无线电工程2025，Vol.55Issue(3)：642-648,7.DOI:10.3969/j.issn.1003-3106.2025.03.022
基于CPO-SVM模型的目标表面发射率求解方法
Solution of Target Surface Emissivity Based on CPO-SVM Model
摘要
Abstract
In order to improve the solution accuracy of target surface emissivity,the radiant brightness data are obtained by building a radiant brightness test system,and the Crested Porcupine Optimizer-Support Vector Machines(CPO-SVM)infrared radiant brightness regression model is established based on the CPO-SVM network to complete the smoothing output of the radiant brightness outliers,and to achieve a method of solving the emissivity of the target surface.To address the difficulty in selecting initial parameters of the SVM,the CPO algorithm is applied to optimize the penalty factor and kernel parameters of SVM,taking the wavelength as input and the radiant brightness data as output,and then iteratively searching for optimal parameters of the model,and then solving for the emissivity after the training of test samples with these optimal parameters.Comparing with the standard test sample with a known emissivity of 0.48,results show that the maximum relative error of the optimized model is 2.2%,and comparing with the results of traditional SVM and PSO-SVM algorithms,the maximum relative error of the CPO-SVM model has been reduced by 1.4%and 0.2%,respectively,providing an effective method for solving the emissivity of the target surface.关键词
红外发射率/冠豪猪优化算法/支持向量机/红外辐射亮度Key words
infrared emissivity/CPO algorithm/SVM/infrared radiant brightness分类
信息技术与安全科学引用本文复制引用
傅莉,陈广明,席剑辉,胡为..基于CPO-SVM模型的目标表面发射率求解方法[J].无线电工程,2025,55(3):642-648,7.基金项目
电磁信息控制与效应全国重点实验室开放基金项目(20230201)Open Fund Project of National Key Laboratory of Elec-tromagnetic Information Control and Effect(20230201) （20230201）
