opencv视频流目标检测

1.采用opencv深度学习模块调用mobilenet_ssd模型进行目标检测，模型可以直接移植到移动端使用，毕竟模型不是很重，直接放线下测试代码：以供参考

import numpy as np
import cv2

deploy_prototxt = 'MobileNet-SSD/deploy.prototxt'
model = 'MobileNet-SSD/mobilenet_iter_73000.caffemodel'
confidence_default = 0.2

CLASSES = ["background", "aeroplane", "bicycle", "bird", "boat",
           "bottle", "bus", "car", "cat", "chair", "cow", "diningtable",
           "dog", "horse", "motorbike", "person", "pottedplant", "sheep",
           "sofa", "train", "tvmonitor"]
COLORS = np.random.uniform(0, 255, size=(len(CLASSES), 3))

print("[info] loading model...")
net = cv2.dnn.readNetFromCaffe(deploy_prototxt, model)

# 打开摄像头
camera = cv2.VideoCapture('video/2c2b6d119306711be3bd7e52f7f27ca7.mp4')
print(camera)
# cv2.namedWindow('Dynamic')
width = int(camera.get(cv2.CAP_PROP_FRAME_WIDTH))
height = int(camera.get(cv2.CAP_PROP_FRAME_HEIGHT))
print(height)
fps = camera.get(cv2.CAP_PROP_FPS)
fourcc = cv2.VideoWriter_fourcc(*"mp4v")
writer = cv2.VideoWriter("video/1.mp4", fourcc, fps, (width, height))
while (True):
    # 读取一帧图像
    ret, frame = camera.read()
    # 判断图片读取成功？
    if ret:
        # gray_img = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
        # 物体检测
        blob = cv2.dnn.blobFromImage(frame, 0.007843, (width, height), 127.5)

        print("[info] computing object detections...")
        net.setInput(blob)
        detections = net.forward()

        for i in np.arange(0, detections.shape[2]):
            confidence = detections[0, 0, i, 2]
            if confidence > confidence_default:
                idx = int(detections[0, 0, i, 1])
                box = detections[0, 0, i, 3:7] * np.array([width, height, width, height])
                (startX, startY, endX, endY) = box.astype("int")

                label = "{}: {:.2f}%".format(CLASSES[idx], confidence * 100)
                print("[info] {}".format(label))
                cv2.rectangle(frame, (startX - 15, startY), (endX + 15, endY), COLORS[idx], 2)
                y = startY - 15 if startY - 15 > 15 else startY + 15
                cv2.putText(frame, label, (startX, y), cv2.FONT_HERSHEY_SIMPLEX, 0.5, COLORS[idx], 2)

        writer.write(frame)
        # 如果按下q键则退出
        if cv2.waitKey(100) & 0xff == ord('q'):
            break

camera.release()
writer.release()
cv2.destroyAllWindows()

代码中使用到的MobileNet_SSD可以在这里下载：提取码：2zsx

2.采用opencv做人脸检测不过效果奇差

import cv2


def DynamicDetect():
    '''
    打开摄像头，读取帧，检测帧中的人脸，扫描检测到的人脸中的眼睛，对人脸绘制蓝色的矩形框，对人眼绘制绿色的矩形框
    '''
    # 创建一个级联分类器 加载一个 .xml 分类器文件. 它既可以是Haar特征也可以是LBP特征的分类器.
    face_cascade = cv2.CascadeClassifier('./haarcascade_frontalface_default.xml')
    face_cascade.load('/home/work/test/opencv/data/haarcascades/haarcascade_frontalface_default.xml')

    # 打开摄像头
    camera = cv2.VideoCapture('video/2c2b6d119306711be3bd7e52f7f27ca7.mp4')
    print(camera)
    # cv2.namedWindow('Dynamic')
    width = int(camera.get(cv2.CAP_PROP_FRAME_WIDTH))
    height = int(camera.get(cv2.CAP_PROP_FRAME_HEIGHT))
    print(height)
    fps = camera.get(cv2.CAP_PROP_FPS)
    fourcc = cv2.VideoWriter_fourcc(*"mp4v")
    writer = cv2.VideoWriter("video/2.mp4", fourcc, fps, (width, height))
    while (True):
        # 读取一帧图像
        ret, frame = camera.read()
        # 判断图片读取成功？
        if ret:
            gray_img = cv2.cvtColor(frame, cv2.COLOR_BGR2GRAY)
            # 人脸检测
            faces = face_cascade.detectMultiScale(gray_img, 1.3, 5)
            for (x, y, w, h) in faces:
                # 在原图像上绘制矩形
                label = "face"
                print("[info] {}".format(label))
                y = x - 15 if y - 15 > 15 else y + 15
                cv2.putText(frame, label, (x, y), cv2.FONT_HERSHEY_SIMPLEX, 0.5, 2)
                cv2.rectangle(frame, (x, y), (x + w, y + h), (255, 0, 0), 2)
            writer.write(frame)
            # cv2.imshow('Dynamic', frame)
            # 如果按下q键则退出
            if cv2.waitKey(100) & 0xff == ord('q'):
                break

    camera.release()
    writer.release()
    cv2.destroyAllWindows()


if __name__ == '__main__':
    DynamicDetect()

代码中用到的xml文件可以在opencv安装目录中找到，一定要用绝对路径，不然会出错。。