Optimization of Multi-Channel Optoelectronic Recognition System with Simulated Annealing
碩士 === 元智大學 === 光電工程研究所 === 99 === In this study, we utilize the convergence of simulated annealing to optimize the reference function generated by minimum average correlation energy. Based on the RGB color space, we improve the multi-channel image recognition system with optimal reference functions...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | en_US |
| Published: |
2011
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/25798675717882678781 |
| Summary: | 碩士 === 元智大學 === 光電工程研究所 === 99 === In this study, we utilize the convergence of simulated annealing to optimize the reference function generated by minimum average correlation energy. Based on the RGB color space, we improve the multi-channel image recognition system with optimal reference functions. We use the images with different angles as input training patterns, to generate the reference functions by each color channel separated. The reference functions are also modified by multilevel quantitative analysis, and then we use them as the initial solutions of simulated annealing for optimization. We adopt Mach-Zehnder joint transform correlator for the recognition of polychromatic images in multiple channels. The technique with optimal reference functions for optical recognition is feasible.
|
|---|