A5059966127- Infrared Target Detection Methodologies
- Advanced Image Fusion Techniques
- Image Enhancement Techniques
- Advanced Image Processing Techniques
- Advanced Neural Network Applications
- Visual Attention and Saliency Detection
- Spacecraft Design and Technology
- Space Satellite Systems and Control
- Image and Video Quality Assessment
- Remote-Sensing Image Classification
- Remote Sensing and LiDAR Applications
National University of Defense Technology
2018-2024
Since remote sensing images are captured from the top of target, such as a satellite or plane platform, ship targets can be presented at any orientation. When detecting using horizontal bounding boxes, there will background clutter in box. This makes it harder to detect and find its precise location, especially when close proximity staying shore. To solve these problems, this paper proposes deep learning algorithm multiscale rotated box target complex obtain location orientation information...
Convolutional Neural Networks (CNNs) have achieved remarkable results in the field of infrared image enhancement. However, research on visual perception mechanism and objective evaluation indicators for enhanced images is still not in-depth enough. To make subjective more consistent, this paper uses a perceptual metric to evaluate enhancement effect images. The mimics early conversion process human system normalized Laplacian pyramid distance (NLPD) between original scene radiance effect....
We proposed a saliency based algorithm to detect the ground mobile targets, such as plane and vehicle, in images. The combines bottom-up top-down mode targets. Firstly, mode, extracts low level image features, intensity, standard deviation Gabor calculate difference between current pixel around it use of pixel; Then, HOG features vehicle targets train SVM classifier, which can learn high knowledge target. When testing on new image, classifier uses predict possibility appearance target each...
Abstract A real-time method for estimating the mass center of low earth orbit (LEO) remote sensing satellite in is introduced this paper. LEO requires highly on calculating position its center, due to need precise determination and attitude control. This article offers an analysis factors that play a leading role change centroid orbit, which are consuming fuel, circulation satellite-to-ground data transmission antenna rotation solar arrays. The parameters defined, formula computing proposed.