# Industry Image Classification Based on Stochastic Configuration Networks and Multi-Scale Feature Analysis

## TL;DR

This paper introduces a new image classification method using stochastic configuration networks and multi-scale features to improve accuracy in industrial image recognition.

## Contribution

The novelty lies in combining stochastic configuration networks with multi-scale feature extraction for industrial image classification.

## Key findings

- Multi-scale features extracted using deep 2DSCN improve feature informativeness.
- The proposed method achieves higher classification accuracy compared to existing methods.
- Experiments on handwritten digits and steel strip images validate the method's effectiveness.

## Abstract

For industry image data, this paper proposes an image classification method based on stochastic configuration networks and multi-scale feature extraction. The multi-scale features are extracted from images of different scales using deep 2DSCN, and the hidden features of multiple layers are also connected together to obtain more informational features. The integrated features are fed into SCNs to learn a classifier which improves the recognition rate for different categories. In the experiments, a handwritten digit database and an industry hot-rolled steel strip database are used, and the comparison results demonstrate the proposed method can effectively improve the classification accuracy.