Diverse Region-Based CNN for Hyperspectral Image Classification

By Mengmeng Zhang, Qian Du. IEEE Transactions on Image Processing, 2018

533 citations recorded by the source corpus.

Convolutional neural network (CNN) is of great interest in machine learning and has demonstrated excellent performance in hyperspectral image classification. In this paper, we propose a classification framework, called diverse region-based CNN, which can encode semantic context-aware representation to obtain promising features. With merging a diverse set of discriminative appearance factors, the resulting CNN-based representation exhibits spatial-spectral context sensitivity that is essential for accurate pixel classification. The proposed method exploiting diverse region-based inputs to learn contextual interactional features is expected to have more discriminative power. The joint representation containing rich spectral and spatial information is then fed to a fully connected network and the label of each pixel vector is predicted by a softmax layer. Experimental results with widely used hyperspectral image data sets demonstrate that the proposed method can surpass any other conventional deep learning-based classifiers and other state-of-the-art classifiers.

Subfields: Atmospheric Science, Biomedical Engineering, Media Technology

View Paper: https://www.semanticscholar.org/paper/b4891152b7291a204d30a4786f25ea6846635c4b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Atmospheric Science

In Biomedical Engineering

In Media Technology

Citations in the corpus

A Survey of Convolutional Neural Networks: Analysis, Applications, and Prospects (2020, methodology): “In addition, CNN can be deployed in medical image classification [123], [124], traffic scenes related classification [125], [126], and so on [127], [128].”
HSI-BERT: Hyperspectral Image Classification Using the Bidirectional Encoder Representation From Transformers (2020, methodology): “For better estimation of parameters and classification performance, the same data augmentation strategy introduced in DRCNN [2] is applied, which augments the training data by a factor of two through flipping and adds small Gaussian noise to the original HSI data.”
Cascaded Recurrent Neural Networks for Hyperspectral Image Classification (2017, background): “Different from them, convolutional neural networks (CNNs) have local connection and weight sharing properties, thus largely reducing the number of training parameters [32]–[34].”
Hyperspectral Image Classification—Traditional to Deep Models: A Survey for Future Prospects (2021, methodology): “for classification tasks [132]–[134] however, many works have also utilized SVM [135], [136] for this purpose.”
Multiscale Dynamic Graph Convolutional Network for Hyperspectral Image Classification (2019, methodology): “However, the aforementioned methods are all based on the handcrafted spectral-spatial features [14] which heavily depend on professional expertise and are quite empirical.”
Deep Learning for Hyperspectral Image Classification: An Overview (2019, methodology): “This method of augmenting data is widely used in [56], [108], [98], [80], and [85].”
Deep Learning Meets Hyperspectral Image Analysis: A Multidisciplinary Review (2019, background): “Instead, in [57] authors proposed a Diverse-region-based CNN (DR-CNN), which uses a joint representation from diverse regions in the proposed CNN framework, simultaneously taking advantage of both spectral and spatial features.”