AIIA_flood Dataset

The dataset for the flood binary segmentation task comprises 720 images consolidated from two sources: D.MALIAN_1 and BRK_1 trials. It is structured into distinct directories for each source (dmalian_1 and brk_1), each containing standard train and val splits with separate folders for images (.jpg) and labels (.png). The annotation masks are binary, where pixels are labeled as 0 for background and 1 for floodwater. The total split consists of 428 training images (198 from D.MALIAN_1 and 230 from BRK_1) and 292 validation images (140 from D.MALIAN_1 and 152 from BRK_1), resulting in an approximate 60% - 40% training-validation distribution.

Mora information about the dataset can be found here.