Target tracking method fusing template matching and grey prediction

Abstract

The invention discloses a target tracking method fusing template matching and grey prediction. The method comprises the steps of firstly predicting the position of a target in an image by using a grey prediction GM(1,1) model; and secondly calculating the definite position of the target by using a template matching method. The position prediction greatly reduces a search range, so that the target tracking method meets a real-timeliness requirement. In a matching process, a sequential similarity detection method with an adaptive threshold is used, and the threshold is compared with an absolute error sum, so that the calculation amount is reduced and the threshold well adapts to the matching process. Finally, a weighted template updating method is adopted for changes of target scale and environment, so that the robustness of the tracking method is improved.

Description

technical field

[0001] The invention belongs to computer vision tracking technology, in particular to a target tracking method combining template matching and gray prediction. Background technique

[0002] Object tracking is a basic core technology in the field of computer vision and the basis for various subsequent high-level visual processing. Target tracking technology has a very broad application prospect in many civil and military fields, such as video surveillance, intelligent transportation, missile precision guidance and so on. Template matching is a basic algorithm in image processing, pattern recognition and target tracking. This algorithm has the advantages of simple operation and low requirements on image quality, but it also has the disadvantages of large amount of calculation and poor real-time performance.

[0003] In recent years, there are mainly two methods to solve the problem of large amount of calculation and poor real-time performance of template matc...

Images