How to save Detected Class Name in Text file Instead of Label

Hamad_Younis · March 27, 2023, 4:58am

I have a question on this please YOLOv5, as I have been searching all over google and cannot seem to get a fruitful response.

Can I get the same output using this command?

python3 detect.py --weights /Users/yolov5/runs/train/exp6/weights/best.pt  --img 640 --conf 0.25 --iou-thres  0.10 --source '/Users/vids50.mp4' --line-thickness 1 --save-txt

I want to get the output in text file when detect.py runs… The output will be a text file or csv file.

as demonstrate here:

#      xmin    ymin    xmax   ymax  confidence  class    name
      749.50   43.50  1148.0  704.5    0.874023      0  person
     114.75  195.75  1095.0  708.0    0.624512      0  person
     986.00  304.00  1028.0  420.0    0.286865     27     tie

Or it can be like the following in text file

#    
  0  person   749.50   43.50  1148.0  704.5    
  0  person  114.75  195.75  1095.0  708.0    
  27  tie        986.00  304.00  1028.0  420.0

*** IF yes!! may I request your guidance to this example, please?*** I’ve been struggling with this expression for some weeks now.

I am getting the --saved txt files but not in the about format…please help
Thanx for acknowledging my digital requests.

Mohamed · March 27, 2023, 11:14pm

@Hamad_Younis Yes, using the --save-txt flag, as you provided in your first snippet, is how you save the txt file.

github.com

ultralytics/yolov5/blob/master/detect.py#L219


      
              # 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():
              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=ROOT / 'data/images', 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')

github.com

ultralytics/yolov5/blob/master/detect.py#L164


      
          # 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 + '\n')
          
          
    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:

Hamad_Younis · March 28, 2023, 4:16am

@Mohamed But --save-txt file only save the labels not the classname. Can you modified the code so it can also save the class name

Mohamed · March 28, 2023, 4:45am

I can take a look at that portion. Of note, the first value in each line of the file is the class ID index value based on your obj.names file.

It will just require accessing the index for the class label name in obj.names and mapping it (as the key) to the class label name (value) in a variable that is of type, dictionary.

Then use the value (class label name) from the key:value pairs to map the index to the label name before appending it to the end of the line being written for the txt file.

note, the key is the class ID index value, and the file uses 0-base indexing (0, 1, 2, etc.)
make the keys as type string

The classes are listed alphabetically in that obj.names file, too. So if you know all of your class names in your trained model, you can make the dictionary without even viewing obj.names

Hamad_Younis · March 28, 2023, 4:57am

@Mohamed Thanks

Topic		Replies	Views
How to save Time of the object when it detected in Video using YoloV5 or YoloV8 Community Help split-after-upload , bugs , feature-request , formats , export	14	8241	March 31, 2023
How to save time in .txt file when the object is Detected in Video using YoloV5 Community Help bugs , feature-request , segmentation , export	1	847	March 28, 2023
How to output the class label distribution if I export my dataset as YOLO v5 PyTorch format? Community Help	1	229	May 12, 2022
YOLOv9 Labeling Community Help	1	87	April 29, 2024
Get unusal yaml file and wrong label text files Community Help	10	493	January 29, 2024

How to save Detected Class Name in Text file Instead of Label

Related topics