37 datasets found

The ISPRS 2D semantic labeling benchmark (Vaihingen) dataset is used for evaluating the proposed method.

The LoveDA dataset consists of 5,987 high-quality optical remote sensing images with a resolution of 0.3 meters per pixel.

The ISPRS Vaihingen dataset contains 33 high-quality images with topographical information, each with an average resolution of 2494×2064 pixels.

Four widely-used remote sensing datasets are considered for evaluating the efficacy of the proposed approach. Sample images of these datasets are provided in Fig. 5.

PlanetScope data

Sentinel-2 data

The Zurich Summer dataset is a benchmark for semantic segmentation in remote sensing. It contains aerial images with 8 urban classes.

The UCM dataset is a benchmark for scene classification in remote sensing. It contains aerial images with 21 land-use classes.

The UAE-RS dataset is a benchmark for black-box adversarial attacks in remote sensing. It contains adversarial examples generated using the Mixup-Attack and Mixcut-Attack methods.

Bag-of-visual-words and spatial extensions for land-use classification

Deep semantic understanding of high resolution remote sensing image

Exploring models and data for remote sensing image caption generation

Remote Sensing Image Captioning Dataset (RSICD) and UCM-captions dataset for remote sensing image captioning

The AID dataset is a benchmark for scene classification in remote sensing. It contains aerial images with 30 scene types.

The Vaihingen dataset consists of 1440 scenes with a size of 250×250 pixels. Each scene is a colour-infrared (CIR) true orthophoto and a height grid (digital surface model; DSM)...

Sentinel 2 multispectral data was acquired using Sentinel 2 satellite over Central State farm in Hissar, Haryana (India).

DAIS hyperspectral data used in this study was acquired using DAIS 7915 instrument through an airborne campaign on 29th June 2000 covering La Mancha Alta region south of Madrid,...

AVIRIS-NG hyperspectral data used in this study was acquired on 8 February, 2016 through an aerial campaign over Anand district in Gujarat, Ahmedabad (India).

Four remote sensing datasets were used for classification with both small and large training samples.

Self-supervised remote sensing images change detection at pixel-level.