CNN边缘提取在冰雹预测中的应用  

Study on CNN Edge Extraction in Hail Prediction

作　　者：李冰洁[1] 李国东[1] LI Bing-jie LI Guo-dong(School of Applied Mathematics, Xinjiang University of Finance and Economics, Xinjiang Urumqi, 830012)

机构地区：[1]新疆财经大学应用数学学院,新疆乌鲁木齐830012

出　　处：《淮阴师范学院学报（自然科学版）》2016年第3期228-232,共5页Journal of Huaiyin Teachers College;Natural Science Edition

基　　金：国家自然科学基金项目(11461063);国家社科基金项目(14BTJ021);教育部人文社会科学基金项目(13YJAZH040);新疆维吾尔自治区高校科研计划项目(XJEDU2013I26)

摘　　要：冰雹是一种破坏性较为严重的自然灾害之一.结合细胞神经网络理论对雹云进行了预测研究,选取新疆几个地区的冰雹云层图像进行分析,提供了一种简单的利用细胞神经网络边缘探测方法来判别云层图像是否降雹.在MATLAB程序进行边缘探测得到图像对应的像素矩阵数据,将矩阵数据处理得到每个图像的像素强度,像素强度小于0.8的是降雹云层,大于0.9的是无雹云层.经验证细胞神经网络理论在冰雹的预测识别中是有效的.The hail is one of the most serious a devastating natural disaster. Bring huge losses to agriculture, architecture, communication, electric power, transportation and people＇s life and property. Chinese hail disas- ters occur frequently state, we need as soon as possible to prevent hail. This paper combines.the theory of cel- lular neural network prediction research of hail cloud. Selection of hail clouds image in several areas of Xin- jiang are analyzed, and Provides a simple use of cellular neural network edge detection method to identify whether the image of hail cloud. Edge detection of image data corresponding to the pixel matrix in the MAT- LAB program, the matrix data processing to get the pixel intensity of each image, pixel intensity is less than 0. 8 is hail clouds, greater than 0. 9 is no hail cloud. Results show that using cell neural network theory in the prediction of hail identification is effective

关 键 词：细胞神经网络 冰雹预报 边缘检测 MATLAB 

分 类 号：P412.25[天文地球—大气科学及气象学]