Labeled dataset for training despeckling filters for SAR imagery
When training Artificial Intelligence and Deep Learning models, especially by using Supervised Learning techniques, a labeled dataset is required to have an input with data and its corresponding labeled output data. In the case of images, for classification, segmentation, or other processing tasks,...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Elsevier
2024-04-01
|
| Series: | Data in Brief |
| Subjects: | |
| Online Access: | http://www.sciencedirect.com/science/article/pii/S2352340924000398 |
| _version_ | 1797256377269223424 |
|---|---|
| author | Rubén Darío Vásquez-Salazar Ahmed Alejandro Cardona-Mesa Luis Gómez Carlos M. Travieso-González Andrés F. Garavito-González Esteban Vásquez-Cano |
| author_facet | Rubén Darío Vásquez-Salazar Ahmed Alejandro Cardona-Mesa Luis Gómez Carlos M. Travieso-González Andrés F. Garavito-González Esteban Vásquez-Cano |
| author_sort | Rubén Darío Vásquez-Salazar |
| collection | DOAJ |
| description | When training Artificial Intelligence and Deep Learning models, especially by using Supervised Learning techniques, a labeled dataset is required to have an input with data and its corresponding labeled output data. In the case of images, for classification, segmentation, or other processing tasks, a pair of images is required in the same sense, one image as an input (the noisy image) and the desired (the denoised image) one as an output. For SAR despeckling applications, the common approach is to have a set of optical images that then are corrupted with synthetic noise, since there is no ground truth available. The corrupted image is considered the input and the optical one is the noiseless one (ground truth). In this paper, we provide a dataset based on actual SAR images. The ground truth was obtained from SAR images of Sentinel 1 of the same region in different instants of time and then they were processed and merged into one single image that serves as the output of the dataset. Every SAR image (noisy and ground truth) was split into 1600 images of 512 × 512 pixels, so a total of 3200 images were obtained. The dataset was also split into 3000 for training and 200 for validation, all of them available in four labeled folders. |
| first_indexed | 2024-03-08T11:26:15Z |
| format | Article |
| id | doaj.art-e2a066663b234cba80c4a232087ff3bd |
| institution | Directory Open Access Journal |
| issn | 2352-3409 |
| language | English |
| last_indexed | 2024-04-24T22:20:46Z |
| publishDate | 2024-04-01 |
| publisher | Elsevier |
| record_format | Article |
| series | Data in Brief |
| spelling | doaj.art-e2a066663b234cba80c4a232087ff3bd2024-03-20T06:09:35ZengElsevierData in Brief2352-34092024-04-0153110065Labeled dataset for training despeckling filters for SAR imageryRubén Darío Vásquez-Salazar0Ahmed Alejandro Cardona-Mesa1Luis Gómez2Carlos M. Travieso-González3Andrés F. Garavito-González4Esteban Vásquez-Cano5Faculty of Engineering, Politécnico Colombiano Jaime Isaza Cadavid, Medellín, 48th Av, 7-151, Colombia; Corresponding author.Faculty of Engineering, Institución Universitaria Digital de Antioquia, Medellín, 55th Av, 42-90, ColombiaElectronic Engineering and Automatic Department, IUCES, Universidad de Las Palmas de Gran Canaria, Las Palmas de Gran Canaria, SpainSignals and Communications Department, IDeTIC, Universidad de Las Palmas de Gran Canaria, SpainFaculty of Engineering, Politécnico Colombiano Jaime Isaza Cadavid, Medellín, 48th Av, 7-151, ColombiaFaculty of Engineering, Politécnico Colombiano Jaime Isaza Cadavid, Medellín, 48th Av, 7-151, ColombiaWhen training Artificial Intelligence and Deep Learning models, especially by using Supervised Learning techniques, a labeled dataset is required to have an input with data and its corresponding labeled output data. In the case of images, for classification, segmentation, or other processing tasks, a pair of images is required in the same sense, one image as an input (the noisy image) and the desired (the denoised image) one as an output. For SAR despeckling applications, the common approach is to have a set of optical images that then are corrupted with synthetic noise, since there is no ground truth available. The corrupted image is considered the input and the optical one is the noiseless one (ground truth). In this paper, we provide a dataset based on actual SAR images. The ground truth was obtained from SAR images of Sentinel 1 of the same region in different instants of time and then they were processed and merged into one single image that serves as the output of the dataset. Every SAR image (noisy and ground truth) was split into 1600 images of 512 × 512 pixels, so a total of 3200 images were obtained. The dataset was also split into 3000 for training and 200 for validation, all of them available in four labeled folders.http://www.sciencedirect.com/science/article/pii/S2352340924000398SpeckleSynthetic Aperture Radar (SAR)Image denoisingSupervised learningLabeled dataset |
| spellingShingle | Rubén Darío Vásquez-Salazar Ahmed Alejandro Cardona-Mesa Luis Gómez Carlos M. Travieso-González Andrés F. Garavito-González Esteban Vásquez-Cano Labeled dataset for training despeckling filters for SAR imagery Data in Brief Speckle Synthetic Aperture Radar (SAR) Image denoising Supervised learning Labeled dataset |
| title | Labeled dataset for training despeckling filters for SAR imagery |
| title_full | Labeled dataset for training despeckling filters for SAR imagery |
| title_fullStr | Labeled dataset for training despeckling filters for SAR imagery |
| title_full_unstemmed | Labeled dataset for training despeckling filters for SAR imagery |
| title_short | Labeled dataset for training despeckling filters for SAR imagery |
| title_sort | labeled dataset for training despeckling filters for sar imagery |
| topic | Speckle Synthetic Aperture Radar (SAR) Image denoising Supervised learning Labeled dataset |
| url | http://www.sciencedirect.com/science/article/pii/S2352340924000398 |
| work_keys_str_mv | AT rubendariovasquezsalazar labeleddatasetfortrainingdespecklingfiltersforsarimagery AT ahmedalejandrocardonamesa labeleddatasetfortrainingdespecklingfiltersforsarimagery AT luisgomez labeleddatasetfortrainingdespecklingfiltersforsarimagery AT carlosmtraviesogonzalez labeleddatasetfortrainingdespecklingfiltersforsarimagery AT andresfgaravitogonzalez labeleddatasetfortrainingdespecklingfiltersforsarimagery AT estebanvasquezcano labeleddatasetfortrainingdespecklingfiltersforsarimagery |