The Research Of The Method Of The Well-facilitated Farmland Identification Based On The High-resolution Remote Sensing Imagine

Farmland is the carrier of agricultural development,because all activities closely related to agriculture are attached to it.Farmland provides human with food security,and it is also a fundamental agricultural resources.With the rapidly increase and development of population and the urbanization,there is a huge needs in construction land.At the same time,because of some uncontrollable factors such as natural disasters,a large number of available lands reduce and the quality of the lands also reduced rapidly.Facing the shortage of grain caused by the reduction of the quality and quantity,we must guarantee the food production by improving the quality of our lands and provide plenty of food for humans to make sure the development of human society.Recently,as we strongly advocate the construction of well-facilitated farmland,it’s area has a trend of the multiple growth.In order to achieve the better guide of the construction of well-facilitated farmland,the steadily increase and the safety guarantee in grain production,we should monitor and manage it properly and reasonably.Nowadays,the method of judging the well-facilitated farmland lacks unified understanding,which still needs more research.Remote sensing technology has the advantages of short revisit cycle and low cost,which all play a unique role in the investigation,evaluation,monitoring and management of agricultural production.This thesis aims at extracting the remote sensing image information and forming the well-facilitated farmland index with the remote sensing technology and the object-oriented classification.In the end the thesis comes up with the method to identify the well-facilitated farmland based on high resolution remote sensing.What specific contents the thesis has is as follows:multi-source data of farmland to classify the remote sensing image and extract the information.Secondly,combined the general principles with other relevant regulations and standards on the well-facilitated farmland,the thesis forms the comprehensive index system accorded with the well-facilitated farmland,consisting of three factors,a total of 28 indicators.According to exact situation of the study area,the thesis screens and simplifies the comprehensive index,choosing a suitable indicators that consists of three factors,a total of five indicators.Then it forms the final identification index system which is suitable for the study area by normalizing each index and making use of the order of influencing factors sorting and AHP to endow weight for each index.Finally,using the identification index and the index weight,the thesis calculates the result extracted from the study area and forms the comprehensive identification index system.Then the thesis tests the accuracy of identification index by combining the sample data with the result of index,with an accuracy of 80.78%.As a result,the method of the well-facilitated farmland identification suggested in this thesis has high accuracy and reliability,which can be provide as the important reference for the well-facilitated farmland identification.