Kernel sparse representation-based fast remote sensing target detection and recognition method

Abstract

The invention discloses a kernel sparse representation-based fast remote sensing target detection and recognition method. The method includes the following steps that: S1, four RGB characteristic channels are created; S2, the four-phase Fourier transformation of the four characteristic channels of a given image is calculated, a phase spectrum is extracted, the images of the four characteristic channels are reestablished through inverse Fourier transformation, and a saliency map can be generated; S3, binaryzation division is performed on the saliency map obtained in the step S2, and candidate regions of interest are extracted; S4, a search box is scanned through an effective sub-window search algorithm, so that image blocks to be detected are obtained, so that a remote sensing target image block training set is obtained; S5, SIFT features are extracted from the remote sensing target image block training set, and a sparse dictionary is generated; S6, a spatial pyramid is adopted to map the SIFT features; S7, kernel sparse representation is obtained; S8, the kernel sparse representation is solved; S9, the space pyramid vector representation of a target is performed; and S10, a linear support vector machine classification algorithm is used in combination to complete a recognition task.

Description

technical field [0001] The invention relates to the field of remote sensing image analysis, in particular to a fast remote sensing object detection and recognition method based on kernel sparse representation. Background technique [0002] Due to the complexity of remote sensing images, it is a difficult research direction to detect multiple types of objects of interest in the entire large-format remote sensing images. In the field of computer vision, the BOVW model has been widely studied and applied to image classification and pattern recognition tasks. In recent years, the BOVW model has also been introduced into the field of remote sensing target detection and recognition, and has achieved good performance. However, the methods proposed so far neither preprocess the large-format remote sensing images, that is, predict the target area of ​​interest, before implementing the detection and recognition algorithm, nor do they fully capture the similarity between the extracted...

AI Technical Summary

Problems solved by technology

[0003] However, the above methods neither preprocess the large-format remote sensing images, that is, predict the target area of ​​interest, before implementing the detection and recognition algorithm, nor do they fully capture the similarity between the extracted features in the feature encoding stage.

This respectively means that a large number of meaningless redundant calculations and information loss have a negative impact on the performance of detection and recognition

More importantly, the above methods or models are all proposed for a specific target, which limits their scope of application to a large extent.

Images