Identification of the Citrus Greening Disease Using Spectral and Textural Features Based on Hyperspectral Imaging.

In this paper we discussed the application of spectral and textural features in identifying early stage of the citrus greening disease (Huanglongbing or HLB). A total of 176 hyperspectral images of citrus leaves (60 for healthy, 60 for HLB-infected and 56 for zinc-deficient) were captured by using a near-ground hyperspectral imaging system. Regions of interest (ROI) were extracted manually from the part of pathological changes in the images to calculate the average reflectance spectra of each sample as the sample spectra, ranging from 396 to 1 010 nm. The dimensions of the sample spectra were reduced with the algorithms of principal component analysis (PCA) and successive projection analysis (SPA). Classification models were built with the original spectra and candidate variables, the first four PCs selected by PCA and a set of wavelengths (630.5, 679.4, 749.4 and 899.9 nm) selected by SPA. The results based on a classifier of least square-support vector machine (LS-SVM) showed that the classification models built with the candidate variables selected by PCA and SPA had a better performance, achieving 89.7% and 87.4% in terms of average accuracy. In addition, two groups of textural features, extracted from gray images of the four selected wavelengths based on gray-level histogram and gray-level co-occurrence matrix (GLCM), were also used for the classifier. The first ten features ranked by SPA promoted the average accuracy of classifier significantly, achieving 100%, 93.3% and 92.9% for the three class samples respectively. The results of this study indicated that it would be feasible to identify HLB using the image textural features based on selected wavelengths, and it provided a basis for developing a portable HLB detection system with multispectral imaging techniques.

[Detection of citrus greening based on Vis-NIR spectroscopy and spectral feature analysis].

In the present paper we discussed the methods of classification of citrus greening and extraction of spectral features based on the spectral reflectance of four different statuses of citrus leaves (healthy, HLB, iron deficiency and nitrogen deficiency). Between two classes of classification, the values of discriminability of different spectra were calculated to extract spectral features. The greater value of discriminability showed a bigger difference of the two spectra, which means it would be easier to distinguish the two classes. By the Fisher linear discriminant analysis, three classification models (HLB & healthy, HLB & iron deficiency and HLB & nitrogen deficiency) based on the spectral features yielded more than 90% accuracies, which were better than expected. And at last, we discussed the application of the classification tree in multi-class discriminant analysis and spectral features extraction. The models trained based on the original reflectance spectra, first derivative and selected spectral features yielded more than 88% average accuracy, and especially the model based on the spectral features yielded more than 94% average accuracies, which verified the feasibility of detection of citrus greening in multi-class discriminant analysis and the importance of the spectral feature extraction. The results were compared based on classification tree, k-NN and Bayesian classifiers. Adoption of spectral features as input variables was significantly superior to using the original spectrum, which confirmed the validity of spectral feature selection. Spectral features could be used well for developing a multi-spectral imaging system to detect the citrus greening.

High efficient expression of bioactive human BMP-14 in E. coli using SUMO fusion partner.

Bone morphogenetic proteins (BMPs) are cytokines from the TGF-ß superfamily, with important roles during embryonic development and in the induction of bone and cartilage tissue differentiation in the adult body. In this contribution, We report here the application of small ubiquitin-related modifier (SUMO) fusion technology to the expression and purification of human BMP-14. The fusion protein expressed in a soluble form was purified to a purity of 90% by Ni-IDA chromatography. After the SUMO-BMP14 fusion protein was cleaved by the SUMO protease at 30 °C for 1 h, the cleaved sample was re-applied to a Ni-IDA. Finally, about 45 mg recombinant hBMP-14 was obtained from 1 litre bacterial culture with no less than 95% purity. The purified hBMP-14 dimer was over 90% purity and could induce the expression of alkaline phosphatase activity in C2C12 cells in a dose-dependent manner. Thus the SUMO-mediated peptide expression and purification system potentially could be employed for the production of other homodimeric proteins.

[Study of building quantitative analysis model for chlorophyll in winter wheat with reflective spectrum using MSC-ANN algorithm].

Preprocess method of multiplicative scatter correction (MSC) was used to reject noises in the original spectra produced by the environmental physical factor effectively, then the principal components of near-infrared spectroscopy were calculated by nonlinear iterative partial least squares (NIPALS) before building the back propagation artificial neural networks method (BP-ANN), and the numbers of principal components were calculated by the method of cross validation. The calculated principal components were used as the inputs of the artificial neural networks model, and the artificial neural networks model was used to find the relation between chlorophyll in winter wheat and reflective spectrum, which can predict the content of chlorophyll in winter wheat. The correlation coefficient (r) of calibration set was 0.9604, while the standard deviation (SD) and relative standard deviation (RSD) was 0.187 and 5.18% respectively. The correlation coefficient (r) of predicted set was 0.9600, and the standard deviation (SD) and relative standard deviation (RSD) was 0.145 and 4.21% respectively. It means that the MSC-ANN algorithm can reject noises in the original spectra produced by the environmental physical factor effectively and set up an exact model to predict the contents of chlorophyll in living leaves veraciously to replace the classical method and meet the needs of fast analysis of agricultural products.

[Study of quantitative analysis of protein in barley using OSC-PLS algorithm].

Spectra of barley containing vast information were obtained with the dispersion spectrograph. The contents of protein in barley were determined by dispersive near infrared (NIR) spectroscopy. Pretreatment method of orthogonal signal correction (OSC) was used to reject uncorrelated variables in the original spectra before building the partial least squares NIR method (OSC-PLS). The results were compared with the regular PLS model. With the OSC-PLS method, the determination coefficient R2 was 0.901. The correlation coefficient of validation set was 0.971 7. The standard deviation (SD) and relative standard deviation (RSD)were 0.545 0 and 4.2% respectively. Applying OSC-PLS resulted in removal of non-correlated variation in spectra and reduced model's complexity with preserved ability and improved interpretative ability of variation in spectra. It means that the OSC-PLS is a fungible model to predict the contents of protein in barley veraciously to meet the demand of fast analysis of agricultural products.

[Application study of ant colony algorithm in near infrared spectroscopy quantitative analysis].

Ant colony algorithm is a novel bio-inspired optimization algorithm, which simulates the foraging behavior of ants for solving various complex combinatorial optimization problems. The advantages of ant colony algorithm are intelligent search, global optimization, robustness, distributed computation and easy combination with other heuristic method. Near infrared spectroscopy quantitative analysis has been applied in many fields, whereas the key step is building the calibration model of measured data. In the present paper, ant colony algorithm was used to build the quantitative analysis model of Fourier transform near infrared diffuse spectroscopy for protein in cereal. Satisfied results were obtained. For calibration set, the correlation coefficient and relative standard deviation were 0.943 and 3.41%, respectively, while for prediction set, the correlation coefficient and relative standard deviation were 0.913 and 4.67%, respectively.

[Estimations of chlorophyll and water contents in live leaf of winter wheat with reflectance spectroscopy].

For estimating the yield of wheat, and for crop growth monitoring by using hyperspectral remote sensing, it is very important to quantitatively determine the chlorophyll and water contents in live leaf of winter wheat. In the range of 350-1650 nm, the reflectance spectrum of winter wheat's leaf in different growth periods were scanned by ASD FieldSpec Pro FR portable spectrometer and LI-COR 1800 integrating spheres. Partial least squares method was used to develop the quantitative analysis models for chlorophyll and water contents with reflectance spectroscopy. The model of chlorophyll content with reflectance spectroscopy was built in the range of 400-750 nm, and the results show that the correlation coefficient between the estimated values and the real values is 0.898, and relative standard deviation is 13.6%. The model of water content with reflectance spectroscopy was built in the range of 1400-1600 nm, and the results indicate that the correlation coefficient between the estimated values and the real values is 0.999, and relative standard deviation is 0.3%. These results are satisfying in agricultural production.

[Building artificial neural networks model on portable NIR integrity wheat component measuring apparatus].

The quantitative analysis model of protein in integrity wheat was built by three layers back propagation artificial neural networks for portable near infrared (NIR) integrity wheat component measuring apparatus. The structure diagram of integrity wheat component measuring apparatus, light route structure of apparatus and the spectrum of integrity wheat were given in the present paper. The theory of artificial neural network was briefly introduced and the results of quantitative analysis model of protein were given. For calibration set and prediction set, the correlation coefficient was 0.90 and 0.96 respectively; the relative standard deviation is 3.77% and 4.46% respectively. Because of the influence of light route structure, electrical circuit, and integrity sample forms on the measuring apparatus, some nonlinearity exists between the spectral parameters and chemical values. The results of artificial neural networks nonlinear model were superior to linear model.

[Simultaneous determination of glucose, fructose and sucrose in aqueous solution by short-wavelength near infrared spectroscopy].

The calibration model for simultaneous determination of glucose, fructose and sucrose in aqueous solution was built by partial least squares and short-wavelength near infrared spectroscopy (800-1,100 nm). Twenty five samples in calibration set and 9 samples in prediction set were designed by orthogonal design. Building models from calibration set and validation for prediction set obtained the better results. For concentrations of glucose, fructose and sucrose in the ranges of 12.23-61.14 mg x mL(-1), 12.50-62.50 mg x mL(-1) and 12.09-60.44 mg x mL(-1), the relative standard deviations in the calibration set are 1.43%, 4.51% and 1.59% respectively, and the relative standard deviations in the prediction set are 3.40%, 3.73% and 2.80% respectively. The advantages of the method are simple and effective, with low cost for the simultaneous determination of multi-component system. It may be applied in practice easily.

[Development of portable LED-based NIR integrity wheat component measuring apparatus].

A integrity wheat component quickly measuring apparatus which based on NIR LED was developed, by using new chip and designing software. While providing a lower cost apparatus, it also provides quick answer speed and smaller size for fieldwork of wheat's components measuring. The apparatus is made up of NIR LED, interference filter, lens, Si photronic detector, microprocessor system and calibration model. Compared with other instrument, the apparatus has many advantages, such as compact space, simple structure, low power waste, and anti-shake. This paper introduces the design of the apparatus, and predicts the concentration of wheat protein. The apparatus can realize non-damage measurement of wheat components' concentration on fieldwork.