Tech Talk
summary
Describes building image pipelines that turn terabytes per minute of experimental images into answers: microCT segmentation of ceramic matrix composites and carbon textiles using statistical region merging and morphology to separate matrix, fibers and coatings; a scattering-pattern search over ~4 million GIXS images using piCyber and deep feature vectors from Inception-ResNet v2; an automated cervical cytology system born from a 2,000‑image challenge that uses bilateral enhancement, distance-based nucleus assignment, Python/Azure scaling and moves from counts to abnormality staging.
AI-written summary · about
chapters (11)
- 0:00Intro
- 1:50Inspection of 1TB/min?
- 4:31Image Analysis Challenges
- 6:37Machine Learning and Metrology for Materials Analyze microstructure of materials to advance manufacturing
- 8:48Abstract patterns
- 12:14Women & Data Science image Analysis for Healthcare Improvement
- 16:27Cervical Cancer
- 16:43Where is the cervix?
- 18:38IEEE Int. Symposium on Biomedical Imaging 2014 Overlapping Cervical Cytology Image Segmentation Challenge
- 19:57Super Pixel Voronoi Diagram for Cells
- 24:14Cultivating next generation of scientists Research and development through collaboration with and support to the next generation of scientists and engineers around the world
related talks
19mMy Journey In Tech
43mTopological data analysis: How to quantify "holes" in your data and why?
25mEverything but maps with spatial tools
37mUsing Python to visualize and analyze ion beam experiment data at a particle accelerator facility
46mData Discussion - AI for Good
17mGeoAI: a feature extraction tool set using Tensor Flow and Keras...
01h 07mPhysics at the LHC: Data Analysis and Global Network Challenges with Exascale Data
28mSome Current Computational Challenges in Bioinformatics