Xinjiang Weather Modification Office

Xinjiang, China

Xinjiang Weather Modification Office

Xinjiang, China
SEARCH FILTERS
Time filter
Source Type

Fuxin Z.,Population and Family Planning Bureau of Tancheng County | Guodong L.,Xinjiang University of Finance and Economics | Wenxia X.,Xinjiang Weather Modification Office
Proceedings - 2016 International Conference on Smart City and Systems Engineering, ICSCSE 2016 | Year: 2016

At present, the desertification disaster is one of the most serious environmental problems in Xinjiang. In order to reduce the huge losses of desertification disaster to the economic development of Xinjiang, at the same time in order to stability and unity of the people of all ethnic groups in Xinjiang, it is necessary to forecast the future trend of desertification changes. In this paper, taking Ruoqiang basin as an example, we establish desertification prediction model. First, use exponential smoothing model to carry on the numerical simulation for desertification land, the trend of desertification of land, non desertification land area in the area in 2000-2011 years, Then, we examine forecasts by the MAPE, Desertification of land's MAPE value is 1.96%, have land desertification the land that having a trend of desertification's MAPE value is 7.13%, The predicted result achieves high accuracy prediction effect. Secondly, use exponential smoothing model to predict the land desertification, the desertification trend of land, non desertification land in 2012-2025 years, we again proved that exponential smoothing model has a huge advantage in the prediction of desertification trends. Finally draws the conclusion: The exponential smoothing model can predict the change trend of desertification land, having the trend of desertification land, non desertification land in the future. The research can provide a powerful reference and guidance for the prevention and control of desertification in the future, but also has theoretical meaning for desertification research in the future. © 2016 IEEE.


Xiaoting W.,Xinjiang University of Finance and Economics | Guodong L.,Xinjiang University of Finance and Economics | Shuo L.,Xinjiang University of Finance and Economics | Wenxia X.,Xinjiang Weather Modification Office
Proceedings - 2016 International Conference on Smart City and Systems Engineering, ICSCSE 2016 | Year: 2016

The hail is a serious disaster weather of hail cloud with sudden, moving rapidly, the history of life is short, etc. characteristics. These characteristics make the artificial hail forecast is very difficult, so the research is now a hot spot. This paper designs the hail cloud recognition model, the probability of discrimination method, according to the measure analysis and extraction of hail cloud and hail cloud image data, analysis of hail cloud images and non hail cloud image feature data. The measurement and analysis of the extraction of the relevant statistics as the largest probability discriminant method of variable data, tectonic discriminant function. Using this model to Xinjiang Aksu area samples were detected. Know the discriminant method for hail cloud recognition non hail cloud images with higher resolution, and judging the lower error rate. The experimental results show that, through the measure analysis of weather radar reflectivity image of echo intensity and the maximum probability discriminant method combining structure to identify hail cloud model and has a good effect. © 2016 IEEE.


Wenxia X.,Xinjiang Weather Modification Office | Guodong L.,Xinjiang University of Finance and Economics | Jun T.,Aksu Weather Modification Office
Proceedings - 2016 International Conference on Smart City and Systems Engineering, ICSCSE 2016 | Year: 2016

One Belt And One Road strategy has brought new opportunities for agricultural development in Xinjiang. Xinjiang should seize the opportunity to speed up the agricultural development. Hailstorm will brought huge losses to agricultural production in Xinjiang. This paper put forward hail cloud identifying new method which combine the radar image intensity and Support Vector Machine (Support Vector Machine, SVM). Extract the green, yellow and red area of the heavy rain and hail cloud, the ratio of yellow green area and the ratio of red yellow area, constitute two-dimensional recognition vector, training SVM model. Compared this method with the existing SVM model [1], the hail cloud prediction accuracy increased by 8.25%, which is equivalent to reducing 18003.7 acres of agricultural losses in Aksu area in 2011. It shows that the new method proposed in this paper can effectively prevent the happening of hailstorm of Aksu area, and reduce the influence of the hailstorm agricultural development in Aksu region, and the smooth implementation of the strategy of ' One Belt And One Road '. © 2016 IEEE.


Li G.,Xinjiang University of Finance and Economics | Yang B.,Xinjiang University | Pu Y.,Xinjiang University | Xu W.,Xinjiang Weather Modification Office
International Journal of Pattern Recognition and Artificial Intelligence | Year: 2017

The problem of the secure transmission of digital image has paid more and more attention to the network and this paper designs a special image encryption scheme. Image encryption scheme is designed based on the hyper chaos of generalized five-order Henon mapping and five-order cellular neural network (CNN) system. Firstly, the chaotic sequence (Formula presented.), which is regarded as the initial conditions of CNN system, is generated by the five-order generalized Henon mapping. Then another chaotic sequence (Formula presented.) is produced by the CNN system. At last, the cipher image is generated by the transformation of random sequence (Formula presented.) and the original image. Toward the end, the paper makes the simulation experiment and draws a conclusion that the algorithm of image encryption has strong attack resistance, good safety, and suitable to spread in the network through analyzing the statistical characteristics of the image information entropy, correlation and histogram as well as the key space and the sensitivity. © 2017 World Scientific Publishing Company


Guodong L.,Xinjiang University of Finance and Economics | Wenxia X.,Xinjiang Weather Modification Office | Bing Y.,Xinjiang University of Finance and Economics | Awudong B.,Xinjiang University of Finance and Economics | Xiaojuan C.,Harbin Normal University
Cluster Computing | Year: 2016

A hailstorm forecast method is proposed in the paper. Edge Detect Cellular Neural Network (EDCNN) method is used to extract the edge of cloud radar images. We have detected the texture of the cloud images. Then the texture image has been processed with wavelet transform. The hail data information from the image has been found. We will get approximate detail coefficients, level detail coefficients, vertical detail coefficients, diagonal detail coefficients, and reconstructed coefficients. Construct hail cloud life feature vector matrix to explain the problem. Found the corresponding rules through the five coefficients. At last, through the simulation experiment achieve the purpose of hail forecast. A feature vector of hail cloud life has been constructed, the rules of hail had been found from the feature vector. And Compared with the contour variance of the hail cloud inner and outer, we will find this paper puts forward the method is more effective. The conclusion is rationality according the simulation experiment verifies. © 2016 Springer Science+Business Media New York


Li G.,Xinjiang University of Finance and Economics | Xu W.,Xinjiang Weather Modification Office | Xu W.,Chengdu University of Information Technology | Wang X.,Xinjiang Weather Modification Office
Journal of Convergence Information Technology | Year: 2012

Hail identification is one of the most important steps for weather forecast. In this paper, we first present a robustness design theorem for the edgegray detection cellular neural network (EDGE CNN). Then we process some cloud radar images by the EDGE CNN, and the veins of the radar image had been pickup from the figure. The polynomial fitting has been used to analysis the veins figure. Detailed experimental results show that the proposed hail cloudy identification scheme based on CNN can provide more accurate performance to diagnose the cloudy.


Li G.,Xinjiang University of Finance and Economics | Xu W.,Chengdu University of Information Technology | Wang Y.,Xinjiang Weather Modification Office
Lecture Notes in Electrical Engineering | Year: 2013

In this paper, In terms of the data of stratiform could microphysical structure from airborne particle measurement system (PMS) in 1986-1996 in Xinjiang province. It is a way to detect the time for work in artificial precipitation by the total volume particle. We use traditional way to find the total rain in the cloudy. And then we find the relationship between the time and the total. It will be fit for work when the number is rising and the mean of the rain is more. So we can find the top point in the image, it is the best time to work on artificial rain. © 2013 Springer Science+Business Media.


Guodong L.,Xinjiang University of Finance and Economics | Wenxia X.,Xinjiang Weather Modification Office | Wenxia X.,Chengdu University of Information Technology | Xu W.,Xinjiang Weather Modification Office | Liming D.,University of Regina
Advances in Information Sciences and Service Sciences | Year: 2012

Hail identification is one of the most important steps for weather forecast. In this paper, we had provided a new way to diagnose the cloud that will hail or not. The work had been combined the Cellular Neural Network (CNN) theory and polynomial theory. First present is that a robustness design theorem for the Edge gray Detection Cellular Neural Network (EDGE CNN) had been mentioned. Second, we process some cloud radar images by the EDGE CNN, the veins of the radar image had been pickup from the figure. At last, the polynomial fitting has been used to analysis the veins figure. The parameter of the polynomial will indicate the cloud hail or not. Our findings show that CNN and Polynomal should provide a useful tool to diagnose the hail cloudy.


Li B.,Xinjiang University of Finance and Economics | Xu W.,Xinjiang Weather Modification Office | Li G.,Xinjiang University of Finance and Economics | Wangxu,Xinjiang Weather Modification Office
Proceedings - 2015 International Conference on Intelligent Transportation, Big Data and Smart City, ICITBS 2015 | Year: 2015

Using Cellular Neural Network edge detection to the image of hail and non hail get the features of extraction edge of the image, get the main feature of image.Two-dimensional wavelet transform for edge detection, get the four discriminant features of each image; With feature data analysis the difference of hail and non hail image ; Using the Fisher discriminant method to construct the discriminant model, Select the xinjiang region of radar chart analysis. Results show that the method used in this article can be a clear distinguish hail and non hail, and the discriminant error is small, with good results. © 2016 IEEE.


Tong Z.,Xinjiang University of Finance and Economics | Li G.-D.,Xinjiang University of Finance and Economics | Xu W.-X.,Xinjiang Weather Modification Office
Proceedings - 2014 5th International Conference on Intelligent Systems Design and Engineering Applications, ISDEA 2014 | Year: 2014

This paper argues that the best effect to remove Gaussian noise is to use wiener filtering, and to remove salt & pepper noise to use median filtering will get a better effect. By using the correlation index, and through the original image adding noise and removing noise, it calculates the correlation index of the removal-noise image of the original image is better than traditional methods which used as average filtering and median filtering and wiener filtering to delete the noise of an image. Specifically, this research paper puts forward two results: one is to provide the add noise image first and then to remove the noise, and then to use CNN to detect the image edge, the other is to provide the noise image first by using CNN to detect edge and then to remove the noise. Via these two results compared with the result of the original image edge detection, the conclusion will be as following: in order to avoiding the impact of noise bring to an image, before the image edge detect, one must deal with the noise first. © 2014 IEEE.

Loading Xinjiang Weather Modification Office collaborators
Loading Xinjiang Weather Modification Office collaborators