基于DNA遗传萤火虫优化的盲均衡与图像盲恢复算法

发布时间：2018-05-18 17:57

本文选题：盲均衡算法 + 小波变换　；参考：《南京信息工程大学》2017年硕士论文

【摘要】：图像在信道传输过程中,未知点扩展函数和噪声等干扰因素对图像质量有重要影响,因此,利用盲均衡算法对接收到的图像进行有效恢复是值得探究的问题。然而,盲均衡算法存在局部收敛甚至发散的不足,故本文将盲均衡算法及DNA遗传算法和萤火虫算法相结合,对图像盲恢复进行研究。研究内容主要包括以下几个方面:1.定义了 DNA遗传萤火虫算法。针对萤火虫算法对优秀个体的依赖程度高等缺点,通过DNA遗传算法提高萤火虫种群基因,从而提高了求解精度。将DNA遗传算法和萤火虫算法这两种算法结合起来,定义了 DNA遗传萤火虫算法,该算法利用DNA遗传算法的优点对萤火虫进行优化,使其找到的极值点更加准确,提高了寻优能力。2.研究了基于DNA遗传萤火虫优化的盲均衡算法。针对常模盲均衡算法收敛速度慢、易于陷入局部极值等缺点,通过DNA遗传萤火虫算法优化萤火虫种群,提高了该算法的全局搜索能力,获得了代价函数的全局极值,从而提高了均衡效果,克服了常模盲均衡的缺点。3.提出了基于新型DNA遗传萤火虫优化的小波盲均衡算法。针对现有传统DNA遗传算法的缺陷,定义了新型交叉和变异操作,获得了新型DNA遗传萤火虫算法。该算法提高了遗传性能,保证了萤火虫种群质量,优化了正交小波变换盲均衡器的性能。4.提出了基于新型DNA遗传萤火虫优化的二维图像小波盲恢复算法。对于图像信号来说,用一维均衡器进行均衡,在降维和升维处理过程中,会使一部分信息丢失,因此本研究直接对二维图像进行盲恢复,从而提出了基于新型DNA遗传萤火虫优化的二维图像小波盲恢复算法。该算法减少了信息的缺失,提高了复原图像的质量。
[Abstract]:In the process of image transmission, the unknown point spread function and noise have an important effect on the image quality. Therefore, it is worth exploring to use blind equalization algorithm to recover the received image effectively. However, the blind equalization algorithm has the shortcoming of local convergence and even divergence. Therefore, the blind equalization algorithm, DNA genetic algorithm and firefly algorithm are combined in this paper to study the blind image restoration. The research mainly includes the following aspects: 1. DNA genetic algorithm for fireflies is defined. Aiming at the high dependence degree of the firefly algorithm on the excellent individuals, the DNA genetic algorithm is used to improve the gene of the firefly population so as to improve the accuracy of the solution. The DNA genetic algorithm and the firefly algorithm are combined to define the DNA genetic firefly algorithm. The algorithm uses the advantages of DNA genetic algorithm to optimize the firefly and make it more accurate. Improved the ability to find the best. 2. A blind equalization algorithm based on DNA genetic firefly optimization is studied. Aiming at the shortcomings of the blind equalization algorithm, such as slow convergence speed and easy to fall into local extremum, the DNA genetic firefly algorithm is used to optimize the population of fireflies, and the global searching ability of the algorithm is improved, and the global extremum of the cost function is obtained. Thus, the equalization effect is improved and the disadvantage of blind equalization of norm is overcome. A new blind equalization algorithm based on DNA genetic firefly optimization is proposed. Aiming at the defects of traditional DNA genetic algorithm, new crossover and mutation operations are defined, and a new DNA genetic firefly algorithm is obtained. The algorithm improves the genetic performance, ensures the quality of the firefly population, and optimizes the performance of the orthogonal wavelet transform blind equalizer. A new blind restoration algorithm based on DNA genetic firefly optimization for two-dimensional image is presented in this paper. For image signal equalization with one-dimensional equalizer will make part of the information lose in the process of reducing and raising dimension. Therefore this study directly performs blind restoration of two-dimensional images. Thus, a new DNA genetic firefly optimization algorithm based on wavelet blind restoration of two-dimensional images is proposed. The algorithm reduces the missing information and improves the quality of reconstructed image.
【学位授予单位】：南京信息工程大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TP18;TP391.41

【参考文献】