import cv2 
import numpy as np 
import mediapipe as mp
import time

mpHands = mp.solutions.hands
hands = mpHands.Hands()  # 设置参数，详见 hands.py 中的 __init__
mpDraw = mp.solutions.drawing_utils  # 将检测出的手上的标记点连接起来
# 定义时间用于后边的fps计算
pTime = 0
cTime = 0

prototxt_path = "weights/deploy.prototxt.txt" 
model_path = "weights/res10_300x300_ssd_iter_140000_fp16.caffemodel" 
model = cv2.dnn.readNetFromCaffe(prototxt_path, model_path) 
cap = cv2.VideoCapture(0, cv2.CAP_DSHOW)
while True: 
    _, image = cap.read() 
    h, w = image.shape[:2] 
    blob = cv2.dnn.blobFromImage(image, 1.0, (300, 300), (104.0, 177.0, 123.0)) 
    model.setInput(blob) 
    output = np.squeeze(model.forward()) 
    font_scale = 1.0 
    for i in range(0, output.shape[0]): 
        confidence = output[i, 2] 
        if confidence > 0.5: 
            box = output[i, 3:7] * np.array([w, h, w, h]) 
            start_x, start_y, end_x, end_y = box.astype(np.int) 
            cv2.rectangle(image, (start_x, start_y), (end_x, end_y), color=(255, 0, 0), thickness=2) 
            cv2.putText(image, f"{confidence*100:.2f}%", (start_x, start_y-5), cv2.FONT_HERSHEY_SIMPLEX, font_scale, (255, 0, 0), 2) 
    imgRGB = cv2.cvtColor(image, cv2.COLOR_BGR2RGB)  # 将BGR格式图像转换为RGB
    results = hands.process(imgRGB)  # 对输入图像进行处理，探索图像中是否有手
    # print(results.multi_hand_landmarks)  # 如果有手，输出手所有0~20个标记点的比例坐标，如果没有，输出None
    if results.multi_hand_landmarks:
        for handLms in results.multi_hand_landmarks:  # 捕捉画面中的每一只手
            for id, lm in enumerate(handLms.landmark):
                # print(id, lm)
                h, w, c = image.shape
                cx, cy = int(lm.x * w), int(lm.y * h)  # 根据比例还原出每一个标记点的像素坐标
                print(id, cx, cy)  # 根据手上标记点的id打印出其相应所在图像中中的像素位置
                if id == 4:  # 可以根据手上标记点的id获得任意id对应的标记点的信息
                    cv2.circle(image, (cx, cy), 10, (255, 0, 255), cv2.FILLED)  # 这里加粗强调了大拇指上的一个标记点

            mpDraw.draw_landmarks(image, handLms, mpHands.HAND_CONNECTIONS)  # 给画面中的每一只手进行标点、连线的操作
    # 得到fps
    cTime = time.time()
    fps = 1/(cTime - pTime)
    pTime = cTime
    # 在画面上显示fps
    cv2.putText(image, 'FPS:' + str(int(fps)), (10, 70), cv2.FONT_ITALIC, 1, (0, 0, 255), 3)
    cv2.imshow("image", image) 
    if cv2.waitKey(1) == ord("q"): 
        break 
    # 点击窗口X按钮关闭窗口
    if cv2.getWindowProperty('image',cv2.WND_PROP_VISIBLE) < 1:        
        break 
cv2.destroyAllWindows() 
cap.release()