Simulated InSAR building dataset for CVCMFF Net

5
1 rating - Please login to submit your rating.

Abstract 

The simulated InSAR building dataset contains 312 simulated SAR image pairs generated from 39 different building models. Each building model is simulated at 8 viewing-angles. The sample number is 216 of the train set and is 96 of the test set. Each simulated InSAR sample contains three channels: master SAR image, slave SAR image, and interferometric phase image. This dataset serves the CVCMFF Net for building semantic segmentation of InSAR images.

Instructions: 

1. The simulated InSAR building dataset contains 312 simulated SAR image pairs generated from 39 different building models. Each building model is simulated at 8 viewing-angles. The sample number is 216 of the train set and is 96 of the test set. Each simulated InSAR sample contains three channels: master SAR image, slave SAR image, and interferometric phase image.

2. In the data set, the complex-valued SAR images are divided into real channel images and imaginary channel images, which are stored separately. Then, they will be recombined into complex-valued in the CVCMFF Net.

3. Value of all the image pixel is normalized to 0 ~ 255.

4. For one sample, image names of different channels are the same and consistent with the naming of the ground truth label.

5. Configuration of the ground truth label is: 0 = shadow; 1 = ground or single scattering; 2 = layover or secondary scattering.