回传数据解析，兼容v5和v10
This commit is contained in:
jiajie555
359
bakeup/pipeline.py
Normal file
359
bakeup/pipeline.py
Normal file
@ -0,0 +1,359 @@
|
||||
# -*- coding: utf-8 -*-
|
||||
"""
|
||||
Created on Sun Sep 29 08:59:21 2024
|
||||
|
||||
@author: ym
|
||||
"""
|
||||
import os
|
||||
# import sys
|
||||
import cv2
|
||||
import pickle
|
||||
import numpy as np
|
||||
from pathlib import Path
|
||||
from scipy.spatial.distance import cdist
|
||||
from track_reid import yolo_resnet_tracker, yolov10_resnet_tracker
|
||||
|
||||
from tracking.dotrack.dotracks_back import doBackTracks
|
||||
from tracking.dotrack.dotracks_front import doFrontTracks
|
||||
from tracking.utils.drawtracks import plot_frameID_y2, draw_all_trajectories
|
||||
from utils.getsource import get_image_pairs, get_video_pairs
|
||||
from tracking.utils.read_data import read_similar
|
||||
|
||||
|
||||
def save_subimgs(imgdict, boxes, spath, ctype, featdict = None):
|
||||
'''
|
||||
当前 box 特征和该轨迹前一个 box 特征的相似度，可用于和跟踪序列中的相似度进行比较
|
||||
'''
|
||||
boxes = boxes[np.argsort(boxes[:, 7])]
|
||||
for i in range(len(boxes)):
|
||||
simi = None
|
||||
tid, fid, bid = int(boxes[i, 4]), int(boxes[i, 7]), int(boxes[i, 8])
|
||||
|
||||
if i>0:
|
||||
_, fid0, bid0 = int(boxes[i-1, 4]), int(boxes[i-1, 7]), int(boxes[i-1, 8])
|
||||
if f"{fid0}_{bid0}" in featdict.keys() and f"{fid}_{bid}" in featdict.keys():
|
||||
feat0 = featdict[f"{fid0}_{bid0}"]
|
||||
feat1 = featdict[f"{fid}_{bid}"]
|
||||
simi = 1 - np.maximum(0.0, cdist(feat0[None, :], feat1[None, :], "cosine"))[0][0]
|
||||
|
||||
img = imgdict[f"{fid}_{bid}"]
|
||||
imgpath = spath / f"{ctype}_tid{tid}-{fid}-{bid}.png"
|
||||
if simi is not None:
|
||||
imgpath = spath / f"{ctype}_tid{tid}-{fid}-{bid}_sim{simi:.2f}.png"
|
||||
|
||||
cv2.imwrite(imgpath, img)
|
||||
|
||||
|
||||
def save_subimgs_1(imgdict, boxes, spath, ctype, simidict = None):
|
||||
'''
|
||||
当前 box 特征和该轨迹 smooth_feat 特征的相似度, yolo_resnet_tracker 函数中，
|
||||
采用该方式记录特征相似度
|
||||
'''
|
||||
for i in range(len(boxes)):
|
||||
tid, fid, bid = int(boxes[i, 4]), int(boxes[i, 7]), int(boxes[i, 8])
|
||||
|
||||
key = f"{fid}_{bid}"
|
||||
img = imgdict[key]
|
||||
imgpath = spath / f"{ctype}_tid{tid}-{fid}-{bid}.png"
|
||||
if simidict is not None and key in simidict.keys():
|
||||
imgpath = spath / f"{ctype}_tid{tid}-{fid}-{bid}_sim{simidict[key]:.2f}.png"
|
||||
|
||||
cv2.imwrite(imgpath, img)
|
||||
|
||||
|
||||
def pipeline(
|
||||
eventpath,
|
||||
savepath,
|
||||
SourceType,
|
||||
weights,
|
||||
YoloVersion="V5"
|
||||
):
|
||||
'''
|
||||
eventpath: 单个事件的存储路径
|
||||
|
||||
'''
|
||||
optdict = {}
|
||||
optdict["weights"] = weights
|
||||
|
||||
if SourceType == "video":
|
||||
vpaths = get_video_pairs(eventpath)
|
||||
elif SourceType == "image":
|
||||
vpaths = get_image_pairs(eventpath)
|
||||
event_tracks = []
|
||||
|
||||
## 构造购物事件字典
|
||||
evtname = Path(eventpath).stem
|
||||
barcode = evtname.split('_')[-1] if len(evtname.split('_'))>=2 \
|
||||
and len(evtname.split('_')[-1])>=8 \
|
||||
and evtname.split('_')[-1].isdigit() else ''
|
||||
'''事件结果存储文件夹'''
|
||||
if not savepath:
|
||||
savepath = Path(__file__).resolve().parents[0] / "events_result"
|
||||
|
||||
savepath_pipeline = Path(savepath) / Path("Yolos_Tracking") / evtname
|
||||
|
||||
|
||||
"""ShoppingDict pickle 文件保存地址 """
|
||||
savepath_spdict = Path(savepath) / "ShoppingDict_pkfile"
|
||||
if not savepath_spdict.exists():
|
||||
savepath_spdict.mkdir(parents=True, exist_ok=True)
|
||||
pf_path = Path(savepath_spdict) / Path(str(evtname)+".pickle")
|
||||
|
||||
# if pf_path.exists():
|
||||
# print(f"Pickle file have saved: {evtname}.pickle")
|
||||
# return
|
||||
|
||||
'''====================== 构造 ShoppingDict 模块 ======================='''
|
||||
ShoppingDict = {"eventPath": eventpath,
|
||||
"eventName": evtname,
|
||||
"barcode": barcode,
|
||||
"eventType": '', # "input", "output", "other"
|
||||
"frontCamera": {},
|
||||
"backCamera": {},
|
||||
"one2n": [] #
|
||||
}
|
||||
yrtDict = {}
|
||||
|
||||
|
||||
procpath = Path(eventpath).joinpath('process.data')
|
||||
if procpath.is_file():
|
||||
SimiDict = read_similar(procpath)
|
||||
ShoppingDict["one2n"] = SimiDict['one2n']
|
||||
|
||||
|
||||
for vpath in vpaths:
|
||||
'''================= 1. 构造相机事件字典 ================='''
|
||||
CameraEvent = {"cameraType": '', # "front", "back"
|
||||
"videoPath": '',
|
||||
"imagePaths": [],
|
||||
"yoloResnetTracker": [],
|
||||
"tracking": [],
|
||||
}
|
||||
|
||||
if isinstance(vpath, list):
|
||||
CameraEvent["imagePaths"] = vpath
|
||||
bname = os.path.basename(vpath[0])
|
||||
if not isinstance(vpath, list):
|
||||
CameraEvent["videoPath"] = vpath
|
||||
bname = os.path.basename(vpath).split('.')[0]
|
||||
if bname.split('_')[0] == "0" or bname.find('back')>=0:
|
||||
CameraEvent["cameraType"] = "back"
|
||||
if bname.split('_')[0] == "1" or bname.find('front')>=0:
|
||||
CameraEvent["cameraType"] = "front"
|
||||
|
||||
'''================= 2. 事件结果存储文件夹 ================='''
|
||||
if isinstance(vpath, list):
|
||||
savepath_pipeline_imgs = savepath_pipeline / Path("images")
|
||||
else:
|
||||
savepath_pipeline_imgs = savepath_pipeline / Path(str(Path(vpath).stem))
|
||||
|
||||
if not savepath_pipeline_imgs.exists():
|
||||
savepath_pipeline_imgs.mkdir(parents=True, exist_ok=True)
|
||||
|
||||
savepath_pipeline_subimgs = savepath_pipeline / Path("subimgs")
|
||||
if not savepath_pipeline_subimgs.exists():
|
||||
savepath_pipeline_subimgs.mkdir(parents=True, exist_ok=True)
|
||||
|
||||
|
||||
'''================= 3. Yolo + Resnet + Tracker ================='''
|
||||
optdict["source"] = vpath
|
||||
optdict["save_dir"] = savepath_pipeline_imgs
|
||||
optdict["is_save_img"] = True
|
||||
optdict["is_save_video"] = True
|
||||
|
||||
|
||||
if YoloVersion == "V5":
|
||||
yrtOut = yolo_resnet_tracker(**optdict)
|
||||
elif YoloVersion == "V10":
|
||||
yrtOut = yolov10_resnet_tracker(**optdict)
|
||||
|
||||
|
||||
yrtOut_save = []
|
||||
for frdict in yrtOut:
|
||||
fr_dict = {}
|
||||
for k, v in frdict.items():
|
||||
if k != "imgs":
|
||||
fr_dict[k]=v
|
||||
yrtOut_save.append(fr_dict)
|
||||
CameraEvent["yoloResnetTracker"] = yrtOut_save
|
||||
|
||||
# CameraEvent["yoloResnetTracker"] = yrtOut
|
||||
|
||||
'''================= 4. tracking ================='''
|
||||
'''(1) 生成用于 tracking 模块的 boxes、feats'''
|
||||
bboxes = np.empty((0, 6), dtype=np.float64)
|
||||
trackerboxes = np.empty((0, 9), dtype=np.float64)
|
||||
trackefeats = {}
|
||||
for frameDict in yrtOut:
|
||||
tboxes = frameDict["tboxes"]
|
||||
ffeats = frameDict["feats"]
|
||||
|
||||
boxes = frameDict["bboxes"]
|
||||
bboxes = np.concatenate((bboxes, np.array(boxes)), axis=0)
|
||||
trackerboxes = np.concatenate((trackerboxes, np.array(tboxes)), axis=0)
|
||||
for i in range(len(tboxes)):
|
||||
fid, bid = int(tboxes[i, 7]), int(tboxes[i, 8])
|
||||
trackefeats.update({f"{fid}_{bid}": ffeats[f"{fid}_{bid}"]})
|
||||
|
||||
|
||||
'''(2) tracking, 后摄'''
|
||||
if CameraEvent["cameraType"] == "back":
|
||||
vts = doBackTracks(trackerboxes, trackefeats)
|
||||
vts.classify()
|
||||
event_tracks.append(("back", vts))
|
||||
|
||||
CameraEvent["tracking"] = vts
|
||||
ShoppingDict["backCamera"] = CameraEvent
|
||||
|
||||
yrtDict["backyrt"] = yrtOut
|
||||
|
||||
'''(2) tracking, 前摄'''
|
||||
if CameraEvent["cameraType"] == "front":
|
||||
vts = doFrontTracks(trackerboxes, trackefeats)
|
||||
vts.classify()
|
||||
event_tracks.append(("front", vts))
|
||||
|
||||
CameraEvent["tracking"] = vts
|
||||
ShoppingDict["frontCamera"] = CameraEvent
|
||||
|
||||
yrtDict["frontyrt"] = yrtOut
|
||||
|
||||
'''========================== 保存模块 ================================='''
|
||||
'''(1) 保存 ShoppingDict 事件'''
|
||||
with open(str(pf_path), 'wb') as f:
|
||||
pickle.dump(ShoppingDict, f)
|
||||
|
||||
'''(2) 保存 Tracking 输出的运动轨迹子图，并记录相似度'''
|
||||
for CamerType, vts in event_tracks:
|
||||
if len(vts.tracks)==0: continue
|
||||
if CamerType == 'front':
|
||||
# yolos = ShoppingDict["frontCamera"]["yoloResnetTracker"]
|
||||
|
||||
yolos = yrtDict["frontyrt"]
|
||||
ctype = 1
|
||||
if CamerType == 'back':
|
||||
# yolos = ShoppingDict["backCamera"]["yoloResnetTracker"]
|
||||
|
||||
yolos = yrtDict["backyrt"]
|
||||
ctype = 0
|
||||
|
||||
imgdict, featdict, simidict = {}, {}, {}
|
||||
for y in yolos:
|
||||
imgdict.update(y["imgs"])
|
||||
featdict.update(y["feats"])
|
||||
simidict.update(y["featsimi"])
|
||||
|
||||
for track in vts.Residual:
|
||||
if isinstance(track, np.ndarray):
|
||||
save_subimgs(imgdict, track, savepath_pipeline_subimgs, ctype, featdict)
|
||||
else:
|
||||
save_subimgs(imgdict, track.slt_boxes, savepath_pipeline_subimgs, ctype, featdict)
|
||||
|
||||
'''(3) 轨迹显示与保存'''
|
||||
illus = [None, None]
|
||||
for CamerType, vts in event_tracks:
|
||||
if len(vts.tracks)==0: continue
|
||||
|
||||
if CamerType == 'front':
|
||||
edgeline = cv2.imread("./tracking/shopcart/cart_tempt/board_ftmp_line.png")
|
||||
|
||||
h, w = edgeline.shape[:2]
|
||||
# nh, nw = h//2, w//2
|
||||
# edgeline = cv2.resize(edgeline, (nw, nh), interpolation=cv2.INTER_AREA)
|
||||
|
||||
img_tracking = draw_all_trajectories(vts, edgeline, savepath_pipeline, CamerType, draw5p=True)
|
||||
illus[0] = img_tracking
|
||||
|
||||
plt = plot_frameID_y2(vts)
|
||||
plt.savefig(os.path.join(savepath_pipeline, "front_y2.png"))
|
||||
|
||||
if CamerType == 'back':
|
||||
edgeline = cv2.imread("./tracking/shopcart/cart_tempt/edgeline.png")
|
||||
|
||||
h, w = edgeline.shape[:2]
|
||||
# nh, nw = h//2, w//2
|
||||
# edgeline = cv2.resize(edgeline, (nw, nh), interpolation=cv2.INTER_AREA)
|
||||
|
||||
img_tracking = draw_all_trajectories(vts, edgeline, savepath_pipeline, CamerType, draw5p=True)
|
||||
illus[1] = img_tracking
|
||||
|
||||
illus = [im for im in illus if im is not None]
|
||||
if len(illus):
|
||||
img_cat = np.concatenate(illus, axis = 1)
|
||||
if len(illus)==2:
|
||||
H, W = img_cat.shape[:2]
|
||||
cv2.line(img_cat, (int(W/2), 0), (int(W/2), int(H)), (128, 128, 255), 3)
|
||||
|
||||
trajpath = os.path.join(savepath_pipeline, "trajectory.png")
|
||||
cv2.imwrite(trajpath, img_cat)
|
||||
|
||||
def execute_pipeline(evtdir = r"D:\datasets\ym\后台数据\unzip",
|
||||
source_type = "video", # video, image,
|
||||
save_path = r"D:\work\result_pipeline",
|
||||
yolo_ver = "V10", # V10, V5
|
||||
|
||||
weight_yolo_v5 = r'./ckpts/best_cls10_0906.pt' ,
|
||||
weight_yolo_v10 = r'./ckpts/best_v10s_width0375_1205.pt',
|
||||
k=0
|
||||
):
|
||||
'''
|
||||
运行函数 pipeline()，遍历事件文件夹，每个文件夹是一个事件
|
||||
'''
|
||||
parmDict = {}
|
||||
parmDict["SourceType"] = source_type
|
||||
parmDict["savepath"] = save_path
|
||||
parmDict["YoloVersion"] = yolo_ver
|
||||
if parmDict["YoloVersion"] == "V5":
|
||||
parmDict["weights"] = weight_yolo_v5
|
||||
elif parmDict["YoloVersion"] == "V10":
|
||||
parmDict["weights"] = weight_yolo_v10
|
||||
|
||||
evtdir = Path(evtdir)
|
||||
errEvents = []
|
||||
for item in evtdir.iterdir():
|
||||
if item.is_dir():
|
||||
item = evtdir/Path("20250310-175352-741")
|
||||
parmDict["eventpath"] = item
|
||||
pipeline(**parmDict)
|
||||
# try:
|
||||
# pipeline(**parmDict)
|
||||
# except Exception as e:
|
||||
# errEvents.append(str(item))
|
||||
k+=1
|
||||
if k==1:
|
||||
break
|
||||
|
||||
errfile = os.path.join(parmDict["savepath"], 'error_events.txt')
|
||||
with open(errfile, 'w', encoding='utf-8') as f:
|
||||
for line in errEvents:
|
||||
f.write(line + '\n')
|
||||
|
||||
if __name__ == "__main__":
|
||||
execute_pipeline()
|
||||
|
||||
# spath_v10 = r"D:\work\result_pipeline_v10"
|
||||
# spath_v5 = r"D:\work\result_pipeline_v5"
|
||||
# execute_pipeline(save_path=spath_v10, yolo_ver="V10")
|
||||
# execute_pipeline(save_path=spath_v5, yolo_ver="V5")
|
||||
|
||||
datapath = r'/home/wqg/dataset/test_dataset/base_dataset/single_event/source/'
|
||||
savepath = r'/home/wqg/dataset/pipeline/contrast/single_event_V5'
|
||||
|
||||
|
||||
|
||||
|
||||
execute_pipeline(evtdir = datapath,
|
||||
DataType = "raw", # raw, pkl
|
||||
kk=1,
|
||||
source_type = "video", # video, image,
|
||||
save_path = savepath,
|
||||
yolo_ver = "V10", # V10, V5
|
||||
weight_yolo_v5 = r'./ckpts/best_cls10_0906.pt' ,
|
||||
weight_yolo_v10 = r'./ckpts/best_v10s_width0375_1205.pt',
|
||||
saveimages = False
|
||||
)
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
629
bakeup/track_reid_20240515.py
Normal file
629
bakeup/track_reid_20240515.py
Normal file
@ -0,0 +1,629 @@
|
||||
# YOLOv5 🚀 by Ultralytics, AGPL-3.0 license
|
||||
"""
|
||||
Run YOLOv5 detection inference on images, videos, directories, globs, YouTube, webcam, streams, etc.
|
||||
|
||||
Usage - sources:
|
||||
$ python detect.py --weights yolov5s.pt --source 0 # webcam
|
||||
img.jpg # image
|
||||
vid.mp4 # video
|
||||
screen # screenshot
|
||||
path/ # directory
|
||||
list.txt # list of images
|
||||
list.streams # list of streams
|
||||
'path/*.jpg' # glob
|
||||
'https://youtu.be/Zgi9g1ksQHc' # YouTube
|
||||
'rtsp://example.com/media.mp4' # RTSP, RTMP, HTTP stream
|
||||
|
||||
Usage - formats:
|
||||
$ python detect.py --weights yolov5s.pt # PyTorch
|
||||
yolov5s.torchscript # TorchScript
|
||||
yolov5s.onnx # ONNX Runtime or OpenCV DNN with --dnn
|
||||
yolov5s_openvino_model # 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
|
||||
yolov5s_paddle_model # PaddlePaddle
|
||||
"""
|
||||
|
||||
import argparse
|
||||
import csv
|
||||
import os
|
||||
import platform
|
||||
import sys
|
||||
from pathlib import Path
|
||||
import glob
|
||||
import numpy as np
|
||||
import pickle
|
||||
import torch
|
||||
|
||||
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.dataloaders import IMG_FORMATS, VID_FORMATS, LoadImages, LoadScreenshots, LoadStreams
|
||||
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)
|
||||
# =============================================================================
|
||||
|
||||
# tracker_yaml = r"./tracking/trackers/cfg/botsort.yaml"
|
||||
|
||||
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
|
||||
|
||||
|
||||
@smart_inference_mode()
|
||||
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
|
||||
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
|
||||
):
|
||||
source = str(source)
|
||||
# filename = os.path.split(source)[-1]
|
||||
|
||||
save_img = not nosave and not source.endswith('.txt') # save inference images
|
||||
is_file = Path(source).suffix[1:] in (IMG_FORMATS + VID_FORMATS)
|
||||
is_url = source.lower().startswith(('rtsp://', 'rtmp://', 'http://', 'https://'))
|
||||
webcam = source.isnumeric() or source.endswith('.streams') or (is_url and not is_file)
|
||||
screenshot = source.lower().startswith('screen')
|
||||
if is_url and is_file:
|
||||
source = check_file(source) # download
|
||||
|
||||
save_dir = Path(project) / Path(source).stem
|
||||
if save_dir.exists():
|
||||
print(Path(source).stem)
|
||||
# return
|
||||
|
||||
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
|
||||
else:
|
||||
save_dir.mkdir(parents=True, exist_ok=True)
|
||||
|
||||
# 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
|
||||
|
||||
# Dataloader
|
||||
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]
|
||||
|
||||
handpose = hand_pose()
|
||||
handlocals_dict = {}
|
||||
|
||||
boxes_and_imgs = []
|
||||
track_boxes = np.empty((0, 9), dtype = np.float32)
|
||||
det_boxes = np.empty((0, 9), dtype = np.float32)
|
||||
|
||||
|
||||
features_dict = {}
|
||||
for path, im, im0s, vid_cap, s in dataset:
|
||||
if save_img and 'imgshow' not in locals().keys():
|
||||
imgshow = im0s.copy()
|
||||
|
||||
## ============================= tracking 功能只处理视频，writed by WQG
|
||||
if dataset.mode == 'image':
|
||||
continue
|
||||
|
||||
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)
|
||||
|
||||
# Second-stage classifier (optional)
|
||||
# pred = utils.general.apply_classifier(pred, classifier_model, im, im0s)
|
||||
|
||||
# Define the path for the CSV file
|
||||
# csv_path = save_dir / 'predictions.csv'
|
||||
|
||||
# Create or append to the CSV file
|
||||
# def write_to_csv(image_name, prediction, confidence):
|
||||
# data = {'Image Name': image_name, 'Prediction': prediction, 'Confidence': confidence}
|
||||
# with open(csv_path, mode='a', newline='') as f:
|
||||
# writer = csv.DictWriter(f, fieldnames=data.keys())
|
||||
# if not csv_path.is_file():
|
||||
# writer.writeheader()
|
||||
# writer.writerow(data)
|
||||
|
||||
# Process predictions
|
||||
for i, det in enumerate(pred): # per image
|
||||
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)
|
||||
|
||||
im0_ant = im0.copy()
|
||||
|
||||
p = Path(p) # to Path
|
||||
save_path = str(save_dir / p.name) # im.jpg
|
||||
s += '%gx%g ' % im.shape[2:] # print string
|
||||
|
||||
annotator = Annotator(im0_ant, 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()
|
||||
|
||||
det = det.cpu().numpy()
|
||||
det = np.concatenate([det[:, :4], np.arange(nd).reshape(-1, 1), det[:, 4:]], axis=-1)
|
||||
|
||||
'''FeatFlag为相对于上一帧boxes，当前boxes是否为静止的标志。'''
|
||||
# def static_estimate(box1, box2, TH1=8, TH2=12):
|
||||
# dij_abs = max(np.abs(box1 - box2))
|
||||
# dij_euc = max([np.linalg.norm((box1[:2] - box2[:2])),
|
||||
# np.linalg.norm((box1[2:4] - box2[2:4]))
|
||||
# ])
|
||||
# if dij_abs < TH1 and dij_euc < TH2:
|
||||
# return True
|
||||
# else:
|
||||
# return False
|
||||
|
||||
# FeatFlag = [-1] * nd
|
||||
# if len(boxes_and_imgs):
|
||||
# detj = boxes_and_imgs[-1][0]
|
||||
# frmj = boxes_and_imgs[-1][-1]
|
||||
# for ii in range(nd):
|
||||
# ## flag 中保存的是box索引
|
||||
# condt1 = frame-frmj==1
|
||||
# flag = [idx for jj, idx in enumerate(detj[:, 4]) if condt1 and static_estimate(det[ii, :4], detj[jj, :4])]
|
||||
# if len(flag) == 1:
|
||||
# FeatFlag[ii] = flag[0]
|
||||
|
||||
|
||||
boxes_and_imgs.append((det, im0, frame))
|
||||
|
||||
## ================================================================ writed by WQG
|
||||
'''tracks: [x1, y1, x2, y2, track_id, score, cls, frame_index, box_index]
|
||||
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)
|
||||
|
||||
# detbox = [tlwh2tlbr(x._tlwh).tolist() + [x.track_id, x.score, x.cls, x.frame_id, x.idx]
|
||||
# for x in tracker.tracked_stracks if x.is_activated]
|
||||
if len(tracks):
|
||||
'''
|
||||
tracks: [x1, y1, x2, y2, track_id, score, cls, frame_index, box_index]
|
||||
0 1 2 3 4 5 6 7 8
|
||||
'''
|
||||
tracks[:, 7] = dataset.frame
|
||||
|
||||
|
||||
'''================== 1. 提取手势位置 ==================='''
|
||||
# idx_0 = tracks[:, 6].astype(np.int_) == 0
|
||||
# hn = 0
|
||||
# for j, index in enumerate(idx_0):
|
||||
# if index:
|
||||
# track = tracks[j, :]
|
||||
# hand_local, imgshow = handpose.get_hand_local(track, im0)
|
||||
# handlocals_dict.update({int(track[7]): {int(track[8]): hand_local}})
|
||||
|
||||
# # '''yoloV5和手势检测的召回率并不一直，用hand_local代替tracks中手部的(x1, y1, x2, y2)，会使得两种坐标方式混淆'''
|
||||
# # if hand_local: tracks[j, :4] = hand_local
|
||||
|
||||
# hn += 1
|
||||
# cv2.imwrite(f"D:\DeepLearning\yolov5\hands\images\{Path(source).stem}_{int(track[7])}_{hn}.png", imgshow)
|
||||
|
||||
|
||||
'''================== 2. 存储轨迹信息 ==================='''
|
||||
track_boxes = np.concatenate([track_boxes, tracks], axis=0)
|
||||
|
||||
# det_boxes = np.concatenate([det_boxes, detbox], axis=0)
|
||||
'''================== 3. 存储轨迹 REID 特征 ============='''
|
||||
|
||||
def crop_img(track, image):
|
||||
tlbr = track.tlwh_to_tlbr(track._tlwh).astype(np.int_)
|
||||
|
||||
H, W = image.shape[:2]
|
||||
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])
|
||||
|
||||
img = image[tlbr[1]:tlbr[3], tlbr[0]:tlbr[2], :]
|
||||
# cv2.imwrite(f"./runs/imgs/{int(track.idx)}.png", img)
|
||||
return img
|
||||
|
||||
feat_dict_1 = {f'{int(x.idx)}_img': crop_img(x, im0) for x in tracker.tracked_stracks if x.is_activated}
|
||||
feat_dict = {int(x.idx): x.curr_feat for x in tracker.tracked_stracks if x.is_activated}
|
||||
feat_dict.update(feat_dict_1)
|
||||
|
||||
features_dict.update({int(dataset.frame): feat_dict})
|
||||
|
||||
# det_anno = tracks.copy()
|
||||
# else:
|
||||
# idmark = -1 * np.ones([det.shape[0], 1])
|
||||
# det_anno = np.concatenate([det[:,:4], idmark, det[:, 4:]], axis=1)
|
||||
|
||||
|
||||
|
||||
|
||||
for *xyxy, id, conf, cls, fid, bid 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()
|
||||
save_path_img, ext = os.path.splitext(save_path)
|
||||
if save_img:
|
||||
if dataset.mode == 'image':
|
||||
imgpath = save_path_img + f"_{dataset}.png"
|
||||
else:
|
||||
imgpath = save_path_img + f"_{dataset.frame}.png"
|
||||
|
||||
cv2.imwrite(Path(imgpath), im0)
|
||||
|
||||
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}{'' if len(det) else '(no detections), '}{dt[1].dt * 1E3:.1f}ms")
|
||||
|
||||
|
||||
## ======================================================================== written by WQG
|
||||
## track_boxes: Array, [x1, y1, x2, y2, track_id, score, cls, frame_index, box_id]
|
||||
|
||||
'''上面保存了检测结果是视频和图像，以下还保存五种类型的数据'''
|
||||
filename = os.path.split(save_path_img)[-1]
|
||||
# file, ext = os.path.splitext(filename)
|
||||
# =============================================================================
|
||||
# fileElements = filename.split('_')
|
||||
# if len(fileElements) == 6 and len(fileElements[3])==1:
|
||||
# barcode = fileElements[1]
|
||||
# camera = fileElements[3]
|
||||
# elif len(fileElements) == 7 and len(fileElements[3])==1:
|
||||
# barcode = fileElements[2]
|
||||
# camera = fileElements[4]
|
||||
# else:
|
||||
# barcode = ''
|
||||
# camera = ''
|
||||
# =============================================================================
|
||||
|
||||
|
||||
'''======================== 1. save in './run/detect/' ===================='''
|
||||
if source.find("front") >= 0:
|
||||
carttemp = cv2.imread("./tracking/shopcart/cart_tempt/board_ftmp_line.png")
|
||||
else:
|
||||
carttemp = cv2.imread("./tracking/shopcart/cart_tempt/edgeline.png")
|
||||
|
||||
|
||||
imgshow = drawtracks(track_boxes, carttemp)
|
||||
showpath_1 = save_path_img + "_show.png"
|
||||
cv2.imwrite(Path(showpath_1), imgshow)
|
||||
|
||||
|
||||
'''======================== 2. save boxes and raw images =================='''
|
||||
# boxes_imgs_dir = Path('./tracking/data/boxes_imgs/')
|
||||
# if not boxes_imgs_dir.exists():
|
||||
# boxes_imgs_dir.mkdir(parents=True, exist_ok=True)
|
||||
# boxes_imgs_path = boxes_imgs_dir.joinpath(f'{filename}.pkl')
|
||||
# with open(boxes_imgs_path, 'wb') as file:
|
||||
# pickle.dump(boxes_and_imgs, file)
|
||||
|
||||
'''======================== 3. save tracks data ==========================='''
|
||||
tracks_dir = Path('./tracking/data/tracks/')
|
||||
if not tracks_dir.exists():
|
||||
tracks_dir.mkdir(parents=True, exist_ok=True)
|
||||
tracks_path = tracks_dir.joinpath(filename + "_track.npy")
|
||||
np.save(tracks_path, track_boxes)
|
||||
|
||||
detect_path = tracks_dir.joinpath(filename + "_detect.npy")
|
||||
np.save(detect_path, det_boxes)
|
||||
|
||||
'''======================== 4. save reid features data ===================='''
|
||||
feats_dir = Path('./tracking/data/trackfeats/')
|
||||
if not feats_dir.exists():
|
||||
feats_dir.mkdir(parents=True, exist_ok=True)
|
||||
feats_path = feats_dir.joinpath(f'{filename}.pkl')
|
||||
with open(feats_path, 'wb') as file:
|
||||
pickle.dump(features_dict, file)
|
||||
|
||||
'''======================== 5. save hand_local data =================='''
|
||||
# handlocal_dir = Path('./tracking/data/handlocal/')
|
||||
# if not handlocal_dir.exists():
|
||||
# handlocal_dir.mkdir(parents=True, exist_ok=True)
|
||||
# handlocal_path = handlocal_dir.joinpath(f'{filename}.pkl')
|
||||
# with open(handlocal_path, 'wb') as file:
|
||||
# pickle.dump(handlocals_dict, file)
|
||||
|
||||
|
||||
|
||||
|
||||
# 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)
|
||||
|
||||
|
||||
def parse_opt():
|
||||
modelpath = ROOT / 'ckpts/best_yolov5m_250000.pt' # 'ckpts/best_15000_0908.pt', 'ckpts/yolov5s.pt', 'ckpts/best_20000_cls30.pt'
|
||||
|
||||
'''datapath为视频文件目录或视频文件'''
|
||||
datapath = r"D:/datasets/ym/videos/标记视频/" # ROOT/'data/videos', ROOT/'data/images' images
|
||||
# datapath = r"D:\datasets\ym\highvalue\videos"
|
||||
# datapath = r"D:/dcheng/videos/"
|
||||
# modelpath = ROOT / 'ckpts/yolov5s.pt'
|
||||
|
||||
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=datapath, 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 main_loop_folders(opt):
|
||||
check_requirements(ROOT / 'requirements.txt', exclude=('tensorboard', 'thop'))
|
||||
|
||||
# path1 = r"D:\datasets\ym\videos\标记视频"
|
||||
|
||||
path2 = r"D:\datasets\ym\永辉双摄视频\加购_前摄\videos_front"
|
||||
# path3 = r"D:\datasets\ym\永辉双摄视频\加购_后摄\videos_back"
|
||||
# path4 = r"D:\datasets\ym\永辉双摄视频\退购_前摄\videos_front"
|
||||
# path5 = r"D:\datasets\ym\永辉双摄视频\退购_后摄\videos_back"
|
||||
|
||||
path6 = r"D:\datasets\ym\测试数据20240328\front"
|
||||
path7 = r"D:\datasets\ym\测试数据20240328\back"
|
||||
|
||||
'''列表paths内的元素为视频文件夹，该文件夹下元素为视频文件'''
|
||||
paths = [path2, path7] # [path1, path2, path3, path4, path5]
|
||||
|
||||
optdict = vars(opt)
|
||||
k1, k2 = 0, 0
|
||||
for p in paths:
|
||||
files = []
|
||||
if os.path.isdir(p):
|
||||
files.extend(sorted(glob.glob(os.path.join(p, '*.*'))))
|
||||
for file in files:
|
||||
file = r"D:\datasets\ym\测试数据20240328\front\112954521-7dd5ddad-922a-427b-b59e-a593e95e6ff4_front.mp4"
|
||||
|
||||
optdict["source"] = file
|
||||
run(**optdict)
|
||||
|
||||
k2 += 1
|
||||
if k2 == 1:
|
||||
break
|
||||
|
||||
elif os.path.isfile(p):
|
||||
run(**optdict)
|
||||
|
||||
k1 += 1
|
||||
if k1 == 1:
|
||||
break
|
||||
|
||||
def find_files_in_nested_dirs(root_dir):
|
||||
all_files = []
|
||||
extensions = ['.mp4']
|
||||
for dirpath, dirnames, filenames in os.walk(root_dir):
|
||||
for filename in filenames:
|
||||
file, ext = os.path.splitext(filename)
|
||||
if ext in extensions:
|
||||
all_files.append(os.path.join(dirpath, filename))
|
||||
return all_files
|
||||
|
||||
print('=======')
|
||||
|
||||
def main(opt):
|
||||
check_requirements(ROOT / 'requirements.txt', exclude=('tensorboard', 'thop'))
|
||||
|
||||
p = r"D:\datasets\ym\永辉测试数据_202404\20240402"
|
||||
optdict = vars(opt)
|
||||
files = []
|
||||
k = 0
|
||||
|
||||
all_files = find_files_in_nested_dirs(p)
|
||||
|
||||
if os.path.isdir(p):
|
||||
files.extend(sorted(glob.glob(os.path.join(p, '*.*'))))
|
||||
for file in files:
|
||||
optdict["source"] = file
|
||||
run(**optdict)
|
||||
|
||||
k += 1
|
||||
if k == 2:
|
||||
break
|
||||
elif os.path.isfile(p):
|
||||
run(**vars(opt))
|
||||
|
||||
def main_loop(opt):
|
||||
check_requirements(ROOT / 'requirements.txt', exclude=('tensorboard', 'thop'))
|
||||
|
||||
optdict = vars(opt)
|
||||
|
||||
# p = r"D:\datasets\ym\永辉测试数据_比对"
|
||||
p = r"D:\datasets\ym\广告板遮挡测试\8"
|
||||
# p = r"D:\datasets\ym\videos\标记视频"
|
||||
# p = r"D:\datasets\ym\实验室测试"
|
||||
|
||||
k = 0
|
||||
if os.path.isdir(p):
|
||||
files = find_files_in_nested_dirs(p)
|
||||
|
||||
files = [r"D:\datasets\ym\videos\标记视频\test_20240402-173935_6920152400975_back_174037372.mp4",
|
||||
r"D:\datasets\ym\videos\标记视频\test_20240402-173935_6920152400975_front_174037379.mp4"
|
||||
]
|
||||
|
||||
files = [r"D:\datasets\ym\广告板遮挡测试\8\2500441577966_20240508-175946_front_addGood_70f75407b7ae_155_17788571404.mp4"]
|
||||
for file in files:
|
||||
optdict["source"] = file
|
||||
run(**optdict)
|
||||
|
||||
k += 1
|
||||
if k == 2:
|
||||
break
|
||||
elif os.path.isfile(p):
|
||||
optdict["source"] = p
|
||||
run(**vars(opt))
|
||||
|
||||
|
||||
|
||||
|
||||
if __name__ == '__main__':
|
||||
opt = parse_opt()
|
||||
|
||||
# main_loop_folders(opt)
|
||||
# main(opt)
|
||||
main_loop(opt)
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
|
||||
Reference in New Issue
Block a user