Considering that there is no single full reference image quality assessment method that could give the best performance in all situations, some multi-method fusion metrics were proposed. Machine learning techniques ar...Considering that there is no single full reference image quality assessment method that could give the best performance in all situations, some multi-method fusion metrics were proposed. Machine learning techniques are often involved in such multi-method fusion metrics so that its output would be more consistent with human visual perceptions. On the other hand, the robustness and generalization ability of these multi-method fusion metrics are questioned because of the scarce of images with mean opinion scores. In order to comprehensively validate whether or not the generalization ability of such multi-method fusion IQA metrics are satisfying, we construct a new image database which contains up to 60 reference images. The newly built image database is then used to test the generalization ability of different multi-method fusion IQA metrics. Cross database validation experiment indicates that in our new image database, the performances of all the multi-method fusion IQA metrics have no statistical significant different with some single-method IQA metrics such as FSIM and MAD. In the end, a thorough analysis is given to explain why the performance of multi-method fusion IQA framework drop significantly in cross database validation.展开更多
The Tianshan orogenic belt is a major part of the southern Central Asian Orogenic Belt（CAOB）,extending from west to east for over 2500 km through Uzbekistan,Tajikistan,Kyrgyzstan and Kazakhstan to Xinjiang in NW Chi...The Tianshan orogenic belt is a major part of the southern Central Asian Orogenic Belt（CAOB）,extending from west to east for over 2500 km through Uzbekistan,Tajikistan,Kyrgyzstan and Kazakhstan to Xinjiang in NW China,and contains the record of multi-phase tectonothermal evolution.Till now.展开更多
目的针对飞机货舱配载方案评估中多属性决策的复杂性,以及现有评估方法在权重确定上过度依赖专家经验导致的主观偏差,或单纯依赖客观数据忽略决策者偏好的局限性,提出一种融合主观先验与客观数据驱动的混合赋权评估模型,以提供更为合理...目的针对飞机货舱配载方案评估中多属性决策的复杂性,以及现有评估方法在权重确定上过度依赖专家经验导致的主观偏差,或单纯依赖客观数据忽略决策者偏好的局限性,提出一种融合主观先验与客观数据驱动的混合赋权评估模型,以提供更为合理、可靠的配载方案择优决策支持。方法首先,引入大型语言模型(Large language model,LLM),构建“虚拟专家委员会”,通过精心设计的提示词工程,获取多维度、多情境下的主观权重。其次,针对传统熵权法对数据分布敏感、难以有效区分指标优劣等问题,提出一种改进的数据预处理熵权法(Improved data preprocessing entropy weighting method,IDPEW),该方法结合指标值的辨识度和信息熵的均衡性来确定客观权重。最后,将LLM生成的主观权重与IDPEW计算的客观权重进行加权组合,构建综合评价函数,对飞机货舱配载方案进行全面评估和排序。结果实验结果表明,LLM模拟专家意见时最关注“装载率”(主观权重0.2250),而IDPEW方法从数据中识别出“横向不平衡度”最具区分力(客观权重0.2481)。混合赋权模型(α=0.5)有效平衡了主客观偏好,在24个方案中精准识别出综合性能最优的方案,验证了模型在复杂情境下的稳定性。结论创新性地利用LLM低成本构建“虚拟专家”获取先验知识,并通过耦合指标辨识度与均衡性的IDPEW方法,提升了客观赋权精度。该模型克服了单一赋权的局限,为飞机货舱配载方案的科学评估提供了一种兼具可解释性和实用性的新范式。展开更多
与单阈值分割相比,多阈值彩色图像分割具有更高的分割精度,对复杂的彩色图像分割有着较好的效果.但是由于阈值增多导致计算量增大,整体算法的运算时间增加.针对此问题,提出了一种改进的多元宇宙优化算法对多阈值彩色图像分割算法进行优...与单阈值分割相比,多阈值彩色图像分割具有更高的分割精度,对复杂的彩色图像分割有着较好的效果.但是由于阈值增多导致计算量增大,整体算法的运算时间增加.针对此问题,提出了一种改进的多元宇宙优化算法对多阈值彩色图像分割算法进行优化.首先,对多元宇宙优化算法进行改进,引入樽海鞘优化算法中的收敛因子,提高算法的寻优能力;然后,选取多阈值大津法作为图像分割算法,将其作为优化算法的适应度函数;最后,通过对标准数学公式的仿真实验,以及选取四幅伯克利大学图像库图像进行实验分析,实验结果表明该算法能够对图像进行精确分割,在PSNR(Peak Signal to Noise Ratio)和FSIM(Feature Similarity Index)两个指标上均优于其他算法,提高了分割精度.展开更多
基金supported by “the Fundamental Research Funds for the Central Universities” No.2018CUCTJ081
文摘Considering that there is no single full reference image quality assessment method that could give the best performance in all situations, some multi-method fusion metrics were proposed. Machine learning techniques are often involved in such multi-method fusion metrics so that its output would be more consistent with human visual perceptions. On the other hand, the robustness and generalization ability of these multi-method fusion metrics are questioned because of the scarce of images with mean opinion scores. In order to comprehensively validate whether or not the generalization ability of such multi-method fusion IQA metrics are satisfying, we construct a new image database which contains up to 60 reference images. The newly built image database is then used to test the generalization ability of different multi-method fusion IQA metrics. Cross database validation experiment indicates that in our new image database, the performances of all the multi-method fusion IQA metrics have no statistical significant different with some single-method IQA metrics such as FSIM and MAD. In the end, a thorough analysis is given to explain why the performance of multi-method fusion IQA framework drop significantly in cross database validation.
基金supported by the Major Basic Research Project of the Ministry of Science and Technology of China(Grant No.2014CB448000)National Science Foundation of China(Grant Nos..41473053 and 41573045)a grant of Chinese Ministry of Land and Resources(Grant No.201211074–05)
文摘The Tianshan orogenic belt is a major part of the southern Central Asian Orogenic Belt（CAOB）,extending from west to east for over 2500 km through Uzbekistan,Tajikistan,Kyrgyzstan and Kazakhstan to Xinjiang in NW China,and contains the record of multi-phase tectonothermal evolution.Till now.
文摘目的针对飞机货舱配载方案评估中多属性决策的复杂性,以及现有评估方法在权重确定上过度依赖专家经验导致的主观偏差,或单纯依赖客观数据忽略决策者偏好的局限性,提出一种融合主观先验与客观数据驱动的混合赋权评估模型,以提供更为合理、可靠的配载方案择优决策支持。方法首先,引入大型语言模型(Large language model,LLM),构建“虚拟专家委员会”,通过精心设计的提示词工程,获取多维度、多情境下的主观权重。其次,针对传统熵权法对数据分布敏感、难以有效区分指标优劣等问题,提出一种改进的数据预处理熵权法(Improved data preprocessing entropy weighting method,IDPEW),该方法结合指标值的辨识度和信息熵的均衡性来确定客观权重。最后,将LLM生成的主观权重与IDPEW计算的客观权重进行加权组合,构建综合评价函数,对飞机货舱配载方案进行全面评估和排序。结果实验结果表明,LLM模拟专家意见时最关注“装载率”(主观权重0.2250),而IDPEW方法从数据中识别出“横向不平衡度”最具区分力(客观权重0.2481)。混合赋权模型(α=0.5)有效平衡了主客观偏好,在24个方案中精准识别出综合性能最优的方案,验证了模型在复杂情境下的稳定性。结论创新性地利用LLM低成本构建“虚拟专家”获取先验知识,并通过耦合指标辨识度与均衡性的IDPEW方法,提升了客观赋权精度。该模型克服了单一赋权的局限,为飞机货舱配载方案的科学评估提供了一种兼具可解释性和实用性的新范式。
文摘与单阈值分割相比,多阈值彩色图像分割具有更高的分割精度,对复杂的彩色图像分割有着较好的效果.但是由于阈值增多导致计算量增大,整体算法的运算时间增加.针对此问题,提出了一种改进的多元宇宙优化算法对多阈值彩色图像分割算法进行优化.首先,对多元宇宙优化算法进行改进,引入樽海鞘优化算法中的收敛因子,提高算法的寻优能力;然后,选取多阈值大津法作为图像分割算法,将其作为优化算法的适应度函数;最后,通过对标准数学公式的仿真实验,以及选取四幅伯克利大学图像库图像进行实验分析,实验结果表明该算法能够对图像进行精确分割,在PSNR(Peak Signal to Noise Ratio)和FSIM(Feature Similarity Index)两个指标上均优于其他算法,提高了分割精度.
