[Measurement and analysis on complex refraction indices of pear pollen in infrared band].

Pollen is an important part of bioaerosols, and its complex refractive index is a crucial parameter for study on optical characteristics and detection, identification of bioaerosols. The reflection spectra of pear pollen within the 2. 5 - 15µm waveband were measured by squash method. Based on the measured data, the complex refractive index of pear pollen within the wave-band of 2. 5 to 15 µm was calculated by using Kramers-Kroning (K-K) relation, and calculation deviation about incident angle and different reflectivities at high and low frequencies.were analyzed. The results indicate that 18 degrees angle of incidence and different reflectivities at high and low frequencies have little effect on the results, and it is practicable to calculate the complex refractive index of pollen based on its reflection spectral data. The data of complex refractive index of pollen have some reference value for optical characteristics of pollen, detection and identification of bioaerosols.

[Classification of hyperspectral imagery based on ant colony compositely optimizing SVM in spatial and spectral features].

A novel classification algorithm of hyperspectral imagery based on ant colony compositely optimizing support vector machine in spatial and spectral features was proposed. Two types of virtual ants searched for the bands combination with the maximum class separation distance and heterogeneous samples in spatial and spectral features alternately. The optimal characteristic bands were extracted, and bands redundancy of hyperspectral imagery decreased. The heterogeneous samples were eliminated form the training samples, and the distribution of samples was optimized in feature space. The hyperspectral imagery and training samples which had been optimized were used in classification algorithm of support vector machine, so that the class separation distance was extended and the accuracy of classification was improved. Experimental results demonstrate that the proposed algorithm, which acquires an overall accuracy 95.45% and Kappa coefficient 0.925 2, can obtain greater accuracy than traditional hyperspectral image classification algorithms.