Hyper-spectral image ground object recognition method based on sparse kernel representation (SKR)

Abstract

The invention discloses a hyper-spectral image ground object recognition method based on a sparse kernel representation (SKR), which mainly solves the defects that the recognition time is long and the recognition accuracy is not high when the dimensions of a sample are decreased to be low in the existing method. The method comprises the recognition steps of: firstly, using the spectral vectors ofknown tags in a hyper-spectral image as a dictionary for sparse coding, wherein the spectral vectors of the known tags are arranged by classifications and the spectral vector samples of all unknown tags form a test sample set; secondary, using a neighbor method to construct a central sample matrix, respectively mapping test samples and the dictionary to a feature space by constructing a sparse kernel function to obtain the mapped dictionary and the mapped test samples, and conducting line normalization to the mapped dictionary; and finally using the normalized dictionary to conduct sparse coding to the mapped test samples and judging the classifications of the test samples through an error discriminant. The hyper-spectral image ground object recognition method disclosed by the invention has the advantages that the recognition accuracy can be ensured to be high, the ground object recognition of the hyper-spectral image can be completed rapidly at the same time and the subsequent processing of the recognized ground objects is facilitated.

Description

technical field [0001] The invention belongs to the technical field of remote sensing image processing, and relates to a ground object recognition method. The method can use hyperspectral images to analyze complex landforms and ground objects, and determine different types of similar ground objects. Background technique [0002] Hyperspectral image object recognition refers to the use of hyperspectral images to analyze complex landforms and objects to determine the type of objects. Hyperspectral imagery is a massive amount of data generated by multispectral remote sensing imaging equipment, which contains both spatial information and rich spectral information of ground objects. Each point in the image can be described by a high-dimensional spectral vector composed of spectral information of many spectral segments, and various types of ground objects can be identified by using these spectral vectors. [0003] At present, hyperspectral image object recognition methods can be ...

AI Technical Summary

Problems solved by technology

[0003] At present, hyperspectral image object recognition methods can be divided into two categories: unsupervised and supervised. Unsupervised methods do not need to know the labels of any samples in advance, such as the K-means algorithm, but the recognition rate is often low. And it is easily affected by the selection of the initial center point; supervised methods need to know the labels of some samples, such as support vector machine SVM, although the recognition rate of this method has improved, it often requires a complicated training process

With the recent rise of sparse representation, some scholars have proposed the idea of ​​using the sparse representation of samples for classification, such as the sparse representation classification algorithm SR, which does not require a complicated training process, but the recognition is not accurate enough

The kernel sparse representation classification algorithm KSR is an improvement to SR, which can improve the recognition accuracy, but the space and time complexity of calculating the kernel function is high, which often consumes a lot of computer resources and recognition time, which affects the recognized features. Subsequent processing of

Images