Vehicle detection tracking method based on color feature nonlinear dimension reduction

Abstract

The invention discloses a vehicle detection tracking method based on color feature nonlinear dimensionality reduction. The method comprises the steps of collecting the road vehicle driving video datathrough video shooting; extracting the LBP features; converting the RGB space into a CN space through color conversion, and extracting the color features; carrying out feature fusion on the LBP features and the color features; enabling the input characteristics to return to target Gaussian distribution, and training a related filter in an on line manner; adding a scale pool to enable the size of atracking frame to be matched with the target vehicle; calculating a correlation filter response value so as to determine a new position of the target vehicle; repeating the above steps until the video is finished; and connecting the central positions of the tracked vehicle obtained by each frame, and drawing the track of the tracked vehicle in the video. According to the invention, through the LLE dimension reduction and in combination with the LBP features, the accuracy of the tracking algorithm can be improved.

Description

technical field [0001] The invention belongs to the technical field of visual tracking, and in particular relates to a vehicle detection and tracking method based on nonlinear dimensionality reduction of color features. Background technique [0002] Visual tracking is one of the basic research topics in the field of computer vision, and has a wide range of applications in the fields of video surveillance, robotics, human-computer interaction, and unmanned driving. Despite great progress in the past decade, model-free tracking remains a challenge due to illumination changes, geometric deformations, partial occlusions, fast motion, and background clutter. Correlation filters were introduced to vision problems and have been used in many ways. According to the convolution theorem, correlation in the time domain corresponds to element-wise multiplication in the frequency domain. Therefore, the essential idea of ​​kernel correlation filtering is to calculate the correlation in t...

AI Technical Summary

Problems solved by technology

[0005] (1) Use simple linear interpolation to update the model. When the appearance of the target is scaled up or down, the tracking algorithm is prone to drift;

[0006] (2) Use the correlation filtering method of the Histogram of Oriented Gradient (HOG), which has no size and rotation invariance;

[0007] (3) Use feature detection algorithms such as color naming, which cannot detect features such as lines and textures well

Images