Downloads

Vessel segmentation

Introduction

For several reasons, the segmentation of the vascular system is one of the most researched problems of retinal image analysis: on the one hand, it can be used for the detection of various other anatomical parts; on the other hand, the geometry of the vessels can be used to infer on various diseases, like hypertension. We have developed a novel technique for the accurate segmentation of the vasculature. The method is described in the paper (1).

Example

To have an impression on what the software does, see the following input and output images.

Input Output

Files

The implementation of the method described in (1) is available in binary format compiled for Ubuntu 17.04 AMD64. The usage of the software is included in the README files:

References

  1. Kovács, G., & Hajdu, A. (2016). A Self-Calibrating Approach for the Segmentation of Retinal Vessels by Template Matching and Contour Reconstruction. Medical Image Analysis, 29(4), 24–46, (IF=4.565). https://doi.org/10.1016/j.media.2015.12.003

Sample codes for books on parallel and OpenCL programming

Sample codes for the books (1), (2), (3) are available below:

References

  1. Kovács, G. (2013). Párhuzamos programozási eszközök és összetett alkalmazásaik [Parallel programming and its complex applications] (p. 321). Typotex.
  2. Kovács, G. (2013). OpenCL [Hungarian] (p. 361). Typotex.
  3. Kovács, G. (2013). OpenCL [English] (p. 355). Typotex.

openip - The open source image processing library

Introduction

The openip open source image processing library is a lightweight image processin library for academic and industrial applications. Detailed descriptions can be found in the paper (1).

Files

References

  1. Kovács, G., Iván, J. I., Pányik, A., & Fazekas, A. (2010). The OpenIP open source image processing library. In Proc. of ACM Multimedia 2010 (pp. 1489–1492). October 25–29, Florence, Italy.