Image complexity judgment-based scene classification method for aerial remote sensing images

Abstract

The invention discloses an image complexity judgment-based scene classification method for aerial remote sensing images, and belongs to the technical field of image processing. The method comprises the steps of extracting complexity features of the aerial remote sensing images, and selecting multiple to-be-processed aerial remote sensing images to form training samples; forming a test sample set by using the rest of the aerial remote sensing images; manually classifying and tagging a training sample set according to the complexity, and performing combination with classifiers after multi-core mapping to obtain three image complexity classifiers; extracting the complexity features of the aerial remote sensing images A in the test set, and inputting the complexity features to the three classifiers, thereby obtaining a complexity judgment result of the aerial remote sensing images A, wherein a corresponding type when the hinge loss is minimum serves as a complexity type of the images; andaccording to the complexity judgment result of the aerial remote sensing images A, performing scene classification on the aerial remote sensing images A by using a proper method. According to the method, the complexity of the aerial remote sensing images can be effectively judged, thereby efficiently and accurately realizing scene classification of the aerial remote sensing images.

Description

technical field [0001] The invention belongs to the technical field of image processing, and in particular relates to an aerial remote sensing image scene classification method based on image complexity judgment. Background technique [0002] In recent years, with the continuous progress of the domestic aviation, electronics and information industries, UAV-related technologies are also developing rapidly, which makes the types of aerial remote sensing images richer, the quality is improved, and the application is more extensive. Aerial remote sensing images are the main means of obtaining basic geographic information, the basis for surveying and mapping, and the main source of data for relevant departments to obtain original surface information. Environmental protection and scientific research are closely related. [0003] Scene classification of images is an important branch in computer vision. For aerial remote sensing images, the result of scene classification represent...

AI Technical Summary

Problems solved by technology

[0008] The present invention is aimed at different aerial remote sensing images, the content of which is quite different, and a single method may generate large time redundancy when classifying simple images, and may not be accurate enough when processing complex images

Images