A Method for Determining the Scene Category of Remote Sensing Images

Abstract

The invention relates to the fields of remote sensing image processing, image coding, machine learning, image scene classification, etc., and proposes a method for determining the scene category of remote sensing images based on Gaussian kernel density estimation saliency preprocessing and local constraint linear coding. With the continuous improvement of the resolution of satellite remote sensing images and aerial remote sensing images, more useful data and information can be obtained from remote sensing images; The important link in the processing, that is, how to determine the scene classification of remote sensing image attribution is particularly important. The method of the invention uses Gaussian kernel density estimation to preprocess the original image for saliency detection, and introduces local constraint linear coding to represent the image, and finally uses a support vector machine for image classification, combining the results of saliency detection and image processing. An encoding method that improves the performance of remote sensing image classification.

Description

technical field [0001] The invention relates to the fields of satellite remote sensing image processing, image data coding, machine learning, image scene classification, etc., and proposes a method based on Gaussian kernel density estimation saliency preprocessing and local constrained linear coding to determine the scene category of remote sensing image belonging. The field of computer application technology. Background technique [0002] Remote Sensing Image (Remote Sensing Image) refers to the film or photo that records various ground features through geomagnetic wave observation, mainly divided into aerial photos and satellite photos. With the continuous improvement of the resolution of satellite remote sensing images and aerial remote sensing images, more useful data and information can be obtained from remote sensing images. With the introduction of the concept of "Digital Earth", more and more civilian applications need to use remote sensing images, including resourc...

AI Technical Summary

Problems solved by technology

[0006] Although the image classification performance has been greatly improved after using the LLC method, psychological and biological research results have shown that there is an attention mechanism in the visual system of humans and primates, allowing them to focus on specific objects in the field of vision. areas, objects, and the current LLC method does not reflect this mechanism

Images