The Active Segmentation platform for ImageJ (ASP/IJ) was developed in the scope of GSOC 2016 - 2021. The plugin provides a general-purpose environment that allows biologists and other domain experts to use transparently state-of-the-art techniques in machine learning to achieve excellent image segmentation and classification. ImageJ is a public-domain Java image processing program extensively used in life and material sciences. The program was designed with an open architecture that provides extensibility via plugins. Weka (Waikato Environment for Knowledge Analysis) is a collection of machine learning algorithms for data mining tasks. Weka contains tools for data pre-processing, classification, regression, clustering, association rules, and visualization. It is also well-suited for developing new machine-learning schemes. The algorithms can be applied directly to a dataset or called from your Java.
The project idea: The Weka library offers some advanced visualization and analysis functionality. The student will develop a new visualization and reporting panel within ASP/IJ, exposing the Weka visualization and analysis functions.
● Fix existing issues and bugs
● SQL database design
● GUI implementation and integration
Minimal set of deliverables
● Requirement specification - Prepared by the candidate after understanding the functionality.
● System Design - Detailed plan for development of the plugin and test cases.
● Implementation and testing - Details of implementation and testing of the platform.
Skill level: Intermediate
Required skills: Java, Machine learning.
Time commitment: Half-time (175 h)
Lead mentor: Dimiter Prodanov (email@example.com), INCF Belgian Node
- ImageJ: https://imagej.nih.gov/
- Weka https://www.cs.waikato.ac.nz/ml/weka/
- Active Segmentation : GitHub - sumit3203/ACTIVESEGMENTATION: Active Segmentation Project
Backup mentors: Sumit Vohra, ZIB, Berlin, Germany, Purva Chaudhari, Vishwakarma Institute Technology (backups)
Tech keywords: TBD
Hello everyone, my name is Pakhi Banchalia. I find this problem statement quite interesting and I would like to contribute to this project as I have some experience with Java and some knowledge of Machine Learning. I have also worked on data visualization previously. Can anyone guide me on how to begin with this project? @arnab1896
Hi @Pakhi_Banchalia , nice to hear from you
As mentioned in the project idea itself, there are quite a few links and resources shared (like the project websites and other github link). Please go through them and try to come up with an idea of how “you” will implement the project. In the meantime, please give @dprodanov and the other mentors (@sumit.3203 @Purva-Chaudhari ) some time to reply.
Happy to help in case of more queries.
Also, please remember, that the more pointed and specific queries you come up with, the better mentors will be able to give you feedback. So, please go ahead and try installing the modules/exploring them in the backend repository.
Hello @arnab1896 @dprodanov @sumit.3203 @Purva-Chaudhari . I successfully installed ImageJ and set-up ActiveSegmentation in it. I went through the different features in ActiveSegmentation to get an idea about them. I was not able to understand some of them and I could not find any documentation for the same. Would it be possible to have a meeting so that I can get a clear picture about ActiveSegmentation?
Great news! You can try the segmentation functionality. We have a manually segmented dataset that you can try out.
regarding the documentation please check first the tutorial on YouTube.
Thank you @dprodanov, I will go through it and get back to you.
I also wanted to ask if there is any communication channel which I can join.
We have a Slack channel but we use it mainly when GSOC projects start.
I went through the visualization tools offered by weka library which can be implemented along with active segmentation. I had a query whether the project requires visualizing the segmented image for a better representation or visualizing the features of the segmented image for data analysis with the help of various tools.
Thanks and regards
It is better to focus on the feature visualization. The segmentation output is already visualized by an overlay layer on top of the raw image.
After segmenting the ISIBI 2012 dataset image, I tried visualizing the features of the segmented image using Weka and this is the output I achieved.
@dprodanov I wanted to know whether my approach is correct and if this is the visualization that needs to be implemented within Active Segmentation.
It is not clear what you are showing. You’d better visualize the feature space for a given pixel of the image. It could be useful to do it around edges and corners.
@dprodanov These are some of the visualization techniques I found in Weka that are not currently present in Active Segmentation and could potentially be added:
If there are any more techniques that could be implemented, please suggest.
They all look interesting.
I like the last one best because it gives some idea about the feature separation.
@dprodanov could you please review the proposal draft that I have mailed to you and provide any suggestions.