Multi-channel Features Combined with Gray Features for Remote Sensing Image Classification

Authors

Khaled Hasan, Electrical Engineering Department, Faculty of Engineering at Shoubra, Benha University, Cairo, EgyptFollow
Hussien Harb, Computer Department, National Telecommunications Institute, Cairo, Egypt
Lamiaa A. Elrefaei, Electrical Engineering Department, Faculty of Engineering at Shoubra, Benha University, Cairo, Egypt
Hala M. Abdel-Kader, Electrical Engineering Department, Faculty of Engineering at Shoubra, Benha University, Cairo, Egypt

Corresponding Author

Khaled Hasan

Subject Area

Electronics and Communication Engineering

Article Type

Original Study

Abstract

Image classification in remote sensing is crucial for various applications like deforestation monitoring and Land Use/Land Cover (LULC) mapping. While grayscale images offer simplicity, recent advancements have enhanced classification by integrating spatial and relative data. However, relying solely on grayscale images overlooks valuable color information from different channels. To overcome this, researchers have explored two strategies: combining diverse features and utilizing multi-channel features. These methods merge spatial information and exploit shared features across channels. Despite their improved performance compared to grayscale, they face increased dimensionality, necessitating dimensionality reduction techniques. In our study, we propose a novel approach that integrates spatial information into multichannel features without dimensionality reduction, while also incorporating grayscale features. Specifically, we extract features such as Scale Invariant Feature Transform 3 Channels (SIFT3CH), Speed Up Robust Feature 3 Channels (SURF3CH), and color histogram features from multi-channel images, alongside features like SIFT, SURF, Local Binary Pattern (LBP), Gray Level Cooccurrence Matrix (GLCM), and Haralick features from grayscale images. These features are used to train and test three nonparametric classifiers: K-Nearest Neighbors (KNN), Support Vector Machines (SVM), and Random Forests (RF). Additionally, various feature combinations are tested with the SVM classifier. Experimental results demonstrate that our proposed features achieve comparable performance to other multi-channel features, with SVM consistently outperforming KNN and RF classifiers. By combining our features with grayscale features, we achieve higher classification accuracy of 92.78%. These findings underscore the potential of integrating multi-channel features with spatial information and grayscale features to enhance remote sensing image classification.

Keywords

classification; Spatial information; multi-channel features; texture features; Fusion

Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 License.

Recommended Citation

Hasan, Khaled; Harb, Hussien; Elrefaei, Lamiaa A.; and Abdel-Kader, Hala M. (2025) "Multi-channel Features Combined with Gray Features for Remote Sensing Image Classification," Mansoura Engineering Journal: Vol. 49 : Iss. 5 , Article 17.
Available at: https://doi.org/10.58491/2735-4202.3244