Remote sensing image vegetation classification method based on gradient scale interval change rule operator

Abstract

The invention discloses a remote sensing image vegetation classification method based on a gradient scale interval change rule operator. According to the present invention, the gradient scale intervalchange rule operator is established and can describe the characteristics of different vegetations in a remote sensing image from a non-directional specific brightness interval and various scales, andthen the characteristics are learned by using a support vector machine model to obtain a classification model, and the classification result of the remote sensing image can be obtained by using the gradient scale interval change rule operator and the classification model. By utilizing the method, the influence of a specific scale and a specific direction on classification can be avoided, the characteristics of the vegetation on the remote sensing image are displayed according to the interval characteristics of the specific brightness pixel of the vegetation on the image and the change characteristics of the scale, and further the precision of the remote sensing image vegetation classification is improved by utilizing the characteristics.

Description

Technical field:[0001]The invention discloses a remote sensing image vegetation classification method based on a gradient scale interval change rule operator, which belongs to the technical field of remote sensing images.Background technique:[0002]Remote sensing images can obtain vegetation distribution data in a large area. These distributions can well reflect agricultural production, environment and vegetation protection, the number of specific plant populations, and the vegetation succession in specific areas. These conditions are formulated The data basis for socio-economic development and environmental protection strategies, so it is of great value to obtain vegetation types in a specific surface area through remote sensing images. Automatic identification of vegetation types in remote sensing images is the most important way to quickly obtain these data.[0003]Automated remote sensing image vegetation classification currently mainly uses two categories: the first category is th...

AI Technical Summary

Problems solved by technology

[0003] Currently, there are two main methods used for automatic remote sensing image vegetation classification: the first is the traditional shallow intelligent classification method, which uses auxiliary texture algorithms (such as gray-level co-occurrence matrix) to obtain the texture of vegetation in remote sensing images. features, and then use shallow classification models (such as neural networks, support vector machines, and decision trees) to classify images; such methods have low computational complexity and are easy to implement, but a key problem with this method is that The shooting of remote sensing images will be affected by terrain, sensors, vegetation distribution direction, and light direction. Each texture algorithm needs to conduct a large number of experiments on scale and direction parameters; and there may also be textures displayed by different ground objects in the same image. Inconsistent scales make it impossible to find better parameters, so it is difficult for this type of method to obtain good classification results

The second category is the deep learning method, which introduces the convolutional neural network architecture to classify vegetation. This type of method can obtain better classification results, but on the one hand, the series of deep learning methods require a large number of sample supports, and the vegetation area itself is small in many areas. It is difficult to support a large sample size, and the deep learning algorithm cannot play a role when there are few samples; on the other hand, the convolution is also affected by the input scale. If there are inconsistent texture scales displayed by different ground objects in the same image, the same The choice of convolution scale for deep learning neural network will also encounter difficulties

Images