Quaternion based speeded up robust features (SURF) description method and system for color image

Abstract

The invention relates to a quaternion based speeded up robust features (SURF) description method for a color image. The method comprises the following steps of: inputting a color image; showing color information of each pixel point in the color image with pure quaternion to obtain a pure quaternion matrix for representing the color image; looking for a main direction of each feature point; forming a rectangular region by taking the feature point as a center along the main direction of each feature point, and calculating the feature vector of each feature point; and converting the feature vector of each feature point into the feature point of a pure quaternion norm; therefore, the quaternion based SURF description for the color image is achieved. With the quaternion based SURF description method for the color image, disclosed by the invention, the features of the color image can be described.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to a quaternion-based color image SURF feature description method and system. Background technique [0002] As the underlying feature in image processing, the visual features of images have been widely used in the fields of object recognition, image retrieval, object tracking, image matching, and image classification. The research on local visual features of images is mainly divided into three parts: feature extraction, feature description and feature performance analysis. For these three parts, researchers have done a lot of research. Someone summarized various visual features in recent years from a physical point of view, including the relationship between edges, corners and spots, and divided the detection algorithms of various visual features into different categories, and gave a detailed description of representative algorithms. Aiming at the feature detection of the...

AI Technical Summary

Problems solved by technology

Aiming at the defect that the classic SURF algorithm cannot detect symmetric objects, a new symmetric SURF descriptor is proposed to enrich the performance of the SURF algorithm. Through mirror transformation, all possible symmetric matching pairs are detected and applied to the vehicle detection and license plate recognition

[0003] Although the above algorithms can optimize the performance of the SURF algorithm to a certain extent, they all analyze the local features of the grayscale image.

For the description of local features of color images, no corresponding method is given.

Images