The 3rd International Workshop on
Physics Based Vision meets Deep Learning (PBDL)

Access the Challenge Dataset

Challenge Dataset can be downloaded from here:   Training and Testing Set

The Dataset is for this chanllenge and research purpose only. Commercial use is not permitted.

Content of The Downloaded Package

In the dataset, you may find a folder for training data and a folder for testing data.

Train
In the training dataset folder, you will find all images, each of which is named with an index. In the folder, you will find five files for each image, which are:

< ∗.hsd >,   < rgb ∗ .jpg >,   < rgb ∗ _gray.png >,   < rgb ∗ .png >,

They are: 1. the hyperspectral data file, 2. The RGB visualization of the hyperspectral image, 3. Semantic labelling coded from 0 to 18 as per Table 1, 4. Visualization of the semantic labelling by overlay with the RGB image. The ∗ is the filename which is the same as the folder name.

Test
In the testing dataset, you will find 200 groups of images. The same as training dataset, each of the group is named with a file index. In each of the group, you will find two files, which are:

< ∗.hsd >,   < rgb ∗ .jpg > .

They are: 1. the hyperspectral data file, 2. The RGB visualization of the hyperspectral image. The ∗ is the filename which is the same as the folder name.

Table 1. A summary of dataset label


Read Hyperspectral Cubic Dataset
In the dataset package, you will find a MATLAB file named readHSD.m, which is the MATLAB code to read the hyperspectral data. The data format is similar as RGB images, the different is the file has 128 channels rather than 3 channels.