# YOLOv5 by Ultralytics, GPL-3.0 license
import argparse
import os
import platform
importsys
from pathlib import Path
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
#Import modules under relative paths
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.plots import Annotator, colors, save_one_box
from utils.torch_utils import select_device, smart_inference_mode
@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)
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
vid_stride=1, # video frame-rate stride
):
#The first part of the run function: judge the incoming source parameter
source = str(source)
save_img = not nosave and not source.endswith('.txt') # save inference images, save inference images
is_file = Path(source).suffix[1:] in (IMG_FORMATS + VID_FORMATS) # Picture or video
is_url = source.lower().startswith(('rtsp://', 'rtmp://', 'http://', 'https://')) # Judge Is it a network flow address?
webcam = source.isnumeric() or source.endswith('.txt') or (is_url and not is_file) #Camera address/txt file path/network stream address
screenshot = source.lower().startswith('screen')
if is_url and is_file:
source = check_file(source) # download, download pictures or videos through the network stream address
#Directories
#Part 2: Create a new folder to save the results
save_dir = increment_path(Path(project) / name, exist_ok=exist_ok) # increment run, get the incremental path
(save_dir / 'labels' if save_txt else save_dir).mkdir(parents=True, exist_ok=True) # make dir, create a folder based on the incremental path
# Load model
#Part 3: Load model weights
device = select_device(device)
model = DetectMultiBackend(weights, device=device, dnn=dnn, data=data, fp16=half) # Select a deep learning framework from the multi-backend for inference and load the model
"""
device is used for training
data is the label name file path
super().__init__()
# Get the pre-training weight file path
w = str(weights[0] if isinstance(weights, list) else weights)
# Determine the model type pytorch
pt, jit, onnx, xml, engine, coreml, saved_model, pb, tflite, edgetpu, tfjs, paddle, triton = self._model_type(w)
fp16 &= pt or jit or onnx or engine # FP16
nhwc = coreml or saved_model or pb or tflite or edgetpu # BHWC formats (vs torch BCWH)
#yolov5 detects small targets, medium targets, and large targets in the three dimensions of stride 8, 16, and 32 respectively.
stride = 32 #default stride, 32 times downsampling (convolution)
cuda = torch.cuda.is_available() and device.type != 'cpu' # use CUDA
if not (pt or triton):
w = attempt_download(w) # download if not local
if pt: # PyTorch
model = attempt_load(weights if isinstance(weights, list) else w, device=device, inplace=True, fuse=fuse)
stride = max(int(model.stride.max()), 32) # model stride
names = model.module.names if hasattr(model, 'module') else model.names # get class names
model.half() if fp16 else model.float()
self.model = model # explicitly assign for to(), cpu(), cuda(), half()
"""
stride, names, pt = model.stride, model.names, model.pt # Model stride, category names that can be detected by the model, and whether the model is pytorch
imgsz = check_img_size(imgsz, s=stride) # check image size, the step size is generally 32, determine whether imgsz is a multiple of 32, 640*640
#Dataloader
#Part 4: Load images to be predicted
bs = 1 #batch_size, parameters are passed to the model one batch at a time, and one image is input each time
if webcam:
view_img = check_imshow(warn=True)
dataset = LoadStreams(source, img_size=imgsz, stride=stride, auto=pt, vid_stride=vid_stride)
bs = len(dataset)
elif screenshot:
dataset = LoadScreenshots(source, img_size=imgsz, stride=stride, auto=pt)
else:
dataset = LoadImages(source, img_size=imgsz, stride=stride, auto=pt, vid_stride=vid_stride)
# The returned dataset is a dataset object, including the path list of the data to be inferred, the classification and total number of pictures and videos, the step size, etc.
vid_path, vid_writer = [None] * bs, [None] * bs #Video path, video author
"""
LoadImagesAnnotation
def __init__(self, path, img_size=640, stride=32, auto=True, transforms=None, vid_stride=1):
files = []
for p in sorted(path) if isinstance(path, (list, tuple)) else [path]:
p = str(Path(p).resolve())
if '*' in p:
files.extend(sorted(glob.glob(p, recursive=True))) # glob
elif os.path.isdir(p):
files.extend(sorted(glob.glob(os.path.join(p, '*.*')))) # dir
elif os.path.isfile(p):
files.append(p) # files
else:
raise FileNotFoundError(f'{p} does not exist')
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"""
#Part 5: Model inference process
# Input the image into the model, generate prediction results, and draw the detection frame
# Run inference
model.warmup(imgsz=(1 if pt or model.triton else bs, 3, *imgsz)) # warmup
seen, windows, dt = 0, [], (Profile(), Profile(), Profile())
# Object iterator, first call the __iter__ method of the dataset object, the returned object is used as the iteration object, each loop calls the __next__ method, and the return value is passed to path, im, im0s, vid_cap, s
# path is the path of the file in the loop, im0s is the original image, im is the scaled image (640*640), s is the printed string information
# Each loop executes the next method: opencv imports the original image and scales the image (long side scaling, short side zero padding)
for path, im, im0s, vid_cap, s in dataset:
# Preprocessing
with dt[0]:
im = torch.from_numpy(im).to(model.device) # Convert the numpy format image into torch format and put it into the gpu
im = im.half() if model.fp16 else im.float() # uint8 to fp16/32
im /= 255 # 0 - 255 to 0.0 - 1.0 #Picture pixel value normalization
if len(im.shape) == 3:
im = im[None] # expand for batch dim, increase the batch dimension, how many image files are processed at one time
# Inference, prediction
with dt[1]:
visualize = increment_path(save_dir / Path(path).stem, mkdir=True) if visualize else False
pred = model(im, augment=augment, visualize=visualize) # Perform model prediction on the image, the return value is torch.Size([1,18900,85])
# 1 is an image, 18900 is the detected a priori frame, for each a priori frame: 80 is the probability of each category, 4 coordinate information, 1 confidence
# augment data enhancement, visuallize visualization (save the trained feature map), both of which are false by default
#NMS
with dt[2]:
pred = non_max_suppression(pred, conf_thres, iou_thres, classes, agnostic_nms, max_det=max_det)
# Non-maximum suppression: Filter the predicted a priori box and category according to the confidence threshold and iou threshold. For example, after filtering, it is [1,5,6], 6 is four coordinates, and two category probabilities
# Second-stage classifier (optional)
# pred = utils.general.apply_classifier(pred, classifier_model, im, im0s)
# Process predictions
for i, det in enumerate(pred): # per image
seen + = 1 # count, add one for each image processed
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
save_path = str(save_dir / p.name) # im.jpg # Save image path
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, print out the image size string
gn = torch.tensor(im0.shape)[[1, 0, 1, 0]] # normalization gain whwh, save the original image size
imc = im0.copy() if save_crop else im0 # for save_crop, crop the detected object and save it separately
annotator = Annotator(im0, line_width=line_thickness, example=str(names)) # Draw a border
iflen(det):
# Rescale boxes from img_size to im0 size
det[:, :4] = scale_boxes(im.shape[2:], det[:, :4], im0.shape).round()
# Print results
for c in det[:, 5].unique():
n = (det[:, 5] == c).sum() # detections per class
s + = f"{n} {names[int(c)]}{'s' * (n > 1)}, " # add to string
# Write results
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(f'{txt_path}.txt', 'a') as f:
f.write(('%g ' * len(line)).rstrip() % line + '\\
')
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)
#Stream results
im0 = annotator.result() # Get the painted picture
if view_img:
if platform.system() == 'Linux' and p not in windows:
windows.append(p)
cv2.namedWindow(str(p), cv2.WINDOW_NORMAL | cv2.WINDOW_KEEPRATIO) # allow window resize (Linux)
cv2.resizeWindow(str(p), im0.shape[1], im0.shape[0])
cv2.imshow(str(p), im0)
cv2.waitKey(1) # 1 millisecond
# Save results (image with detections)
if save_img: # Save the image
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}{'' if len(det) else '(no detections), '}{dt[1].dt * 1E3:.1f}ms")
# Part 6: Print output information
# 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"\\
{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)
#ROOT / 'data/QQ screen recording 20230803142136.mp4'
# 'runs/train/exp2/weights/best.pt'
def parse_opt():
parser = argparse.ArgumentParser()
parser.add_argument('--weights', nargs=' + ', type=str, default=ROOT / 'yolov5s.pt', help='model path or triton URL')
parser.add_argument('--source', type=str, default='0', help='file/dir/URL/glob/screen/0(webcam)')
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-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 # [640,640]
print_args(vars(opt))# Print a dictionary object of attributes and attribute values
return opt
def main(opt):
run(**vars(opt))
if __name__ == "__main__":
opt = parse_opt()
main(opt)