Research Experience on Image Analysis

Image Analysis Experience

This section briefly describes research experience of the Image Analysis team involved in the project. The team is lead by Dr. Vassili Kovalev, Head of Biomedical Image Analysis Department of the United Institute of Informatics, Belarus National Academy of Sciences. He is also playing the role of Principal Investigator of the TB project which Portal you are currently in.

Dr. Kovalev has been working in the field of Image Processing, Image Analysis, and Pattern Recognition since the late 80th. There were a number of research projects accomplished during these 20+ years with rather broad range of corresponding R&D topics. However, the main contribution to the field is perhaps the general idea of multi-sort, multi-dimensional co-occurrence matrices describing the image content. The suggested approach is very flexible and is proven to be very suitable for a range of image characterization, image analysis, image classification, and image retrieval tasks. On the top, it was found to be equally-well suited for 2D and 3D shapes (plain contours and volumetric meshes), 2D and 3D textures as well as for color, multi-spectral and other kinds of biomedical images.

Interested colleagues professionally working in the field of biomedical image analysis as well as occasional Portal visitors are invited to look for more details given below.

Generalized multisort co-occurrence matrices

Although the general notion of multi-sort co-occurrence matrices was introduced in early 90th and first published in English in 1996 , it is worth to start with the following paper which introduces them in the context of 3D MRI textures (but not in rather outdated 2D shape problem, more computer-vision 3D meshes, nor relatively specific light microscope histology/cytology color image domain):

Kovalev V.A., Kruggel F., Gertz H.-J., and von Cramon D.Y. Three-dimensional Texture Analysis of MRI Brain Datasets, IEEE Transactions on Medical Imaging, Vol. 20, No. 5, May, pp. 424-433, 2001.

article_icon A New Method for Quantification of Age-Related Brain Changes