We are developing AI/CI-based enhanced heuristic and metaheuristic optimization, computer vision algorithms, which enable us to achieve good quality solutions within reasonable computation time. We pay attention to seven fundamental, theoretical aspects (factors) of heuristic algorithms: intensification, diversification, hybridization, self-similarity (and hierarchy), learning, (self-)adaptivity (self-tuning), and inspirational sources from nature (real physical world). Note that most of the modern heuristic algorithms are high-level universal guiding strategies (metaheuristics) oriented at different types of practical problems, rather than pure heuristics tailored for particular problems. Furthermore, we aim to apply the created algorithms to the actual practical areas such as multimedia technologies and virtual/augmented reality as well as in other related modern technological areas. The developed algorithms would be useful as prototypes of specialized computer-aided design software for both the scientists and practitioners working in different areas (like education, media/advertising, fashion, healthcare, entertainment, tourism, retail, architecture, etc.). Computer vision systems are used also in many fields, including manufacturing, medical, traffic monitoring, security systems, etc. Computer vision methods allow automatic analysis of important information in the processing of large data flows. Artificial intelligence algorithms in this case are an integral part of solving computer vision problems.
Research topics:
- Design and implementation of heuristic and metaheuristic optimization algorithms
- Hybridized, evolutionary-population-based heuristic algorithms
- Hierarchicity-based (self-similar) heuristic algorithms
- Nature-inspired heuristic algorithms
- Machine-learning based algorithms
- Digital colors and color halftones
- Digital image processing
- Computer vision and Image classification
- Using metaheuristic techniques for solving optimization problems arising in manufacturing
- Graph-based clustering algorithms
