School_Zone_V2/implement.py at master · CodeofO/School_Zone_V2 · 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
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
# !pip install ultralytics==8.0.88 # 실시간은 8.0.88 버전에서 가능

import cv2
from ultralytics import YOLO
import os
from multiprocessing import Process
from utils import video_1_load_video, video_2_yolo_1, gps_1_find_nearest_school, led_1, alram_2_result_analysis, led_2_beep


'''
NOTE : module pipeline

utils.py의 모듈들을 순서에 맞게 배치
(각 모듈들의 설명은 utils 페이지에서 확인)

'''
source_path = os.getcwd()
window_name = 'combined'

def implement(source_path, source_1, source_2, latitude, longitude):

    print('✅ source_path : ', source_path)

    cap1, video_info_1 = video_1_load_video(source_path, source_1)
    cap2, video_info_2 = video_1_load_video(source_path, source_2)
    ori_w_1, ori_h_1 = video_info_1.resolution_wh
    ori_w_2, ori_h_2 = video_info_2.resolution_wh

    model1 = YOLO(('yolov8l.pt'))
    model2 = YOLO(('yolov8l.pt'))

    frame_size=(500, 500)
    frame_h = frame_size[0] # 세로
    frame_w = frame_size[1] # 가로

    frame_size_led = (500, 700)
    frame_h_led = frame_size_led[0]
    frame_w_led = frame_size_led[1]

    video_out_path = os.path.join(source_path, 'output', f'{source_1}_{source_2}.mp4')
    cap_out = cv2.VideoWriter(video_out_path,
                            cv2.VideoWriter_fourcc(*'MP4V'),
                            cap1.get(cv2.CAP_PROP_FPS),
                            (frame_w * 2, frame_h))

    car_path = os.path.join(source_path, 'led_source', 'car.png')
    person_path = os.path.join(source_path, 'led_source', 'person.png')
    warn_path = os.path.join(source_path, 'led_source', 'warn.png')
    crash_path = os.path.join(source_path, 'led_source', 'crash.png')

    car_img = cv2.imread(car_path)
    person_img = cv2.imread(person_path)
    crash_img = cv2.imread(crash_path)
    warn_img = cv2.imread(warn_path)

    imogi_ratio_w = int(frame_w_led * 0.3)
    imogi_ratio_h = int(frame_h_led * 0.5)

    car_img = cv2.resize(car_img, (imogi_ratio_w, imogi_ratio_h)) # w : 320, h : 480
    person_img = cv2.resize(person_img, (imogi_ratio_w, imogi_ratio_h))

    crash_img = cv2.resize(crash_img, (int(frame_w_led * 0.2), int(frame_h_led * 0.4)))
    warn_img = cv2.resize(warn_img, (int(frame_w_led * 1), int(frame_h_led * 0.4))) # (800, 320)

    imgs = [car_img, person_img, crash_img, warn_img]

    warning_count_down = 0
    alraming_time = 100
    c = 0
    warning_count = 5
    warning_text1 = ''
    warning_text2 = ''
    real_warning_text1 = ''
    real_warning_text2 = ''
    before_track_infos1 = []
    before_track_infos2 = []
    detect_type='weight'
    weight = 4

    school_info = gps_1_find_nearest_school(latitude, longitude, source_path)

    while True:

        ret1, frame1 = cap1.read()
        ret2, frame2 = cap2.read()

        result1, result2 = video_2_yolo_1(model1, model2, frame1, frame2)

        video_1 = [result1, ori_w_1, ori_h_1, video_info_1, model1, frame1, before_track_infos1, real_warning_text1]
        video_2 = [result2, ori_w_2, ori_h_2, video_info_2, model2, frame2, before_track_infos2, real_warning_text2]
        utils = [detect_type, warning_count, weight, alraming_time, c, warning_count_down]


        cap_out, real_warning_text_set, before_track_infos_set, text_source_set, zone_set = alram_2_result_analysis(video_1, video_2, cap_out, utils, frame_size)

        real_warning_text1 = real_warning_text_set[0]
        real_warning_text2 = real_warning_text_set[1]

        before_track_infos1 = before_track_infos_set[0]
        before_track_infos2 = before_track_infos_set[1]

        utils_led = [alraming_time, c, warning_count_down]
        text_set = [warning_text1, warning_text2]

        real_warning_text_set, warning_count_down = led_1(source_path, real_warning_text1, real_warning_text2, imgs, school_info, frame_size_led, text_source_set, zone_set, utils_led, text_set)

        real_warning_text1 = real_warning_text_set[0]
        real_warning_text2 = real_warning_text_set[1]

        if (warning_count_down > 0) & (0 == (warning_count_down % 5)):
            p_led = Process(target=led_2_beep(source_path))
            p_led.start()

        # 키 입력을 1밀리초 대기하고, 'q' 키가 눌리면 종료
        if cv2.waitKey(1) & 0xFF == ord('q'):
            break

    # 반복문이 종료되면, 모든 창을 닫고 비디오 캡처를 해제
    cv2.destroyAllWindows()
    cap1.release()
    cap2.release()
    cap_out.release()


##############################################################################

source_1 = 'blur_person.mp4'
source_2 = 'blur_car.mp4'


# 초등학교 위경도 좌표 입력
latitude = 35.00000
longitude = 129.00000

implement(source_path, source_1, source_2, latitude, longitude)