Skip to content

IceStreams/GID-MTL

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

7 Commits
 
 
 
 
 
 

Repository files navigation

Pixel Scene Classification Dataset (PSC)

PSC5 (GID-MTL5) and PSC15 (GID-MTL15) datasets

These datasets are used for multi-task learning of simultanelously conduct scene classificton and pixel-level land-cover classification. Every image in each dataset contains pixel-level land-cover labels and image-level scene label.

The image is cropped from GID datasets [1] and set to a size of 224×224 pixels. The categories of the scene are the same as the categories of pixel-level annotations, except that pixel-level annotations further contain an undefined category.

PSC5 contains five land-cover categories. PSC15 contains 15n land-cover categories. The contained categories, the number of images per category, and some sample images of PSC5 are shown in Fig.1. The same information for PSC15 is shown in Fig. 2.


fig1 Fig.1 Sample images of PSC5. Each group displays an image corresponding to a scene category, and the corresponding segmentation map. The annotations below list the category, the number of scene images contained in the category, and the color of the pixel-level label.


fig2 Fig.2 Sample images of PSC15.


[1] Tong X, Xia G, Lu Q, Shen H, Li S, You S, Zhang L (2020) Land-cover classification with high-resolution remote sensing images using transferable deep models. Remote Sensing of Environment 237:111322.

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published