Considering the flexible attitude maneuver and the narrow field of view of agile Earth observation satellite(AEOS)together,a comprehensive task clustering(CTC)is proposed to improve the observation scheduling problem ...Considering the flexible attitude maneuver and the narrow field of view of agile Earth observation satellite(AEOS)together,a comprehensive task clustering(CTC)is proposed to improve the observation scheduling problem for AEOS(OSPFAS).Since the observation scheduling problem for AEOS with comprehensive task clustering(OSWCTC)is a dynamic combination optimization problem,two optimization objectives,the loss rate(LR)of the image quality and the energy consumption(EC),are proposed to format OSWCTC as a bi-objective optimization model.Harnessing the power of an adaptive large neighborhood search(ALNS)algorithm with a nondominated sorting genetic algorithm II(NSGA-II),a bi-objective optimization algorithm,ALNS+NSGA-II,is developed to solve OSWCTC.Based on the existing instances,the efficiency of ALNS+NSGA-II is analyzed from several aspects,meanwhile,results of extensive computational experiments are presented which disclose that OSPFAS considering CTC produces superior outcomes.展开更多
针对全波段光谱技术的生鲜猪肉综合品质快速无损分类存在光谱数据量大、样本数量较少时分类准确率较低等缺点。该文提出了一种基于偏最小二乘(partial least squares,PLS)投影分析算法和支持向量机的生鲜猪肉综合品质分类器。利用基于...针对全波段光谱技术的生鲜猪肉综合品质快速无损分类存在光谱数据量大、样本数量较少时分类准确率较低等缺点。该文提出了一种基于偏最小二乘(partial least squares,PLS)投影分析算法和支持向量机的生鲜猪肉综合品质分类器。利用基于偏最小二乘投影分析算法对全波段光谱数据进行数据降维,选取了13个特征波长。利用粒子群优化算法优化支持向量机惩罚参数和径向基核函数参数,优化后二者最优为4.939和0.01。利用选取的特征波长和优化后的参数建立了生鲜猪肉综合品质支持向量分类器。研究结果表明,分类器对训练集中白肌肉(pale,soft and exudative,PSE)、正常肉(reddish-pink,firm and non-exudative,RFN)和黑干肉(dark,firm and dry,DFD)的回判识别率分别为为88.46%、94.11%和92.31%;测试集中PSE、RFN和DFD预测正确率分别为84.62%、94.11%和84.62%。该分类器满足模型简单、预测准确率高等优点,为生鲜猪肉综合品质在线分级提供参考。展开更多
文摘Considering the flexible attitude maneuver and the narrow field of view of agile Earth observation satellite(AEOS)together,a comprehensive task clustering(CTC)is proposed to improve the observation scheduling problem for AEOS(OSPFAS).Since the observation scheduling problem for AEOS with comprehensive task clustering(OSWCTC)is a dynamic combination optimization problem,two optimization objectives,the loss rate(LR)of the image quality and the energy consumption(EC),are proposed to format OSWCTC as a bi-objective optimization model.Harnessing the power of an adaptive large neighborhood search(ALNS)algorithm with a nondominated sorting genetic algorithm II(NSGA-II),a bi-objective optimization algorithm,ALNS+NSGA-II,is developed to solve OSWCTC.Based on the existing instances,the efficiency of ALNS+NSGA-II is analyzed from several aspects,meanwhile,results of extensive computational experiments are presented which disclose that OSPFAS considering CTC produces superior outcomes.
文摘针对全波段光谱技术的生鲜猪肉综合品质快速无损分类存在光谱数据量大、样本数量较少时分类准确率较低等缺点。该文提出了一种基于偏最小二乘(partial least squares,PLS)投影分析算法和支持向量机的生鲜猪肉综合品质分类器。利用基于偏最小二乘投影分析算法对全波段光谱数据进行数据降维,选取了13个特征波长。利用粒子群优化算法优化支持向量机惩罚参数和径向基核函数参数,优化后二者最优为4.939和0.01。利用选取的特征波长和优化后的参数建立了生鲜猪肉综合品质支持向量分类器。研究结果表明,分类器对训练集中白肌肉(pale,soft and exudative,PSE)、正常肉(reddish-pink,firm and non-exudative,RFN)和黑干肉(dark,firm and dry,DFD)的回判识别率分别为为88.46%、94.11%和92.31%;测试集中PSE、RFN和DFD预测正确率分别为84.62%、94.11%和84.62%。该分类器满足模型简单、预测准确率高等优点,为生鲜猪肉综合品质在线分级提供参考。
