基于多模态遗传优化的非参数方法多曲线识别算法的研究-计算机应用技术专业论文.docxVIP

基于多模态遗传优化的非参数方法多曲线识别算法的研究 Research of Multi-nonparametric Curve Recognition Algorithm Based on Multi-modal Genetic Optimization ABSTRACT Hough transform can recognize curve in binary image with break points and a large number of noises, while curve function or shape is must given before recognition when using Hough transform. But in actual images, the curve function or shape is not a priori knowledge before recognition in most cases and the curve can not be recognized by Hough transform. Its recognition can be seen a kind of permutation-based optimization problem. Based on deep study of genetic algorithm, use of genetic optimization approaching identify non-parametric curve. In actual images, there are also two curves or even more than two curves. This problem-solving can be taken by multi-modal genetic optimization to identify multi-curve. The main research contexts are as follows: Introducing parameter curve and non-parametric curve’s concept approaches using Hough transform to recognize parameter curve. And researching human filtering properties, reached the critical noise of the human filtering. Its purpose is to calibration for the multi-curve recognition. Multi-modal genetic algorithm has some problems to achieve the global optimal solution and the most possible local optimal solutions, this dissertation propose an improved algorithm, using similar cross-cutting and cross-operator algorithm. The new algorithm gives the theory, flow chart and simulation experiment. Based on row-column code niche technology applied to recognize the non-parametric curve and achieved good result. Using multi-modal genetic optimization algorithm to recognize multi non-parametric curves. In a binary image two or even more non-parametric curves were extracted by using niche strategy to retain the multi-elite of the visually optimal curves. KEY WORDS: Hough transform, Genetic algorithm, Multi-modal genetic optimization, Niche technology, Non-parametric curve recognition ii 原创性声明 本人郑重声