基于自学习框架的红外场景仿真效果评价
|
摘要
针对红外场景仿真效果评估困难的问题,提出了基于自学习框架的评价方法。首先,从仿真图像与实际图像视觉对比的角度提出了面向图像视觉相似的红外场景仿真效果评价指标体系,用于量化评价过程;其次,提出以极限学习机(extreme learning machine,ELM)为核心建立评估模型,建立包括蒙特卡罗样本仿真、ELM评估网络及自更新仿真样本评估模型等3部分在内的自学习框架来生成仿真样本、强化对ELM的训练;最后,针对实际样本数量较少的问题,在此框架基础上提出了包括样本评定、自学习、评估模型测试3个阶段在内的仿真图像相似性评估方法,实现了从样本生成到评估过程的自动化。实验结果表明提出的自学习框架能够显著提高评估模型的正确率,而且训练后的评估模型适用性强,可独立自主进行红外场景仿真效果评估。
In order to reduce the influence of subjective cognition on infrared camouflage simulation evaluation,an evaluation method based on self-learning framework is proposed.Firstly,from the perspective of visual comparation between the simulation image and its corresponding actual image to measure the simulation results,focusing on visual similarity,the infrared camouflage simulation evaluation index system is established.Secondly,an evaluation model is established with extreme learning machine(ELM),and a self-learning framework is proposed to generate simulation samples and strengthen the training of ELM,including MonteCarlo simulation,ELM evaluation network and a self-updating evaluation model for simulation samples,Thirdly,due to the small number of actual samples,a new method of evaluating the similarity of simulation images,including three stages of the sample evaluation,self-learning and network evaluation is puts forward on the basis of the artificial evaluation experiment.Realizing the automation from samples generation to evaluation.Experimental results not only confirm that the proposed self-learning framework can significantly improve the accuracy of the evaluation model,but also show that the practiced evaluation model can be used to evaluate the simulation effects of infrared camouflage independently with strong applicability.
In order to reduce the influence of subjective cognition on infrared camouflage simulation evaluation,an evaluation method based on self-learning framework is proposed.Firstly,from the perspective of visual comparation between the simulation image and its corresponding actual image to measure the simulation results,focusing on visual similarity,the infrared camouflage simulation evaluation index system is established.Secondly,an evaluation model is established with extreme learning machine(ELM),and a self-learning framework is proposed to generate simulation samples and strengthen the training of ELM,including MonteCarlo simulation,ELM evaluation network and a self-updating evaluation model for simulation samples,Thirdly,due to the small number of actual samples,a new method of evaluating the similarity of simulation images,including three stages of the sample evaluation,self-learning and network evaluation is puts forward on the basis of the artificial evaluation experiment.Realizing the automation from samples generation to evaluation.Experimental results not only confirm that the proposed self-learning framework can significantly improve the accuracy of the evaluation model,but also show that the practiced evaluation model can be used to evaluate the simulation effects of infrared camouflage independently with strong applicability.
引文
[1]杨妹,杨山亮,许霄,等.面向高层辅助决策的作战分析仿真系统框架[J].系统工程与电子技术,2016,38(6):1440-1449.YANG M,YANG S L,XU X,et al.Framework of operational simulation in high level decision-making system[J].Systems Engineering and Electronics,2016,38(6):1440-1449.
[2]焦松,李伟,杨明.基于经验模态分解和灰色关联度分析的仿真模型验证方法[J].系统工程与电子技术,2013,35(12):2613-2618.JIAO S,LI W,YANG M.Validation of simulation models based on empirical modal decomposition and grey relevance analysis[J].Systems Engineering and Electronics,2013,35(12):2613-2618.
[3]王霞,汪昊,徐超,等.红外场景仿真技术发展综述[J].红外技术,2015,37(7):537-543.WANG X,WANG H,XU C,et al.Overview on development of infrared scene simulation[J].Infrared Technology,2015,37(7):537-543.
[4]方浩,李艾华,潘玉龙,等.面向视觉相似的红外伪装仿真效果评估[J].兵工学报,2017,38(2):351-357.FANG H,LI A H,PAN Y L,et al.The effect evaluation of infrared camouflage simulation system based on visual similarity[J].Acta Armamentarii,2017,38(2):351-357.
[5]句彦伟,张燕.IASR图像质量评估方法[J].系统工程与电子技术,2015,37(2):297-303.JU Y W,ZHANG Y.Research on ISAR image quality evaluation[J].Systems Engineering and Electronics,2015,37(2):297-303.
[6]侯旺,梅凤华,陈国军,等.基于背景最佳滤波尺度的红外图像复杂度评价准则[J].物理学报,2015,64(23):234202.HOU W,MEI F H,CHENG G J,et al.An evaluation criterion of infrared image complexity based on background optimal filter scale[J].Acta Physica Sinica,2015,64(23):234202.
[7]娄树理,韩艳丽,任建存,等.基于结构相似度和保真度的红外成像传感器仿真评价方法研究[J].红外,2015,30(2):30-35.LOU S L,HAN Y L,REN J C,et al.Research on evaluation method of simulation of infrared imaging sensor based on structural similarity and fidelity[J].Infrared,2015,30(2):30-35.
[8]查亚兵,唐见兵.基于主观综合评判的作战仿真可信度评估方法研究[J].国防科技大学学报,2010,32(6):153-157.ZHA Y B,TANG J B.Research on a credibility evaluation method for warfare simulation based on the subject integrate judgment[J].Journal of National University of Defense Technology,2010,32(6):153-157.
[9]吕绪良,林伟,许卫东,等.应用迭代自组织数据分析技术模糊聚类法划分伪装效能等级[J].兵工学报,2005,26(5):681-684.LU X L,LIN W,XU W D,et al.Applying the fuzzy clustering analysis of ISODATA to the classification of camouflage effectiveness[J].Acta Armamentarii,2005,26(5):681-684.
[10]张忠.仿真系统可信度评估方法研究[D].哈尔滨:哈尔滨工业大学,2014.ZHANG Z.Research on evaluation method for credibility of simulation system[D].Harbin:Harbin Institute of Technology,2014.
[11]曾艳阳,贾盼盼.基于改进灰色聚类的三维视景效果评估方法[J].系统仿真学报,2015,27(10):2497-2501.ZENG Y Y,JIA P P.Evaluation method of dimensional visual effects based on improved grey clustering[J].Journal of System Simulation,2015,27(10):2497-2501.
[12]MARTENS J,PUT F,KERRE E.A fuzzy-neural resem-blance approach to validate simulation models[J].Soft Computing,2007,11(3):299-307.
[13]李宏宁,白廷柱,曹峰梅,等.用于场景仿真的红外成像模型及其有效性分析[J].红外与毫米波学报,2010,29(2):57-62.LI H N,BAI T Z,CAO F M,et al.Infrared imaging model for scene simulation and its validation[J].Journal of Infrared and Millimeter Waves,2010,29(2):57-62.
[14]JOHNSTON L,SCHRUBEN L,YANG A,et.al.Establishing the credibility of a biotech simulation model[C]∥Proc.of the Winter Simulation Conference,2008:822-826.
[15]KLEIJNEN J P C,SARGENT R G.A methodology for fitting and validating meta-models in simulations[J].European Journal of Operational Research,2000,120(1):14-29.
[16]HUANG G,HUANG G B,SONG S J,et.al.Trends in extreme learning machines:a review[J].Neural Networks,2015,61:32-48.
[17]ZHU W,LEVINSON S.Edge orientation-based multi-view object recognition[C]∥Proc.of the IEEE 15th International Conference on Pattern Recognition,2000:936-939.
[18]ENGELKE U,KAPRYKOWSKY H,ZEPERNICK H,et al.Visual attention in quality assessment[J].IEEE Signal Processing Magazine,2011,28(6):50-59.
[19]ZHANG L,SHEN Y,LI H Y.VSI:a visual saliency-induced index for perceptual image quality assessment[J].IEEE Trans.on Image Processing,2014,23(10):4270-4281.
[20]ZHANG L,GU Z Y,LI H Y.SDSP:a novel saliency detection method by combining simple priors[C]∥Proc.of the IEEE International Conference on Image Process,2013:171-175.
[2]焦松,李伟,杨明.基于经验模态分解和灰色关联度分析的仿真模型验证方法[J].系统工程与电子技术,2013,35(12):2613-2618.JIAO S,LI W,YANG M.Validation of simulation models based on empirical modal decomposition and grey relevance analysis[J].Systems Engineering and Electronics,2013,35(12):2613-2618.
[3]王霞,汪昊,徐超,等.红外场景仿真技术发展综述[J].红外技术,2015,37(7):537-543.WANG X,WANG H,XU C,et al.Overview on development of infrared scene simulation[J].Infrared Technology,2015,37(7):537-543.
[4]方浩,李艾华,潘玉龙,等.面向视觉相似的红外伪装仿真效果评估[J].兵工学报,2017,38(2):351-357.FANG H,LI A H,PAN Y L,et al.The effect evaluation of infrared camouflage simulation system based on visual similarity[J].Acta Armamentarii,2017,38(2):351-357.
[5]句彦伟,张燕.IASR图像质量评估方法[J].系统工程与电子技术,2015,37(2):297-303.JU Y W,ZHANG Y.Research on ISAR image quality evaluation[J].Systems Engineering and Electronics,2015,37(2):297-303.
[6]侯旺,梅凤华,陈国军,等.基于背景最佳滤波尺度的红外图像复杂度评价准则[J].物理学报,2015,64(23):234202.HOU W,MEI F H,CHENG G J,et al.An evaluation criterion of infrared image complexity based on background optimal filter scale[J].Acta Physica Sinica,2015,64(23):234202.
[7]娄树理,韩艳丽,任建存,等.基于结构相似度和保真度的红外成像传感器仿真评价方法研究[J].红外,2015,30(2):30-35.LOU S L,HAN Y L,REN J C,et al.Research on evaluation method of simulation of infrared imaging sensor based on structural similarity and fidelity[J].Infrared,2015,30(2):30-35.
[8]查亚兵,唐见兵.基于主观综合评判的作战仿真可信度评估方法研究[J].国防科技大学学报,2010,32(6):153-157.ZHA Y B,TANG J B.Research on a credibility evaluation method for warfare simulation based on the subject integrate judgment[J].Journal of National University of Defense Technology,2010,32(6):153-157.
[9]吕绪良,林伟,许卫东,等.应用迭代自组织数据分析技术模糊聚类法划分伪装效能等级[J].兵工学报,2005,26(5):681-684.LU X L,LIN W,XU W D,et al.Applying the fuzzy clustering analysis of ISODATA to the classification of camouflage effectiveness[J].Acta Armamentarii,2005,26(5):681-684.
[10]张忠.仿真系统可信度评估方法研究[D].哈尔滨:哈尔滨工业大学,2014.ZHANG Z.Research on evaluation method for credibility of simulation system[D].Harbin:Harbin Institute of Technology,2014.
[11]曾艳阳,贾盼盼.基于改进灰色聚类的三维视景效果评估方法[J].系统仿真学报,2015,27(10):2497-2501.ZENG Y Y,JIA P P.Evaluation method of dimensional visual effects based on improved grey clustering[J].Journal of System Simulation,2015,27(10):2497-2501.
[12]MARTENS J,PUT F,KERRE E.A fuzzy-neural resem-blance approach to validate simulation models[J].Soft Computing,2007,11(3):299-307.
[13]李宏宁,白廷柱,曹峰梅,等.用于场景仿真的红外成像模型及其有效性分析[J].红外与毫米波学报,2010,29(2):57-62.LI H N,BAI T Z,CAO F M,et al.Infrared imaging model for scene simulation and its validation[J].Journal of Infrared and Millimeter Waves,2010,29(2):57-62.
[14]JOHNSTON L,SCHRUBEN L,YANG A,et.al.Establishing the credibility of a biotech simulation model[C]∥Proc.of the Winter Simulation Conference,2008:822-826.
[15]KLEIJNEN J P C,SARGENT R G.A methodology for fitting and validating meta-models in simulations[J].European Journal of Operational Research,2000,120(1):14-29.
[16]HUANG G,HUANG G B,SONG S J,et.al.Trends in extreme learning machines:a review[J].Neural Networks,2015,61:32-48.
[17]ZHU W,LEVINSON S.Edge orientation-based multi-view object recognition[C]∥Proc.of the IEEE 15th International Conference on Pattern Recognition,2000:936-939.
[18]ENGELKE U,KAPRYKOWSKY H,ZEPERNICK H,et al.Visual attention in quality assessment[J].IEEE Signal Processing Magazine,2011,28(6):50-59.
[19]ZHANG L,SHEN Y,LI H Y.VSI:a visual saliency-induced index for perceptual image quality assessment[J].IEEE Trans.on Image Processing,2014,23(10):4270-4281.
[20]ZHANG L,GU Z Y,LI H Y.SDSP:a novel saliency detection method by combining simple priors[C]∥Proc.of the IEEE International Conference on Image Process,2013:171-175.