Articles on kruzaeva's Blog

Week 11 Chek-in

kruzaeva@gmail.com (kruzaeva) — Thu, 22 Aug 2019 07:34:58 +0000

What did you do this week?

Documentation is written (Crystallinity map GUI, Clustering GUI, Crystallinity map + Clustering Jupyter Notebook)

Default frame view is reorganized (added new dropdown "ROI", code looks more consistent and nice)

What is coming up next?

Final evaluation

Did you get stuck anywhere?

Week 10 Chek-in

kruzaeva@gmail.com (kruzaeva) — Mon, 12 Aug 2019 09:13:24 +0000

What did you do this week?

New parameter (minimal distance between peaks) was added in GUI for region clustering and UDF for feature vector generation.

Writing of Jupiter notebook for non-zero order peaks detection and region clustering

What is coming up next?

I am planning to finish documentation writing this week.

Did you get stuck anywhere?

Not yet.

Week 9 Chek-in

kruzaeva@gmail.com (kruzaeva) — Mon, 05 Aug 2019 09:05:15 +0000

What did you do this week?

Documentation writing. The chapter for Crystallinity map calculation is written.

What is coming up next?

Chapter for Region clustering, Jupiter notebook for feature vector generation, non-zero order peaks detection and region clustering.

Did you get stuck anywhere?

Not yet.

Week 9 Chek-in

kruzaeva@gmail.com (kruzaeva) — Tue, 30 Jul 2019 08:50:18 +0000

What did you do this week?

Submitting of Clustering GUI and rectangular ROI, tests writing. All tests and cheks for Clustering GUI have passed.

What is coming up next?

Documentation writing.

Did you get stuck anywhere?

Not yet.

Week 8 Chek-in

kruzaeva@gmail.com (kruzaeva) — Mon, 22 Jul 2019 12:07:40 +0000

What did you do this week?

Restructuring Default Frame mode. Availiable modes: Average over disk, Average over rectangle, SD over disk, SD over rectangle.

Submitting of GUI for Clustering based on the positions of non-zero order diffraction peaks.

What is coming up next?

Fixing bugs and documentation writing.

Did you get stuck anywhere?

Not yet.

Week 7 Chek-in

kruzaeva@gmail.com (kruzaeva) — Tue, 16 Jul 2019 08:42:18 +0000

What did you do this week?

Creating a GUI for Standard Deviation (SD) of frames and SD of frames over ROI. ROI now can have rectangular shape.

Building of the first working prototype for Clustering based on the positions of non-zero order diffraction peaks.

What is coming up next?

Fixing bugs and submission of region clustering analysis, documentation writing.

Did you get stuck anywhere?

Not yet.

Week 6 Chek-in

kruzaeva@gmail.com (kruzaeva) — Mon, 08 Jul 2019 11:18:29 +0000

What did you do this week?

GUIs for Fluctuation EM (Standard deviation (SD) over ring) and Crystallinity map (Sum over ring in Fourier space with possibility to mask out zero-order diffraction peak as an option) generation were submitted.

FFT pick frame mode and FFT average mode were also submitted as a part of FFT analysis (Crystallinity map generation + FFT view modes)

What is coming up next?

Region clustering, based on non-zero diffraction peaks positions, SD frame mode integration, documentation writing.

Did you get stuck anywhere?

The interface of the GUI for region clustering is not clear yet, but this issue can be solved later, after first working prototype

Week 5 Chek-in

kruzaeva@gmail.com (kruzaeva) — Mon, 01 Jul 2019 12:07:05 +0000

What did you do this week?

Working on GUI for crystallinity map generation.

Developing of GUI for Fluctuation EM analysis (calculation of standard deviation over ring).

Small bug in blo format reader was fixed.

What is coming up next?

Finish integration of GUI for crystallinity map generation, documentation writing.

Did you get stuck anywhere?

Checkbox for disabling of zero-order peak deleting (lack of experience with React)

Week4 Check-In

kruzaeva@gmail.com (kruzaeva) — Wed, 26 Jun 2019 07:38:04 +0000

What did you do this week?

Submission of UDF for feature vector forming based on presence and positions of non-zero order diffraction peaks.

Building of the first prototype of GUI for crystallinity map generation in restructured LiberTEM environment for GUIs.

What is coming up next?

Writing documentation for UDFs for LiberTEM web site. Finish integration of GUI for crystallinity map generation. Fixing bugs in UDFs.

Did you get stuck anywhere?

Week3 Check-In

kruzaeva@gmail.com (kruzaeva) — Mon, 17 Jun 2019 08:47:12 +0000

What did you do this week?

I submitted UDF for crystallinity map generation, fixed some bugs in UDFs I submitted during last weeks.

I got familiar with JavaScript and React. First attempts to build the first prototype of GUI for crystallinity map generation.

What is coming up next?

Writing documentation for UDFs for LiberTEM web site. Finish integration of GUI for crystallinity map generation

Did you get stuck anywhere?

Current LiberTEM internal structure does not allow to apply 2 different masks (in real and Fourier space) in parallel in GUI.

So, once restructuring will be finished, I'll start the integration of GUI.

Week 2 Check-In

kruzaeva@gmail.com (kruzaeva) — Tue, 11 Jun 2019 08:45:15 +0000

What did you do this week?

During this week I developed, integrated and tested UDF for forming of a feature vector, based on the presence of non-zero order diffraction peaks.

Also I got familiar with LiberTEM GUI and had first attempts to integrate GUI for crystallinity map generation.

What is coming up next?

Finish integration of GUI for crystallinity map generation

Did you get stuck anywhere?

Applying 2 different masks (in real and Fourier space) in GUI

Introduction + bonding period

kruzaeva@gmail.com (kruzaeva) — Tue, 04 Jun 2019 15:56:16 +0000

My name is Ruzaeva Karina and I am a master student in RWTH Aachen, I study Electrical Engineering and sphere of my interest includes image/signal processing and data science.

During Google Summer of Code I am planning to integrate the algorithms for 4D STEM datasets processing in LiberTEM and make them accessible for material scientists without programming knowledge. I will focus on algorithms, which are applicable for phase-change materials (PCM), because they are some of the most promising materials for data-storage applications.

During bonding period I got familiar with GUI basics and made some literature and libraries overview.

Week 1 Check-In

kruzaeva@gmail.com (kruzaeva) — Mon, 03 Jun 2019 09:05:00 +0000

What did I do this week?

During my firs week, I developed user-defined function (UDF) for "crystallinity mapping" in LiberTEM for phase change materials and wrote some tests. This UDF helps to easily distinguish the amorphous and crystalline regions of the specimen.

The data which can be processed with the UDF is 4D-STEM datasets (2D scan domain and 2D detector domain). The algorithm I used is based on Integration over a ring of frames in Fourier space after deleting of outliers and frames preprocessing.

What is coming up next?

Implementing of a UDF for feature vector forming based on blob (non-zero order diffraction peaks) finding, this feature vector can be used for region clustering

Did you get stuck anywhere?

No specific problems appeared, just lack of experience with Github could make me a little slower.