I took part at the projects of Computer Vision Group, Brno University of Technology. These projects were realized as semestral project in computer vision subject, so nothing very valueable, but topics of these projects are really interesting, so I put them on this page ...
You can see my another computer vision projects:
Leucocyte Segmentation and Classification
Self-Locating by Pattern Tracking
The aim of this project was to identify different types of crops, outline them and count the number of them. Particular objects could be in touch and also overlapped, which made segmentation more difficult. Final algorithm uses several methods like thresholding, adaptive thresholding, watershed, template matching, etc. Because analysis are performed on the high resolution image, time domain is an important point of view. Because of it, the methods are consequently applied to image from simple and fast ones to the complex and slow ones. Finally, the time spent to analyse an image varies from 0.1 s to 10 s (in Matlab) according to the complexity of image.
Here you can find some papers about solving out this interesting task. Documentation is available only in Czech language.
size: 452.74kB, last modified: 19.08.2014