2021SC@SDUSC山东大学软件学院软件工程应用与实践——yolov5代码分析——第十三篇——detect.py（2）

2021SC@SDUSC

目录

正式推理

---

请不要忽视代码中的注释

正式推理

 for path, img, im0s, vid_cap in dataset:
        t1 = time_sync()
# 处理每一张图片/视频的格式
        if onnx:
            img = img.astype('float32')
        else:
            img = torch.from_numpy(img).to(device)
            img = img.half() if half else img.float()  # uint8 to fp16/32
        img = img / 255.0  # 0 - 255 to 0.0 - 1.0
        if len(img.shape) == 3:
            img = img[None]  # expand for batch dim
        t2 = time_sync()
        dt[0] += t2 - t1

nms出去多余的框

# NMS
        pred = non_max_suppression(pred, conf_thres, iou_thres, classes, agnostic_nms, max_det=max_det)
        dt[2] += time_sync() - t3
# Apply NMS  进行NMS
        # conf_thres: 置信度阈值
        # iou_thres: iou阈值
        # classes: 是否只保留特定的类别 默认为None
        # agnostic_nms: 进行nms是否也去除不同类别之间的框
        # max_det: 每张图片的最大目标个数 默认1000
        # pred: [num_obj, 6] = [5, 6]   这里的预测信息pred还是相对于 img_size(640) 的

考虑进行二次分类

 # Second-stage classifier (optional)
# Apply Classifier  如果需要二次分类 就进行二次分类  一般是不需要的
        if classify:
            pred = apply_classifier(pred, modelc, img, im0s)

后续保存或者打印预测信息

# Process predictions
# 对每张图片进行处理  将pred(相对img_size 640)映射回原图img0 size
        for i, det in enumerate(pred):  # per image
            seen += 1
            if webcam:  # batch_size >= 1
 # 如果输入源是webcam（网页）则batch_size>=1 取出dataset中的一张图片
                p, s, im0, frame = path[i], f'{i}: ', im0s[i].copy(), dataset.count
            else:
# 但是大部分我们一般都是从LoadImages流读取本都文件中的照片或者视频 所以batch_size=1
                # p: 当前图片/视频的绝对路径 如 F:\yolo_v5\yolov5-U\data\images\bus.jpg
                # s: 输出信息 初始为 ''
                # im0: 原始图片 letterbox + pad 之前的图片
                # frame: 初始为0  可能是当前图片属于视频中的第几帧？

                p, s, im0, frame = path, '', im0s.copy(), getattr(dataset, 'frame', 0)

# 当前图片路径 如 F:\yolo_v5\yolov5-U\data\images\bus.jpg
            p = Path(p)  # to Path
# 图片/视频的保存路径save_path 如 runs\\detect\\exp8\\bus.jpg
            save_path = str(save_dir / p.name)  # img.jpg
# txt文件(保存预测框坐标)保存路径 如 runs\\detect\\exp8\\labels\\bus
            txt_path = str(save_dir / 'labels' / p.stem) + ('' if dataset.mode == 'image' else f'_{frame}')  # img.txt
# print string  输出信息  图片shape (w, h)
            s += '%gx%g ' % img.shape[2:]  # print string
#  normalization gain gn = [w, h, w, h]  用于后面的归一化
            gn = torch.tensor(im0.shape)[[1, 0, 1, 0]]  # normalization gain whwh
# imc: for save_crop 在save_crop中使用
            imc = im0.copy() if save_crop else im0  # for save_crop
            annotator = Annotator(im0, line_width=line_thickness, pil=not ascii)
            if len(det):
                # Rescale boxes from img_size to im0 size
 # 将预测信息（相对img_size 640）映射回原图 img0 size
                det[:, :4] = scale_coords(img.shape[2:], det[:, :4], im0.shape).round()

                # Print results
 # 输出信息s + 检测到的各个类别的目标个数
                for c in det[:, -1].unique():
 # 将每个图片的预测信息分别存入save_dir/labels下的xxx.txt中 每行: class_id+score+xywh
                    n = (det[:, -1] == c).sum()  # detections per class
                    s += f"{n} {names[int(c)]}{'s' * (n > 1)}, "  # add to string

                # Write results
# 保存预测信息: txt、img0上画框、crop_img
                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')

# 在原图上画框 + 将预测到的目标剪切出来 保存成图片 保存在save_dir/crops下
                    if save_img or save_crop or view_img:  # Add bbox to image
                        c = int(cls)  # integer class
                        label = None if hide_labels else (names[c] if hide_conf else f'{names[c]} {conf:.2f}')
                        annotator.box_label(xyxy, label, color=colors(c, True))
                        if save_crop:
                            save_one_box(xyxy, imc, file=save_dir / 'crops' / names[c] / f'{p.stem}.jpg', BGR=True)

            # Print time (inference-only)
 # 打印前向传播 + NMS 花费的时间
            print(f'{s}Done. ({t3 - t2:.3f}s)')

            # Stream results
            im0 = annotator.result()
# 是否需要显示我们预测后的结果  img0(此时已将pred结果可视化到了img0中)
            if view_img:
                cv2.imshow(str(p), im0)
                cv2.waitKey(1)  # 1 millisecond

            # Save results (image with detections)
# 是否需要保存图片或视频（检测后的图片/视频 里面已经被我们画好了框的） img0
            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 += '.mp4'
                        vid_writer[i] = cv2.VideoWriter(save_path, cv2.VideoWriter_fourcc(*'mp4v'), fps, (w, h))
                    vid_writer[i].write(im0)

推理结束，保存结果，打印信息

# Print results
# 保存预测的label信息 xywh等   save_txt
    t = tuple(x / seen * 1E3 for x in dt)  # speeds per image
    print(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 ''
        print(f"Results saved to {colorstr('bold', save_dir)}{s}")
    if update:
# strip_optimizer函数将optimizer从ckpt中删除  更新模型
        strip_optimizer(weights)  # update model (to fix SourceChangeWarning)