A small target detection approach based on independent component analysis for hyperspectral data is put forward. In this algorithm, firstly the fast independent component analysis(FICA) is used to collect target infor...A small target detection approach based on independent component analysis for hyperspectral data is put forward. In this algorithm, firstly the fast independent component analysis(FICA) is used to collect target information hided in high-dimensional data and projects them into low-dimensional space.Secondly, the feature images are selected with kurtosis .At last, small targets are extracted with histogram image segmentation which has been labeled by skewness.展开更多
Computer simulation was carried out on fiber length and width for plantation-grown Chinesewhite poplar (Populus tomentosa Carr. clone) and plantation-grown poplar I-72 (P. x eurumericana (Dode)Guiner cv.). Skewness an...Computer simulation was carried out on fiber length and width for plantation-grown Chinesewhite poplar (Populus tomentosa Carr. clone) and plantation-grown poplar I-72 (P. x eurumericana (Dode)Guiner cv.). Skewness and kurtosis of measured results exhibited that distributions of the fiber length andwidth departured from normal distribution. Three-parameter Weibull density function was used in thisinvestigation and the corresponding program was written with Turbo C. The results showed that profiles ofsimulated length and width histograms were similar to ones of measured histograms, and that there was apretty good agreement between simulated and measured means of fiber length and width. There was a littleinfluence on the simulated means from seed used in random number generator and number of simulatedvariables. That indicated that the simulation was steady when the seed and the number were altered. Differenthistograms can be obtained with different values of the location, the shape, and the scale parameter correspondingto different values of the minimum, the mean, and the standard deviation for fiber length and width. Thesimulation presented here can be used as a tool for the studies on the variations in fiber morphology.展开更多
基金Funded by the National 863 Program of China (No.2002AA783050)
文摘A small target detection approach based on independent component analysis for hyperspectral data is put forward. In this algorithm, firstly the fast independent component analysis(FICA) is used to collect target information hided in high-dimensional data and projects them into low-dimensional space.Secondly, the feature images are selected with kurtosis .At last, small targets are extracted with histogram image segmentation which has been labeled by skewness.
文摘Computer simulation was carried out on fiber length and width for plantation-grown Chinesewhite poplar (Populus tomentosa Carr. clone) and plantation-grown poplar I-72 (P. x eurumericana (Dode)Guiner cv.). Skewness and kurtosis of measured results exhibited that distributions of the fiber length andwidth departured from normal distribution. Three-parameter Weibull density function was used in thisinvestigation and the corresponding program was written with Turbo C. The results showed that profiles ofsimulated length and width histograms were similar to ones of measured histograms, and that there was apretty good agreement between simulated and measured means of fiber length and width. There was a littleinfluence on the simulated means from seed used in random number generator and number of simulatedvariables. That indicated that the simulation was steady when the seed and the number were altered. Differenthistograms can be obtained with different values of the location, the shape, and the scale parameter correspondingto different values of the minimum, the mean, and the standard deviation for fiber length and width. Thesimulation presented here can be used as a tool for the studies on the variations in fiber morphology.
