VR-2026-Spring/track.py at main · futurerealitylab/VR-2026-Spring · 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
import math
import cv2
import numpy as np

from collections import deque
import imutils

import requests
import urllib.request

from ultralytics import YOLO

device_ID = 0
timestep = 0
num_users = 4

cam_w = 640
cam_h = 480
cam_m = math.sqrt(cam_w**2+cam_h**2)

focal_length = 0.0036
fov = 65
qr_l_real = 0.09525

cam_l = focal_length*math.tan(fov/2*math.pi/180)*2

detector = cv2.QRCodeDetector()

model = YOLO("yolov8s.pt")
objs = ["dining table","bench","chair"]

# mode = ["qr_code","object_detection"]
mode = ["qr_code"]

# Stereocamera debug
vid = cv2.VideoCapture(0)

while(True):
    #frame: 1280x960
    #req = urllib.request.urlopen('http://192.168.1.189/640x480.jpg')
    #arr = np.asarray(bytearray(req.read()), dtype=np.uint8)
    #frame = cv2.imdecode(arr, -1)
    _, frame = vid.read()
    frame = cv2.resize(frame[:,0:1280,:], dsize=(640,480), interpolation=cv2.INTER_CUBIC)
    #frame = cv2.flip(frame, -1)

    #====================QR CODE DETECTION====================#
    if "qr_code" in mode:
        try:
            data, bbox, _ = detector.detectAndDecode(frame)
        except:
            data = None
            bbox = None

        if data:
            box = bbox.flatten().tolist()
            point_1 = [box[0],box[1]]
            point_2 = [box[2],box[3]]
            point_3 = [box[4],box[5]]
            point_4 = [box[6],box[7]]
            center = [(point_1[0]+point_2[0]+point_3[0]+point_4[0])/2,(point_1[1]+point_2[1]+point_3[1]+point_4[1])/2]
            turn_x = (center[0]-cam_m/2)/(cam_m/2)*fov*math.pi/180
            turn_y = (center[1]-cam_m/2)/(cam_m/2)*fov*math.pi/180

            qr_l = (math.dist(point_1,point_2)+math.dist(point_3,point_4))/2
            distance = focal_length*(qr_l_real/(cam_l*qr_l/cam_m))
            qrinfo = {'ID':device_ID,'data':data,'angle':[turn_x,turn_y],'distance':distance}
        else:
            qrinfo = {'ID':device_ID,'data':"no qrcode",'angle':[-1,-1],'distance':-1}
        res = requests.post('http://localhost:2024/qrcode-python', json=qrinfo)
    #====================QR CODE DETECTION====================#

    #==================GREEN BALL DETECTION===================#
    if "green_ball" in mode:
        hsv = cv2.cvtColor(frame, cv2.COLOR_BGR2HSV)
        #minimum and maximum hsv value for green color
        gmin = np.array([25,52,72])
        gmax = np.array([102,255,255])

        g_clamp = cv2.inRange(hsv,gmin,gmax)
        g_clamp = cv2.erode(g_clamp, None, iterations=2)
        g_clamp = cv2.dilate(g_clamp, None, iterations=2)

        cnts = cv2.findContours(g_clamp.copy(), cv2.RETR_EXTERNAL,cv2.CHAIN_APPROX_SIMPLE)
        cnts = imutils.grab_contours(cnts)
        center = (-1,-1)

        position = []
        radii = []

        if len(cnts) > num_users:
            cnts = sorted(cnts, key=cv2.contourArea)
            for i in range(num_users):
                c = cnts[-i]
                ((x, y), radius) = cv2.minEnclosingCircle(c)
                if radius > 10:
                    position.append(x)
                    position.append(y)
                    radii.append(radius)

        if len(radii) > 0:
            index = -1
            rmax = -1
            for i in range(0,len(radii)):
                #print(distances[i])
                if radii[i] > rmax:
                    rmax = radii[i]
                    index = i
            gbinfo = {'ID':device_ID,'radius':rmax,'position':[position[index*2],position[index*2+1]]}
        else:
            gbinfo = {'ID':device_ID,'radius':-1,'position':[]}
        res = requests.post('http://localhost:2024/greenball-python', json=gbinfo)
    #==================GREEN BALL DETECTION===================#

    #=====================OBJECT DETECTION====================#
    if "object_detection" in mode:
        results = model.predict([frame],save=False)
        if len(results[0].boxes.cls) > 0:
            objinfo = {}
            for i in range(len(results[0].boxes.cls)):
                if results[0].names[int(results[0].boxes.cls[i].item())] in objs:
                    x1 = float(results[0].boxes.xyxy[i][0])
                    x2 = float(results[0].boxes.xyxy[i][2])
                    y1 = float(results[0].boxes.xyxy[i][1])
                    y2 = float(results[0].boxes.xyxy[i][3])
                    center = [(x1+x2)/2,(y1+y2)/2]
                    turn_x = (center[0]-cam_m/2)/(cam_m/2)*fov*math.pi/180
                    turn_y = (center[1]-cam_m/2)/(cam_m/2)*fov*math.pi/180
                    objinfo[results[0].names[int(results[0].boxes.cls[i].item())]] = [turn_x,turn_y]
            if objinfo:
                res = requests.post('http://localhost:2024/obj-python', json=objinfo)

    #=====================OBJECT DETECTION====================#
    timestep += 1
    cv2.imshow("frame",frame)
    cv2.waitKey(1)