Advanced Texture Classification Using Convolutional Deep Long Short-Term Memory for Enhanced Accuracy and Efficiency

DOI

10.2991/978-94-6463-738-0_64How to use a DOI?

Keywords

Texture classification; Computer Vision; Convolutional Neural Network; Long Short Term Memory; Two-Dimensional Haar Wavelet Transform

Abstract

The texture is the basic factor in an image, representing each layer of the image. Hence, texture analysis is considered asa fundamental process for various computer vision operations including segmentation, image retrieval, and image recognition. Several current applications demand an effective texture classification with higher accuracy. In this context, texture descriptors are designed for texture classification, same time these descriptors encounter difficulties due to time consumption, and overfittingleading to lower classification accuracy. Thus, this paper presents an effective and robust texture classification method namely Convolutional Deep Long Short-Term Memory (CDLSTM). In this paper, two advanced methods namely Convolutional Neural Network (CNN) and Deep Long Short Term Memory (DLSTM) are amalgamated to classify texture. The initial step of the proposed approach is to acquire the images through various datasets including CUReT, KTH-TIPS, and UIUC. The collected data from these three datasets are pre-processed by applying Geometric Transformations, Noise Reduction, Scaling, and Illumination Correction. Using these pre-processed images, the effective Two-Dimensional Haar Wavelet Transform (2DHWT) carried out feature extraction. The 2DHWT method manages detailed texture information by decomposing the images into several frequency bands. Utilizing the extracted features, the proposed CDLSTM approach handles the sequential data for effective texture classification. Following this, the efficiency of the developed 2DHWT method is tested using three datasets. In this, five evaluation measures are exploited such as precision, kappa score, F1-score, recall, and accuracy. The overall effectiveness of the developed CDLSTM method affirms that it is the finest texture classification solution with a higher accuracy of 98.1% and less processing time.

Copyright

© 2025 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Supriya Bagewadi
AU  - Sachinkumar Veerashetty
PY  - 2025
DA  - 2025/06/22
TI  - Advanced Texture Classification Using Convolutional Deep Long Short-Term Memory for Enhanced Accuracy and Efficiency
BT  - Proceedings of the International Conference on Advances and Applications in Artificial Intelligence (ICAAAI 2025)
PB  - Atlantis Press
SP  - 817
EP  - 833
SN  - 1951-6851
UR  - https://doi.org/10.2991/978-94-6463-738-0_64
DO  - 10.2991/978-94-6463-738-0_64
ID  - Bagewadi2025
ER  -