-
Notifications
You must be signed in to change notification settings - Fork 0
/
detect_fruits.py
134 lines (96 loc) · 4.05 KB
/
detect_fruits.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
import cv2
import json
import click
import numpy as np
from tqdm import tqdm
from pathlib import Path
from typing import Dict
class Fruit:
def __init__(self, img, img_hsv, name):
self.img = img
self.img_hsv = img_hsv
self.name = name
self.mask = None
self.found_objects = 0
def create_mask(self, hsv_lower, hsv_upper):
"""Create a mask with given hsv threshold values"""
for lower, upper in zip(hsv_lower, hsv_upper):
if self.mask is not None:
self.mask = cv2.bitwise_or(self.mask, cv2.inRange(self.img_hsv, lower, upper))
else:
self.mask = cv2.inRange(self.img_hsv, hsv_lower[0], hsv_upper[0])
return self.mask
def get_rect(self, mask):
"""
Finds contours on a given mask, finds rectangles around these contours
and returns dimensions of found_objects rectangles
"""
conts, _hierarchy = cv2.findContours(mask.copy(), cv2.RETR_EXTERNAL, cv2.CHAIN_APPROX_SIMPLE)
rects = []
for cont in conts:
rect = cv2.boundingRect(cont)
x, y, w, h = rect
if w < 70 or h < 70:
continue
rects.append(rect)
self.found_objects += 1
return rects
def draw_rect(self, rects):
"""
Draws a rectangle around found objects on original image
Parameters
----------
opencv object containing rectangles from function get_rect
Returns
-------
None
"""
if rects:
for rect in rects:
x, y, w, h = rect
cv2.rectangle(self.img, (x, y), (x + w, y + h), (0, 255, 0), thickness=2)
cv2.putText(self.img, self.name, (x, y), cv2.FONT_HERSHEY_SIMPLEX, fontScale=1, color=(0, 255, 255),
thickness=2)
cv2.imshow(f"Mask of {self.name.upper()}", self.mask)
def detect_fruits(img_path: str) -> Dict[str, int]:
"""Fruit detection function
Parameters
----------
img_path : str
Path to processed image.
Returns
-------
Dict[str, int]
Dictionary with quantity of each fruit.
"""
img = cv2.imread(img_path, cv2.IMREAD_COLOR)
img_res = cv2.resize(img, dsize=None, fx=0.15, fy=0.15, interpolation=cv2.INTER_CUBIC)
img_hsv = cv2.cvtColor(img_res, cv2.COLOR_BGR2HSV)
apple = Fruit(img=img_res, img_hsv=img_hsv, name="apple")
banana = Fruit(img=img_res, img_hsv=img_hsv, name="banana")
orange = Fruit(img=img_res, img_hsv=img_hsv, name="orange")
apple_mask = apple.create_mask(hsv_lower=np.array([[0, 50, 40], [0, 75, 75], [130, 25, 40]]),
hsv_upper=np.array([[9, 220, 255], [18, 215, 150], [180, 200, 230]]))
banana_mask = banana.create_mask(hsv_lower=np.array([[20, 90, 120]]),
hsv_upper=np.array([[30, 255, 250]]))
orange_mask = orange.create_mask(hsv_lower=np.array([[10, 200, 130]]),
hsv_upper=np.array([[18, 255, 255]]))
apple_rects = apple.get_rect(apple_mask)
banana_rects = banana.get_rect(banana_mask)
orange_rects = orange.get_rect(orange_mask)
return {'apple': apple.found_objects, 'banana': banana.found_objects, 'orange': orange.found_objects}
@click.command()
@click.option('-p', '--data_path', help='Path to data directory', type=click.Path(exists=True, file_okay=False,
path_type=Path), required=True)
@click.option('-o', '--output_file_path', help='Path to output file', type=click.Path(dir_okay=False, path_type=Path),
required=True)
def main(data_path: Path, output_file_path: Path):
img_list = data_path.glob('*.jpg')
results = {}
for img_path in tqdm(sorted(img_list)):
fruits = detect_fruits(str(img_path))
results[img_path.name] = fruits
with open(output_file_path, 'w') as ofp:
json.dump(results, ofp)
if __name__ == '__main__':
main()