Related Experiment Video
Updated: Jul 19, 2025
14:09
Localization and Relative Quantification of Carbon Nanotubes in Cells with Multispectral Imaging Flow Cytometry
Published on: December 12, 2013
6.2K
Coal Flow Foreign Body Classification Based on ESCBAM and Multi-Channel Feature Fusion
Qiqi Kou1, Haohui Ma2, Jinyang Xu2
1School of Computer Science and Technology, China University of Mining and Technology, Xuzhou 221116, China.
Sensors (Basel, Switzerland)
|August 12, 2023
Summary
This study introduces an efficient deep learning network for classifying foreign bodies on belt conveyors. The novel approach enhances accuracy and processing speed while reducing computational complexity for improved coal transportation safety.
Area of Science:
- Engineering
- Computer Science
- Artificial Intelligence
Background:
- Foreign bodies in belt conveyors cause significant operational issues like scratching, tearing, and plugging.
- Existing classification networks suffer from large parameters, high computational complexity, low accuracy, and slow processing speeds.
Purpose of the Study:
- To develop a novel deep learning network for effective foreign body classification in coal belt conveyors.
- To address the limitations of current methods by improving accuracy, speed, and computational efficiency.
Main Methods:
- Proposed a novel network integrating a multi-channel feature fusion strategy for enhanced feature utilization.
- Developed an information fusion network using depthwise separable convolution and an improved residual network structure.
- Introduced a novel Enhanced Spatial and Channel Attention (ESCBAM) mechanism to improve image context understanding and feature performance.
Main Results:
- The proposed method significantly reduces parameters and computational complexity.
- Achieved high classification accuracy and fast processing speeds.
- Effectively classifies foreign bodies on belt conveyors, demonstrating superior performance.
Conclusions:
- The novel network offers a computationally efficient and accurate solution for foreign body detection in coal belt conveyors.
- The ESCBAM attention mechanism and multi-channel feature fusion contribute to improved network performance.
- This approach enhances the safety and efficiency of coal transportation systems.
