A5025487107- Advanced Image Processing Techniques
- Advanced Image Fusion Techniques
- Geochemistry and Geologic Mapping
- Image and Signal Denoising Methods
- Image Processing Techniques and Applications
- Geophysics and Gravity Measurements
- Remote-Sensing Image Classification
- Satellite Image Processing and Photogrammetry
- Space Satellite Systems and Control
- Inertial Sensor and Navigation
- Advanced Vision and Imaging
Ecole Mohammadia d'Ingénieurs
2022-2023
Mohammed V University
2022-2023
In this paper we present a study of deep learning models for single image super-resolution (SISR), through the some latest methods used in neural networks super-resolution, exploring many and proposed. Moreover, presents global complete technical benchmark state-of-the-art machine algorithms based on reference metrics (PSNR SSIM) field visualization perception. This involved 53 different tested 7 datasets established as vision domain (Set5, Set14, BSD100, Urban100, DIV2K, Manga109, DIV8K)....
The super-resolution of images has seen remarkable progress, especially with the use deep learning models.This technique allows having a better-quality image from one or more low-resolution versions.Super-resolution, therefore, aims at enriching lowresolution additional pixel density and high-frequency detail.This paper presents comprehensive empirical study based on systematic review learningbased models for single (SISR), exploring set techniques offered by technology used SISR.In this...
Open access in space remote sensing has allowed easy to satellite imagery; however, high-resolution imagery is not given everyone, but only those who master technology.Thus, this paper presents a new approach for improving the quality of Sentinel-2 images by super-resolution exploiting deep learning techniques.In context, work proposes generic solution that improves spatial resolution from 10m 2.5m (scaling factor 4) taking into account constraints volumetry and dependence between spectral...
Open access in space remote sensing has allowed easy to satellite imagery; however, high-resolution imagery is not given everyone, but only those who master technology. Thus, this paper presents a new approach for improving the quality of Sentinel-2 images by super-resolution exploiting deep learning techniques. In context, work proposes generic solution that improves spatial resolution from 10m 2.5m (scaling factor 4) taking into account constraints volumetry and dependence between spectral...
The use of machine learning models, particularly deep for the analysis remote sensing products, especially multispectral satellite images, has recently experienced exponential development. Therefore, this article will present a protocol processing images by through latest methods used in neural networks computer vision, exploring all and proposed. In study, we main adapted to form an efficient protocol. Our methodology proceeds with systematic concepts testing applicability contribution...