Hyperspectral Image Classification with Optimized Compressed Synergic Deep Convolution Neural Network with Aquila Optimization

Abstract

The classification technology of hyperspectral images (HSI) consists of many contiguous spectral bands that are often utilized for a various Earth observation activities, such as surveillance, detection, and identification. The incorporation of both spectral and spatial characteristics is necessary for improved classification accuracy. In the classification of hyperspectral images, deep learning has gained significant traction. This research analyzes how to accurately classify new HSI from limited samples with labels. A novel deep-learning-based categorization based on feature extraction and classification is designed for this purpose. Initial extraction of spectral and spatial information is followed by spectral and spatial information integration to generate fused features. The classification challenge is completed using a compressed synergic deep convolution neural network with Aquila optimization (CSDCNN-AO) model constructed by utilising a novel optimization technique known as the Aquila Optimizer (AO). The HSI, the Kennedy Space Center (KSC), the Indian Pines (IP) dataset, the Houston U (HU) dataset, and the Salinas Scene (SS) dataset are used for experiment assessment. The sequence testing on these four HSI-classified datasets demonstrate that our innovative framework outperforms the conventional technique on common evaluation measures such as average accuracy (AA), overall accuracy (OA), and Kappa coefficient (k). In addition, it significantly reduces training time and computational cost, resulting in enhanced training stability, maximum performance, and remarkable training accuracy.

Enhancing Hyper-Spectral Image Classification with Reinforcement Learning and Advanced Multi-Objective Binary Grey Wolf Optimization

“…GWO is especially valuable because it balances exploration and exploitation: It explores the search space to avoid local minima and exploits the best solutions to converge upon an optimal set of bands [24][25][26]. By doing so, it eliminates unnecessary bands, thereby reducing computational load, improving classification accuracy, and enhancing the overall efficiency of the HS image processing [17][18][19].…”

“…Reddy et al [17] developed a compressed synergic deep convolution neural network enhanced with Aquila optimization (CSDCNN-AO) for HS image classification, which incorporates a novel optimization technique known as AO. Their experimental evaluation encompassed four datasets.…”

“…It can exponentially increase the computational burden, as the algorithms have to process a larger volume of data. This not only slows down the analysis but also requires more computational resources, which may not be feasible in all application scenarios [15][16][17]. Moreover, the existence of noise and closely linked bands may conceal the actual signal, complicating the task of classifiers in precisely differentiating between various classes.…”

“…This process not only simplifies the classifier's model by reducing the complexity of the feature space but also enhances the classifier's ability to make accurate predictions by focusing on the most relevant information. Additionally, the removal of redundant bands leads to faster processing times and less demanding storage requirements, facilitating more efficient and scalable applications of HS imaging in diverse fields [17][18][19].…”

Workload-Aware Performance Tuning for Multimodel Databases Based on Deep Reinforcement Learning

“…However, there are still some limitations to the above methods. It is difcult for them to achieve optimal performance under limited training samples, and they are prone to overftting [45][46][47].…”