2019/12/06 by Shivangi Dwivedi, Dwivedi, Shivangi, Murari Mandal +4
Computer Science · Engineering · #Advanced Image and Video Retrieval Techniques #Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences #FOS: Electrical engineering #Image and Video Processing (eess.IV) #Remote-Sensing Image Classification #Video Surveillance and Tracking Methods #electronic engineering #information engineering
paper · pdf · doi:10.48550/arxiv.1912.03000
openalex publication_date 2019/12/06 · openalex created_date 2020/10/08 · openalex updated_date 2026/07/28
In this paper we propose a novel 3D CNN network with localized residual\nconnections for hyperspectral image classification. Our work chalks a\ncomparative study with the existing methods employed for abstracting deeper\nfeatures and propose a model which incorporates residual features from multiple\nstages in the network. The proposed architecture processes individual\nspatiospectral feature rich cubes from hyperspectral images through 3D\nconvolutional layers. The residual connections result in improved performance\ndue to assimilation of both low-level and high-level features. We conduct\nexperiments over Pavia University and Pavia Center dataset for performance\nanalysis. We compare our method with two recent state-of-the-art methods for\nhyperspectral image classification method. The proposed network outperforms the\nexisting approaches by a good margin.\n