目标跟踪实验(计算机视觉）通过python 语言编程设计跟踪算法，实现目标行人的选择，跟踪。促进学生理解并掌握相关跟踪算法，培养学生的编程能力。_行人追踪算法

一、实验目的

通过python 语言编程设计跟踪算法，实现目标行人的选择，跟踪。促进学生理解并掌握相关跟踪算法，培养学生的编程能力。

二、实验硬、软件环境

PC、windows10系统,Visual Studio Code编辑器，opencv视觉库，numpy库, matplolib库等。

三、实验内容及步骤

（一）实验内容 

编程实现目标行人的跟踪。

• 实验过程

1. 读取视频或打开摄像头进行实时录像。

tracker_types = ['BOOSTING', 'MIL','KCF', 'TLD', 'MEDIANFLOW', 'MOSSE', 'CSRT']
 
    tracker_type = tracker_types[6]
 
    if tracker_type == 'BOOSTING':
 
        tracker = cv2.legacy.TrackerBoosting_create()
 
    if tracker_type == 'MIL':
 
        tracker = cv2.TrackerMIL_create()
 
    if tracker_type == 'KCF':
 
        tracker = cv2.TrackerKCF_create()
 
    if tracker_type == 'TLD':
 
        tracker = cv2.legacy.TrackerTLD_create()
 
    if tracker_type == 'MEDIANFLOW':
 
        tracker = cv2.legacy.TrackerMedianFlow_create()
 
    if tracker_type == "CSRT":
 
        tracker = cv2.legacy.TrackerCSRT_create()
 
    if tracker_type == "MOSSE":
 
        tracker = cv2.legacy.TrackerMOSSE_create()
 
video = cv2.VideoCapture("D:/XiaoStudy/JiSuanJiShiJue/jisuanjishijue/shiyan3/vtest.mp4")
 
    if not video.isOpened():
 
        print("Could not open video")
 
        sys.exit()
 
    ok, frame = video.read()
 
    if not ok:
 
        print('Cannot read video file')
 
        sys.exit()

1. 人工选择跟踪目标

bbox = (287, 23, 86, 320)

    bbox = cv2.selectROI(frame, False)

    ok = tracker.init(frame, bbox)

目标行人选择算法，即人工选择方框的方法。

3. 选择合适的跟踪算法对目标进行跟踪。

while True:

        # Read a new frame

        ok, frame = video.read()

        if not ok:

            break

        timer = cv2.getTickCount()

        ok, bbox = tracker.update(frame)

        fps = cv2.getTickFrequency() / (cv2.getTickCount() - timer);

        if ok:

            p1 = (int(bbox[0]), int(bbox[1]))

            p2 = (int(bbox[0] + bbox[2]), int(bbox[1] + bbox[3]))

            cv2.rectangle(frame, p1, p2, (255,0,0), 2, 1)

        else :

            cv2.putText(frame, "Tracking failure detected", (100,80), cv2.FONT_HERSHEY_SIMPLEX, 0.75,(0,0,255),2)

        cv2.putText(frame, tracker_type + " Tracker", (100,20), cv2.FONT_HERSHEY_SIMPLEX, 0.75, (50,170,50),2);

        cv2.putText(frame, "FPS : " + str(int(fps)), (100,50), cv2.FONT_HERSHEY_SIMPLEX, 0.75, (50,170,50), 2);

4. 记录目标行人跟踪过程中的跟踪目标图像。

记录的图像，行人遮挡时的图像，进行对比分析

附（记得cmd安装pip install -i https://pypi.tuna.tsinghua.edu.cn/simple opencv-contrib-python）

源代码（记得该视频路径）：

import cv2
import sys
import argparse
 
if __name__ == '__main__' :
 
    # Set up tracker.
    # Instead of MIL, you can also use
 
    tracker_types = ['BOOSTING', 'MIL','KCF', 'TLD', 'MEDIANFLOW', 'MOSSE', 'CSRT']
    tracker_type = tracker_types[6]
 
 ####OpenCv的contrib modules在4.5.1之后不再支持 部分跟踪方法
 # #迁移到cv2.legacy.Multitracker_create。加上“.legacy”即可
    if tracker_type == 'BOOSTING':
        tracker = cv2.legacy.TrackerBoosting_create()
    if tracker_type == 'MIL':
        tracker = cv2.TrackerMIL_create()
    if tracker_type == 'KCF':
        tracker = cv2.TrackerKCF_create()
    if tracker_type == 'TLD':
        tracker = cv2.legacy.TrackerTLD_create()
    if tracker_type == 'MEDIANFLOW':
        tracker = cv2.legacy.TrackerMedianFlow_create()
    if tracker_type == "CSRT":
        tracker = cv2.legacy.TrackerCSRT_create()
    if tracker_type == "MOSSE":
        tracker = cv2.legacy.TrackerMOSSE_create()
    # Read video 获取视频
    video = cv2.VideoCapture("D:/XiaoStudy/JiSuanJiShiJue/jisuanjishijue/shiyan3/vtest.mp4")
 
    # Exit if video not opened.
    if not video.isOpened():
        print("Could not open video")
        sys.exit()
 
    # Read first frame.
    ok, frame = video.read()
    if not ok:
        print('Cannot read video file')
        sys.exit()
    
    # Define an initial bounding box 定义初始化的边界框
    bbox = (287, 23, 86, 320)
 
    # Uncomment the line below to select a different bounding box 取消下面一行的注释，选择一个不同的边界框架
    bbox = cv2.selectROI(frame, False)
 
    # Initialize tracker with first frame and bounding box 用第一帧和包围框初始化跟踪边界框架
    ok = tracker.init(frame, bbox)
 
    while True:
        # Read a new frame
        ok, frame = video.read()
        if not ok:
            break
        
        # Start timer 开始计时
        timer = cv2.getTickCount()
 
        # Update tracker 更新跟踪信号
        ok, bbox = tracker.update(frame)
 
        # Calculate Frames per second (FPS) 计算每秒帧数(FPS) 
        fps = cv2.getTickFrequency() / (cv2.getTickCount() - timer);
 
        # Draw bounding box 绘制边界框
        if ok:
            # Tracking success 跟踪成功
            p1 = (int(bbox[0]), int(bbox[1]))
            p2 = (int(bbox[0] + bbox[2]), int(bbox[1] + bbox[3]))
            cv2.rectangle(frame, p1, p2, (255,0,0), 2, 1)
        else :
            # Tracking failure 跟踪失败
            cv2.putText(frame, "Tracking failure detected", (100,80), cv2.FONT_HERSHEY_SIMPLEX, 0.75,(0,0,255),2)
 
        # Display tracker type on frame 在帧上显示跟踪器类型
        cv2.putText(frame, tracker_type + " Tracker", (100,20), cv2.FONT_HERSHEY_SIMPLEX, 0.75, (50,170,50),2);
    
        # Display FPS on frame 帧显示FPS
        cv2.putText(frame, "FPS : " + str(int(fps)), (100,50), cv2.FONT_HERSHEY_SIMPLEX, 0.75, (50,170,50), 2);
 
 
        # Display result 显示结果
        cv2.imshow("Tracking", frame)
 
        # Exit if ESC pressed 
        k = cv2.waitKey(1) & 0xff
        if k == 27 : break