Moving target robust detection method under area array camera rotary scanning

Abstract

The present invention provides a robust method of moving target detection under an area array camera rotary scanning condition. The method comprises the characteristic matching, the distortion compensation, the cylindrical projection model and background modeling, and the target detection. According to the present invention, a camera equation under a camera scanning work mode is established, the equation is linearized, a linear detection problem is transformed into a parameter space by utilizing the point-line duality of an image space and the parameter space and adopting the Hough transformation, so that the rapid robust estimation of the equation parameters is realized, and further, the background motion and the image distortion compensation are realized simultaneously. On the basis, a camera image is projected in a cylindrical background model to establish a panorama cylindrical model. The robust method of moving target detection under the area array camera rotary scanning condition provided by the present invention can eliminate the influence of the background motion and the image distortion on the moving target detection under the premise of guaranteeing the real-time performance, and can extract the moving targets rapidly and accurately.

Description

technical field [0001] The invention belongs to the technical field of image processing, relates to background modeling and target detection, and provides a robust detection method for moving targets under rotational scanning of an area array camera. Background technique [0002] In recent years, with the development of computer vision, moving target detection technology has been widely used in military, environmental monitoring and other fields. However, most of the current research on moving target detection is based on static platforms, and further research is still needed on moving target detection under moving platforms. [0003] Line scan cameras are usually used to obtain high-speed, high-precision panoramic images of large scenes. However, line array cameras have high requirements for platform stability, are expensive to manufacture, and require continuous scanning line by line when acquiring images, and cannot achieve "staring" monitoring of targets like area array...

AI Technical Summary

Problems solved by technology

Among them, the inter-frame difference method and the background subtraction method have many improved algorithms, which can better detect the target and the calculation is simple, but if no motion compensation is performed, these methods can only be applied to static platforms

The optical flow method can be applied to the target detection under the moving platform, but its computational complexity is high, and it is difficult to achieve real-time

In addition, none of the above methods considered the non-uniform distortion caused by the planar imaging mechanism

Therefore, how to realize the real-time and robust detection of moving targets under the moving platform still faces a huge challenge.

Images