Texture Image Classification Using DWT and AlexNet Feature Fusion (DWT_AlexNet_DNN)

Forum topic · 小凯 · 2026-09-01

Summary

This paper proposes DWT_AlexNet_DNN, a hybrid feature fusion framework for texture image classification that combines Discrete Wavelet Transform (DWT) features with deep features extracted by AlexNet and feeds the fused representation to a deep neural network. The work addresses a key limitation of both handcrafted and deep approaches: handcrafted features capture local texture characteristics but have limited capability to represent complex visual patterns, while deep learning models learn discriminative representations automatically but may not fully exploit the multiscale spatial-frequency information inherent in texture images. Texture image classification is important for computer vision applications including industrial inspection, medical image analysis, remote sensing, and object recognition. The paper is authored by Arun D. Kulkarni and is available on arXiv (2608.28524).

Overview

This post introduces a computer vision paper by Arun D. Kulkarni, published on arXiv: 2608.28524.

Key points

Topic: Texture image classification, which plays a significant role in computer vision applications including industrial inspection, medical image analysis, remote sensing, and object recognition.
Problem: Handcrafted features can capture local texture characteristics but may have limited capability to represent complex visual patterns. Conversely, deep learning models automatically learn discriminative representations but may not fully exploit the multiscale spatial-frequency information inherent in texture images.
Proposed method: A hybrid feature fusion framework, termed DWT_AlexNet_DNN, which combines Discrete Wavelet Transform (DWT) features with deep features extracted using AlexNet for texture image classification.

Abstract (original)

> Texture image classification plays a significant role in computer vision applications, including industrial inspection, medical image analysis, remote sensing, and object recognition. Handcrafted features can capture local texture characteristics but may have limited capability to represent complex visual patterns. In contrast, deep learning models automatically learn discriminative representations but may not fully exploit the multiscale spatial-frequency information inherent in texture images. This paper proposes a hybrid feature fusion framework, termed DWT_AlexNet_DNN, which combines Discrete Wavelet Transform (DWT) features with deep features extracted using AlexNet for texture image classification.

Paper link: https://arxiv.org/abs/2608.28524