-
Notifications
You must be signed in to change notification settings - Fork 0
/
adi.py
201 lines (171 loc) · 7.44 KB
/
adi.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
import cv2
import numpy as np
from typing import *
import os
class MotionADI(object):
# limit = batas ADI yang di proses || fd = batas frame untuk mereset put-text deteksi
def __init__(self, thresh = 0.5, limit = 10, fdn = 20, adi_path = "result", **kwargs):
self.fdn = fdn
self.thresh = thresh
self.limit = limit
self.frames: List = []
self.idx = 0
self.motion_idx = []
self.motion_frames = []
self.font = cv2.FONT_HERSHEY_SIMPLEX
self.is_detected = False
self.detected_id = 0
self.adi_path = adi_path
self.adi_id = 0
self.human_detector = HumanDetector()
def _segmentation(self, frame):
gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
# r = 500.0 / gray.shape[1]
# dim = (500, int(gray.shape[0] * r))
# gray = cv2.resize(gray, dim, interpolation=cv2.INTER_AREA)
# cv2.imshow('grayscale',gray)
ret, thresh = cv2.threshold(gray, 0, 255, cv2.THRESH_BINARY + cv2.THRESH_OTSU)
op_kernel = np.ones((5, 5), np.uint8)
opening = cv2.morphologyEx(thresh, cv2.MORPH_OPEN, op_kernel)
mean_kernel = np.ones((5, 5), np.uint8) / np.mean(opening)
mean = cv2.filter2D(opening, -1, mean_kernel)
output = mean
return output
# Mengkombinsai setiap frame yang ada gerakan per 20 frame
def combine_motion_frame(self):
fshape = self.motion_frames[-1].shape
image = np.zeros(fshape)
for frame in self.motion_frames:
image = np.add(image, frame)
return image
def filter(self, frame):
# Menghitung jumlah frame
seg = self._segmentation(frame)
self.frames.append(seg)
self._put_text(frame, f'Frame Number: {str(self.idx)}', loc=(10, 40))
if len(self.frames) > 0:
abs_diff = cv2.absdiff(self.frames[self.idx],self.frames[self.idx - 1])
motion = np.mean(abs_diff) > self.thresh
# mengecek jika ada gerakan dan itu gerakan manusia maka sistem mengeluarkan "Human Detection"
if motion and self.human_detector.detect(frame) :
# Looping id jika ada gerakan
self.motion_idx.append(self.idx)
if len(self.motion_frames) < self.limit:
# Mengumpulkan Frame
self.motion_frames.append(abs_diff)
return False, abs_diff
else :
# Meyimpan hasil 10 frame ketika ada gerakan
image_adi = self.combine_motion_frame()
path = os.path.join(self.adi_path, f'frame_{self.adi_id}.jpg')
cv2.imwrite(path, image_adi)
self.adi_id += 1
self.motion_frames = []
self.motion_idx = []
self.is_detected = True
self.detected_id = self.idx
return True, image_adi
else:
return False, abs_diff
else:
return False, np.zeros(frame.shape)
# Penambahan/Looping frame id
def increment_id(self):
self.idx = self.idx + 1
def show_detection(self, frame):
# Jika terdeteksi munculin text
if self.is_detected and self.human_detector.detect(frame):
self._put_text(frame, "Human Motion: Detected")
# Jika sudah lebih dari fdn maka akan kembali ke tidak terdeteksi jika tidak ada gerakan
if self.idx - self.detected_id >= self.fdn:
self.is_detected = False
self.detected_id = 0
else:
# Jika tidak terdeteksi munculin text
self._put_text(frame, "Human Motion: Undetected")
def _put_text(self, frame, text, loc=(10, 20)):
cv2.putText(frame, text, loc, self.font, 0.7, (0, 0, 255), 2, cv2.LINE_AA)
class HumanDetector():
def __init__(self, upper_detect = False, face_detect = False):
self.upper_detect = upper_detect
self.face_detect = face_detect
self.build()
def build(self):
self.person_cascade = cv2.CascadeClassifier(os.path.join('data/haarcascade_fullbody.xml'))
if self.upper_detect:
self.upper_cascade = cv2.CascadeClassifier(os.path.join('data/haarcascade_upperbody.xml'))
if self.face_detect:
self.face_cascade = cv2.CascadeClassifier(os.path.join('data/haarcascade_frontalface_default.xml'))
def detect(self, frame):
# mengecek dengan haar cascade clasifier apaka itu manusia atau bukan
gray = cv2.cvtColor(frame, cv2.COLOR_RGB2GRAY)
human = self.person_cascade.detectMultiScale(gray)
if len(human) > 0:
if self.upper_detect:
upper_body = self.upper_cascade.detectMultiScale(gray)
if len(upper_body) > 0:
return True
if self.face_detect:
face = self.face_cascade.detectMultiScale(gray)
if len(face) > 0:
return True
if self.face_detect and self.upper_detect:
upper_body = self.upper_cascade.detectMultiScale(gray)
face = self.face_cascade.detectMultiScale(gray)
if len(face) > 0 and len(upper_body) > 0:
return True
return True
else:
return False
# person_cascade = cv2.CascadeClassifier(os.path.join('data/haarcascade_fullbody.xml'))
# upper_body_cascade = cv2.CascadeClassifier(os.path.join('data/haarcascade_upperbody.xml'))
# lower_body_cascade = cv2.CascadeClassifier(os.path.join('data/haarcascade_lowerbody.xml'))
# face_cascade = cv2.CascadeClassifier(os.path.join('data/haarcascade_frontalface_default.xml'))
# def tester(frame):
#
#
# gray = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
# human = face_cascade.detectMultiScale(gray)
#
# for (x, y, w, h) in human:
# # for whole body detetction
# cv2.rectangle(frame, (x, y), (x + w, y + h), (0, 255, 0), 3)
# roi_gray = gray[y:y + h, x:x + w]
# roi_color = frame[y:y + h, x:x + w]
#
# # for upper body detection
# # upper_body = upper_body_cascade.detectMultiScale(roi_gray)
# # for (ux, uy, uw, uh) in upper_body:
# # cv2.rectangle(roi_color, (ux, uy), (ux + uw, uy + uh), (0, 0, 255), 2)
#
# # for lower body detection
# # lower_body = lower_body_cascade.detectMultiScale(roi_gray)
# # for (lx, ly, lw, lh) in lower_body:
# # cv2.rectangle(roi_color, (lx, ly), (lx + lw, ly + lh), (255, 0, 0), 2)
#
# # for face detection
# face = face_cascade.detectMultiScale(roi_gray)
# for (fx, fy, fw, fh) in face:
# cv2.rectangle(roi_color, (fx, fy), (fx + fw, fy + fh), (120, 230, 0), 4)
cap = cv2.VideoCapture("video/1.mp4")
madi = MotionADI(thresh = 0.3, fdn = 10)
while (True):
# Capture frame-by-frame
ret, frame = cap.read()
mot_detect, output = madi.filter(frame)
madi.show_detection(frame)
madi.increment_id()
# tester(frame)
r = 800.0 / frame.shape[1]
dim = (800, int(frame.shape[0] * r))
frame = cv2.resize(frame, dim, interpolation=cv2.INTER_AREA)
r = 500.0 / output.shape[1]
dim = (500, int(output.shape[0] * r))
output = cv2.resize(output, dim, interpolation=cv2.INTER_AREA)
cv2.imshow('motion', output)
cv2.imshow('frame', frame)
if cv2.waitKey(1) & 0xFF == ord('q'):
break
# When everything done, release the capture
cap.release()
cv2.destroyAllWindows()