Скачать или смотреть Ground Target Classification from SAR Image with the Pixel Complement for Target Shape

Sponsored by IEEE Sensors Council (https://ieee-sensors.org/)

Title: Ground Target Classification from SAR Image with the Pixel Complement for Target Shape

Author: Hongliang Zhu{1}, Minyi Hon{2}, Wat Wong{1}, Rocky Leung{3}, Nang Lin{4}, Kinley Lin{1}

Affiliation: {1}Chinese University of Hong Kong, Hong Kong; {1}Chinese University of Hong Kong, China; {2}University of Minnesota, United States; {3}University of Tokyo, Japan; {4}Washington University in St. Louis, United States

Abstract: This paper introduced a novel method to reconstruct the missing part of the ground target in the original SAR image, which is occluded by the shadow. We utilize a simulated optical image of the top aspect viewed ground target with SAR’s depression angle to match the ground target in the original SAR image. Then, the pixel correlation between two regions in both images could be established, which is used to predict the pixel value for the missing part of the ground target in the original SAR image. Finally, we make up the pixel for the missing part of the target in the SAR image by the pixel prediction equation.

IEEE Sensors Conferences (https://ieee-sensors.org/conferences/)
IEEE Sensors Journal (https://ieee-sensors.org/sensors-jour...)
IEEE Sensors Letters (https://ieee-sensors.org/sensors-lett...)
IEEE Internet of Things Journal (https://ieee-iotj.org/)

IEEE SENSORS conference proceedings (https://ieeexplore.ieee.org/xpl/conho...)