site-surveillance-cv/discarded_during_research.txt at main · bigsaso/site-surveillance-cv · GitHub

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
## Haar Cascades
# haar_cascade_frontfaces = cv2.CascadeClassifier('haarcascade/haarcascade_frontalface_alt_tree.xml')
# haar_cascade_profilefaces = cv2.CascadeClassifier('haarcascade/haarcascade_profileface.xml')
# haar_cascade_fullbodies = cv2.CascadeClassifier('haarcascade/haarcascade_fullbody.xml')
# haar_cascade_upperbodies = cv2.CascadeClassifier('haarcascade/haarcascade_upperbody.xml')
# haar_cascade_lowerbodies = cv2.CascadeClassifier('haarcascade/haarcascade_lowerbody.xml')
# frontfaces = haar_cascade_frontfaces.detectMultiScale(gaussian_blur, 1.1, 2)
# profilefaces = haar_cascade_profilefaces.detectMultiScale(gaussian_blur, 1.1, 4)
# fullbodies = haar_cascade_fullbodies.detectMultiScale(gaussian_blur, 1.1, 4)
# upperbodies = haar_cascade_upperbodies.detectMultiScale(gaussian_blur, 1.1, 4)
# lowerbodies = haar_cascade_lowerbodies.detectMultiScale(gaussian_blur, 1.1, 4)

## Draw rectangles for HAAR
# This part was removed to get rid of false positives
# for (x, y, w, h) in frontfaces:
#     cv2.rectangle(image, (x, y), (x + w, y + h), red, 2)
# for (x, y, w, h) in profilefaces:
#     cv2.rectangle(image, (x, y), (x + w, y + h), red, 2)
# for (x, y, w, h) in fullbodies:
#     cv2.rectangle(image, (x, y), (x + w, y + h), blue, 2)
# for (x, y, w, h) in upperbodies:
#     cv2.rectangle(image, (x, y), (x + w, y + h), orange, 2)
# for (x, y, w, h) in lowerbodies:
#     cv2.rectangle(image, (x, y), (x + w, y + h), purple, 2)

## Draw bounding boxes for YOLO detections -- used to draw rectangles around all objects
# for i in indices.flatten():
#     box = boxes[i]
#     x, y, w, h = box[0], box[1], box[2], box[3]
#     label = str(classes[class_ids[i]])
#     color = (0, 255, 0)
#     cv2.rectangle(image, (x, y), (x + w, y + h), color, 2)
#     cv2.putText(image, label, (x, y + 30), cv2.FONT_HERSHEY_PLAIN, 3, color, 3)

# Counting -- moved to its own variable
# Number of boxes
# frontface_count = len(frontfaces)
# profileface_count = len(profilefaces)
# fullbody_count = len(fullbodies)
# upperbody_count = len(upperbodies)
# lowerbody_count = len(lowerbodies)
# yolo_count = len(indices)

# print(f"Front faces detected: {frontface_count}")
# print(f"Profile faces detected: {profileface_count}")
# print(f"Fullbodies detected: {fullbody_count}")
# print(f"Upperbodies detected: {upperbody_count}")
# print(f"Lowerbodies detected: {lowerbody_count}")

# Median blur
median_blur = cv2.medianBlur(gray, 5)
# Bilateral filter
bilateral_filter = cv2.bilateralFilter(gray, 9, 75, 75)
saveImage(median_blur, 'median_blur.png')
saveImage(bilateral_filter, 'bilateral_filter.png')