Image Quality Enhancement and Detection of Small and Dense Objects in Industrial Recycling Processes

Oussama Messai, Abbass Zein-Eddine, Abdelouahid Bentamou, Mickaël Picq, Nicolas Duquesne +2 more
9/1/2025
cs.CVeess.IV

Official code: o-messai/SDOOD (4 stars, last commit 2025-01-01). 1 more implementation below.

Abstract

This paper tackles two key challenges: detecting small, dense, and overlapping objects (a major hurdle in computer vision) and improving the quality of noisy images, especially those encountered in industrial environments. [1, 2]. Our focus is on evaluating methods built on supervised deep learning. We perform an analysis of these methods, using a newly developed dataset comprising over 10k images and 120k instances. By evaluating their performance, accuracy, and computational efficiency, we identify the most reliable detection systems and highlight the specific challenges they address in industrial applications. This paper also examines the use of deep learning models to improve image quality in noisy industrial environments. We introduce a lightweight model based on a fully connected convolutional network. Additionally, we suggest potential future directions for further enhancing the effectiveness of the model. The repository of the dataset and proposed model can be found at: https://github.com/o-messai/SDOOD, https://github.com/o-messai/DDSRNet

View on arXivView PDF

Code Implementations(2)

o-messai/SDOODOfficial100%
40Oct 21, 20241 years ago
o-messai/DDSRNetOfficial100%
MIT

DDSRNet (A Deep Model for Denoising and Super-Resolution) is a new two-stage learning architecture designed for simultaneous image denoising and super-resolution.

40PythonMar 30, 20251 years agoMIT
anomaly-detectiondeep-learningdenoising-imagesimage-processingloss-functions+3 more

Similar Papers

Small Object Detection: A Comprehensive Survey on Challenges, Techniques and Real-World Applications

85% match2025

Real-Time Object Detection in Occluded Environment with Background Cluttering Effects Using Deep Learning

84% match2024

Noisy Label Processing for Classification: A Survey

84% match1 repo2024

ContextMix: A context-aware data augmentation method for industrial visual inspection systems

83% match1 repo12024

Can KAN Work? Exploring the Potential of Kolmogorov-Arnold Networks in Computer Vision

82% match2024

Rapid Salient Object Detection with Difference Convolutional Neural Networks

82% match1 repo252025

Cite this paper

@article{messai2025image,
  title  = {Image Quality Enhancement and Detection of Small and Dense Objects in Industrial Recycling Processes},
  author = {Oussama Messai and Abbass Zein-Eddine and Abdelouahid Bentamou and Mickaël Picq and Nicolas Duquesne and Stéphane Puydarrieux and Yann Gavet},
  year   = {2025},
  eprint = {2509.01332},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2509.01332v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2509.01332.svg)](https://codeofpaper.com/paper/2509.01332)

Discussion