Xu S.-W.,Chinese Academy of Agricultural Sciences |
Xu S.-W.,Key Laboratory of Agricultural Information Service Technology |
Li G.-Q.,Chinese Academy of Agricultural Sciences |
Li G.-Q.,Key Laboratory of Agricultural Information Service Technology |
And 2 more authors.
Journal of Integrative Agriculture | Year: 2015
The primary goal of Chinese agricultural development is to guarantee national food security and supply of major agricultural products. Hence, the scientific work on agricultural monitoring and early warning as well as agricultural outlook must be strengthened. In this study, we develop the China Agricultural Monitoring and Early-warning System (CAMES) on the basis of a comparative study of domestic and international agricultural outlook models. The system is a dynamic and multi-market partial equilibrium model that integrates biological mechanisms with economic mechanisms. This system, which includes 11 categories of 953 kinds of agricultural products, could dynamically project agricultural market supply and demand, assess food security, and conduct scenario analysis at different spatial levels, time scale levels, and macro-micro levels. Based on the CAMES, the production, consumption, and trade of the major agricultural products in China over the next decade are projected. The following conclusions are drawn: i) The production of major agricultural products will continue to grow steadily, mainly because of the increase in yield. ii) The growth of agricultural consumption will be slightly higher than that of agricultural production. Meanwhile, a high self-sufficiency rate is expected for cereals such as rice, wheat, and maize, with the rate being stable at around 97%. iii) Agricultural trade will continue to thrive. The growth of soybean and milk imports will slow down, but the growth of traditional agricultural exports such as vegetables and fruits is expected to continue. © 2015 Chinese Academy of Agricultural Sciences.
Sun J.,Jiangsu University |
Sun J.,Key Laboratory of Agricultural Information Service Technology |
Zi L.X.,Jiangsu University |
Zhang X.,Jiangsu University |
And 3 more authors.
International Agricultural Engineering Journal | Year: 2015
It has practical values to classify greengrocery seeds efficiently, accurately and automatically. In this paper, greengrocery seeds were taken as samples, and four varieties of seeds were bought from market such as Dwarf ji Suzhou Qing, Green collar Wuta CAI, Green Collar May Slow and Qibao vegetables. FieldSpec®3 Spectrometer was used to collect near infrared spectroscopy of four varieties of greengrocery seeds. Firstly, multiplicative scatter correction (MSC) was used to preprocess the raw data. Then the principal component analysis (PCA) was used to reduce the dimensionality of the preprocessed data. Finally, grid search algorithm, genetic algorithm, and particle swarm optimization algorithm were introduced to support vector machine's (SVM, Support Vector Machine) modeling process, and the penalty parameter c and the optimal kernel function parameters g of the SVM model were searched automatically based on the algorithm in order to construct the optimal classification model. The several questions were studied that whether the raw data were preprocessed, whether PCA dimensionality reduction was made and whether to optimize the parameters of the SVM and which optimization methods was chosen. From experiment results, it can be seen that, the model based on the data processed by the MSC+PCA and the grid search parameter optimization method is the best, whose modeling time was 1.6 s and the total accuracy rate reached 99.59%. The results showed that, near infrared spectroscopy technology can be used to distinguish the greengrocery seeds quickly, accurately, and non-destructively.
Jun S.,Jiangsu University |
Jun S.,Key Laboratory of Agricultural Information Service Technology |
Shuying J.,Jiangsu University |
Meixia Z.,Jiangsu University |
And 3 more authors.
Journal of Residuals Science and Technology | Year: 2016
Nonstandard use of pesticides often causes poisoning incidents of silkworm, which is a serious threat to the development of sericulture industry. In view of this, it is very urgent to study new non-destructive testing methods that can detect pesticide residues in mulberry leaves rapidly and accurately. In this paper, six groups of mulberry leaves (144 mulberry leaves in total), on which chlorpyrifos pesticide of six different concentrations had been sprayed respectively, were chosen as experimental samples, and their hyperspectral images in 390-1050 nm were acquired by hyperspectral imaging devices. The region of interest from hyperspectral image was selected, and five sensitive wavelengths including 561.25, 680.86, 706.58, 714.32, and 724.66 nm were determined by correlation coefficients between pesticides residues and spectral reflectances. Further, based on multiple linear regression (MLR) and support vector regression (SVR), the prediction models of pesticide residues in mulberry leaves were established respectively to fit the experimental data. The results showed that the root mean square error (RMSE) and coefficient of determination (R2) of prediction set of MLR model were 47.165 and 0.637 respectively, and those of prediction set of SVR model were 27.719 and 0.874 respectively. Therefore, hyperspectral imaging technology together with SVR prediction model could accurately detect the pesticide residues in mulberry leaves. © 2016 DEStech Publications, Inc.
Sun J.,Jiangsu University |
Sun J.,Key Laboratory of Agricultural Information Service Technology |
Jin X.,Jiangsu University |
Mao H.,Jiangsu University |
And 4 more authors.
Nongye Gongcheng Xuebao/Transactions of the Chinese Society of Agricultural Engineering | Year: 2014
In this study, we developed a fast and non-destructive technology for the prediction of the nitrogen content in lettuce leaves based on hyperspectral imaging technology which contains abundant spectral and spatial information in an object. First, hyperspectral images of lettuce leaves in the visible and near infrared (390-1050 nm) regions were acquired by the hyperspectral imaging system, and then the corresponding nitrogen content in the lettuce leaves were obtained by Kjeldahl method successively. The binary mask image was successfully determined by the method of dividing the image of very high reflectance intensity by the image of very low reflectance intensity with a certain threshold, and ROI (Region of Interest) in the sample of lettuce leaf was determined by removing the regions of noise using the acquired binary mask image. As the hyperspectral imaging technology provided much more information including spectral and spatial information for all the samples of lettuce leaves, and in which some information is noisy and redundant. In fact, this leads to the difficulty of meeting the needs for fast and efficient detection of some objects. So it is very hard to be directly used for on-line industrial application in our daily life. Therefore, effective selection of several characteristic wavelengths is necessary for the hyperspectral images. In this paper, the initial investigation was carried out by using a principal component analysis (PCA) to identify a number of potential characteristic wavelengths (662.9 nm, 711.7 nm, 735.0 nm, 934.6 nm) according to the weight coefficient distribution curve of the first three principal component images (PC1, PC2, PC3) under the full wavelengths. Both spectral data and texture data based on a co-occurrence matrix were extracted from the four characteristic wavelength images on the ROI, and the texture data of the first three principal component images were also extracted simultaneously. Then spectral data from four characteristic wavelength images, texture data (from four characteristic wavelength images, from the first three principle component images) and the combined data were utilized to develop different SVR (support vector machine regression) models to predict the nitrogen content in the lettuce leaves respectively. According to the performance of all the SVR models in the calibration set and the prediction set, the experiment results show that, from the calibration performance index, the model based on spectral data combined texture data from four characteristic wavelength images is the best with a coefficient of determination (RC 2 =0.996) and the root-mean-square error (RMSEC) of 0.034. From the prediction performance index, however the model based merely on spectral data is the best with a coefficient of determination (RP 2=0.86) and the root-mean-square error (RMSEP) of 0.22. This study provides valuable information for rapid and non-destructive nitrogen detection in crops.
Xu L.,Chinese Academy of Agricultural Sciences |
Xu L.,Key Laboratory of Agricultural Information Service Technology |
Zhang Q.,Chinese Academy of Agricultural Sciences |
Zhang Q.,Key Laboratory of Agricultural Information Service Technology |
And 10 more authors.
Intelligent Systems and Decision Making for Risk Analysis and Crisis Response - Proceedings of the 4th International Conference on Risk Analysis and Crisis Response, RACR 2013 | Year: 2013
Drought catastrophe risk assessment is imperative for the steady development of agriculture under the context of global climate change, and meanwhile, it is an urgent scientific issue need to be solved in risk assessment discipline. This paper developed the methodology of drought catastrophe risk assessment, which can be shown as the process of crop loss collection, Monte Carlo simulation, the Generalized Extreme Value distribution (GEV) fitting, and risk calculation. Data on crop loss were collected based on hectares covered by drought disaster, hectares affected by drought disaster, and hectares destroyed by drought disaster using the standard equation. Monte Carlo simulation based on appropriate distribution was used to expand sample size to overcome the insufficiency of crop loss data. Block Maxima Model (BMM) approach based on the extreme value theory was for modeling the generalized Extreme Value Distribution (GEV) of drought catastrophe loss, and then drought catastrophe risk at the provincial scale in China was calculated. The Type III Extreme distribution (Weibull) has a weighted advantage of modeling drought catastrophe risk for grain production. The impact of drought catastrophe to grain production in China was serious, and very high risk of drought catastrophe mainly occurs in the northeast regions of China. Given the scenario of suffering once-in-a-century drought disaster, for majority of the major-producing provinces in China, the probability of 15% reduction of grain output is more than 90%. Our findings can provide multifaceted information about drought catastrophe risk that can help to guide management of drought catastrophe. © 2013 Taylor & Francis Group.
Zhu M.S.,Chinese Academy of Agricultural Sciences |
Zhang J.H.,Key Laboratory of Agricultural information Service Technology
Applied Mechanics and Materials | Year: 2014
Accurate prediction of agricultural prices is beneficial to properly guide the circulation of agricultural products and agricultural production and realize the equilibrium of supply and demand of agricultural area. On the basis of wavelet neural network, this paper, choosing tomato prices as study object, tomato retail price data from ten collection sites in Hebei province from January 1, 2013 to December 30, 2013 as samples, builds the tomato price time series prediction model to test price model. As the results show, model prediction error rate is smaller than 0.01, and the correlation (R2) of predicted value and actual value is 0.908, showing that the model could accurately predict tomatoes price movements. The establishment of the model will provide technical support for tomato market monitoring and early warning and references for related policies. © (2014) Trans Tech Publications, Switzerland.