Reliability enhancement testing(RET) is an accelerated testing which hastens the performance degradation process to surface its inherent defects of design and manufacture. It is an important hypothesis that the degrad...Reliability enhancement testing(RET) is an accelerated testing which hastens the performance degradation process to surface its inherent defects of design and manufacture. It is an important hypothesis that the degradation mechanism of the RET is the same as the one of the normal stress condition. In order to check the consistency of two mechanisms, we conduct two enhancement tests with a missile servo system as an object of the study, and preprocess two sets of test data to establish the accelerated degradation models regarding the temperature change rate that is assumed to be the main applied stress of the servo system during the natural storage. Based on the accelerated degradation models and natural storage profile of the servo system, we provide and demonstrate a procedure to check the consistency of two mechanisms by checking the correlation and difference of two sets of degradation data. The results indicate that the two degradation mechanisms are significantly consistent with each other.展开更多
为了得到优质的融合图像,提出了一种改进的拉普拉斯能量和(New Sum of Modified Laplacian,NSML)多聚焦图像融合算法。该算法在传统SML计算每个像素点的变步长拉普拉斯算子值仅有的水平和垂直方向的基础上,增加了斜对角线上的四个方向...为了得到优质的融合图像,提出了一种改进的拉普拉斯能量和(New Sum of Modified Laplacian,NSML)多聚焦图像融合算法。该算法在传统SML计算每个像素点的变步长拉普拉斯算子值仅有的水平和垂直方向的基础上,增加了斜对角线上的四个方向。同时通过分析NSML算法的计算过程,发现存在大量的重复计算,从而提出了基于积分图像的快速NSML图像融合方法。该方法通过简化NSML的计算过程,大大减少了图像融合处理过程消耗的时间,提高了图像融合的效率。实验结果表明,快速NSML方法在达到极佳融合图像质量的同时,提升了算法的实时性。展开更多
基金supported by the Natural Science Foundation of Hunan Province(2018JJ2282)
文摘Reliability enhancement testing(RET) is an accelerated testing which hastens the performance degradation process to surface its inherent defects of design and manufacture. It is an important hypothesis that the degradation mechanism of the RET is the same as the one of the normal stress condition. In order to check the consistency of two mechanisms, we conduct two enhancement tests with a missile servo system as an object of the study, and preprocess two sets of test data to establish the accelerated degradation models regarding the temperature change rate that is assumed to be the main applied stress of the servo system during the natural storage. Based on the accelerated degradation models and natural storage profile of the servo system, we provide and demonstrate a procedure to check the consistency of two mechanisms by checking the correlation and difference of two sets of degradation data. The results indicate that the two degradation mechanisms are significantly consistent with each other.
文摘为了得到优质的融合图像,提出了一种改进的拉普拉斯能量和(New Sum of Modified Laplacian,NSML)多聚焦图像融合算法。该算法在传统SML计算每个像素点的变步长拉普拉斯算子值仅有的水平和垂直方向的基础上,增加了斜对角线上的四个方向。同时通过分析NSML算法的计算过程,发现存在大量的重复计算,从而提出了基于积分图像的快速NSML图像融合方法。该方法通过简化NSML的计算过程,大大减少了图像融合处理过程消耗的时间,提高了图像融合的效率。实验结果表明,快速NSML方法在达到极佳融合图像质量的同时,提升了算法的实时性。
