Speed measurement method based on monocular vision system

Abstract

The invention discloses a speed measurement method based on a monocular vision system, and belongs to the technical field of machine vision, and the speed measurement method specifically comprises the following steps: (1) carrying out background modeling by using a Gaussian mixture model; (2) acquiring high-precision pixel displacement of the target by using a parameter-optimized template matching method; (3) accurately acquiring a pixel size corresponding to the target to be measured by using an improved two-dimensional measurement model method; (4) calculating the actual speed according to the proportional relation; according to the method, the Gaussian mixture model is fused based on template matching, a foreground extraction effect which is more stable than that of traditional background difference can be obtained by automatically updating a background model for a complex background and the Gaussian mixture model, and then parameter setting of template matching is optimized, so that the matching success rate is further improved; and finally, the pixel size corresponding to the target is accurately obtained by using an improved two-dimensional measurement model method, the actual speed is converted through the proportional relation between the physical size of the target and the pixel size, and the visual speed measurement precision and robustness can be improved.

Description

technical field [0001] The invention relates to the technical field of machine vision, in particular to a speed measurement method based on a monocular vision system. Background technique [0002] With the development of machine vision technology, vision-based target speed measurement has more and more extensive application and development space in the fields of intelligent transportation, industrial production and military; traditional speed measurement methods such as radar speed measurement, ground-sensing coil speed measurement and laser speed measurement need Precise sensor equipment is installed in advance, and there are strict requirements for the surrounding environment during measurement; in contrast, vision-based speed measurement has the advantages of simple equipment installation, low cost, and relatively loose application conditions; but the disadvantage is that it is easily affected by complex backgrounds. Therefore, how to optimize the speed measurement algori...

AI Technical Summary

Problems solved by technology

[0002] With the development of machine vision technology, vision-based target speed measurement has more and more extensive application and development space in the fields of intelligent transportation, industrial production and military; traditional speed measurement methods such as radar speed measurement, ground sensing coil speed measurement and laser speed measurement need to Precise sensor equipment is installed in advance, and there are strict requirements on the surrounding environment during measurement; in contrast, vision-based speed measurement has the advantages of simple equipment installation, low cost, and relatively loose application conditions; but the disadvantage is that it is vulnerable to complex backgrounds. and illumination changes, so how to optimize the speed measurement algorithm to improve the robustness and accuracy of the visual speed measurement algorithm has become a popular research direction; the vision-based speed measurement method uses the camera to collect the moving image of the target in a non-contact manner, through the target The detection algorithm recognizes the target from each frame of image, and calculates the displacement of the target between two frames to achieve the purpose of speed measurement; the mainstream moving target detection algorithm includes: frame difference method, background difference method and optical flow method; the operation of optical flow method is complex, It is susceptible to light interference and has few practical applications; the frame difference method is widely used and is not easily disturbed by the environment, but it is not suitable for targets with slow detection speeds; the current improved frame difference methods include three-frame difference method, five-frame difference method, etc. , but the "hole" phenomenon is still difficult to eliminate; the background difference method can accurately extract moving objects, but it is sensitive to changes in the background environment; the improved background difference method separates the background by introducing pixel variation coefficients on the basis of background modeling, but it still cannot To achieve accurate detection; therefore, it is particularly important to invent a speed measurement method based on a monocular vision system

Images