forked from Future-Gadgets-Lab/FOS-Fire-SOS-System
-
Notifications
You must be signed in to change notification settings - Fork 0
/
tracking_trial5.py
131 lines (117 loc) · 6.13 KB
/
tracking_trial5.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
import time
from imutils.object_detection import non_max_suppression
import numpy as np
import imutils
import sys
import os
#import left
#import right
#import ahead
import cv2
def Output(img):
cv2.imwrite("/Users/ishani/Documents/Celestini/Phase-II/Workspace/PedestrianOnly/output.png",img)
return 0
def pedestrian():
vid='peopleCounter.avi'
cap = cv2.VideoCapture(vid)
hog = cv2.HOGDescriptor()
hog.setSVMDetector(cv2.HOGDescriptor_getDefaultPeopleDetector())
#face_cascade = cv2.CascadeClassifier('/Users/ishani/Documents/Celestini/Phase-II/Workspace/VehiclesOnly/trial2.xml')
start_time = time.time()
tracker_types = ['BOOSTING', 'MIL','KCF', 'TLD', 'MEDIANFLOW', 'GOTURN']
tracker_type = tracker_types[2]
fno = 0
count = 0
hold = 0
skip_frames = 3
tracking = 0
while True:
ret, frames = cap.read()
if not(tracking == 1):
for i in range(skip_frames):
ret, frames = cap.read()
if ret == False:
break
else:
fno = fno+1
frameY, frameX, frameD = frames.shape
# frames = cv2.resize(frames,(int(frameX/2), int(frameY/2)))
frameY, frameX, frameD = frames.shape
frames = frames[(int)(frameY*0.40):(int)(frameY*0.75), (int)(frameX*0.2):(int)(frameX*0.8)]
frameY, frameX, frameD = frames.shape
return_array = []
if fno == 1:
# print("fx = ",frameX, "fy = ",frameY)
# output = "OutputVideo.mp4"
# fourcc = cv2.VideoWriter_fourcc(*'MPEG')
# out = cv2.VideoWriter(output, fourcc, 20.0, (frameX, frameY))
gray = cv2.cvtColor(frames, cv2.COLOR_BGR2GRAY)
# (rects, weights) = hog.detectMultiScale(gray, hitThreshold = 0.3 ,winStride=(8, 8), padding=(24, 24), scale=1.05)#1.01
(rects, weights) = hog.detectMultiScale(gray, hitThreshold = 0.3 ,winStride=(4, 4), padding=(24, 24), scale=1.1)
rectss = np.array([[xC, yC, xC + wC, yC + hC] for (xC, yC, wC, hC) in rects])
pick = non_max_suppression(rectss, probs=None, overlapThresh=0.65)
for (xA, yA, xB, yB) in pick:
pad_w, pad_h = int(0.152*(xB-xA)), int(0.152*(yB-yA))
cv2.rectangle(frames, (xA+pad_w, yA+pad_h), (xB-pad_w, yB-pad_h), (0, 0, 255), 2)
elif (not(fno == 1) and len(rects)==0) or count == 5:
count = 0
tracking = 0
gray = cv2.cvtColor(frames, cv2.COLOR_BGR2GRAY)
(rects, weights) = hog.detectMultiScale(gray, hitThreshold = 0.3 ,winStride=(8, 8), padding=(24, 24), scale=1.05)
rectss = np.array([[xC, yC, xC + wC, yC + hC] for (xC, yC, wC, hC) in rects])
pick = non_max_suppression(rectss, probs=None, overlapThresh=0.65)
for (xA, yA, xB, yB) in pick:
pad_w, pad_h = int(0.152*(xB-xA)), int(0.152*(yB-yA))
cv2.rectangle(frames, (xA+pad_w, yA+pad_h), (xB-pad_w, yB-pad_h), (0, 0, 255), 2)
elif not(len(rects)==0) and count==0:
count = 1
tracking = 1
tracker = []
init_tracker = []
bbox = []
roi_array = np.array([[x, y, x + w, y + h] for (x, y, w, h) in rects])
roi_array = non_max_suppression(roi_array, probs=None, overlapThresh=0.65)
i = 0
# rects = roi_array
for (a,b,c,d) in roi_array:
# sss = frames[rects[i][1] : rects[i][3], rects[i][0] + rects[i][1]:rects[i][2] + rects[i][3]]
# rois.append(sss)
track = cv2.TrackerKCF_create()
tracker.append(track)
bbox.append((a,b,c-a,d-b))
okay = tracker[i].init(frames, bbox[i])
init_tracker.append(okay)
i = i + 1
i = 0
elif not(count==0):
count = count + 1
i = 0
## gray = cv2.cvtColor(frames, cv2.COLOR_BGR2GRAY)
## (rects1, weights) = hog.detectMultiScale(gray, hitThreshold = 0.3 ,winStride=(8, 8), padding=(24, 24), scale=1.05)
## rects1 = np.array([[x, y, x + w, y + h] for (x, y, w, h) in rects1])
## rects1 = non_max_suppression(rects1, probs=None, overlapThresh=0.65)
## if not(len(rects1)==len(roi_array)):
## count = 10
## hold = 1
for i in range(len(bbox)):
## if hold == 1:
## hold = 0
## break
ok, bbox[i] = tracker[i].update(frames)
if ok :
p1 = (int(bbox[i][0]), int(bbox[i][1]))
p2 = (int(bbox[i][0] + bbox[i][2]), int(bbox[i][1] + bbox[i][3]))
pad_w, pad_h = int(0.152*bbox[i][2]), int(0.152*bbox[i][3])
send1 = int(( (bbox[i][0]+pad_w)+(frameX*0.2) )*2)
send2 = int(( (bbox[i][1]+pad_h)+(frameY*0.40))*2)
send3 = int(((bbox[i][2]-pad_w)+(frameX*0.2) )*2)
send4 = int(((bbox[i][3]-pad_h)+(frameY*0.40) )*2)
return_array.append((send1,send2,send3,send4))
cv2.rectangle(frames, (int(bbox[i][0]+pad_w), int(bbox[i][1]+pad_h)), (int(bbox[i][0] + bbox[i][2]-pad_w), int(bbox[i][1] + bbox[i][3]-pad_h)), (0,0,0), 2, 1)
# out.write(frames)
cv2.imshow("JAI MATA DI!",frames)
if cv2.waitKey(33) == 27:
break
#fileP.close()
#cv2.destroyAllWindows()
pedestrian()