Image feature extraction method capable of being relied on for long time

Abstract

The invention discloses an image feature extraction method capable of being relied on for a long time, and the method comprises the following steps: analyzing an overlapping region of images continuously shot by a camera through estimating the motion state of the camera, and carrying out the feature extraction of the overlapping region; constructing a pyramid layer, extracting angular points on the pyramid layer layer-by-layer, selecting the number of the angular points according to the image resolution, wherein the larger the image resolution is, the more the number of the angular points needing to be extracted is; adjusting the mean value and variance of the Gaussian function through the motion state to achieve the effect of controlling the threshold distribution of the extracted features, and thus controlling the distribution of the extracted features on the image. According to the invention, the universality property is good; the extracted features come from the overlapping region,and the feature matching is the feature of the needed part of the image, so that the extracted features can be used for feature matching, and the feature utilization rate is high; the method can be accurately used for image feature extraction of intelligent video search in target recognition of pedestrians and the like while complete autonomous extraction is achieved, and the method is easier toimplement in engineering.

Description

technical field [0001] The invention relates to the technical field of image processing, in particular to an image feature extraction method that can be relied upon for a long time. Background technique [0002] With the rapid development of multimedia technology, more and more image data appear in life, how to process images quickly and effectively becomes particularly important. In image processing, the extracted features are usually used to match between images, and then the positional relationship between two frames of images is obtained through the matching relationship. Image features mainly include image color features, texture features, shape features, and spatial relationships. Features are the best way to describe patterns. We usually think that each dimension of features can describe patterns from different perspectives. Ideally, the dimensions are complementary and complete. The main purpose of feature extraction is dimensionality reduction. The main idea of ​​...

AI Technical Summary

Problems solved by technology

However, when the image is captured by the camera in fast motion, the features extracted directly by traditional methods on each frame of image have a short utilization time, and there are few reusable features. In fact, the features of different parts of the image exist at different times when the camera is in motion. different validity period

[0003] Commonly used feature extraction methods have the following disadvantages when applied to camera motion positioning: when the camera is moving, the overlapping degree of the images captured by the camera is small, which leads to the fact that most of the feature points are useless in feature matching, while extracting features point takes a lot of time

The traditional feature extraction method is to extract features indiscriminately from the image, and will not adjust the position of the extracted features according to the difference in the overlapping area, so the utilization rate of the extracted features is very low

Images