Color-information-based scale invariant feature point describing and matching method

Abstract

The invention discloses a color-information-based scale invariant feature point describing and matching method which comprises the following steps: firstly, carrying out Gaussian convolution on R, G and B chrominance components in digital color images respectively; according to the coordinates, direction, scale and other information of feature points, determining the position and structure of feature point descriptors; calculating the average value of the R, G and B chrominance components of subregions in the concentric circle structure of the descriptor, and taking each average value as one-dimensional element of the feature vector to construct a feature vector; according to the distance between the subregion and the feature points, multiplying all-dimensional feature vector elements by Gaussian weight; respectively carrying out normalization treatment on the feature vector element belonging to the same one chrominance component; sequentially calculating the feature vectors of all feature points, and constructing feature vector space of the images; and finally calculating the distance between every two feature vectors and matching corresponding feature points in the feature vector space of the two images.

Description

technical field [0001] The invention belongs to a general image data processing technology, in particular to a scale-invariant feature point description and matching method based on color information. Background technique [0002] Image feature point is a local image feature that can be robust to various image changes. After extracting the feature points, it is necessary to use the descriptor, and use the local image information around the feature points to obtain the feature vector of each feature point through calculation, and form the feature vector space of the image. Matching the same feature points in two images is to match the corresponding feature vectors in the feature vector space of the two images. [0003] Scale-invariant feature points can not only adapt to image changes such as translation, rotation and noise, but also have strong robustness to image scale changes (image scaling). Therefore, the description method corresponding to the scale-invariant feature ...

AI Technical Summary

Problems solved by technology

This matching method is essentially an extension of the SIFT descriptor, but the dimension of the feature vector is greatly increased, and the real-time performance of the matching is greatly affected.

[0008] Summarizing the relevant literature and research results of image feature point description and matching published at home and abroad, there are the following deficiencies: (1) Although the feature point descriptor based on differential calculation has a low dimension of feature vectors and a small amount of matching calculation, it is more sensitive to image noise. Sensitive; (2) SIFT descriptors and other description methods based on the distribution of image information in a specific area have strong robustness, but due to the complex structure of the descriptor and the high dimension of the feature vector, the amount of calculation for matching is often large; (3) The descriptor based on the color histogram is less robust to image changes; (4) The color component of the image is treated as a grayscale image, and the generated descriptor is based on image grayscale information such as SIFT. The dimensionality of the feature vector will be multiplied, and the matching calculation will be more

Images