Attributes | Values |
---|
rdf:type
| |
Description
| - The correspondence problem in the field of stereo vision has been attracting a lot of efforts amongst computer scientists recently. Although the epipolar geometry simplified the search for corresponding pixels in stereo images considerably, many challenging difficulties are still left. In this paper it is proposed the novel image segmentation technique designated for consequent exploitation by segment-based stereo matching algorithm. Segment-based methods arise from assumption that the scene can be divided into segments having homogenous color where each segment from one stereo image coincide with correspondent segment in the other image. In comparison with correlation-based and feature-based methods it tends to have better performance in textureless regions, while dense disparity map is being produced. The new approach is based on the use of standard inter-pixel Euclidean distance utilization, which is enhanced by hue similarity and minimal size of segments criteria. All three conditions can be applied to both 4-connected and 8-connected pixel neighborhood, nevertheless only latter mentioned is used in this article. Segmentation results are shown in some standard stereo image sets, where the accuracy and robustness of proposed algorithm is presented. Finally, the discussion about some significant pros and cons is led and the direction of consequent research is outlined.
- The correspondence problem in the field of stereo vision has been attracting a lot of efforts amongst computer scientists recently. Although the epipolar geometry simplified the search for corresponding pixels in stereo images considerably, many challenging difficulties are still left. In this paper it is proposed the novel image segmentation technique designated for consequent exploitation by segment-based stereo matching algorithm. Segment-based methods arise from assumption that the scene can be divided into segments having homogenous color where each segment from one stereo image coincide with correspondent segment in the other image. In comparison with correlation-based and feature-based methods it tends to have better performance in textureless regions, while dense disparity map is being produced. The new approach is based on the use of standard inter-pixel Euclidean distance utilization, which is enhanced by hue similarity and minimal size of segments criteria. All three conditions can be applied to both 4-connected and 8-connected pixel neighborhood, nevertheless only latter mentioned is used in this article. Segmentation results are shown in some standard stereo image sets, where the accuracy and robustness of proposed algorithm is presented. Finally, the discussion about some significant pros and cons is led and the direction of consequent research is outlined. (en)
|
Title
| - Color Segmentation for Segment-based Stereo Matching
- Color Segmentation for Segment-based Stereo Matching (en)
|
skos:prefLabel
| - Color Segmentation for Segment-based Stereo Matching
- Color Segmentation for Segment-based Stereo Matching (en)
|
skos:notation
| - RIV/70883521:28140/12:43868422!RIV13-MSM-28140___
|
http://linked.open...avai/riv/aktivita
| |
http://linked.open...avai/riv/aktivity
| |
http://linked.open...vai/riv/dodaniDat
| |
http://linked.open...aciTvurceVysledku
| |
http://linked.open.../riv/druhVysledku
| |
http://linked.open...iv/duvernostUdaju
| |
http://linked.open...titaPredkladatele
| |
http://linked.open...dnocenehoVysledku
| |
http://linked.open...ai/riv/idVysledku
| - RIV/70883521:28140/12:43868422
|
http://linked.open...riv/jazykVysledku
| |
http://linked.open.../riv/klicovaSlova
| - Computer Vision, Segment-based Stereo Matching, Color Segmentation (en)
|
http://linked.open.../riv/klicoveSlovo
| |
http://linked.open...ontrolniKodProRIV
| |
http://linked.open...v/mistoKonaniAkce
| |
http://linked.open...i/riv/mistoVydani
| |
http://linked.open...i/riv/nazevZdroje
| - Proceedings of International Conference on Innovative Technologies IN-TECH
|
http://linked.open...in/vavai/riv/obor
| |
http://linked.open...ichTvurcuVysledku
| |
http://linked.open...cetTvurcuVysledku
| |
http://linked.open...vavai/riv/projekt
| |
http://linked.open...UplatneniVysledku
| |
http://linked.open...iv/tvurceVysledku
| - Prokop, Roman
- Beneda, Martin
|
http://linked.open...vavai/riv/typAkce
| |
http://linked.open.../riv/zahajeniAkce
| |
number of pages
| |
http://purl.org/ne...btex#hasPublisher
| - Faculty of Engineering University of Rijeka
|
https://schema.org/isbn
| |
http://localhost/t...ganizacniJednotka
| |