为保证您获得更好的学习体验,请使用电脑阅读学习,本专栏对手机用户并不友好!

本文将以detec.py文件为主,带你从头开始逐一追踪代码,了解detect运行流程。

目录

detect.py:

common.py(models):

    DetectMultiBackend:       (line279)

datasets.py(utils):

       LoadImages:                      (line178)

augmentations.py(utils):

       letterbox:                                  (line91)

plots.py(utils):

       Annotator:                               (line68)


detect.py:

FILE                当前文件绝对路径

ROOT              整个yolov5项目的路径(多数情况下在文件的下载转移更新时已存在包导入时无法查找,则可查看该路径是否正确)

Parse_opt        定参,返回opt(存储所有参数信息)

Main                 ①检测requirement中依赖包

                         ②执行Run

Run

Yolov5实现小目标检测_Yolov5实现小目标检测

                        1.判断source传入数据

                                ①is_file: 判断输入图片格式是否在设定格式中(dataset.py

                                ②webcam:false

                         2. Directories,新建保存结果文件夹

# Directories
    save_dir = increment_path(Path(project) / name, exist_ok=exist_ok)  # increment run
    (save_dir / 'labels' if save_txt else save_dir).mkdir(parents=True, exist_ok=True)  # make dir

                         3. Load model,加载模型

# Load model
    device = select_device(device)
    model = DetectMultiBackend(weights, device=device, dnn=dnn, data=data, fp16=half)
    stride, names, pt = model.stride, model.names, model.pt
    imgsz = check_img_size(imgsz, s=stride)  # check image size

                                ①device:选择设备,摄像头、GPU、CPU等

                                ②model:(weight,coco.yaml)显示后端框架(pytorch、TorchScri等)

 DetectMultiBackend(common.py)

                                ③加载模型数据

                                ④imgsz保证图片尺寸为32的倍数,不是则自动计算出32倍数尺寸

                         4. Dataloader,加载待预测图片

# Dataloader
    if webcam:
        view_img = check_imshow()
        cudnn.benchmark = True  # set True to speed up constant image size inference
        dataset = LoadStreams(source, img_size=imgsz, stride=stride, auto=pt)
        bs = len(dataset)  # batch_size
    else:
        dataset = LoadImages(source, img_size=imgsz, stride=stride, auto=pt)
        bs = 1  # batch_size
    vid_path, vid_writer = [None] * bs, [None] * bs

LoadImages(datasets.py)

                         5. Run inference,输入模型推理产生推理结果画出识别框:

                                初始化:

                                       Warmup:传入一张空图片到GPU预热

LoadImages):

                                       im: 图片numpy转pytroch支持的格式

                                              /=255:归一化

                                       扩张维度

                                Inference: 预测

                                       visualize(默认false):若为true,保存推断过程特征图

                                       pred:                         检测框

                                                augment:可对推断做数据增强,但降低模型运行速度

                                                [1,18900,85]:85指4个坐标信息,1个置信度,80个类别概率

                                NMS:   非极大值过滤

                                       pred: 1,5,6: 6指4坐标,1置信度,1类别

                                Process:

                                       det: [5,6],5个矩形框, 6指4坐标,1置信度,1类别

                                       seen:计数器

                                       save_path:图片保存路径

                                       txt_path:默认不保存txt文件

                                       s:

                                       gn:获得原图宽高,保存txt时有用

                                       imc:判断是否把检测框裁剪保存

plots.py):原图绘制

                                       if(det):   画框

                                              det[]       从调整图中坐标映射回原图

                                              遍历所有框:n统计所有框->s打印信息

                                       write results:选择保存方式

                                              add bbox to image(默认选择):

                                                     label:hide_labels、hide_conf(detect参数)是否打印

                                                            标签、置信度

                                                     annotator.box_label:画框

                                                     save_crop:默认false,是否保存截取检测框

                                       stream:       (view_img)展示结果

                                       save_img:    保存图片

                         6. Print results,打印输出结果

# Print results
    t = tuple(x / seen * 1E3 for x in dt)  # speeds per image
    LOGGER.info(f'Speed: %.1fms pre-process, %.1fms inference, %.1fms NMS per image at shape {(1, 3, *imgsz)}' % t)
    if save_txt or save_img:
        s = f"\n{len(list(save_dir.glob('labels/*.txt')))} labels saved to {save_dir / 'labels'}" if save_txt else ''
        LOGGER.info(f"Results saved to {colorstr('bold', save_dir)}{s}")
    if update:
        strip_optimizer(weights)  # update model (to fix SourceChangeWarning)

                                       t:  统计预测每张图片平均时间

                                              seen:预测图片数量,dt每张图片所用时间

                                              LOGGER.info:日志

detect.py源码注释:

(有小改动,可对比自身本地项目文件阅读)

# YOLOv5 🚀 by Ultralytics, GPL-3.0 license
"""
Run inference on images, videos, directories, streams, etc.

Usage - sources:
    $ python path/to/detect.py --weights yolov5s.pt --source 0              # webcam
                                                             img.jpg        # image
                                                             vid.mp4        # video
                                                             path/          # directory
                                                             path/*.jpg     # glob
                                                             'https://youtu.be/Zgi9g1ksQHc'  # YouTube
                                                             'rtsp://example.com/media.mp4'  # RTSP, RTMP, HTTP stream

Usage - formats:
    $ python path/to/detect.py --weights yolov5s.pt                 # PyTorch
                                         yolov5s.torchscript        # TorchScript
                                         yolov5s.onnx               # ONNX Runtime or OpenCV DNN with --dnn
                                         yolov5s.xml                # OpenVINO
                                         yolov5s.engine             # TensorRT
                                         yolov5s.mlmodel            # CoreML (MacOS-only)
                                         yolov5s_saved_model        # TensorFlow SavedModel
                                         yolov5s.pb                 # TensorFlow GraphDef
                                         yolov5s.tflite             # TensorFlow Lite
                                         yolov5s_edgetpu.tflite     # TensorFlow Edge TPU
"""

import argparse
import os
import sys
from pathlib import Path

import cv2
import torch
import torch.backends.cudnn as cudnn

FILE = Path(__file__).resolve()      # 当前文件绝对路径
# 整个yolov5项目的路径
ROOT = FILE.parents[0]  # YOLOv5 root directory
if str(ROOT) not in sys.path:
    sys.path.append(str(ROOT))  # add ROOT to PATH
ROOT = Path(os.path.relpath(ROOT, Path.cwd()))  # relative

from models.common import DetectMultiBackend
from utils.datasets import IMG_FORMATS, VID_FORMATS, LoadImages, LoadStreams
from utils.general import (LOGGER, check_file, check_img_size, check_imshow, check_requirements, colorstr,
                           increment_path, non_max_suppression, print_args, scale_coords, strip_optimizer, xyxy2xywh)
from utils.plots import Annotator, colors, save_one_box
from utils.torch_utils import select_device, time_sync


@torch.no_grad()
def run(weights=ROOT / 'yolov5s.pt',  # model.pt path(s)
        source=ROOT / 'data/images',  # file/dir/URL/glob, 0 for webcam
        data=ROOT / 'data/coco128.yaml',  # dataset.yaml path
        imgsz=(640, 640),  # inference size (height, width)
        conf_thres=0.25,  # confidence threshold
        iou_thres=0.45,  # NMS IOU threshold
        max_det=1000,  # maximum detections per image
        device='',  # cuda device, i.e. 0 or 0,1,2,3 or cpu
        view_img=False,  # show results
        save_txt=False,  # save results to *.txt
        save_conf=False,  # save confidences in --save-txt labels
        save_crop=False,  # save cropped prediction boxes
        nosave=False,  # do not save images/videos
        classes=None,  # filter by class: --class 0, or --class 0 2 3
        agnostic_nms=False,  # class-agnostic NMS
        augment=False,  # augmented inference
        visualize=False,  # visualize features
        update=False,  # update all models
        project=ROOT / 'runs/detect',  # save results to project/name
        name='exp',  # save results to project/name
        exist_ok=False,  # existing project/name ok, do not increment
        line_thickness=3,  # bounding box thickness (pixels)
        hide_labels=False,  # hide labels
        hide_conf=False,  # hide confidences
        half=False,  # use FP16 half-precision inference
        dnn=False,  # use OpenCV DNN for ONNX inference
        ):
    source = str(source)
    save_img = not nosave and not source.endswith('.txt')  # save inference images
    # 判断输入的是否为文件地址且是否包含于相应格式
    is_file = Path(source).suffix[1:] in (IMG_FORMATS + VID_FORMATS)
    is_url = source.lower().startswith(('rtsp://', 'rtmp://', 'http://', 'https://'))   # 判断是否为网络流地址(false)
    # 判断是否传入是0-打开电脑摄像头,默认false
    webcam = source.isnumeric() or source.endswith('.txt') or (is_url and not is_file)
    if is_url and is_file:
        source = check_file(source)  # download

    # Directories
    # 新建保存结果文件夹
    save_dir = increment_path(Path(project) / name, exist_ok=exist_ok)  # increment run
    (save_dir / 'labels' if save_txt else save_dir).mkdir(parents=True, exist_ok=True)  # make dir

    # Load model
    # 加载模型
    device = select_device(device)   # 选择设备,摄像头、GPU、CPU等
    # 显示后端框架(pytorch、TorchScri等)
    model = DetectMultiBackend(weights, device=device, dnn=dnn, data=data, fp16=half)
    stride, names, pt = model.stride, model.names, model.pt         # 加载模型数据
    # 保证图片尺寸为32的倍数,不是则自动计算出32倍数尺寸
    imgsz = check_img_size(imgsz, s=stride)  # check image size

    # Dataloader
    # 加载待预测图片
    if webcam:
        view_img = check_imshow()
        cudnn.benchmark = True  # set True to speed up constant image size inference
        dataset = LoadStreams(source, img_size=imgsz, stride=stride, auto=pt)
        bs = len(dataset)  # batch_size
    else:
        # 初始化
        dataset = LoadImages(source, img_size=imgsz, stride=stride, auto=pt)
        bs = 1  # batch_size
    vid_path, vid_writer = [None] * bs, [None] * bs

    # Run inference
    # 输入模型推理产生推理结果画出识别框:
    # 传入一张空图片到GPU预热
    model.warmup(imgsz=(1 if pt else bs, 3, *imgsz))  # warmup
    dt, seen = [0.0, 0.0, 0.0], 0
    for path, im, im0s, vid_cap, s in dataset:
        t1 = time_sync()
        # 图片numpy转pytroch支持的格式
        im = torch.from_numpy(im).to(device)
        im = im.half() if model.fp16 else im.float()  # uint8 to fp16/32
        # 图片归一化
        im /= 255  # 0 - 255 to 0.0 - 1.0
        # 扩张维度
        if len(im.shape) == 3:
            im = im[None]  # expand for batch dim
        t2 = time_sync()
        dt[0] += t2 - t1

        # Inference
        # 若为true,保存推断过程特征图
        visualize = increment_path(save_dir / Path(path).stem, mkdir=True) if visualize else False
        pred = model(im, augment=augment, visualize=visualize)      # 检测框[1,18900,85]
        # [1,18900,85]指4个坐标信息,1个置信度,80个类别概率
        # augment可对推断做数据增强,但降低模型运行速度
        t3 = time_sync()
        dt[1] += t3 - t2

        # NMS
        # 非极大值过滤
        pred = non_max_suppression(pred, conf_thres, iou_thres, classes, agnostic_nms, max_det=max_det)     # [1,5,6]
        # [1,5,6]: 6指4坐标,1置信度,1类别
        dt[2] += time_sync() - t3

        # Second-stage classifier (optional)
        # pred = utils.general.apply_classifier(pred, classifier_model, im, im0s)

        # Process predictions
        # det:[5,6],5个矩形框, 6指4坐标,1置信度,1类别
        for i, det in enumerate(pred):  # per image
            # 计数器
            seen += 1
            if webcam:  # batch_size >= 1
                p, im0, frame = path[i], im0s[i].copy(), dataset.count
                s += f'{i}: '
            else:
                p, im0, frame = path, im0s.copy(), getattr(dataset, 'frame', 0)

            p = Path(p)  # to Path
            # 图片保存路径
            save_path = str(save_dir / p.name)  # im.jpg
            # 默认不保存txt文件
            txt_path = str(save_dir / 'labels' / p.stem) + ('' if dataset.mode == 'image' else f'_{frame}')  # im.txt
            s += '%gx%g ' % im.shape[2:]  # print string
            # 获得原图宽高,用于保存txt
            gn = torch.tensor(im0.shape)[[1, 0, 1, 0]]  # normalization gain whwh
            # 判断是否把检测框裁剪保存
            imc = im0.copy() if save_crop else im0  # for save_crop
            # 原图绘制
            annotator = Annotator(im0, line_width=line_thickness, example=str(names))
            # 画框
            if len(det):
                # Rescale boxes from img_size to im0 size
                # 从调整图中坐标映射回原图
                det[:, :4] = scale_coords(im.shape[2:], det[:, :4], im0.shape).round()

                # Print results
                # 遍历所有框:n统计所有框用于基于s打印信息
                for c in det[:, -1].unique():
                    n = (det[:, -1] == c).sum()  # detections per class
                    s += f"{n} {names[int(c)]}{'s' * (n > 1)}, "  # add to string

                # Write results
                # 选择保存方式(默认选择第二种,即非txt)
                for *xyxy, conf, cls in reversed(det):
                    if save_txt:  # Write to file
                        xywh = (xyxy2xywh(torch.tensor(xyxy).view(1, 4)) / gn).view(-1).tolist()  # normalized xywh
                        line = (cls, *xywh, conf) if save_conf else (cls, *xywh)  # label format
                        with open(txt_path + '.txt', 'a') as f:
                            f.write(('%g ' * len(line)).rstrip() % line + '\n')

                    if save_img or save_crop or view_img:  # Add bbox to image
                        c = int(cls)  # integer class
                        # 是否打印标签、置信度
                        label = None if hide_labels else (names[c] if hide_conf else f'{names[c]} {conf:.2f}')
                        # 画框
                        annotator.box_label(xyxy, label, color=colors(c, True))
                        # 是否保存截取检测框(默认false)
                        if save_crop:
                            save_one_box(xyxy, imc, file=save_dir / 'crops' / names[c] / f'{p.stem}.jpg', BGR=True)

            # Stream results
            im0 = annotator.result()
            if view_img:
                cv2.imshow(str(p), im0)
                cv2.waitKey(1)  # 1 millisecond

            # Save results (image with detections)
            if save_img:
                if dataset.mode == 'image':
                    cv2.imwrite(save_path, im0)
                else:  # 'video' or 'stream'
                    if vid_path[i] != save_path:  # new video
                        vid_path[i] = save_path
                        if isinstance(vid_writer[i], cv2.VideoWriter):
                            vid_writer[i].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 = str(Path(save_path).with_suffix('.mp4'))  # force *.mp4 suffix on results videos
                        vid_writer[i] = cv2.VideoWriter(save_path, cv2.VideoWriter_fourcc(*'mp4v'), fps, (w, h))
                    vid_writer[i].write(im0)

        # Print time (inference-only)
        LOGGER.info(f'{s}Done. ({t3 - t2:.3f}s)')

    # Print results
    # 统计预测每张图片平均时间,seen即预测图片数量,dt即每张图片所用时间
    t = tuple(x / seen * 1E3 for x in dt)  # speeds per image
    LOGGER.info(f'Speed: %.1fms pre-process, %.1fms inference, %.1fms NMS per image at shape {(1, 3, *imgsz)}' % t)
    if save_txt or save_img:
        s = f"\n{len(list(save_dir.glob('labels/*.txt')))} labels saved to {save_dir / 'labels'}" if save_txt else ''
        LOGGER.info(f"Results saved to {colorstr('bold', save_dir)}{s}")
    if update:
        strip_optimizer(weights)  # update model (to fix SourceChangeWarning)

# 定参,返回opt(存储所有参数信息)
def parse_opt():
    parser = argparse.ArgumentParser()
    parser.add_argument('--weights', nargs='+', type=str, default=ROOT / 'runs/train//exp13/weights/best.pt', help='model path(s)')
    parser.add_argument('--source', type=str, default=ROOT / 'C:/Users/Pictures/Saved Pictures/read11.avi',  help='file/dir/URL/glob, 0 for webcam')
    parser.add_argument('--data', type=str, default=ROOT / 'data/coco.yaml', help='(optional) dataset.yaml path')
    parser.add_argument('--imgsz', '--img', '--img-size', nargs='+', type=int, default=[960, 540], help='inference size h,w')
    parser.add_argument('--conf-thres', type=float, default=0.25, help='confidence threshold')
    parser.add_argument('--iou-thres', type=float, default=0.45, help='NMS IoU threshold')
    parser.add_argument('--max-det', type=int, default=1000, help='maximum detections per image')
    parser.add_argument('--device', default='', help='cuda device, i.e. 0 or 0,1,2,3 or cpu')
    parser.add_argument('--view-img', action='store_true', help='show results')      # 置信度
    parser.add_argument('--save-txt', action='store_true', help='save results to *.txt')
    parser.add_argument('--save-conf', action='store_true', help='save confidences in --save-txt labels')
    parser.add_argument('--save-crop', action='store_true', help='save cropped prediction boxes')
    parser.add_argument('--nosave', action='store_true', help='do not save images/videos')
    parser.add_argument('--classes', nargs='+', type=int, help='filter by class: --classes 0, or --classes 0 2 3')
    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('--visualize', action='store_true', help='visualize features')
    parser.add_argument('--update', action='store_true', help='update all models')
    parser.add_argument('--project', default=ROOT / 'runs/detect', help='save results to project/name')
    parser.add_argument('--name', default='exp', help='save results to project/name')
    parser.add_argument('--exist-ok', action='store_true', help='existing project/name ok, do not increment')
    parser.add_argument('--line-thickness', default=3, type=int, help='bounding box thickness (pixels)')
    parser.add_argument('--hide-labels', default=False, action='store_true', help='hide labels')
    parser.add_argument('--hide-conf', default=False, action='store_true', help='hide confidences')
    parser.add_argument('--half', action='store_true', help='use FP16 half-precision inference')
    parser.add_argument('--dnn', action='store_true', help='use OpenCV DNN for ONNX inference')
    opt = parser.parse_args()
    opt.imgsz *= 2 if len(opt.imgsz) == 1 else 1  # expand
    print_args(FILE.stem, opt)
    return opt


def main(opt):
    # 检测requirement中依赖包
    check_requirements(exclude=('tensorboard', 'thop'))
    run(**vars(opt))


if __name__ == "__main__":
    opt = parse_opt()
    main(opt)

common.py(models):

    DetectMultiBackend:       (line279)

        w                             判断weights是否为list,若是取出第一个值作为传入路径

        model_type             判断模型格式(pt、jit等),执行相应加载模式

        fp16                         半精度推算

        if data                      加载传入文件,获取names

datasets.py(utils):

       LoadImages:                      (line178)

         P:                            根据相对路径获得绝对路径

                                        判断是否带*,是否为文件夹,是否为文件

         images/videos:      获取文件格式,判断图片格式是否包含在规定拓展名中

         nf:                         所有文件数

         count:                   文件中图片计数器,起索引作用

         img0:                    读入初始图

         s:                           字符串,表示输入的是第几张图片,用于后续打印

augmentations.py->letterbox)(需要32倍宽高)

         vid_cap(None):

         Convert:

augmentations.py(utils):

       letterbox:                                  (line91)

         r:                       长边缩放图片,(long/640)

         填充图片:

         if auto:              若auto(默认true)为true,判断图片宽高是否为32倍数,若满足直接读取

plots.py(utils):

       Annotator:                               (line68)

         初始化:

                     If-else:        默认opencv画框

                     box_label:   画框画标签

本文谨以用于代码详解笔记,而非项目点评和项目研发,若本文何处理解有误,劳烦广大读者指正。本人才疏学浅,请多包涵。