting error = %.3f" % (error_sum))
|
|
||||||
|
|
||||||
return (cx, cy, w, h, theta)
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
timedict = {"to_gray": [], "int_img": [], "conv_int": [], "crop": [], "total_cv": []}
|
||||||
|
# I'd like to take into account things like print, end_time - start_time processing time, etc., but it's too much trouble.
|
||||||
|
|
||||||
|
# For measuring total processing time
|
||||||
|
main_start_time = timeit.default_timer()
|
||||||
|
|
||||||
|
rng = np.random.default_rng()
|
||||||
|
cvparam = CvParameters(default_radius, default_step)
|
||||||
|
|
||||||
|
cv_mode = ["first_frame", "radius_adjust", "init", "normal"]
|
||||||
|
now_mode = cv_mode[0]
|
||||||
|
|
||||||
|
radius_cand_list = []
|
||||||
|
|
||||||
|
# response_min=0
|
||||||
|
response_max = None
|
||||||
|
response_list = []
|
||||||
|
|
||||||
|
prev_hsfx = 0
|
||||||
|
prev_hsfy = 0
|
||||||
|
prev_ranx = 0
|
||||||
|
prev_rany = 0
|
||||||
|
|
||||||
def HSRAC(self):
|
def HSRAC(self):
|
||||||
|
|
||||||
|
global now_mode
|
||||||
|
global response_list
|
||||||
|
global radius_cand_list
|
||||||
|
global response_max
|
||||||
|
global skip_autoradius
|
||||||
|
global default_radius
|
||||||
|
global prev_rany
|
||||||
|
global prev_ranx
|
||||||
|
global prev_hsfy
|
||||||
|
global prev_hsfx
|
||||||
|
skip_autoradius = self.settings.gui_skip_autoradius
|
||||||
default_radius = self.settings.gui_HSF_radius
|
default_radius = self.settings.gui_HSF_radius
|
||||||
auto_radius_range = (default_radius - 10, default_radius + 10) # (10,30)
|
|
||||||
frame = self.current_image_gray
|
frame = self.current_image_gray
|
||||||
if self.now_mode == self.cv_mode[1]:
|
|
||||||
|
|
||||||
|
if now_mode == cv_mode[1]:
|
||||||
prev_res_len = len(self.response_list)
|
prev_res_len = len(response_list)
|
||||||
# adjustment of radius
|
# adjustment of radius
|
||||||
if prev_res_len == 1:
|
if prev_res_len == 1:
|
||||||
# len==1==self.response_list==[default_radius]
|
# len==1==response_list==[default_radius]
|
||||||
self.cvparam.radius = self.auto_radius_range[0]
|
cvparam.radius = auto_radius_range[0]
|
||||||
elif prev_res_len == 2:
|
elif prev_res_len == 2:
|
||||||
# len==2==self.response_list==[default_radius, self.auto_radius_range[0]]
|
# len==2==response_list==[default_radius, auto_radius_range[0]]
|
||||||
self.cvparam.radius = self.auto_radius_range[1]
|
cvparam.radius = auto_radius_range[1]
|
||||||
elif prev_res_len == 3:
|
elif prev_res_len == 3:
|
||||||
# len==3==self.response_list==[default_radius,self.auto_radius_range[0],self.auto_radius_range[1]]
|
# len==3==response_list==[default_radius,auto_radius_range[0],auto_radius_range[1]]
|
||||||
sort_res = sorted(self.response_list, key=lambda x: x[1])[0]
|
sort_res = sorted(response_list, key=lambda x: x[1])[0]
|
||||||
# Extract the radius with the lowest response value
|
# Extract the radius with the lowest response value
|
||||||
if sort_res[0] == default_radius:
|
if sort_res[0] == default_radius:
|
||||||
# If the default value is best, change self.now_mode to init after setting radius to the default value.
|
# If the default value is best, change now_mode to init after setting radius to the default value.
|
||||||
self.cvparam.radius = default_radius
|
cvparam.radius = default_radius
|
||||||
self.now_mode = self.cv_mode[2] if not self.skip_blink_detect else self.cv_mode[3]
|
now_mode = cv_mode[2] if not skip_blink_detect else cv_mode[3]
|
||||||
self.response_list = []
|
response_list = []
|
||||||
elif sort_res[0] == self.auto_radius_range[0]:
|
elif sort_res[0] == auto_radius_range[0]:
|
||||||
self.radius_cand_list = [i for i in range(self.auto_radius_range[0], default_radius, self.default_step[0])][1:]
|
radius_cand_list = [i for i in range(auto_radius_range[0], default_radius, default_step[0])][1:]
|
||||||
# self.default_step is defined separately for xy, but radius is shared by xy, so it may be buggy
|
# default_step is defined separately for xy, but radius is shared by xy, so it may be buggy
|
||||||
# It should be no problem to set it to anything other than self.default_step
|
# It should be no problem to set it to anything other than default_step
|
||||||
self.cvparam.radius = self.radius_cand_list.pop()
|
cvparam.radius = radius_cand_list.pop()
|
||||||
else:
|
else:
|
||||||
self.radius_cand_list = [i for i in range(default_radius, self.auto_radius_range[1], self.default_step[0])][1:]
|
radius_cand_list = [i for i in range(default_radius, auto_radius_range[1], default_step[0])][1:]
|
||||||
# self.default_step is defined separately for xy, but radius is shared by xy, so it may be buggy
|
# default_step is defined separately for xy, but radius is shared by xy, so it may be buggy
|
||||||
# It should be no problem to set it to anything other than self.default_step
|
# It should be no problem to set it to anything other than default_step
|
||||||
self.cvparam.radius = self.radius_cand_list.pop()
|
cvparam.radius = radius_cand_list.pop()
|
||||||
else:
|
else:
|
||||||
# Try the contents of the self.radius_cand_list in order until the self.radius_cand_list runs out
|
# Try the contents of the radius_cand_list in order until the radius_cand_list runs out
|
||||||
# Better make it a binary search.
|
# Better make it a binary search.
|
||||||
if len(self.radius_cand_list) == 0:
|
if len(radius_cand_list) == 0:
|
||||||
sort_res = sorted(self.response_list, key=lambda x: x[1])[0]
|
sort_res = sorted(response_list, key=lambda x: x[1])[0]
|
||||||
self.cvparam.radius = sort_res[0]
|
cvparam.radius = sort_res[0]
|
||||||
self.now_mode = self.cv_mode[2] if not self.skip_blink_detect else self.cv_mode[3]
|
now_mode = cv_mode[2] if not skip_blink_detect else cv_mode[3]
|
||||||
self.response_list = []
|
response_list = []
|
||||||
else:
|
else:
|
||||||
self.cvparam.radius = self.radius_cand_list.pop()
|
cvparam.radius = radius_cand_list.pop()
|
||||||
|
|
||||||
|
radius, pad, step, hsf = cvparam.get_rpsh()
|
||||||
|
|
||||||
|
# For measuring processing time of image processing
|
||||||
|
cv_start_time = timeit.default_timer()
|
||||||
|
|
||||||
|
gray_frame = frame
|
||||||
|
timedict["to_gray"].append(timeit.default_timer() - cv_start_time)
|
||||||
|
|
||||||
|
# Calculate the integral image of the frame
|
||||||
|
int_start_time = timeit.default_timer()
|
||||||
|
# BORDER_CONSTANT is faster than BORDER_REPLICATE There seems to be almost no negative impact when BORDER_CONSTANT is used.
|
||||||
|
frame_pad = cv2.copyMakeBorder(gray_frame, pad, pad, pad, pad, cv2.BORDER_CONSTANT)
|
||||||
|
frame_int = cv2.integral(frame_pad)
|
||||||
|
timedict["int_img"].append(timeit.default_timer() - int_start_time)
|
||||||
|
|
||||||
|
# Convolve the feature with the integral image
|
||||||
|
conv_int_start_time = timeit.default_timer()
|
||||||
|
xy_step = frameint_get_xy_step(frame_int.shape, step, pad, start_offset=None, end_offset=None)
|
||||||
|
frame_conv, response, center_xy = conv_int(frame_int, hsf, step, pad, xy_step)
|
||||||
|
timedict["conv_int"].append(timeit.default_timer() - conv_int_start_time)
|
||||||
|
|
||||||
|
# Define the center point and radius
|
||||||
|
center_x, center_y = center_xy
|
||||||
|
upper_x = center_x + 20
|
||||||
|
lower_x = center_x - 20
|
||||||
|
upper_y = center_y + 20
|
||||||
|
lower_y = center_y - 20
|
||||||
|
|
||||||
|
# Crop the image using the calculated bounds
|
||||||
|
cropped_image = gray_frame[lower_y:upper_y, lower_x:upper_x]
|
||||||
|
frame = cropped_image
|
||||||
|
if now_mode == cv_mode[0] or now_mode == cv_mode[1]:
|
||||||
|
# If mode is first_frame or radius_adjust, record current radius and response
|
||||||
|
response_list.append((radius, response))
|
||||||
|
elif now_mode == cv_mode[2]:
|
||||||
|
# Statistics for blink detection
|
||||||
|
if len(response_list) < blink_init_frames:
|
||||||
|
# Record the average value of cropped_image
|
||||||
|
response_list.append(cv2.mean(cropped_image)[0])
|
||||||
|
else:
|
||||||
|
# Calculate response_max by computing interquartile range, IQR
|
||||||
|
# Change cv_mode to normal
|
||||||
|
response_list = np.array(response_list)
|
||||||
|
# 25%,75%
|
||||||
|
# This value may need to be adjusted depending on the environment.
|
||||||
|
quartile_1, quartile_3 = np.percentile(response_list, [25, 75])
|
||||||
|
iqr = quartile_3 - quartile_1
|
||||||
|
# response_min = quartile_1 - (iqr * 1.5)
|
||||||
|
response_max = quartile_3 + (iqr * 1.5)
|
||||||
|
now_mode = cv_mode[3]
|
||||||
|
else:
|
||||||
|
if 0 in cropped_image.shape:
|
||||||
|
# If shape contains 0, it is not detected well.
|
||||||
|
print("Something's wrong.")
|
||||||
|
else:
|
||||||
|
# If the average value of cropped_image is greater than response_max
|
||||||
|
# (i.e., if the cropimage is whitish
|
||||||
|
if response_max is not None and cv2.mean(cropped_image)[0] > response_max:
|
||||||
|
# blink
|
||||||
|
self.blinkvalue = True
|
||||||
|
print("HSF BLINK")
|
||||||
|
|
||||||
|
# If you want to update response_max. it may be more cost-effective to rewrite response_list in the following way
|
||||||
|
# https://stackoverflow.com/questions/42771110/fastest-way-to-left-cycle-a-numpy-array-like-pop-push-for-a-queue
|
||||||
|
|
||||||
|
|
||||||
|
# the lower the response the better the likelyhood of there being a pupil. you can adujst the radius and steps accordingly
|
||||||
|
print('Kernel response:', response)
|
||||||
|
print('Pixel position:', center_xy)
|
||||||
|
|
||||||
|
|
||||||
|
if now_mode == cv_mode[0]:
|
||||||
|
# Moving from first_frame to the next mode
|
||||||
|
if skip_autoradius and skip_blink_detect:
|
||||||
|
now_mode = cv_mode[3]
|
||||||
|
response_list = []
|
||||||
|
elif skip_autoradius:
|
||||||
|
now_mode = cv_mode[2]
|
||||||
|
response_list = []
|
||||||
|
else:
|
||||||
|
now_mode = cv_mode[1]
|
||||||
|
|
||||||
|
#run ransac on the HSF crop\
|
||||||
|
frame = cropped_image
|
||||||
|
|
||||||
|
# try:
|
||||||
|
kernel = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (3, 3))
|
||||||
|
thresh_add = 10
|
||||||
|
rng = np.random.default_rng()
|
||||||
|
|
||||||
|
f = False
|
||||||
|
|
||||||
|
# Convert the image to grayscale, and set up thresholding. Thresholds here are basically a
|
||||||
|
# low-pass filter that will set any pixel < the threshold value to 0. Thresholding is user
|
||||||
|
# configurable in this utility as we're dealing with variable lighting amounts/placement, as
|
||||||
|
# well as camera positioning and lensing. Therefore everyone's cutoff may be different.
|
||||||
|
#
|
||||||
|
# The goal of thresholding settings is to make sure we can ONLY see the pupil. This is why we
|
||||||
|
# crop the image earlier; it gives us less possible dark area to get confused about in the
|
||||||
|
# next step.
|
||||||
|
|
||||||
|
# For measuring processing time of image processing
|
||||||
|
# Crop first to reduce the amount of data to process.
|
||||||
|
|
||||||
|
#frame = frame[0:len(frame) - 5, :]
|
||||||
|
|
||||||
|
# To reduce the processing data, first convert to 1-channel and then blur.
|
||||||
|
# The processing results were the same when I swapped the order of blurring and 1-channelization.
|
||||||
|
try:
|
||||||
|
frame = cv2.GaussianBlur(frame, (5, 5), 0)
|
||||||
|
except:
|
||||||
|
pass
|
||||||
|
|
||||||
|
# this will need to be adjusted everytime hardware is changed (brightness of IR, Camera postion, etc)m
|
||||||
|
min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(frame)
|
||||||
|
|
||||||
|
maxloc0_hf, maxloc1_hf = int(0.5 * max_loc[0]), int(0.5 * max_loc[1])
|
||||||
|
|
||||||
|
# crop 15% sqare around min_loc
|
||||||
|
# frame = frame[max_loc[1] - maxloc1_hf:max_loc[1] + maxloc1_hf,
|
||||||
|
# max_loc[0] - maxloc0_hf:max_loc[0] + maxloc0_hf]
|
||||||
|
|
||||||
|
threshold_value = min_val + thresh_add
|
||||||
|
_, thresh = cv2.threshold(frame, threshold_value, 255, cv2.THRESH_BINARY)
|
||||||
|
try:
|
||||||
|
opening = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, kernel)
|
||||||
|
closing = cv2.morphologyEx(opening, cv2.MORPH_CLOSE, kernel)
|
||||||
|
th_frame = 255 - closing
|
||||||
|
except:
|
||||||
|
# I want to eliminate try here because try tends to be slow in execution.
|
||||||
|
th_frame = 255 - frame
|
||||||
|
|
||||||
|
|
||||||
|
contours, _ = cv2.findContours(th_frame, cv2.RETR_TREE, cv2.CHAIN_APPROX_NONE)
|
||||||
|
hull = []
|
||||||
|
# This way is faster than contours[i]
|
||||||
|
# But maybe this one is faster. hull = [cv2.convexHull(cnt, False) for cnt in contours]
|
||||||
|
for cnt in contours:
|
||||||
|
hull.append(cv2.convexHull(cnt, False))
|
||||||
|
if not hull:
|
||||||
|
# If empty, go to next loop
|
||||||
|
pass
|
||||||
|
try:
|
||||||
|
|
||||||
|
cnt = sorted(hull, key=cv2.contourArea)
|
||||||
|
maxcnt = cnt[-1]
|
||||||
|
# ellipse = cv2.fitEllipse(maxcnt)
|
||||||
|
ransac_data = fit_rotated_ellipse_ransac(maxcnt.reshape(-1, 2), rng)
|
||||||
|
if ransac_data is None:
|
||||||
|
# ransac_data is None==maxcnt.shape[0]<sample_num
|
||||||
|
# go to next loop
|
||||||
|
print("NODATYA")
|
||||||
|
pass
|
||||||
|
|
||||||
|
crop_start_time = timeit.default_timer()
|
||||||
|
cx, cy, w, h, theta = ransac_data
|
||||||
|
|
||||||
|
csx = frame.shape[0]
|
||||||
|
csy = frame.shape[1]
|
||||||
|
|
||||||
|
cx = center_x - (csx - cx) # we find the difference between the crop size and ransac point, and subtract from the center point from HSF
|
||||||
|
cy = center_y - (csy - cy)
|
||||||
|
|
||||||
|
out_x, out_y = cx, cy
|
||||||
|
prev_hsfx = center_x
|
||||||
|
prev_hsfy = center_y
|
||||||
|
prev_ranx = cx
|
||||||
|
prev_rany = cy
|
||||||
|
cx, cy, w, h = int(cx), int(cy), int(w), int(h)
|
||||||
|
|
||||||
|
cv2.drawContours(frame, contours, -1, (255, 0, 0), 1)
|
||||||
|
cv2.circle(frame, (cx, cy), 2, (0, 0, 255), -1)
|
||||||
|
# cx1, cy1, w1, h1, theta1 = fit_rotated_ellipse(maxcnt.reshape(-1, 2))
|
||||||
|
cv2.ellipse(frame, (cx, cy), (w, h), theta * 180.0 / np.pi, 0.0, 360.0, (50, 250, 200), 1, )
|
||||||
|
cv2.circle(frame, min_loc, 2, (0, 0, 255),-1) # the point of the darkest area in the image
|
||||||
|
self.current_image_gray = frame
|
||||||
|
#img = newImage2[y1:y2, x1:x2]
|
||||||
|
#except:
|
||||||
|
|
||||||
|
# print('R F')
|
||||||
|
# pass
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
try:
|
||||||
|
# print(radius)
|
||||||
|
return out_x, out_y, thresh
|
||||||
|
|
||||||
|
except:
|
||||||
|
xoff = prev_hsfx - prev_ranx
|
||||||
|
yoff = prev_hsfy - prev_rany
|
||||||
|
return (center_x + xoff), (center_y + yoff), thresh
|
||||||
|
|
||||||
|
except:
|
||||||
|
self.current_image_gray = frame #cv2.resize(frame, (150, 150), interpolation = cv2.INTER_AREA)
|
||||||
|
xoff = prev_hsfx - prev_ranx
|
||||||
|
yoff = prev_hsfy - prev_rany
|
||||||
|
return (center_x + xoff), (center_y + yoff), thresh
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
'''
|
||||||
|
|
||||||
|
try:
|
||||||
|
self.failed = 0
|
||||||
|
cv2.circle(frame, (center_x, center_y), 10, (0, 0, 255), -1)
|
||||||
|
return center_x, center_y, frame
|
||||||
|
|
||||||
|
except:
|
||||||
|
self.failed = self.failed + 1
|
||||||
|
return 0, 0, frame
|
||||||
|
'''
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
'''
|
||||||
|
def HSRAC(self):
|
||||||
|
|
||||||
|
global default_radius
|
||||||
|
global auto_radius_range
|
||||||
|
|
||||||
|
global response_list
|
||||||
|
global radius_cand_list
|
||||||
|
|
||||||
|
global response_max
|
||||||
|
global now_mode
|
||||||
|
global cv_mode
|
||||||
|
default_radius = int(self.settings.gui_HSF_radius)
|
||||||
|
auto_radius_range = (default_radius - 10, default_radius + 10) # (10,30)
|
||||||
|
|
||||||
|
|
||||||
|
print(default_radius, cv_mode, now_mode, radius_cand_list, response_max )
|
||||||
|
|
||||||
|
if self.calibration_frame_counter == 0: # if reset triggered, reset all values
|
||||||
|
now_mode = cv_mode[0]
|
||||||
|
radius_cand_list = []
|
||||||
|
# response_min=0
|
||||||
|
response_max = None
|
||||||
|
response_list = []
|
||||||
|
print("RESET HSRAC")
|
||||||
|
|
||||||
|
frame = self.current_image_gray
|
||||||
|
if now_mode == cv_mode[1]:
|
||||||
|
|
||||||
|
|
||||||
|
prev_res_len = len(response_list)
|
||||||
|
# adjustment of radius
|
||||||
|
if prev_res_len == 1:
|
||||||
|
# len==1==response_list==[default_radius]
|
||||||
|
self.cvparam.radius = self.auto_radius_range[0]
|
||||||
|
elif prev_res_len == 2:
|
||||||
|
# len==2==response_list==[default_radius, self.auto_radius_range[0]]
|
||||||
|
self.cvparam.radius = self.auto_radius_range[1]
|
||||||
|
elif prev_res_len == 3:
|
||||||
|
# len==3==response_list==[default_radius,self.auto_radius_range[0],self.auto_radius_range[1]]
|
||||||
|
sort_res = sorted(response_list, key=lambda x: x[1])[0]
|
||||||
|
# Extract the radius with the lowest response value
|
||||||
|
if sort_res[0] == default_radius:
|
||||||
|
# If the default value is best, change now_mode to init after setting radius to the default value.
|
||||||
|
self.cvparam.radius = default_radius
|
||||||
|
now_mode = cv_mode[2] if not self.skip_blink_detect else cv_mode[3]
|
||||||
|
response_list = []
|
||||||
|
elif sort_res[0] == self.auto_radius_range[0]:
|
||||||
|
radius_cand_list = [i for i in range(self.auto_radius_range[0], default_radius, self.default_step[0])][1:]
|
||||||
|
# self.default_step is defined separately for xy, but radius is shared by xy, so it may be buggy
|
||||||
|
# It should be no problem to set it to anything other than self.default_step
|
||||||
|
self.cvparam.radius = radius_cand_list.pop()
|
||||||
|
else:
|
||||||
|
radius_cand_list = [i for i in range(default_radius, self.auto_radius_range[1], self.default_step[0])][1:]
|
||||||
|
# self.default_step is defined separately for xy, but radius is shared by xy, so it may be buggy
|
||||||
|
# It should be no problem to set it to anything other than self.default_step
|
||||||
|
self.cvparam.radius = radius_cand_list.pop()
|
||||||
|
else:
|
||||||
|
# Try the contents of the radius_cand_list in order until the radius_cand_list runs out
|
||||||
|
# Better make it a binary search.
|
||||||
|
if len(radius_cand_list) == 0:
|
||||||
|
sort_res = sorted(response_list, key=lambda x: x[1])[0]
|
||||||
|
self.cvparam.radius = sort_res[0]
|
||||||
|
now_mode = cv_mode[2] if not self.skip_blink_detect else cv_mode[3]
|
||||||
|
response_list = []
|
||||||
|
else:
|
||||||
|
self.cvparam.radius = radius_cand_list.pop()
|
||||||
|
|
||||||
radius, pad, step, hsf = self.cvparam.get_rpsh()
|
radius, pad, step, hsf = self.cvparam.get_rpsh()
|
||||||
|
|
||||||
@ -593,37 +1077,37 @@ def HSRAC(self):
|
|||||||
# Crop the image using the calculated bounds
|
# Crop the image using the calculated bounds
|
||||||
cropped_image = gray_frame[lower_y:upper_y, lower_x:upper_x] # y is 50px, x is 45? why?
|
cropped_image = gray_frame[lower_y:upper_y, lower_x:upper_x] # y is 50px, x is 45? why?
|
||||||
|
|
||||||
if self.now_mode == self.cv_mode[0] or self.now_mode == self.cv_mode[1]:
|
if now_mode == cv_mode[0] or now_mode == cv_mode[1]:
|
||||||
# If mode is first_frame or radius_adjust, record current radius and response
|
# If mode is first_frame or radius_adjust, record current radius and response
|
||||||
self.response_list.append((radius, response))
|
response_list.append((radius, response))
|
||||||
elif self.now_mode == self.cv_mode[2]:
|
elif now_mode == cv_mode[2]:
|
||||||
# Statistics for blink detection
|
# Statistics for blink detection
|
||||||
if len(self.response_list) < self.blink_init_frames:
|
if len(response_list) < self.blink_init_frames:
|
||||||
# Record the average value of cropped_image
|
# Record the average value of cropped_image
|
||||||
self.response_list.append(cv2.mean(cropped_image)[0])
|
response_list.append(cv2.mean(cropped_image)[0])
|
||||||
else:
|
else:
|
||||||
# Calculate self.response_max by computing interquartile range, IQR
|
# Calculate response_max by computing interquartile range, IQR
|
||||||
# Change self.cv_mode to normal
|
# Change cv_mode to normal
|
||||||
self.response_list = np.array(self.response_list)
|
response_list = np.array(response_list)
|
||||||
# 25%,75%
|
# 25%,75%
|
||||||
# This value may need to be adjusted depending on the environment.
|
# This value may need to be adjusted depending on the environment.
|
||||||
quartile_1, quartile_3 = np.percentile(self.response_list, [25, 75])
|
quartile_1, quartile_3 = np.percentile(response_list, [25, 75])
|
||||||
iqr = quartile_3 - quartile_1
|
iqr = quartile_3 - quartile_1
|
||||||
# response_min = quartile_1 - (iqr * 1.5)
|
# response_min = quartile_1 - (iqr * 1.5)
|
||||||
self.response_max = quartile_3 + (iqr * 1.5)
|
response_max = quartile_3 + (iqr * 1.5)
|
||||||
self.now_mode = self.cv_mode[3]
|
now_mode = cv_mode[3]
|
||||||
else:
|
else:
|
||||||
if 0 in cropped_image.shape:
|
if 0 in cropped_image.shape:
|
||||||
# If shape contains 0, it is not detected well.
|
# If shape contains 0, it is not detected well.
|
||||||
print("Something's wrong.")
|
print("Something's wrong.")
|
||||||
else:
|
else:
|
||||||
# If the average value of cropped_image is greater than self.response_max
|
# If the average value of cropped_image is greater than response_max
|
||||||
# (i.e., if the cropimage is whitish
|
# (i.e., if the cropimage is whitish
|
||||||
if self.response_max is not None and cv2.mean(cropped_image)[0] > self.response_max:
|
if response_max is not None and cv2.mean(cropped_image)[0] > response_max:
|
||||||
# blink
|
# blink
|
||||||
|
|
||||||
cv2.circle(frame, (center_x, center_y), 10, (0, 0, 255), -1)
|
cv2.circle(frame, (center_x, center_y), 10, (0, 0, 255), -1)
|
||||||
# If you want to update self.response_max. it may be more cost-effective to rewrite self.response_list in the following way
|
# If you want to update response_max. it may be more cost-effective to rewrite response_list in the following way
|
||||||
# https://stackoverflow.com/questions/42771110/fastest-way-to-left-cycle-a-numpy-array-like-pop-push-for-a-queue
|
# https://stackoverflow.com/questions/42771110/fastest-way-to-left-cycle-a-numpy-array-like-pop-push-for-a-queue
|
||||||
|
|
||||||
except:
|
except:
|
||||||
@ -652,27 +1136,27 @@ def HSRAC(self):
|
|||||||
|
|
||||||
# To reduce the processing data, first convert to 1-channel and then blur.
|
# To reduce the processing data, first convert to 1-channel and then blur.
|
||||||
# The processing results were the same when I swapped the order of blurring and 1-channelization.
|
# The processing results were the same when I swapped the order of blurring and 1-channelization.
|
||||||
frame_gray = cv2.GaussianBlur(frame, (5, 5), 0)
|
frame = cv2.GaussianBlur(frame, (5, 5), 0)
|
||||||
|
|
||||||
|
|
||||||
# this will need to be adjusted everytime hardware is changed (brightness of IR, Camera postion, etc)m
|
# this will need to be adjusted everytime hardware is changed (brightness of IR, Camera postion, etc)m
|
||||||
min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(frame_gray)
|
min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(frame)
|
||||||
|
|
||||||
maxloc0_hf, maxloc1_hf = int(0.5 * max_loc[0]), int(0.5 * max_loc[1])
|
maxloc0_hf, maxloc1_hf = int(0.5 * max_loc[0]), int(0.5 * max_loc[1])
|
||||||
|
|
||||||
# crop 15% sqare around min_loc
|
# crop 15% sqare around min_loc
|
||||||
# frame_gray = frame_gray[max_loc[1] - maxloc1_hf:max_loc[1] + maxloc1_hf,
|
# frame = frame[max_loc[1] - maxloc1_hf:max_loc[1] + maxloc1_hf,
|
||||||
# max_loc[0] - maxloc0_hf:max_loc[0] + maxloc0_hf]
|
# max_loc[0] - maxloc0_hf:max_loc[0] + maxloc0_hf]
|
||||||
|
|
||||||
threshold_value = min_val + thresh_add
|
threshold_value = min_val + thresh_add
|
||||||
_, thresh = cv2.threshold(frame_gray, threshold_value, 255, cv2.THRESH_BINARY)
|
_, thresh = cv2.threshold(frame, threshold_value, 255, cv2.THRESH_BINARY)
|
||||||
try:
|
try:
|
||||||
opening = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, kernel)
|
opening = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, kernel)
|
||||||
closing = cv2.morphologyEx(opening, cv2.MORPH_CLOSE, kernel)
|
closing = cv2.morphologyEx(opening, cv2.MORPH_CLOSE, kernel)
|
||||||
th_frame = 255 - closing
|
th_frame = 255 - closing
|
||||||
except:
|
except:
|
||||||
# I want to eliminate try here because try tends to be slow in execution.
|
# I want to eliminate try here because try tends to be slow in execution.
|
||||||
th_frame = 255 - frame_gray
|
th_frame = 255 - frame
|
||||||
|
|
||||||
|
|
||||||
detect_start_time = timeit.default_timer()
|
detect_start_time = timeit.default_timer()
|
||||||
@ -737,3 +1221,4 @@ def HSRAC(self):
|
|||||||
|
|
||||||
|
|
||||||
|
|
||||||
|
'''
|
||||||
@ -12,9 +12,6 @@ def cal_osc(self, cx, cy):
|
|||||||
self.calibration_frame_counter = None
|
self.calibration_frame_counter = None
|
||||||
self.xoff = cx
|
self.xoff = cx
|
||||||
self.yoff = cy
|
self.yoff = cy
|
||||||
self.now_mode = self.cv_mode[0]
|
|
||||||
self.response_list = []
|
|
||||||
self.response_max = 0
|
|
||||||
if sys.platform.startswith("win"):
|
if sys.platform.startswith("win"):
|
||||||
PlaySound('Audio/compleated.wav', SND_FILENAME | SND_ASYNC)
|
PlaySound('Audio/compleated.wav', SND_FILENAME | SND_ASYNC)
|
||||||
elif self.calibration_frame_counter != None:
|
elif self.calibration_frame_counter != None:
|
||||||
|
|||||||
@ -32,7 +32,7 @@ class SettingsWidget:
|
|||||||
self.gui_min_cutoff = f"-MINCUTOFF{widget_id}-"
|
self.gui_min_cutoff = f"-MINCUTOFF{widget_id}-"
|
||||||
self.gui_eye_falloff = f"-EYEFALLOFF{widget_id}-"
|
self.gui_eye_falloff = f"-EYEFALLOFF{widget_id}-"
|
||||||
self.gui_blink_sync = f"-BLINKSYNC{widget_id}-"
|
self.gui_blink_sync = f"-BLINKSYNC{widget_id}-"
|
||||||
|
self.gui_skip_autoradius = f"-SKIPAUTORADIUS{widget_id}-"
|
||||||
self.gui_HSRACP = f"-HSRACP{widget_id}-"
|
self.gui_HSRACP = f"-HSRACP{widget_id}-"
|
||||||
self.gui_RANSAC3DP = f"-RANSAC3DP{widget_id}-"
|
self.gui_RANSAC3DP = f"-RANSAC3DP{widget_id}-"
|
||||||
self.gui_HSFP = f"-HSFP{widget_id}-"
|
self.gui_HSFP = f"-HSFP{widget_id}-"
|
||||||
@ -177,7 +177,15 @@ class SettingsWidget:
|
|||||||
),
|
),
|
||||||
],
|
],
|
||||||
|
|
||||||
[sg.Text("HSF Radius:", background_color='#424042'),
|
[sg.Checkbox(
|
||||||
|
"HSF: Skip Auto Radius",
|
||||||
|
default=self.config.gui_skip_autoradius,
|
||||||
|
key=self.gui_skip_autoradius,
|
||||||
|
background_color='#424042',
|
||||||
|
tooltip = "To gain more control and possibly better tracking quality of HSF, please disable auto radius to enable manual adjustment.",
|
||||||
|
),
|
||||||
|
|
||||||
|
sg.Text("HSF Radius:", background_color='#424042'),
|
||||||
sg.Slider(
|
sg.Slider(
|
||||||
range=(1, 50),
|
range=(1, 50),
|
||||||
default_value=self.config.gui_HSF_radius,
|
default_value=self.config.gui_HSF_radius,
|
||||||
@ -398,6 +406,11 @@ class SettingsWidget:
|
|||||||
self.config.gui_HSRAC = values[self.gui_HSRAC]
|
self.config.gui_HSRAC = values[self.gui_HSRAC]
|
||||||
changed = True
|
changed = True
|
||||||
|
|
||||||
|
if self.config.gui_skip_autoradius != values[self.gui_skip_autoradius]:
|
||||||
|
self.config.gui_skip_autoradius = values[self.gui_skip_autoradius]
|
||||||
|
changed = True
|
||||||
|
|
||||||
|
|
||||||
if self.config.gui_BLINK != values[self.gui_BLINK]:
|
if self.config.gui_BLINK != values[self.gui_BLINK]:
|
||||||
self.config.gui_BLINK = values[self.gui_BLINK]
|
self.config.gui_BLINK = values[self.gui_BLINK]
|
||||||
changed = True
|
changed = True
|
||||||
|
|||||||
Loading…
Reference in New Issue
Block a user