YOLOv5+单目实现三维跟踪(python)

news2024/11/24 15:23:44

YOLOv5+单目跟踪(python)

  • 1. 目标跟踪
  • 2. 测距模块
    • 2.1 设置测距模块
    • 2.2 添加测距
  • 3. 主代码
  • 4. 实验效果

相关链接
1. YOLOv5+单目测距(python)
2. YOLOv7+单目测距(python)
3. YOLOv7+单目跟踪(python)
4. 具体效果已在Bilibili发布,点击跳转

工程代码在文章末尾

1. 目标跟踪

用yolov5实现跟踪步骤比较简单,去官网下载deepsort源码,这里有个版本对应关系
DeepSort v3.0 ~YOLOv5 v5.0
DeepSort v4.0 ~ YOLOv5 v6.1

在这里插入图片描述

下载完DeepSort之后去YOLO官网下载相应的YOLO版本,然后把下载的YOLO拖进DeepSort文件夹里,并把YOLO文件夹改名为yolov5,接下来把环境装好,然后运行代码 track.py ,此时如果不出问题就完成了普通检测
也可以用终端运行命令python track.py --source 1.mp4 --show-vid --save-vid --yolo_weights yolov5/weights/yolov5s.pt
请添加图片描述

2. 测距模块

2.1 设置测距模块

测距部分之前已经写过了,具体见这篇文章,我们在Deepsort文件夹里创建一个名为distance.py的文件,或者直接把测距那篇文章里的distance.py文件拖进来也可以
distance.py

foc = 1990.0        # 镜头焦距
real_hight_person = 66.9   # 行人高度
real_hight_car = 57.08      # 轿车高度

# 自定义函数,单目测距
def person_distance(h):
    dis_inch = (real_hight_person * foc) / (h - 2)
    dis_cm = dis_inch * 2.54
    dis_cm = int(dis_cm)
    dis_m = dis_cm/100
    return dis_m

def car_distance(h):
    dis_inch = (real_hight_car * foc) / (h - 2)
    dis_cm = dis_inch * 2.54
    dis_cm = int(dis_cm)
    dis_m = dis_cm/100
    return dis_m

2.2 添加测距

接下来调用测距代码到主代码 track.py 文件中,先在代码开头导入库,添加

from distance import person_distance,car_distance

与测距那篇文章不同,由于跟踪代码自带画框,我们只需要将测距模块写进画框这里,具体如下(注释部分是我添加修改的)

def draw_boxes(img, bbox, identities=None, offset=(0, 0)):
    for i, box in enumerate(bbox):
        x1, y1, x2, y2 = [int(i) for i in box]
        h = y2 - y1   # 计算边框像素点高度
        dis_m = person_distance(h)  # 调用函数,计算行人实际距离
        x1 += offset[0]
        x2 += offset[0]
        y1 += offset[1]
        y2 += offset[1]
        # box text and bar
        id = int(identities[i]) if identities is not None else 0
        color = compute_color_for_labels(id)
        label = '{}{:d}'.format("", id)
        label = label + "  "+"dis:"+str(dis_m)+"m"    # 将距离添加进标签
        t_size = cv2.getTextSize(label, cv2.FONT_HERSHEY_PLAIN, 0, 1)[0]  
        cv2.rectangle(img, (x1, y1), (x2, y2), color, 1)  
        cv2.rectangle(img, (x1, y1), (x1 + t_size[0] + 3, y1 + t_size[1] + 4), color, -1)
        cv2.putText(img, label, (x1, y1 +
                t_size[1] + 1), cv2.FONT_HERSHEY_PLAIN, 1, [255, 255, 255], 2) 
    return img

3. 主代码

import sys
import threading
import random

from yolov5.utils.plots import plot_one_box

sys.path.insert(0, './yolov5')

from yolov5.utils.google_utils import attempt_download
from yolov5.models.experimental import attempt_load
from yolov5.utils.datasets import LoadImages, LoadStreams
from yolov5.utils.general import check_img_size, non_max_suppression, scale_coords, \
    check_imshow,xyxy2xywh
from yolov5.utils.torch_utils import select_device, time_synchronized
from deep_sort_pytorch.utils.parser import get_config
from deep_sort_pytorch.deep_sort import DeepSort
import argparse
import os
import platform
import shutil
import time
from pathlib import Path
import cv2
import torch
import torch.backends.cudnn as cudnn

from yolov5.utils.plots import plot_one_box
palette = (2 ** 11 - 1, 2 ** 15 - 1, 2 ** 20 - 1)
from distance import person_distance,car_distance

def xyxy_to_xywh(*xyxy):
    """" Calculates the relative bounding box from absolute pixel values. """
    bbox_left = min([xyxy[0].item(), xyxy[2].item()])
    bbox_top = min([xyxy[1].item(), xyxy[3].item()])
    bbox_w = abs(xyxy[0].item() - xyxy[2].item())
    bbox_h = abs(xyxy[1].item() - xyxy[3].item())
    x_c = (bbox_left + bbox_w / 2)
    y_c = (bbox_top + bbox_h / 2)
    w = bbox_w
    h = bbox_h
    return x_c, y_c, w, h

def xyxy_to_tlwh(bbox_xyxy):
    tlwh_bboxs = []
    for i, box in enumerate(bbox_xyxy):
        x1, y1, x2, y2 = [int(i) for i in box]
        top = x1
        left = y1
        w = int(x2 - x1)
        h = int(y2 - y1)
        tlwh_obj = [top, left, w, h]
        tlwh_bboxs.append(tlwh_obj)
    return tlwh_bboxs


def compute_color_for_labels(label):
    """
    Simple function that adds fixed color depending on the class
    """
    color = [int((p * (label ** 2 - label + 1)) % 255) for p in palette]
    return tuple(color)

def draw_boxes(img, bbox, identities=None, offset=(0, 0)):
    for i, box in enumerate(bbox):
        x1, y1, x2, y2 = [int(i) for i in box]
        h = y2 - y1
        dis_m = person_distance(h)  # 调用函数,计算行人实际高度
        x1 += offset[0]
        x2 += offset[0]
        y1 += offset[1]
        y2 += offset[1]
        # box text and bar
        id = int(identities[i]) if identities is not None else 0
        color = compute_color_for_labels(id)
        label = '{}{:d}'.format("", id)
        label = label + "  "+"dis:"+str(dis_m)+"m"
        t_size = cv2.getTextSize(label, cv2.FONT_HERSHEY_PLAIN, 0, 1)[0]  #修改字符,原设置: 2,2
        cv2.rectangle(img, (x1, y1), (x2, y2), color, 1)  # 修改线框为1, 原设置:3
        cv2.rectangle(img, (x1, y1), (x1 + t_size[0] + 3, y1 + t_size[1] + 4), color, -1)
        cv2.putText(img, label, (x1, y1 +
                t_size[1] + 1), cv2.FONT_HERSHEY_PLAIN, 1, [255, 255, 255], 2) #修改 2,.,2
    return img


def detect(opt):
    out, source, yolo_weights, deep_sort_weights, show_vid, save_vid, save_txt, imgsz, evaluate = \
        opt.output, opt.source, opt.yolo_weights, opt.deep_sort_weights, opt.show_vid, opt.save_vid, \
            opt.save_txt, opt.img_size, opt.evaluate
    webcam = source == '0' or source.startswith(
        'rtsp') or source.startswith('http') or source.endswith('.txt')


    # initialize deepsort
    cfg = get_config()
    cfg.merge_from_file(opt.config_deepsort)
    attempt_download(deep_sort_weights, repo='mikel-brostrom/Yolov5_DeepSort_Pytorch')
    deepsort = DeepSort(cfg.DEEPSORT.REID_CKPT,
                        max_dist=cfg.DEEPSORT.MAX_DIST, min_confidence=cfg.DEEPSORT.MIN_CONFIDENCE,
                        nms_max_overlap=cfg.DEEPSORT.NMS_MAX_OVERLAP, max_iou_distance=cfg.DEEPSORT.MAX_IOU_DISTANCE,
                        max_age=cfg.DEEPSORT.MAX_AGE, n_init=cfg.DEEPSORT.N_INIT, nn_budget=cfg.DEEPSORT.NN_BUDGET,
                        use_cuda=True)

    # Initialize
    device = select_device(opt.device)

    # The MOT16 evaluation runs multiple inference streams in parallel, each one writing to
    # its own .txt file. Hence, in that case, the output folder is not restored
    if not evaluate:
        if os.path.exists(out):
            pass
            shutil.rmtree(out)  # delete output folder
        os.makedirs(out)  # make new output folder
    half = device.type != 'cpu'  # half precision only supported on CUDA

    # Load model
    model = attempt_load(yolo_weights, map_location=device)  # load FP32 model
    stride = int(model.stride.max())  # model stride
    imgsz = check_img_size(imgsz, s=stride)  # check img_size
    names = model.module.names if hasattr(model, 'module') else model.names  # get class names
    if half:
        model.half()  # to FP16

    # Set Dataloader
    vid_path, vid_writer = None, None
    # Check if environment supports image displays
    if show_vid:
        show_vid = check_imshow()

    if webcam:
        cudnn.benchmark = True  # set True to speed up constant image size inference
        dataset = LoadStreams(source, img_size=imgsz, stride=stride)
    else:
        dataset = LoadImages(source, img_size=imgsz)

    # Get names and colors
    names = model.module.names if hasattr(model, 'module') else model.names
    colors = [[random.randint(0, 255) for _ in range(3)] for _ in names]

    # Run inference
    if device.type != 'cpu':
        model(torch.zeros(1, 3, imgsz, imgsz).to(device).type_as(next(model.parameters())))  # run once
    t0 = time.time()

    save_path = str(Path(out))
    # extract what is in between the last '/' and last '.'
    txt_file_name = source.split('/')[-1].split('.')[0]
    txt_path = str(Path(out)) + '/' + txt_file_name + '.txt'

    for frame_idx, (path, img, im0s, vid_cap) in enumerate(dataset):
        img = torch.from_numpy(img).to(device)
        img = img.half() if half else img.float()  # uint8 to fp16/32
        img /= 255.0  # 0 - 255 to 0.0 - 1.0
        if img.ndimension() == 3:
            img = img.unsqueeze(0)

        # Inference
        t1 = time_synchronized()
        pred = model(img, augment=opt.augment)[0]

        # Apply NMS
        pred = non_max_suppression(
            pred, opt.conf_thres, opt.iou_thres, classes=opt.classes, agnostic=opt.agnostic_nms)
        t2 = time_synchronized()

        # Process detections
        for i, det in enumerate(pred):  # detections per image
            if webcam:  # batch_size >= 1
                p, s, im0 = path[i], '%g: ' % i, im0s[i].copy()
            else:
                p, s, im0 = path, '', im0s

            s += '%gx%g ' % img.shape[2:]  # print string
            save_path = str(Path(out) / Path(p).name)

            if det is not None and len(det):
                # Rescale boxes from img_size to im0 size
                det[:, :4] = scale_coords(
                    img.shape[2:], det[:, :4], im0.shape).round()

                # Print results
                for c in det[:, -1].unique():
                    n = (det[:, -1] == c).sum()  # detections per class
                    s += '%g %ss, ' % (n, names[int(c)])  # add to string

                xywh_bboxs = []
                confs = []

                # Adapt detections to deep sort input format
                for *xyxy, conf, cls in det:
                    # to deep sort format
                    x_c, y_c, bbox_w, bbox_h = xyxy_to_xywh(*xyxy)
                    xywh_obj = [x_c, y_c, bbox_w, bbox_h]
                    xywh_bboxs.append(xywh_obj)
                    confs.append([conf.item()])



                xywhs = torch.Tensor(xywh_bboxs)
                confss = torch.Tensor(confs)

                # pass detections to deepsort
                outputs = deepsort.update(xywhs, confss, im0)

                # draw boxes for visualization
                if len(outputs) > 0:
                    bbox_xyxy = outputs[:, :4]
                    identities = outputs[:, -1]
                    draw_boxes(im0, bbox_xyxy, identities)
                    # to MOT format
                    tlwh_bboxs = xyxy_to_tlwh(bbox_xyxy)

                    # Write MOT compliant results to file
                    if save_txt:
                        for j, (tlwh_bbox, output) in enumerate(zip(tlwh_bboxs, outputs)):
                            bbox_top = tlwh_bbox[0]
                            bbox_left = tlwh_bbox[1]
                            bbox_w = tlwh_bbox[2]
                            bbox_h = tlwh_bbox[3]
                            identity = output[-1]
                            with open(txt_path, 'a') as f:
                                f.write(('%g ' * 10 + '\n') % (frame_idx, identity, bbox_top,
                                                            bbox_left, bbox_w, bbox_h, -1, -1, -1, -1))  # label format

            else:
                deepsort.increment_ages()

            # Print time (inference + NMS)
            print('%sDone. (%.3fs)' % (s, t2 - t1))

            # Stream results
            if show_vid:
                cv2.namedWindow("Webcam", cv2.WINDOW_NORMAL)
                cv2.resizeWindow("Webcam", 1280, 720)
                cv2.moveWindow("Webcam", 0, 100)
                cv2.imshow("Webcam", im0)
                #cv2.imshow(p, im0)
                if cv2.waitKey(1) == ord('q'):  # q to quit
                    raise StopIteration

            # Save results (image with detections)
            if save_vid:
                if vid_path != save_path:  # new video
                    vid_path = save_path
                    if isinstance(vid_writer, cv2.VideoWriter):
                        vid_writer.release()  # release previous video writer
                    if vid_cap:  # video
                        fps = vid_cap.get(cv2.CAP_PROP_FPS)
                        w = int(vid_cap.get(cv2.CAP_PROP_FRAME_WIDTH))
                        h = int(vid_cap.get(cv2.CAP_PROP_FRAME_HEIGHT))
                    else:  # stream
                        fps, w, h = 30, im0.shape[1], im0.shape[0]
                        save_path += '.mp4'

                    vid_writer = cv2.VideoWriter(save_path, cv2.VideoWriter_fourcc(*'mp4v'), fps, (w, h))
                vid_writer.write(im0)

    if save_txt or save_vid:
        print('Results saved to %s' % os.getcwd() + os.sep + out)
        if platform == 'darwin':  # MacOS
            os.system('open ' + save_path)
    print('Done. (%.3fs)' % (time.time() - t0))


if __name__ == '__main__':
    parser = argparse.ArgumentParser()
    parser.add_argument('--yolo_weights', type=str, default='yolov5/weights/yolov5s.pt', help='model.pt path')
    parser.add_argument('--deep_sort_weights', type=str, default='deep_sort_pytorch/deep_sort/deep/checkpoint/ckpt.t7', help='ckpt.t7 path')
    # file/folder, 0 for webcam
    #parser.add_argument('--source', type=str, default='0', help='source')
    parser.add_argument('--source', type=str, default='1.mp4', help='source')
    parser.add_argument('--output', type=str, default='output', help='output folder')  # output folder
    parser.add_argument('--img-size', type=int, default=640, help='inference size (pixels)')
    parser.add_argument('--conf-thres', type=float, default=0.4, help='object confidence threshold')
    parser.add_argument('--iou-thres', type=float, default=0.5, help='IOU threshold for NMS')
    parser.add_argument('--fourcc', type=str, default='mp4v', help='output video codec (verify ffmpeg support)')
    parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu')
    parser.add_argument('--show-vid', action='store_true', help='display tracking video results')
    parser.add_argument('--save-vid', action='store_true', help='save video tracking results')
    parser.add_argument('--save-txt', action='store_true', help='save MOT compliant results to *.txt')
    # class 0 is person, 1 is bycicle, 2 is car... 79 is oven
    #parser.add_argument('--classes', nargs='+', type=int, help='filter by class')
    parser.add_argument('--classes', nargs='+', default=[0], type=int, help='filter by class')
    parser.add_argument('--agnostic-nms', action='store_true', help='class-agnostic NMS')
    parser.add_argument('--augment', action='store_true', help='augmented inference')
    parser.add_argument('--evaluate', action='store_true', help='augmented inference')
    parser.add_argument("--config_deepsort", type=str, default="deep_sort_pytorch/configs/deep_sort.yaml")
    args = parser.parse_args()
    args.img_size = check_img_size(args.img_size)

    with torch.no_grad():
        detect(args)

4. 实验效果

同理,运行 track.py 或者用终端运行命令python track.py --source 1.mp4 --show-vid --save-vid --yolo_weights yolov5/weights/yolov5s.pt

代码打包下载
链接1:https://download.csdn.net/download/qq_45077760/87715776
链接2:https://github.com/up-up-up-up/yolov5_Monocular_ranging

博客主页有更多有关测距的内容

本文来自互联网用户投稿,该文观点仅代表作者本人,不代表本站立场。本站仅提供信息存储空间服务,不拥有所有权,不承担相关法律责任。如若转载,请注明出处:http://www.coloradmin.cn/o/448851.html

如若内容造成侵权/违法违规/事实不符,请联系多彩编程网进行投诉反馈,一经查实,立即删除!

相关文章

C++练级之初级:第四篇

C练级之初级:第四篇 引用 C练级之初级:第四篇1.引用1.1引用的介绍1.2引用的使用场景1.3常引用 2.引用的底层3.引用的与指针的比较 总结 1.引用 1.1引用的介绍 🤔首先还是一个问题,引用是解决C语言什么不足? 指针在&am…

Python自动化sql注入:布尔盲注

在sql注入时,使用python脚本可以大大提高注入效率,这里演示一下编写python脚本实现布尔盲注的基本流程: 演示靶场:sqli-labs 布尔盲注 特点:没有回显没有报错,但根据sql语句正常与否返回不同结果&#x…

新手做电商直播带货怎么和快递合作谈价格

新手做电商直播带货怎么和快递合作谈价格达人带货一般怎样的合作模式?#达人带货 #直播带货 #红人 #百收网 跟快递谈价其实是有方法的,快递的价格不是说不能打下来,就是需要你们多一点点的心机。这个视频我就再给你们补充三个方法,…

HCIP之STP

企业网三层架构 线路冗余—二层网络桥接环路 因为路由器的路由表是由相对完善的计算所得,且存在防环规则;故路由器物理链路上实施备份时,一般不会出现环路;但交换转发数据依赖MAC表(CAM表),该表…

【模式识别4】YOLO目标检测数据集xml格式转txt格式

YOLO目标检测数据集xml格式转txt格式 1. 转换前的xml格式2. xml格式转txt格式代码2.1 源代码2.2 需要修改的地方 3. 转换后的txt格式 代码资源:voc2txt.py 1. 转换前的xml格式 如果我们使用LabelImg工具标注数据集,生成的xml文件如下: xml…

Linux运维:推荐八款Linux远程连接工具

目录 2、XShell 3、SecureCRT 4、PuTTY 5、WindTerm 6、iTerm2 7、MobaXterm 8、Termius 今天给大家推荐八款Linux远程连接工具,非常实用,希望对大家能有所帮助! 1、NxShell NxShell是一款开源的Linux远程管理工具,是我日…

Spring Boot 接口加解密

1. 介绍 在我们日常的Java开发中,免不了和其他系统的业务交互,或者微服务之间的接口调用 如果我们想保证数据传输的安全,对接口出参加密,入参解密。 但是不想写重复代码,我们可以提供一个通用starter,提…

【python数据可视化】基于networkx的10个绘图技巧

一、说明 networkx在02年5月产生,是用python语言编写的软件包,便于用户对复杂网络进行创建、操作和学习。利用networkx可以以标准化和非标准化的数据格式存储网络、生成多种随机网络和经典网络、分析网络结构、建立网络模型、设计新的网络算法、进行网络…

解决java.lang.IllegalStateException: Failed to load ApplicationContext错误~

错误描述: Caused by: org.springframework.beans.factory.BeanDefinitionStoreException: Failed to read candidate component class: file [D:\SSM\SSM_3.22\spring_transaction\target\classes\spring_txAnnotation\Controller\BookController.class]; nested …

装饰器模式C++用法示例

六.装饰器模式 一.装饰器模式1.原理2.适用场景 二.C程序示例1.语法组成2.C示例 一.装饰器模式 1.原理 装饰器模式是一种结构性设计模式,其目的是通过包装对象来扩展其行为而不改变其接口。该模式可以动态地将责任附加到对象上,从而提供一种灵活的替代方…

SQL Server基础 第三章 数据表基本操作(增删改查,不允许保存更改异常!)

往表里插数据我们现在有两种方式第一种是编辑直接修改,第二种是通过查询来修改数据 两种方法的区别 第一种更直接,如果数据量小那么直接改就好了,那如果数据量稍微庞大我们就需要用新建查询来进行表内容的修改了!!&a…

【Linux】进程信号 --- 信号的产生 保存 捕捉递达

被爱情困住的是傻子 文章目录 一、信号的预备知识1.通过生活例子来理解信号2.迁移到进程上来理解信号 二、信号的发送(修改PCB的信号位图)1.通过键盘发送信号(kill指令 和 热键)2.通过系统调用发送信号(kill系统调用 和…

前端技术的miniui 和bootstartp

前端技术的miniui 和bootstartp 目录概述需求: 设计思路实现思路分析1.前端技术boostatrp2.miniui: 参考资料和推荐阅读 Survive by day and develop by night. talk for import biz , show your perfect code,full busy,skip hardness,make a better re…

1686_MATLAB处理Excel文件

全部学习汇总: GreyZhang/g_matlab: MATLAB once used to be my daily tool. After many years when I go back and read my old learning notes I felt maybe I still need it in the future. So, start this repo to keep some of my old learning notes servral …

opencv交叉编译,支持ffmpeg、libjpeg,实现取流及写入jpeg文件

一、准备工作 1、虚拟机交叉编译连 2、在虚拟机目录下创建工作目录,注意不要选择windows共享目录,比如这里选择: /home/cxyazhou/work 3、虚拟机安装cmake cmake-gui pkg-config,直接apt install即可 4、源码下载,需要…

Mongodb的部署

MongoDB下载包获取 https://www.mongodb.com/try/download/community 安装配置 第一步:在服务器 /usr/local 下创建 MongoDB 文件夹,用来存放安装包、服务以及之后 MongoDB 的数据库和日志文件。 cd /usr/local // 进入安装目录 mkdir MongoDB // 创…

【LeetCode: 2369. 检查数组是否存在有效划分 | 暴力递归=>记忆化搜索=>动态规划 | 线性dp】

🚀 算法题 🚀 🌲 算法刷题专栏 | 面试必备算法 | 面试高频算法 🍀 🌲 越难的东西,越要努力坚持,因为它具有很高的价值,算法就是这样✨ 🌲 作者简介:硕风和炜,…

OpenCV(14)-OpenCV4.0中文文档学习2(补充)

相机校准和3D重建 相机校准 标定 findChessboardCorners() 它返回角点和阈值,如果成功找到所有角点,则返回 True。这些角落将按顺序放置(从左到右,从上到下)cornerSubPix()用以寻找图案,找到角点后也可以…

PHREEQC实践建模技术方法

PHREEQC软件为美国地质调查局基于C和C语言开发,面向一般地球化学、水文地球化学模拟的专业软件。该软件历经近30年迭代,功能不断完善,目前已发展成为-以PHREEQC为计算核心(更新至3.7版本),辅以具备二次开发…

【无标题】docker安装jenkins安装流程(二)

配置Publish Over SSH JENKINS_URL/job/goods/build?tokenTOKEN_NAME http://xxxxx:8001/job/goods/job/goods/build?tokenxql1232.提交合并代码触发自动构建 http://xxxxx:8001/buildByToken/build?jobgoods&tokenxql123http://xxxxx:8001/buildByToken/build?jobgood…