Hyperspectral image classification using 3D-2D CNN with multi-scale information extraction and fusion module

Gong, Hang; Mu, Tingkui; Li, Qiuxia; Han, Feng; Tuniyazi, Abudusalamu; Li, Haoyang; Wang, Wenjing; He, Zhiping; Li, Chunlai; Dai, Haishan

doi:10.1117/12.2586803

Cited by 2 publications

(1 citation statement)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Te semisupervised classifcation algorithm based on multilabeled samples and deep learning [55], with labels from both the nearest domain information and training samples [56,57], and nonlabeled samples obtained from self-teaching learning, yields an efective semisupervised hyperspectral image classifcation method [58,59]. Numerous classifcation experiments based on deep learning algorithms on a variety of hyperspectral data found that deep learning algorithms are the optimal classifcation algorithms in most cases [60][61][62][63][64][65][66]. Tis paper's objective was to apply deep learning theory to hyperspectral image classifcation, investigate the hyperspectral image classifcation model combined with a deep learning algorithm, and obtain a preferred method to improve classifcation accuracy by resolving the challenges of the Hughes phenomenon and of extracting the nonlinear features from within image elements.…”

Section: Introductionmentioning

confidence: 99%

Optimization Strategy of a Stacked Autoencoder and Deep Belief Network in a Hyperspectral Remote-Sensing Image Classification Model

Dai

Cheng

Guo

et al. 2023

Discrete Dynamics in Nature and Society

View full text Add to dashboard Cite

Improvements in hyperspectral image technology, diversification methods, and cost reductions have increased the convenience of hyperspectral data acquisitions. However, because of their multiband and multiredundant characteristics, hyperspectral data processing is still complex. Two feature extraction algorithms, the autoencoder (AE) and restricted Boltzmann machine (RBM), were used to optimize the classification model parameters. The optimal classification model was obtained by comparing a stacked autoencoder (SAE) and a deep belief network (DBN). Finally, the SAE was further optimized by adding sparse representation constraints and GPU parallel computation to improve classification accuracy and speed. The research results show that the SAE enhanced by deep learning is superior to the traditional feature extraction algorithm. The optimal classification model based on deep learning, namely, the stacked sparse autoencoder, achieved 93.41% and 94.92% classification accuracy using two experimental datasets. The use of parallel computing increased the model’s training speed by more than seven times, solving the model’s lengthy training time limitation.

show abstract