Topics in Modern Statistics · Statistics
image analysis, image enhancement, iris recognition, convolution, clustering, brain encoding decoding
I learned how to load, view, manipulate, and save imaged using CV2 in Python; preprocess and enhance images; analyze histograms, projections, contours, and other elements of an image to accomplish tasks like segmentation; implement object recognition; apply certain machine learning techniques to classify/match images and/or their features; use SPM fMRI analysis in MATLAB; relate human vision to computer vision, particularly in how images are understood by the computer as matrices with pixel value intensities; understand some theory behind signals, frequency, noise, etc.
image segmentation, signal processing, noise reduction, iris recognition, CNN, object detection
some image analysis techniques but they are very old
The course covers a lot of topics but does not go deep into any of them. Most of the learning happens when you work on the assignments and have to figure things out on your own. The practice questions in class are helpful in putting the concepts into practice.
STATGR5293 was most interesting course for me out of four courses I have taken in this semester. Professor was very knowledgeable and got oportunity to learn about image aalysis. I have learned advance knowledge on deep learning,Supervised and un supervised learning.
I am coming away from this course with demonstrable skills that I can offer to future employers.
I will recommend to my friends and new graduate student to pursue this course due to its practical orientation and applied machine learning. Professor Xiaofu He is great professor and his knowledge is beyond comparision. 1 Columbia University: Arts & Sciences Fall 2021 Course: STATGR5293_001_2021_3-TOPICSINMODERNSTATISTICS : STATGR5293_001_2021_3 - TOPICS IN MODERN STATISTICS Instructor: Xiaofu He
We covered the very basics of what images were, some machine learning techniques like k-means, and some image processing techniques like image segmentation using thresholding. I don't really feel like I learned much since it was mostly review.
Different computer vision models, learned a lot about yolo, iris detection, binary thresholding, face detection, haar, classifying image data over time with SVM and cross fold validation