| Title | Haze Mitigation in High-Resolution Satellite Imagery Using Enhanced Style-Transfer Neural Network and Normalization Across Multiple GPUs |
| Publication Type | Conference Paper |
| Year of Publication | 2021 |
| Authors | Park, Byung H., Chattopadhyay, Somrita, Burgin, John |
| Conference Name | 2021 IEEE International Geoscience and Remote Sensing Symposium IGARSS |
| Date Published | jul |
| Keywords | convolutional neural networks, Cycle-GAN, graphics processing units, Haze Mitigation, Image color analysis, image segmentation, Metrics, Neural networks, neural style transfer, pubcrawl, remote sensing, resilience, Resiliency, Satellites, Scalability, Training, Training data |
| Abstract | Despite recent advances in deep learning approaches, haze mitigation in large satellite images is still a challenging problem. Due to amorphous nature of haze, object detection or image segmentation approaches are not applicable. Also it is practically infeasible to obtain ground truths for training. Bounded memory capacity of GPUs is another constraint that limits the size of image to be processed. In this paper, we propose a style transfer based neural network approach to mitigate haze in a large overhead imagery. The network is trained without paired ground truths; further, perception loss is added to restore vivid colors, enhance contrast and minimize artifacts. The paper also illustrates our use of multiple GPUs in a collective way to produce a single coherent clear image where each GPU dehazes different portions of a large hazy image. |
| DOI | 10.1109/IGARSS47720.2021.9553574 |
| Citation Key | park_haze_2021 |
Haze Mitigation in High-Resolution Satellite Imagery Using Enhanced Style-Transfer Neural Network and Normalization Across Multiple GPUs