Hyperspectral image shadow compensation via cycle-consistent adversarial networks

Abstract Illumination variance and shadows are challenging problems in remote sensing and hyperspectral imaging applications. Shadow compensation can effectively enhance the accuracy of object detection and material classification. Most shadow compensation methods either require preprocessing to detect the shadow region, or extra knowledge collected from additional sensors. Supervised deep learning based methods require paired samples to train the network. To overcome these restrictions, this work proposes an effective cycle-consistent adversarial network for shadow compensation (SC-CycleGAN). This unsupervised method is able to automatically transfer spectra in shadow region to their nonshadow counterparts, without requiring paired training samples and the step of shadow detection. The superiority of the proposed scheme is confirmed with both laboratory-created labeled data and real airborne data.

[1]  Richard J. Murphy,et al.  A Physics-Based Deep Learning Approach to Shadow Invariant Representations of Hyperspectral Images , 2018, IEEE Transactions on Image Processing.

[2]  Ying Liu,et al.  Context-aware attention network for image recognition , 2019, Neural Computing and Applications.

[3]  Xiao Xiang Zhu,et al.  Deep Recurrent Neural Networks for Hyperspectral Image Classification , 2017, IEEE Transactions on Geoscience and Remote Sensing.

[4]  Sen Jia,et al.  Convolutional neural networks for hyperspectral image classification , 2017, Neurocomputing.

[5]  Bo Du,et al.  Binary-Class Collaborative Representation for Target Detection in Hyperspectral Images , 2019, IEEE Geoscience and Remote Sensing Letters.

[6]  Chu Zhang,et al.  Variety Identification of Single Rice Seed Using Hyperspectral Imaging Combined with Convolutional Neural Network , 2018 .

[7]  Abhinav Gupta,et al.  Generative Image Modeling Using Style and Structure Adversarial Networks , 2016, ECCV.

[8]  Qian Du,et al.  GETNET: A General End-to-End 2-D CNN Framework for Hyperspectral Image Change Detection , 2019, IEEE Transactions on Geoscience and Remote Sensing.

[9]  Gang Wang,et al.  Deep Learning-Based Classification of Hyperspectral Data , 2014, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.

[10]  Min Zhao,et al.  Nonlinear Unmixing of Hyperspectral Data via Deep Autoencoder Networks , 2019, IEEE Geoscience and Remote Sensing Letters.

[11]  Zhangquan Shen,et al.  Review of shadow detection and de-shadowing methods in remote sensing , 2013, Chinese Geographical Science.

[12]  Zhenwei Shi,et al.  Multi-scale retinex improvement for nighttime image enhancement , 2014 .

[13]  R. Richter,et al.  De‐shadowing of satellite/airborne imagery , 2005 .

[14]  David A. Landgrebe,et al.  Hyperspectral image data analysis , 2002, IEEE Signal Process. Mag..

[15]  Sabine Süsstrunk,et al.  Automatic and Accurate Shadow Detection Using Near-Infrared Information , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[16]  Bo Du,et al.  Target Dictionary Construction-Based Sparse Representation Hyperspectral Target Detection Methods , 2019, IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing.