From October 24 to November 3 at NSU, with the support of the Mathematical Center in Akademgorodok, a course of lectures "Geometric methods and machine learning methods in modern problems of image processing" will be held.
Modern problems of image processing and methods for their solution based on geometric approaches and machine learning algorithms will be considered. A number of examples of the use of graphic libraries for solving practical problems in computer vision systems are given.
Vladimir A. Klyachin has been working at Volgograd State University since 1992. In 1995 he defended his Ph.D. In 2002, Vladimir was appointed to the position of the head of the Department of Mathematical Analysis and Theory of Functions, after which, from 2003 to the present, he has been the head of the Department of Computer Science and Experimental Mathematics. During his research and teaching activities, Dr. Klyachin developed and delivered many courses.
Course program and schedule:
- October 24, 4:20 pm, room 5273 NSU 鈥� Mathematical models of images. Image processing tasks. Graphic libraries, usage examples.
- October 25, 4:20 pm, room 5251 NSU 鈥� Geometric transformations of images. Threshold clipping, image binarization
- October 26, 4:20 pm, room 5273 NSU 鈥� Image smoothing. Images as functions. Image histograms.
- October 27, 4:20 pm, room 4259 NSU 鈥� Kani's edge detection algorithm. Search for contours. Examples of use in the problem of character segmentation.
- October 28, 4:20 pm, room 5251 NSU 鈥� Search for geometric objects in the image. Hough transformation.
- October 31, 4:20 pm, room 5273 NSU 鈥� The task of detecting singular points in images. Harris corner points.
- November 1, 4:20 pm, room 5251 NSU 鈥� Basic concepts of machine learning. Libraries that implement machine learning algorithms. Probabilistic approach to representation of images. Variational autoencoder.
- November 2, 4:20 pm, room 5273 NSU 鈥� Examples of classification problems in image processing. Identification of objects in the image. Convolutional neural networks.
- November 3, 4:20 pm, room 4259 NSU 鈥� Spatial image analysis. Tasks of 3D reconstruction.
