A method and system for estimating the number of ground object categories in remote sensing images

Abstract

The present invention provides a method and system for estimating the number of object categories in remote sensing images. The method includes: using the fuzzy clustering method to segment the remote sensing image to be segmented to obtain the best fuzzy membership degree and the best fuzzy clustering center; The information entropy and the upper limit of the information entropy of the object category. If it is greater than the upper limit of the information entropy, the split operation is performed, otherwise the Euclidean distance between the best fuzzy clustering centers is calculated; if it is less than a given threshold, the two object categories If it is similar, perform a merge operation; otherwise, it is not similar to obtain the estimation result of the number of object categories and the final segmentation result. The invention can be combined with the traditional image segmentation method to effectively estimate the number of object categories in the remote sensing image. Based on the information entropy to measure the characteristics of the amount of information in the object category, define the split condition, and use the Euclidean distance to describe the difference between different object categories, define the merge condition, and can split the clustering algorithm that cannot be effectively distinguished in the split operation. region to improve segmentation accuracy.

Description

technical field [0001] The invention belongs to the technical field of image processing, in particular to a method and system for estimating the category number of ground objects in remote sensing images. Background technique [0002] Determining the number of categories in an image is an important task in the process of image segmentation. The current category estimation methods are mainly divided into two categories. One is to manually determine the number of categories, but this method is not suitable for high-resolution large-scale images with obvious details; The other is to automatically determine the number of categories through the design method. At present, these methods are designed with the goal of automatically determining the number of image categories and region segmentation at the same time. This design method is relatively simple in principle, but it is highly targeted and not universal. In addition, there are many thresholds to control the change of the numb...

AI Technical Summary

Problems solved by technology

[0002] Determining the number of categories in an image is an important task in the process of image segmentation. The current category estimation methods are mainly divided into two categories. One is to manually determine the number of categories, but this method is not suitable for high-resolution large-scale images with obvious details; The other is to automatically determine the number of categories through the design method. At present, these methods are designed with the goal of automatically determining the number of image categories and region segmentation at the same time. The principle of this design method is relatively simple, but it is highly targeted and not universal. In addition, there are many thresholds for controlling the change of the number of classes, and the interaction between multiple thresholds makes it difficult to obtain ideal estimation results, and requires a large time cost, it is difficult to process large data, and its practicability is not strong

Images