Abstract
Matching model point patterns to observed point patterns is of important concern in machine vision. Conventional search algorithms not only fail to arrive at the optimal match, but are computationally expensive, time consuming, and search the solution space sequentially. This paper presents a fast, inexpensive, algorithmically and operationally parallel Evolutionary Program (EP) for optimal point pattern matching based on a stochastic and heuristic optimization framework. Novel, knowledge-based, genetic operators are defined and are dynamically controlled to achieve 'fast fine tuning' and an optimal global search by efficiently combining the elements of 'gradient descent' and 'random search.' The developed EP algorithm outperforms existing techniques and is robust as it achieves a fast, optimal pattern match even in the presence of high noise and incomplete data sets, with insignificant degradation.
| Original language | English (US) |
|---|---|
| Pages | 1777-1782 |
| Number of pages | 6 |
| State | Published - 1994 |
| Event | Proceedings of the 1994 IEEE International Conference on Neural Networks. Part 1 (of 7) - Orlando, FL, USA Duration: Jun 27 1994 → Jun 29 1994 |
Other
| Other | Proceedings of the 1994 IEEE International Conference on Neural Networks. Part 1 (of 7) |
|---|---|
| City | Orlando, FL, USA |
| Period | 6/27/94 → 6/29/94 |
All Science Journal Classification (ASJC) codes
- Software