Supermarket Product Detection and Recognition: Utilizing Deep Learning with Rectified Imagery
Homography and Hough-based rectification improved grocery detection accuracy on angled supermarket images.
The paper studies supermarket product detection when densely packed shelves are photographed at oblique angles. It supplements object detectors with the Hough transform and homography estimation to rectify images before recognition. Experiments across multiple detectors indicate that rectification improves grocery detection accuracy. The gains are limited by capture angle and object density, and the authors argue for a dataset covering those conditions.
- Angled, densely packed shelf photos degrade grocery object detection.
- Homography estimation and the Hough transform rectify images before detection.
- Rectification improved accuracy across the tested object-detection models.
- Gains depend on capture angle and the number of objects per image.
From the original article
1 min read at arxiv.orgProduct Identification has sprung up to become one of the most challenging problems in the automation of the retail industry. With the new industry 5. 0 standards, automated inventory management, and catalog creation tasks are vitally important.
By Mayank Sah · arxiv.org
A short excerpt; the article and its images belong to the publisher. Summary, key facts and assessment above are AI-written from the article and can be wrong.