混合果蝇优化算法下芯片图像边缘特征增强  

Edge feature enhancement of chip image based on hybrid drosophila optimization algorithm

作　　者：陈闻 陶雪娜 刘青 CHEN Wen;TAO Xuena;LIU Qing(Hubei Central China Technology Development of Electric Power Co.,Ltd.,Wuhan 430000,China)

机构地区：[1]湖北华中电力科技开发有限责任公司,湖北武汉430000

出　　处：《电子设计工程》2025年第6期131-135,共5页Electronic Design Engineering

基　　金：湖北华中电力科技开发有限责任公司第一批科研项目(0322HBDL202107114)。

摘　　要：针对芯片图像边缘点定位不精准,图像边缘特征增强不佳的问题,提出了混合果蝇优化算法下芯片图像边缘特征增强方法。使用果蝇优化算法划分图像邻域,使邻域噪声相互抵消,达到降噪的目的。使用蛙跳算法对每个模因组深度搜索,计算果蝇更新位置的最大值,获取边缘检测结果,达到全局寻优的目的。设计混合果蝇优化算法区间搜索流程,提取待增强标定芯片图像边缘特征,从而获取真实图像边缘特征。实验结果表明,所研究方法获取的坐标与实际坐标的最大误差为0.01 cm,获取的图像边缘完整且清晰,具有良好的芯片图像边缘特征增强效果。In response to the problem of inaccurate positioning of edge points in chip images and poor enhancement of image edge features,a hybrid fruit fly optimization algorithm for chip image edge feature enhancement is proposed.Using the fruit fly optimization algorithm to partition image neighborhoods,ensuring that neighborhood noise cancels out each other,achieving the purpose of noise reduction processing.Use the frog jump algorithm to perform a deep search on each meme group,calculate the maximum updated position of the fruit fly,obtain edge detection results,and achieve global optimization.Design a hybrid fruit fly optimization algorithm interval search process,extract edge features of the chip image to be enhanced and calibrated,and obtain real image edge features.The experimental results show that the maximum error between the coordinates obtained by the research method and the actual coordinates is 0.01 cm.The obtained image edges are complete and clear,and have a good edge feature enhancement effect on chip images.

关 键 词：混合果蝇优化算法 芯片图像 边缘特征增强 蛙跳算法 

分 类 号：TP18[自动化与计算机技术—控制理论与控制工程]