The output is so strange!!

[Puzzled-Hui]

Can you upload the result: mask.tiff?

Because I have spent lots of time running your code and got a strange output mask_dmt, exists many spikes.

I have put the obtained result on the google drive, the link is as follows:
https://drive.google.com/file/d/1HBN6zekfBHKSeNSPf7UflBcT7ULLjUTy/view?usp=sharing

Could you please check it and tell me whether this is correct? If you can provide more samples on 3D data (ISBI13) as you claimed in the paper, I would be very grateful.

[HuXiaoling]

Sorry for the confusion. For 3D volume data, you'll need to use the persistence_ws version to compute the approximate morse structures. In this repo, the output of persistence_ws is saved as cc.mat (connected component matrix). Note that the connected component matrix and binary boundary matrix are convertible.

[Puzzled-Hui]

Sorry for the confusion. For 3D volume data, you'll need to use the persistence_ws version to compute the approximate morse structures. In this repo, the output of persistence_ws is saved as cc.mat (connected component matrix). Note that the connected component matrix and binary boundary matrix are convertible.

Thanks for your reply! However, I have ran the 2D scripts and got a relative strange output, I havenot yet conducted the experiments on the 3D data? Could you kindly provide some concrete 2D and 3D examples for me to have a try?