-
Notifications
You must be signed in to change notification settings - Fork 0
/
faceRecognition.py
309 lines (265 loc) · 12.2 KB
/
faceRecognition.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
# -*- coding: utf-8 -*-
"""
Created on Fri Apr 20 21:08:01 2018
@author: arshd
"""
import numpy as np
import cv2
import math
import pyautogui
from threading import Timer
# Open Camera
capture = cv2.VideoCapture(0)
# capture.set(cv2.CAP_PROP_FRAME_WIDTH, 1000)
# capture.set(cv2.CAP_PROP_FRAME_HEIGHT, 600)
ch = True
#face_cascade = cv2.CascadeClassifier('haarcascade_frontalface_alt.xml')
def calcAngle(v1, v2):
dot = np.dot(v1, v2)
x_modulus = np.linalg.norm(v1)
y_modulus = np.linalg.norm(v2)
cos_angle = (dot / x_modulus) / y_modulus
angle = np.degrees(np.arccos(cos_angle))
return angle
def get_palm_circle(contourPts, origImg):
dist_max = np.zeros((origImg.shape[0], origImg.shape[1]))
for y in range(0, origImg.shape[0], 4):
for x in range(0, origImg.shape[1], 4):
if origImg[y, x]:
dist_max[y, x] = cv2.pointPolygonTest(contourPts, (x, y), True)
min_val, max_val, min_loc, max_loc = cv2.minMaxLoc(dist_max)
return max_loc, max_val
def lineMagnitude(x1, y1, x2, y2):
lineMagnitude = math.sqrt(math.pow((x2 - x1), 2) + math.pow((y2 - y1), 2))
return lineMagnitude
# Calc minimum distance from a point and a line segment (i.e. consecutive vertices in a polyline).
def DistancePointLine(px, py, x1, y1, x2, y2):
# http://local.wasp.uwa.edu.au/~pbourke/geometry/pointline/source.vba
LineMag = lineMagnitude(x1, y1, x2, y2)
if LineMag < 0.00000001:
DistancePointLine = 9999
return DistancePointLine
u1 = (((px - x1) * (x2 - x1)) + ((py - y1) * (y2 - y1)))
u = u1 / (LineMag * LineMag)
if (u < 0.00001) or (u > 1):
# // closest point does not fall within the line segment, take the shorter distance
# // to an endpoint
ix = lineMagnitude(px, py, x1, y1)
iy = lineMagnitude(px, py, x2, y2)
if ix > iy:
DistancePointLine = iy
else:
DistancePointLine = ix
else:
# Intersecting point is on the line, use the formula
ix = x1 + u * (x2 - x1)
iy = y1 + u * (y2 - y1)
DistancePointLine = lineMagnitude(px, py, ix, iy)
return DistancePointLine
def getMaxContour(contoursVec):
# maximumCtrArea=700
# thresholdedCtrs=[]
max_area = 0
maxContour = []
for i in range(len(contoursVec)):
extractedContour = contoursVec[i]
ctrArea = cv2.contourArea(extractedContour)
if(ctrArea > max_area):
max_area = ctrArea
# thresholdedCtrs.append(extractedContour)
maxContour = extractedContour
return maxContour
face_cascade = cv2.CascadeClassifier('haarcascade_frontalface_default.xml')
kernel_square = np.ones((11, 11), np.uint8)
kernel_ellipse = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (5, 5))
kernel_ellipse2 = cv2.getStructuringElement(cv2.MORPH_ELLIPSE, (8, 8))
fireLeftFlag = True
fireRightFlag = True
fireUpFlag = True
fireDownFlag = True
def turnFlagToTrue(ch):
print(ch)
global fireLeftFlag
global fireRightFlag
global fireUpFlag
global fireDownFlag
if(ch == "L"):
fireLeftFlag = True
elif(ch == "R"):
fireRightFlag = True
elif(ch == "U"):
fireUpFlag = True
elif(ch == "D"):
fireDownFlag = True
try:
while capture.isOpened():
# get camera input
ret, frame = capture.read()
# frame = np.fliplr(frame)
blur = cv2.blur(frame, (3, 3))
# skin detection
hsv = cv2.cvtColor(frame, cv2.COLOR_BGR2YCrCb)
skinMask = cv2.inRange(hsv, np.array(
[54, 131, 110]), np.array([163, 157, 135]))
# perform morphology to remove holes
morphedFrame = skinMask
morphedFrame = cv2.dilate(morphedFrame, kernel_ellipse, iterations=1)
morphedFrame = cv2.erode(morphedFrame, kernel_square, iterations=1)
morphedFrame = cv2.dilate(morphedFrame, kernel_ellipse, iterations=1)
morphedFrame = cv2.medianBlur(morphedFrame, 5)
morphedFrame = cv2.erode(morphedFrame, kernel_ellipse, iterations=1)
morphedFrame = cv2.dilate(morphedFrame, kernel_ellipse2, iterations=1)
morphedFrame = cv2.dilate(morphedFrame, kernel_ellipse, iterations=1)
morphedFrame = cv2.medianBlur(morphedFrame, 5)
# find contours
contoursVec, hierarchy = cv2.findContours(morphedFrame.copy(),cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE)
maxContour=getMaxContour(contoursVec);
origContouredFrame=frame.copy()
cv2.drawContours(origContouredFrame, maxContour, -1, (0,122,122), 3)
hull = cv2.convexHull(maxContour)
topMostPoint = []
for ptr in range(hull.shape[0]):
if(np.size(topMostPoint) == 0 or (hull[ptr, 0, 1] < topMostPoint[1])):
topMostPoint = hull[ptr, 0]
# if (np.size(topMostPoint)>0 ) :
# cv2.putText(origContouredFrame,str(topMostPoint),(topMostPoint[0],topMostPoint[1]),cv2.FONT_HERSHEY_COMPLEX_SMALL,1,(255,255,255),1)
gestureThresh = 30
# swipe detection
timeOut = 2
# cv2.line(origContouredFrame,(gestureThresh,1),(gestureThresh,300),[255,255,0],2)
# cv2.line(origContouredFrame,(np.shape(frame)[1]-gestureThresh,1),(np.shape(frame)[1]-gestureThresh,300),[255,255,0],2)
if(topMostPoint[0] < gestureThresh):
cv2.putText(origContouredFrame, "fire left", (23, 59),
cv2.FONT_HERSHEY_COMPLEX_SMALL, 1, (255, 255, 255), 1)
if(fireLeftFlag):
fireLeftFlag = False
pyautogui.press('left')
t = Timer(timeOut, turnFlagToTrue, ["L"])
t.start()
elif(topMostPoint[0] > np.shape(frame)[1]-gestureThresh):
cv2.putText(origContouredFrame, "fire right", (23, 59),
cv2.FONT_HERSHEY_COMPLEX_SMALL, 1, (255, 255, 255), 1)
if(fireRightFlag):
fireRightFlag = False
pyautogui.press('right')
t = Timer(timeOut, turnFlagToTrue, ["R"])
t.start()
# swipe up and down is not working really well now.
# elif(topMostPoint[1]<gestureThresh):
# cv2.putText(origContouredFrame,"fire up",(23,59),cv2.FONT_HERSHEY_COMPLEX_SMALL,1,(255,255,255),1)
# if(fireUpFlag):
# pyautogui.press('up')
# t= Timer(timeOut,turnFlagToTrue,["U"])
# t.start()
# elif(topMostPoint[1]>np.shape(frame)[0]-gestureThresh):
# cv2.putText(origContouredFrame,"fire down",(23,59),cv2.FONT_HERSHEY_COMPLEX_SMALL,1,(255,255,255),1)
# if(fireDownFlag):
# pyautogui.press('down')
# t= Timer(timeOut,turnFlagToTrue,["D"])
# t.start()
# swipe detection ends
cv2.drawContours(origContouredFrame, [hull], -1, (255, 255, 255), 2)
x, y, w, h = cv2.boundingRect(maxContour)
centerCoord, circRadius = get_palm_circle(maxContour, morphedFrame)
cv2.circle(origContouredFrame, (centerCoord[0], centerCoord[1]), int(
circRadius), (0, 255, 0), 1)
origContouredFrame = cv2.rectangle(origContouredFrame, (
centerCoord[0], centerCoord[1]), (centerCoord[0], centerCoord[1]), (125, 255, 0), 12)
cHull2 = cv2.convexHull(maxContour, returnPoints=False)
defects = cv2.convexityDefects(maxContour, cHull2)
for i in range(defects.shape[0]):
s, e, f, d = defects[i, 0]
start = tuple(maxContour[s][0])
end = tuple(maxContour[e][0])
far = tuple(maxContour[f][0])
cv2.circle(origContouredFrame, far, 5, [0, 0, 255], -1)
# limiting region of interest
scaleFactorForROI = 3.5
roiTopLeftRowNum = round(centerCoord[1]-scaleFactorForROI*circRadius)
roiTopLeftColNum = round(centerCoord[0]-scaleFactorForROI*circRadius)
if(roiTopLeftRowNum < 0):
roiTopLeftRowNum = 0
if(roiTopLeftColNum < 0):
roiTopLeftColNum = 0
roiBottomRightRowNum = round(centerCoord[1]+(1.5*circRadius))
roiBottomRightColNum = round(
centerCoord[0]+(scaleFactorForROI*circRadius))
if(roiBottomRightRowNum > np.shape(frame)[0]):
roiBottomRightRowNum = np.shape(frame)[0]
if(roiBottomRightColNum > np.shape(frame)[1]):
roiBottomRightColNum = np.shape(frame)[1]
origContouredFrame = cv2.rectangle(origContouredFrame, (int(roiTopLeftColNum), int(
roiTopLeftRowNum)), (int(roiBottomRightColNum), int(roiBottomRightRowNum)), (0, 255, 0), 2)
roiFrame = morphedFrame[roiTopLeftRowNum:roiBottomRightRowNum,
roiTopLeftColNum:roiBottomRightColNum].copy()
roiFrameColored = frame[roiTopLeftRowNum:roiBottomRightRowNum,
roiTopLeftColNum:roiBottomRightColNum, :].copy()
# contour extraction on ROI
contoursVec, hierarchy = cv2.findContours(roiFrame.copy(),cv2.RETR_TREE,cv2.CHAIN_APPROX_SIMPLE)
maxContour=getMaxContour(contoursVec);
# finding max inscribed circle
maxInscrbdCirclCenterCoord, maxInscrbdCirclRadius = get_palm_circle(
maxContour, roiFrame)
maxInscrbdCirclCenterCoord = (
int(maxInscrbdCirclCenterCoord[0]), int(maxInscrbdCirclCenterCoord[1]))
cv2.circle(roiFrameColored, (int(maxInscrbdCirclCenterCoord[0]), int(
maxInscrbdCirclCenterCoord[1])), int(maxInscrbdCirclRadius), (0, 255, 0), 1)
cv2.circle(roiFrameColored, maxInscrbdCirclCenterCoord,
5, [0, 255, 0], -1)
# finding min enclosed circle
minEnclCircleCenter, minEnclCircleCenterRadius = cv2.minEnclosingCircle(
maxContour)
minEnclCircleCenter = (
int(minEnclCircleCenter[0]), int(minEnclCircleCenter[1]))
cv2.circle(roiFrameColored, (minEnclCircleCenter[0], minEnclCircleCenter[1]), int(
minEnclCircleCenterRadius), (255, 255, 0), 1)
cv2.circle(roiFrameColored, minEnclCircleCenter,
5, [255, 255, 255], -1)
cHull = cv2.convexHull(maxContour)
cv2.drawContours(roiFrameColored, [cHull], -1, (150, 0, 0), 2)
cHull2 = cv2.convexHull(maxContour, returnPoints=False)
defects = cv2.convexityDefects(maxContour, cHull2)
convexPointsArr = []
ctr = 1
for i in range(defects.shape[0]):
s, e, f, d = defects[i, 0]
start = tuple(maxContour[s][0])
end = tuple(maxContour[e][0])
far = tuple(maxContour[f][0])
dist = DistancePointLine(
far[0], far[1], start[0], start[1], end[0], end[1])
if (dist > maxInscrbdCirclRadius and dist < minEnclCircleCenterRadius):
vectSF = [start[0]-far[0], start[1]-far[1]]
vectEF = [end[0]-far[0], end[1]-far[1]]
angleBwVecs = calcAngle(vectSF, vectEF)
if(angleBwVecs < 90):
convexPointsArr.append([start, far, end])
cv2.line(roiFrameColored, start, end, [0, 255, 0], 2)
cv2.line(roiFrameColored, start, far, [0, 255, 255], 2)
cv2.line(roiFrameColored, far, end, [255, 255, 0], 2)
cv2.circle(roiFrameColored, far, 5, [0, 0, 255], -1)
ctr = ctr+1
thresh = 30
# finger count calculation
if (ctr == 1 and abs(minEnclCircleCenter[0]-maxInscrbdCirclCenterCoord[0]) + abs(minEnclCircleCenter[1]-maxInscrbdCirclCenterCoord[1]) < thresh):
cv2.putText(origContouredFrame, "Zero", (0, 50),
cv2.FONT_HERSHEY_COMPLEX, 2, (0, 255, 255), 2)
else:
cv2.putText(origContouredFrame, str(ctr), (0, 50),
cv2.FONT_HERSHEY_COMPLEX, 2, (0, 255, 255), 2)
cv2.imshow("orig", frame)
cv2.imshow("Thresholded", skinMask)
cv2.imshow("morphedFrame", morphedFrame)
cv2.imshow("roiFrameColored", roiFrameColored)
cv2.resizeWindow("roiFrameColored", 400, 400)
cv2.imshow("origContouredFrame", origContouredFrame)
k = cv2.waitKey(5) & 0xFF
if k == 27:
break
capture.release()
cv2.destroyAllWindows()
except Exception as e:
print('there was an error')
print(e)
capture.release()
cv2.destroyAllWindows()