Image array optical flow estimation method for artificial compound eye camera

Abstract

The invention discloses an image array optical flow estimation method for an artificial compound eye camera. The method comprises the steps of detecting an image comprising a moving target in array images according to structural characteristics of used artificial compound eyes; building an optical flow estimation energy function; and optimizing the energy function by utilizing a variational method, thereby finishing optical flow estimation of artificial compound eye images. Compared with the prior art, the method has the remarkable advantages that an artificial compound eye image sequence canbe effectively subjected to the optical flow estimation, and due to constraint items of spatial correlation among sub-eye images, corresponding regions of adjacent sub-eyes can be constrained to obtain a consistent optical flow estimation result, so that the problem of inaccurate image edge optical flow estimation caused by small view field angle of the sub-eyes in the artificial compound eye camera is solved.

Description

Technical field [0001] The present invention relates to the technical field of computer vision, in particular to an image array optical flow estimation method of an artificial compound eye camera. Background technique [0002] Vision is an important means for most organisms to understand and perceive the world. As a widespread visual system, insect compound eyes have the characteristics of small size, clear imaging at close range, large field of view, and sensitivity to moving targets. Although the traditional monocular camera has been able to obtain high-definition images, it is difficult to further reduce the size and focal length of the monocular camera. As a new type of camera, the artificial compound eye camera can have a smaller volume while maintaining the imaging effect comparable to that of a monocular camera by simulating the imaging principle of insect compound eye. Therefore, the artificial compound eye camera has important research value in moving target estimation ...

AI Technical Summary

Problems solved by technology

[0003] Although there are a large number of optical flow estimation methods, there are still many problems in the application of artificial compound eye camera images.

Due to the characteristics of multiple low-resolution imaging channels of the artificial compound eye camera, although there is overlap between adjacent sub-eye images, due to the lack of effective support information, the current optimal optical flow estimation method is used to analyze the artificial compound eye. The optical flow estimation of the image still has the problems of poor image edge estimation results and inconsistent optical flow fields in the corresponding regions of adjacent sub-eyes

Images