I have a question. I have 2 forms of data: one is in .mat format with shape (rows, cols, bands) and the other is AVIRIS data I have downloaded from nasa.gov, both data have no labels and i want to perform anomaly detection on these these images. How do i do so?

Jingtao-Li-CVer / TDD

This is an official implementation for "One-Step Detection Paradigm for Hyperspectral Anomaly Detection via Spectral Deviation Relationship Learning“

74 stars 3 forks source link

I have a question. I have 2 forms of data: one is in .mat format with shape (rows, cols, bands) and the other is AVIRIS data I have downloaded from nasa.gov, both data have no labels and i want to perform anomaly detection on these these images. How do i do so? #7

Closed Potafe closed 4 months ago

Potafe commented 4 months ago

Screenshot 2024-06-25 190954

The above is the folder for my AVIRIS data downloaded open-source. I don't have any prior info on this data and want to perform your TDD HAD on this? How to do so?

Jingtao-Li-CVer commented 4 months ago

Thanks for your interest.

We have added the function to infer without labels as in updated ReadMe.

We use gdal to read the data, and you need to support matlab files yourself in HSIDataset_Inferring_Only.py using scipy package.

Potafe commented 4 months ago

Thank you for your quick response 👍🏾 :)