DayNightClassifier/opencv.py at main · wmapro1001/DayNightClassifier · 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
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
from utils import daynight_helper
import random
import matplotlib.pyplot as plt
import cv2
import numpy as np


training_data_path = 'data/day_night_images/training'
training_data = daynight_helper.load_dataset(training_data_path)
std_training_data = daynight_helper.standardize_inputs(training_data)

img = std_training_data[0][0]
label = std_training_data[0][1]
plt.imshow(img)
print('Shape ', img.shape)
print('label: ', label)

def brightness_value(img):
	hsv_img = cv2.cvtColor(img, cv2.COLOR_RGB2HSV)
	# get 2nd channel values of hsv values
	v_values = np.sum(hsv_img[:,:,2])
	# dimensions
	area = img.shape[0] * img.shape[1]
	avg_brightness = v_values/area
	return avg_brightness

def estimate_label(img, threshold):
	avg_brightness = brightness_value(img)
	predicted_label = 0
	threshold = threshold
	if avg_brightness > threshold:
		predicted_label = 1
	return predicted_label, avg_brightness


def optimize_threshold(img_data, threshold):
	img = img_data[0]
	label = img_data[1]

	threshold = threshold
	pred_label, avg_brightness = estimate_label(img, threshold)

	if pred_label == label:
		return threshold
	else:
		return round((avg_brightness + threshold)/2)


# 120 is initial value
threshold = 120
for i in range(0, len(std_training_data)):
	img_data = std_training_data[i]
	threshold = optimize_threshold(img_data, threshold)
#116
print('threshold = ', threshold)


# test the data
test_data_path = "data/day_night_images/test"
test_data = daynight_helper.load_dataset(test_data_path)
test_data = daynight_helper.standardize_inputs(test_data)

correctly_classfied = []
misclassified = []

for i in range(0, len(test_data)):
	img_data = test_data[i]
	pred, avg_brightness = estimate_label(img_data[0], threshold)
	label = img_data[1]

	if pred == label:
		correctly_classfied.append(img_data)
	else:
		misclassified.append(img_data)

print("Total = ", len(test_data))
print("Correct predictions = ", len(correctly_classfied))
print("Misclassified = ", len(misclassified))
print("Accuracy = ", len(correctly_classfied)/len(test_data))