Hyperspectral Band Selection Based On Ternary Weight Convolutional Neural Network

Feng, Jie; Li, Di; Chen, Jiantong; Zhang, Xiangrong; Tang, Xu; Wu, Xiande

doi:10.1109/igarss.2019.8898889

Cited by 12 publications

(3 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…First, the selection of good features provides a guarantee for later feature mining and feature integration. Jie [22] and others set 11 filters in the first layer of depth convolution to select bands. Although many redundant bands are eliminated, the useful features in bands are simultaneously lost, and the final classification effect is reduced.…”

Section: Related Researchmentioning

confidence: 99%

Small Sample Classification of Hyperspectral Remote Sensing Images Based on Sequential Joint Deeping Learning Model

2020

View full text Add to dashboard Cite

show abstract

Section: Related Researchmentioning

confidence: 99%

Small Sample Classification of Hyperspectral Remote Sensing Images Based on Sequential Joint Deeping Learning Model

2020

View full text Add to dashboard Cite

show abstract

“…Recently, a few studies proposed the effective band selection based on spectral–spatial integrated CNN model. Feng et al 19 combined mathematical method with CNN and pointed out that the importance of band could be differentiated by the use of the hard thresholding function. Liu et al 1 proposed the 2B-CNN (two-branch CNN) model, consisting of 1D CNN and 2D CNN for extracting spectral and spatial features respectively.…”

Section: Introductionmentioning

confidence: 99%

Effective band selection of hyperspectral image by an attention mechanism-based convolutional network

Zheng

Liu

et al. 2022

RSC Adv.

View full text Add to dashboard Cite

show abstract

“…In addition to the common feature selection methods already described, deep learning has also been applied to hyperspectral data processing. In the field of hyperspectral feature selection, a novel ternary weight convolution neural network (TWCNN) was proposed, which uses a depth-wise convolutional layer with 1 × 1 filters as the first layer of the network, and can achieve end-to-end feature selection and classification [33]. Lorenzo et al [34] developed a data-driven hyperspectral band selection algorithm that couples an attention-based convolutional neural network to identify the most information-rich regions in the spectrum.…”

Section: Introductionmentioning

confidence: 99%

Hyperspectral Feature Selection for SOM Prediction Using Deep Reinforcement Learning and Multiple Subset Evaluation Strategies

et al. 2022

View full text Add to dashboard Cite

It has been widely certified that hyperspectral images can be effectively used to monitor soil organic matter (SOM). Though numerous bands reveal more details in spectral features, information redundancy and noise interference also come accordingly. Due to the fact that, nowadays, prevailing dimensionality reduction methods targeted to hyperspectral images fail to make effective band selections, it is hard to capture the spectral features of ground objects quickly and accurately. In this paper, to solve the inefficiency and instability of hyperspectral feature selection, we proposed a feature selection framework named reinforcement learning for feature selection in hyperspectral regression (RLFSR). Specifically, the Markov Decision Process (MDP) was used to simulate the hyperspectral band selection process, and reinforcement learning agents were introduced to improve model performance. Then two spectral feature evaluation methods were introduced to find internal relationships between the hyperspectral features and thus comprehensively evaluate all hyperspectral bands aimed at the soil. The feature selection methods—RLFSR-Net and RLFSR-Cv—were based on pre-trained deep networks and cross-validation, respectively, and achieved excellent results on airborne hyperspectral images from Yitong Manchu Autonomous County in China. The feature subsets achieved the highest accuracy for most inversion models, with inversion R2 values of 0.7506 and 0.7518, respectively. The two proposed methods showed slight differences in spectral feature extraction preferences and hyperspectral feature selection flexibilities in deep reinforcement learning. The experiments showed that the proposed RLFSR framework could better capture the spectral characteristics of SOM than the existing methods.

show abstract

Hyperspectral Band Selection Based On Ternary Weight Convolutional Neural Network

Cited by 12 publications

References 9 publications

Small Sample Classification of Hyperspectral Remote Sensing Images Based on Sequential Joint Deeping Learning Model

Small Sample Classification of Hyperspectral Remote Sensing Images Based on Sequential Joint Deeping Learning Model

Effective band selection of hyperspectral image by an attention mechanism-based convolutional network

Hyperspectral Feature Selection for SOM Prediction Using Deep Reinforcement Learning and Multiple Subset Evaluation Strategies

Contact Info

Product

Resources

About