A Software Platform for Evaluation of Neural Algorithms for Stereo Image Processing
Bogusław Cyganek, Jan Borgosz · 2003
The paper presents a software platform for implementation and evaluation of different neural algorithms for stereo image processing. The described program constitutes a self-contained image evaluation and testing module but also it can be used as a library that can be a part of a bigger image processing software. Thanks to its flexible and object oriented design it can be easily extended to comprise any new neural method. Presented techniques and solutions can also guide other researchers facing similar problems in image processing. These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.