I am to work on a computer vision project where the goal is to count the number of totes in a pile of totes (image shown is an example of what the model might see). Was wondering what the best strategy would be to model this. For example:
1. Would it make sense to label ever individual tote in the image and just count every bounding box?
2. label the entire pile and estimate a count of boxes in the pile by calculating width and height of the bounding box?
Has anyone ever worked on something similar and what tools or models you have used to accomplish this?
In practice, a combination of both approaches may be used depending on the specific requirements of the project. For example, you could use individual tote labeling for accurate counts in smaller or less densely packed piles, while using pile labeling for faster processing of larger or more densely packed piles. As for tools and models, popular object detection frameworks such as YOLO (You Only Look Once), SSD (Single Shot Multibox Detector), or Faster R-CNN (Region-based Convolutional Neural Network) can be used for both approaches. These frameworks provide pre-trained models that can be fine-tuned on your specific dataset to detect and count totes in the images. Additionally, tools like OpenCV and TensorFlow can be used for image processing and deep learning tasks in Python.
- Badges
- Science topic
- Similar topics
- Biological Science
- Physiology
- Physiological Processes
- Stress