Research On Classification And Recognition Method Of Crop In Yellow River Basin

Planting industry in the Yellow River Basin is an important part of China’s agricultural economy.The high-quality development of agricultural economy in the Yellow River Basin supports the stable and sustainable development of China.In the actual agricultural production,it plays an important role in the refinement and precision of the agricultural production management of the Yellow River basin to quickly and accurately obtain the crop planting type information.At present,remote sensing images are widely used in agricultural management.However,the data images provided by different satellite sensors will be affected by the satellite hardware equipment and weather factors,and the data sources will present different spectral information and detail information.In order to realize the classification of crops in the Yellow River Basin,this paper selects Huiji District of Zhengzhou City as the research area,and carries out image fusion and image classification of remote sensing multispectral and panchromatic images,the main contents are as follows:1.I propose to integrate ICA with NSCT to sfigure out of remote sensing image.By NSCT transform,we can get the low-pass and band-pass image coefficients,Then,compounded the two low-pass image coefficients by ICA.The band-pass image coefficients are fused by the neighborhood coefficient difference and information entropy as the standard.Finally,to inverse trsnsformation by NSCT.It is found that the algorithm I proposed is better than ICA and NSCT,and the image performance of spectral and detail information is reformatived.2.For the second question,I proposed deep semantic.For toward to improve the degree of accuracy crop category,this paper proposes that the prior characteristic map with normalized index is combined into the input channel,and the Deep Lab v3+ network feature extraction module added by SEnet is optimized.The results show that it can classify wheat,forest land,rape and buildings,the accuracy of classification come up to 87.02%,it was improved compares with SVM、Seg Net and Deep Lab V3+.The image fusion method and remote sensing image classification model proposed in this paper can effectively classify the crops in the study area,provide technical support for the realization of crop classification in the whole Yellow River Basin,and provide technical support for the accurate and scientific management of agricultural production and the high-quality development of agriculture in the Yellow River Basin.