[Application of hyperspectral fluorescence image technology in detection of early rotten oranges].

Rottenness is most prevalent and devastating disease that threats citrus fruit. Automatic detection of early rottenness can enhance the competitiveness and profitability of the citrus industry. However, there is no efficient automatic detection technology at this time that could detect this disease. The navel orange was selected as research objective. Hyperspectral fluorescence imaging was used to detect early rottenness in orange. Optimum index factor (OIF) method was applied to identify the optimal band combination. 100% detection rate was achieved based on the optimal bands ratio image and two threshold values. The research showed that the proposed method can effectively overcome the affect from florescence effect because stem damage area and stem also can produce florescence under ultraviolet light. This study will lay a foundation for developing multispectral detection system used in on-line detection of early rottenness fruit.

[Advance on application of hyperspectral imaging to nondestructive detection of agricultural products external quality].

Hyperspectral imaging is an emerging technique that integrates conventional imaging and spectroscopy to obtain both spatial and spectral information from a studied object simultaneously. The images data can reflect the external features, surface defects and contamination. The spectra data can analyze physical structure and chemical composition in studied object. Therefore, hyperspectral imaging technology causes more and more attention, and has become a hot research topic recently. In order to track the latest research developments at home and abroad, application of hyperspectral reflectance and fluorescence imaging techniques to nondestructive detection of agricultural products external quality was reviewed, which would provide reference for application of hyperspectral imaging to agriculture.

[Research on the trash content measurement and classification of ginned cotton by using NIR spectroscopy technique].

Near infrared (NIR) spectroscopy was investigated to predict trash content and classify types of ginned cotton by using a fiberoptic in diffuse reflectance mode. Different spectra preprocessing methods were compared, and partial least-squares (PLS) regression was established to predict the trash content of ginned cotton. Discriminant analysis (DA) was used to classify various types of lint and content level of trash. The correlation coefficient r was 0.906 for optimal PLS model using three factors based on first-order derivative spectra, and RMSEC and RMSEP was 0.440 and 0.823 respectively. To classify ginned cotton with and without plant trash, the accuracy rate reached 95.4% using 15 principal components (PCs) via DA, whereas the prediction accuracy rate was only 80.9% for the classification of sample types due to containing foreign fiber, and the classification result for the content level of trash in lint was not good for the samples without any preprocessing. The result indicated that the NIR spectra of sample can be used to predict trash content in ginned cotton, which is often disturbed by type, content and distribution of foreign matters, and the accuracy of some prediction model is unsatisfactory. In order to improve the prediction accuracy, some methods would be applied in future research, such as pretreatment according to acquisition request of solid sample, or using transmission mode.

[Qualitative analysis of fragrant pear class based on near infrared diffuse reflectance spectroscopy].

A method was developed to automatically discriminate the persistent calyx fruit and fruit without calyx of fragrant pear by means of near infrared spectroscopy (NIRS). The prediction performance of different band regions range, different principal component numbers and different preprocessing methods of the spectra (multiplicative signal correction, standard normal variate, and derivative spectra) together with discriminant analysis (DA) was also investigated, and The calibration model was established to classify the different kinds of fragrant pear. The research results for the fragrant pear classification showed that DA calibration models using these parameters with band regions between 9 091 and 4 000 cm(-1) and original spectra are optimal, with the percentage of correct sample classification being 100% and 95% for the calibration and validation set, respectively.

[Identification of tea from different regions using X-ray fluorescence].

The existence of fake tea from non-origin seriously impacts on the credibility of the famous tea. A method was developed to identify tea from difference regions on the basis of the fact that the content of heavy metals in different origin tea is varied by using X-ray fluorescence technique and pattern recognition technique. Samples from different origins were grouped respectively, and their X-ray fluorescence spectra were acquired, and then the principal components of these spectral data were calculated, and the average of the principal components of each group was used as the center of each group. The Mahalanobis distance value between a sample and the center of a group were calculated, when the Mahalanobis distance value reached minimum, the sample was classed to current group, and in this way, a sample was identified. A Niton 792 portable X-ray spectrometer was used to class 120 tea samples from Anji, Jinhua, Hangzhou and Taizhou, in zhejiang province of China. It was found that the spectra between 3 and 13 KeV and the first 4 principal components give enough information for the identification of tea from different regions,and the rate of error was 4.2%.

[Advance in nondestructive detection of fruit internal quality based on hyperspectral imaging].

The present paper reviews the development in the field of hyperspectral imaging technology for nondestructive detection of fruit internal quality in recent years up to the year 2007. With the increasing maturity of hyperspectral imaging technology, decline of cost for its hardware and software, and improvement in hyperspectral image data processing algorithms, hyperspectral imaging technology for fruit quality nondestructive detection has become a hot research topic. In order to track the latest research developments at home and abroad, the fruit internal quality (maturity, firmness, soluble solid content, water content) detection with hyperspectral imaging was reviewed, which would provide reference for Chinese researchers.

Identification of rice seed varieties using neural network.

A digital image analysis algorithm based color and morphological features was developed to identify the six varieties (ey7954, syz3, xs11, xy5968, xy9308, z903) rice seeds which are widely planted in Zhejiang Province. Seven color and fourteen morphological features were used for discriminant analysis. Two hundred and forty kernels used as the training data set and sixty kernels as the test data set in the neural network used to identify rice seed varieties. When the model was tested on the test data set, the identification accuracies were 90.00%, 88.00%, 95.00%, 82.00%, 74.00%, 80.00% for ey7954, syz3, xs11, xy5968, xy9308, z903 respectively.

Firmness evaluation of melon using its vibration characteristic and finite element analysis.

The "Huang gua" melons were measured for their physical properties including firmness and static elastic modulus. The vibrational characteristics of fruits and vegetables are governed by their elastic modulus (firmness), mass, and geometry. Therefore, it is possible to evaluate firmness of fruits and vegetables based on their vibrational characteristics. Analysis of the vibration responses of a fruit is suggested for measuring elastic properties (Firmness) non-destructively. The impulse response method is often used to measure firmness of fruits. The fruit was excited using three types of balls (wooden, steel and rubber) and the vibration is detected by an accelerometer. The Instron device was used to measure the static elastic modulus of the inner, middle and outer portions of melon flesh. Finite element (FE) technique was used to determine the optimum excitation location of the chosen measurement sensor and to analyze the mode shape fruits. Four types of mode shapes (torsional or flexural mode shape, first-type, second-type spherical mode and breathing mode shape) were found. Finite element simulation results agreed well with experimental results. Correlation between the firmness and resonant frequency (r2=0.91) and between the resonant frequency and stiffness factor (r2=0.74) existed. The optimum location and suitable direction for excitation and response measurement on the fruit were suggested.