模糊图像固定跟踪点特征自适应增强算法仿真  

Simulation of Fuzzy Image Fixed Tracking Point Feature Adaptive Enhancement Algorithm

作　　者：钟小莉[1] 谢旻旻[1] ZHONG Xiao-li;XIE Min-min(Computer School,Qinghai Minzu University,Xining Qinghai 810007,China)

机构地区：[1]青海民族大学计算机学院,青海西宁810007

出　　处：《计算机仿真》2023年第12期316-320,共5页Computer Simulation

基　　金：基于立体图像智能分割技术应用的算法研究(2018XJY01)。

摘　　要：为降低外界环境对图像的干扰,提出一种模糊图像固定跟踪点特征自适应增强算法。根据模糊网络得出图像的噪声振幅,均衡不同隶属度函数取值平滑噪声,锐化增强中心像素与邻域像素间亮度差值,通过归一化处理设定合适阈值,用概率密度函数计算图像灰度级,得出均衡后灰度阈值区间,构造全方位多尺度结构进行顶帽细节特征变换,提取图像不同方向、尺度上细节特征,凭借补偿变换简化特征函数,令图像在不同尺度层面中分别完成增强。仿真结果表明,所提算法适用性强,固定点灰度均衡效果好,模糊图像在亮度和色彩上都得到了有效加强。In order to reduce the interference of external environment,an adaptive enhancement algorithm for fixed tracking point feature in fuzzy image was put forward.According to fuzzy network,the noise amplitude of image was obtained.And then,the values of different membership functions were balanced,and the noise was smoothed as well.Moreover,the brightness difference between central pixel and adjacent pixels was sharpened.Meanwhile,the reasonable threshold was set through normalization.Furthermore,probability density functions were used to calculate the gray level of image,thus obtaining the gray threshold interval after equalization.After that,an omni-directional multi-scale structure was constructed for detail feature transformation of top.Finally,the detailed features of the im⁃age in different directions and scales were extracted,and the feature function was simplified by using the compensa⁃tion transformation.Thus,the image can be enhanced at different scales.Simulation results prove that the proposed algorithm has strong applicability and good gray equalization effect of fixed point;In addition,the blurred image is ef⁃fectively strengthened in brightness and color.

关 键 词：模糊图像 自适应增强 归一化处理 概率密度函数 多尺度结构元素 

分 类 号：TP391.9[自动化与计算机技术—计算机应用技术]