# -*- coding: utf-8 -*- """ Created on Fri Oct 18 13:09:42 2024 @author: ym """ import argparse import os import sys import torch from pathlib import Path import numpy as np # from matplotlib.pylab import mpl # mpl.use('Qt5Agg') import matplotlib.pyplot as plt FILE = Path(__file__).resolve() 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.augmentations import letterbox from utils.general import (LOGGER, Profile, check_file, check_img_size, check_imshow, check_requirements, colorstr, cv2, increment_path, non_max_suppression, print_args, scale_boxes, strip_optimizer, xyxy2xywh) from utils.torch_utils import select_device, smart_inference_mode '''集成跟踪模块，输出跟踪结果文件 .npy''' # from ultralytics.engine.results import Boxes # Results # from ultralytics.utils import IterableSimpleNamespace, yaml_load from tracking.utils.plotting import Annotator, colors from tracking.utils import Boxes, IterableSimpleNamespace, yaml_load, boxes_add_fid from tracking.trackers import BOTSORT, BYTETracker from tracking.utils.showtrack import drawtracks from hands.hand_inference import hand_pose # from tracking.trackers.reid.reid_interface import ReIDInterface # from tracking.trackers.reid.config import config as ReIDConfig # ReIDEncoder = ReIDInterface(ReIDConfig) from contrast.feat_extract.config import config as conf from contrast.feat_extract.inference import FeatsInterface ReIDEncoder = FeatsInterface(conf) IMG_FORMATS = 'bmp', 'dng', 'jpeg', 'jpg', 'mpo', 'png', 'tif', 'tiff', 'webp', 'pfm' # include image suffixes VID_FORMATS = 'asf', 'avi', 'gif', 'm4v', 'mkv', 'mov', 'mp4', 'mpeg', 'mpg', 'ts', 'wmv' # include video suffixes '''================== 对图像进行旋转 ================== ''' class LoadImages: # YOLOv5 image/video dataloader, i.e. `python detect.py --source image.jpg/vid.mp4` def __init__(self, files, img_size=640, stride=32, auto=True, transforms=None, vid_stride=1): images = [x for x in files if x.split('.')[-1].lower() in IMG_FORMATS] videos = [x for x in files if x.split('.')[-1].lower() in VID_FORMATS] ni, nv = len(images), len(videos) self.img_size = img_size self.stride = stride self.files = images + videos self.nf = ni + nv # number of files self.video_flag = [False] * ni + [True] * nv self.mode = 'image' self.auto = auto self.transforms = transforms # optional self.vid_stride = vid_stride # video frame-rate stride if any(videos): self._new_video(videos[0]) # new video else: self.cap = None def __iter__(self): self.count = 0 return self def __next__(self): if self.count == self.nf: raise StopIteration path = self.files[self.count] if self.video_flag[self.count]: # Read video self.mode = 'video' for _ in range(self.vid_stride): self.cap.grab() ret_val, im0 = self.cap.retrieve() while not ret_val: self.count += 1 self.cap.release() if self.count == self.nf: # last video raise StopIteration path = self.files[self.count] self._new_video(path) ret_val, im0 = self.cap.read() self.frame += 1 # im0 = self._cv2_rotate(im0) # for use if cv2 autorotation is False s = f'video {self.count + 1}/{self.nf} ({self.frame}/{self.frames}) {path}: ' else: # Read image self.count += 1 im0 = cv2.imread(path) # BGR # image rorate (h, w) = im0.shape[:2] center = (w // 2, h // 2) angle = 90 scale = 1.0 M = cv2.getRotationMatrix2D(center, angle, scale) im0 = cv2.warpAffine(im0, M, (h, w)) assert im0 is not None, f'Image Not Found {path}' s = f'image {self.count}/{self.nf} {path}: ' if self.transforms: im = self.transforms(im0) # transforms else: im = letterbox(im0, self.img_size, stride=self.stride, auto=self.auto)[0] # padded resize im = im.transpose((2, 0, 1))[::-1] # HWC to CHW, BGR to RGB im = np.ascontiguousarray(im) # contiguous return path, im, im0, self.cap, s def _new_video(self, path): # Create a new video capture object self.frame = 0 self.cap = cv2.VideoCapture(path) self.frames = int(self.cap.get(cv2.CAP_PROP_FRAME_COUNT) / self.vid_stride) self.orientation = int(self.cap.get(cv2.CAP_PROP_ORIENTATION_META)) # rotation degrees # self.cap.set(cv2.CAP_PROP_ORIENTATION_AUTO, 0) # disable https://github.com/ultralytics/yolov5/issues/8493 def _cv2_rotate(self, im): # Rotate a cv2 video manually if self.orientation == 0: return cv2.rotate(im, cv2.ROTATE_90_CLOCKWISE) elif self.orientation == 180: return cv2.rotate(im, cv2.ROTATE_90_COUNTERCLOCKWISE) elif self.orientation == 90: return cv2.rotate(im, cv2.ROTATE_180) return im def __len__(self): return self.nf # number of files def inference_image(image, detections): H, W, _ = np.shape(image) imgs = [] batch_patches = [] patches = [] for d in range(np.size(detections, 0)): tlbr = detections[d, :4].astype(np.int_) tlbr[0] = max(0, tlbr[0]) tlbr[1] = max(0, tlbr[1]) tlbr[2] = min(W - 1, tlbr[2]) tlbr[3] = min(H - 1, tlbr[3]) img1 = image[tlbr[1]:tlbr[3], tlbr[0]:tlbr[2], :] img = img1[:, :, ::-1].copy() # the model expects RGB inputs patch = ReIDEncoder.transform(img) imgs.append(img1) # patch = patch.to(device=self.device).half() if str(ReIDEncoder.device) != "cpu": patch = patch.to(device=ReIDEncoder.device).half() else: patch = patch.to(device=ReIDEncoder.device) patches.append(patch) if (d + 1) % ReIDEncoder.batch_size == 0: patches = torch.stack(patches, dim=0) batch_patches.append(patches) patches = [] if len(patches): patches = torch.stack(patches, dim=0) batch_patches.append(patches) features = np.zeros((0, ReIDEncoder.embedding_size)) for patches in batch_patches: pred = ReIDEncoder.model(patches) pred[torch.isinf(pred)] = 1.0 feat = pred.cpu().data.numpy() features = np.vstack((features, feat)) return imgs, features def init_trackers(tracker_yaml = None, bs=1): """ Initialize trackers for object tracking during prediction. """ # tracker_yaml = r"./tracking/trackers/cfg/botsort.yaml" TRACKER_MAP = {'bytetrack': BYTETracker, 'botsort': BOTSORT} cfg = IterableSimpleNamespace(**yaml_load(tracker_yaml)) trackers = [] for _ in range(bs): tracker = TRACKER_MAP[cfg.tracker_type](args=cfg, frame_rate=30) trackers.append(tracker) return trackers def run( weights=ROOT / 'yolov5s.pt', # model path or triton URL source=ROOT / 'data/images', # file/dir/URL/glob/screen/0(webcam) project=ROOT / 'runs/detect', # save results to project/name name='exp', # save results to project/name tracker_yaml = "./tracking/trackers/cfg/botsort.yaml", 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_csv=False, # save results in CSV format save_conf=False, # save confidences in --save-txt labels save_crop=False, # save cropped prediction boxes save_img = True, 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 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 confidencesL half=False, # use FP16 half-precision inference dnn=False, # use OpenCV DNN for ONNX inference vid_stride=1, # video frame-rate stride data=ROOT / 'data/coco128.yaml', # dataset.yaml path ): assert isinstance(source,list), "source must be a list" fulldir, imgname = os.path.split(source[0]) imgbase, ext = os.path.splitext(imgname) # 事件名、相机类型 EventName = fulldir.split('\\')[-2] + "_" + str(Path(fulldir).stem) CamerType = imgbase.split('_')[1] save_dir = Path(project) / Path(EventName) if save_dir.exists(): print(Path(fulldir).stem) # save_dir = increment_path(Path(project) / name, exist_ok=exist_ok) # increment run # save_dir.mkdir(parents=True, exist_ok=True) # make dir else: save_dir.mkdir(parents=True, exist_ok=True) save_path_video = os.path.join(str(save_dir), f"{EventName}_{CamerType}") # 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 bs = 1 # batch_size dataset = LoadImages(source, img_size=imgsz, stride=stride, auto=pt, vid_stride=vid_stride) vid_path, vid_writer = [None] * bs, [None] * bs # Run inference model.warmup(imgsz=(1 if pt or model.triton else bs, 3, *imgsz)) # warmup seen, dt = 0, (Profile(), Profile(), Profile()) tracker = init_trackers(tracker_yaml, bs)[0] track_boxes = np.empty((0, 10), dtype = np.float32) k = 0 for path, im, im0s, vid_cap, s in dataset: # k +=1 # if k==60: # break timeStamp = Path(path).stem.split('_')[2] with dt[0]: im = torch.from_numpy(im).to(model.device) im = im.half() if model.fp16 else im.float() # uint8 to fp16/32 im /= 255 # 0 - 255 to 0.0 - 1.0 if len(im.shape) == 3: im = im[None] # expand for batch dim # Inference with dt[1]: visualize = increment_path(save_dir / Path(path).stem, mkdir=True) if visualize else False pred = model(im, augment=augment, visualize=visualize) # NMS with dt[2]: pred = non_max_suppression(pred, conf_thres, iou_thres, classes, agnostic_nms, max_det=max_det) # Process predictions for i, det in enumerate(pred): # per image seen += 1 im0 = im0s.copy() s += '%gx%g ' % im.shape[2:] # print string annotator = Annotator(im0s, line_width=line_thickness, example=str(names)) nd = len(det) if nd: # Rescale boxes from img_size to im0 size det[:, :4] = scale_boxes(im.shape[2:], det[:, :4], im0.shape).round() '''tracks: [x1, y1, x2, y2, track_id, score, cls, frame_index, box_index, timestamp] 0 1 2 3 4 5 6 7 8 这里，frame_index 也可以用视频的 帧ID 代替, box_index 保持不变 ''' det_tracking = Boxes(det, im0.shape).cpu().numpy() tracks = tracker.update(det_tracking, im0) if len(tracks) == 0: continue tracks[:, 7] = dataset.count stamp = np.ones((len(tracks), 1)) * int(timeStamp) tracks = np.concatenate((tracks, stamp), axis=1) '''================== 1. 存储 dets/subimgs/features Dict =============''' # imgs, features = inference_image(im0, tracks) track_boxes = np.concatenate([track_boxes, tracks], axis=0) for *xyxy, id, conf, cls, fid, bid, t in reversed(tracks): name = ('' if id==-1 else f'id:{int(id)} ') + names[int(cls)] label = None if hide_labels else (name if hide_conf else f'{name} {conf:.2f}') if id >=0 and cls==0: color = colors(int(cls), True) elif id >=0 and cls!=0: color = colors(int(id), True) else: color = colors(19, True) # 19为调色板的最后一个元素 annotator.box_label(xyxy, label, color=color) # Save results (image and video with tracking) im0 = annotator.result() p = Path(path) # to Path if save_img: imgpath = str(save_dir/p.stem) + f"_{dataset.count}.png" cv2.imwrite(Path(imgpath), im0) if vid_path[i] != save_path_video: # new video vid_path[i] = save_path_video if isinstance(vid_writer[i], cv2.VideoWriter): vid_writer[i].release() # release previous video writer fps, w, h = 30, im0.shape[1], im0.shape[0] vpath = str(Path(save_path_video).with_suffix('.mp4')) vid_writer[i] = cv2.VideoWriter(vpath, cv2.VideoWriter_fourcc(*'mp4v'), fps, (w, h)) vid_writer[i].write(im0) LOGGER.info(f"{s}{'' if len(det) else '(no detections), '}{dt[1].dt * 1E3:.1f}ms") for v in vid_writer: v.release() if track_boxes.size == 0: return CamerType, [] save_path_np = os.path.join(str(fulldir), f"{EventName}_{CamerType}") np.save(save_path_np, track_boxes) # Print results t = tuple(x.t / 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[0]) # update model (to fix SourceChangeWarning) return CamerType, track_boxes def parse_opt(): modelpath = ROOT / 'ckpts/best_cls10_0906.pt' # 'ckpts/best_15000_0908.pt', 'ckpts/yolov5s.pt', 'ckpts/best_20000_cls30.pt, best_yolov5m_250000' parser = argparse.ArgumentParser() parser.add_argument('--weights', nargs='+', type=str, default=modelpath, help='model path or triton URL') # 'yolov5s.pt', best_15000_0908.pt parser.add_argument('--source', type=str, default='', help='file/dir/URL/glob/screen/0(webcam)') # images, videos parser.add_argument('--data', type=str, default=ROOT / 'data/coco128.yaml', help='(optional) dataset.yaml path') parser.add_argument('--imgsz', '--img', '--img-size', nargs='+', type=int, default=[640], 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') parser.add_argument('--save-txt', action='store_true', help='save results to *.txt') parser.add_argument('--save-csv', action='store_true', help='save results in CSV format') 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') parser.add_argument('--vid-stride', type=int, default=1, help='video frame-rate stride') opt = parser.parse_args() opt.imgsz *= 2 if len(opt.imgsz) == 1 else 1 # expand print_args(vars(opt)) return opt def run_yolo(eventdir, savedir): opt = parse_opt() optdict = vars(opt) optdict["project"] = savedir optdict["source"] = eventdir run(**vars(opt))