The Impact Of Training Samples On The Accuracy Of Crop Remote Sensing Classification

With the development and maturity of remote sensing technology,it has been widely applicated in various fields,agricultural remote sensing is one of the widely applicated technologies in the field of remote sensing,for agricultural remote sensing,crop remote sensing classification technology is an important aspect of it.Accurate classification of remote sensing images and identification of the types of crops in remote sensing images,not only each classifier is required,but also training samples of the input classifier are required,the effect of the training samples on the classification accuracy is greater than the effect of the classifier on the classification accuracy,the number of and quality of the training samples are the main factors that affect the accuracy of the final classification results.In order to better research and analysis the influence of the number of and quality of the training samples on the classification accuracy,this paper chooses the Helen city in Heilongjiang Province as the research required experimentation area,using Landsat 8remote sensing images as the data source,on the basis of selecting the samples of different quantity and quality,the effects of the number of and quantity of training samples on the classification accuracy were studied respectively by using the maximum likelihood,neural network and support vector machine three kinds of methods,and several experiments on three kinds of classification methods.Finally,using kappa coefficient and the overall classification accuracy to evaluate the classification results.After quantitative analysis,the results show that: 1)in the training sample quality is relatively constant,the degree of response of the same classification method to the same number of training samples as well as the degree of response of the different classification methods to the number of training samples are different,and the classification accuracy has different degree of volatility,with the increase of the numberof training samples,the volatility will decrease,when the number of training samples reaches a certain degree,the mean of classification accuracy will tend to be relatively stable;2)the number of training samples is constant,the same classification methods as well as the different classification methods have different the degree of response to the training samples of the same quality grade;the degree of response of the same classification method to the different training samples quality level is also different.