Deep Feature segmentation model Driven by Hybrid Convolution Network for Hyper Spectral Image Classification

Abstract

Hyperspectral image (HSI) classification can support different applications, such as agriculture, military, city planning, land utilization, and identifying distinct regions. It is treated as a crucial topic in the research community. Recent advancement in convolution neural network (CNN) has shown the unique capability of extracting meaningful feature and classification. However, CNN works with square images with fixed dimensions and cannot extract local information of images having distinct geometric variations with context and content relationships; hence there is a scope for improvement in correctly identifying class boundaries. Encouraged by the facts, we propose an HSI feature segmentation model by the hybrid convolution network (GCNN-RESNET152) for the HSI classification. First, pre-trained CNN on ImageNet is used to obtain the multilayer feature. Second, the 3D discrete wavelet transform image is fed into the graph convolution network GCN model to gain patch-to-patch correlations feature maps. Finally, the features are integrated using the three weighted coefficients concatenation method. Finally, the linear classifier is used to predict the semantic classes of pixel HSI. The proposed model is tested on four benchmark dataset Houston University (HU), Indian pines(IP), Kennedy space station(KSS), and Pavia university(PU). The result is compared with state-of-art algorithms and found to be superior in terms of overall, average, and kappa accuracy. The Overall, average and kappa accuracy achieved for HU: 97.7%, 99.4%, 95.6%, IP: 97.7%, 99.4%, 95.6%, KSS:97.48%,99.68%,96.43%, and PU: 97.7%, 99.4%, 95.6% respectively, which is 5 to 8% more than state of art methods.