Ground object tracking device based on characteristic coupling

Abstract

The invention provides a ground object tracking device based on image key point characteristic coupling. The device comprises a field programmable gate array FPGA and a digital signal processor DSP. The FPGA is used for extracting the image characteristic of an external camera feed image sequence, thus completing adjacent interframe characteristic matching, and transmitting a successful interframe characteristic matching result to the DSP. The cross correlation accurate matching is carried out according to the interframe geometry transformation relation fed by the DSP. The digital signal processor DSP is used for calculating the adjacent image interframe transformation relation according to the characteristic coupling result outputted by the FPGA. According to the ground object tracking device, a complex target tracking method based a characteristic point is completed realized in an embedded FPGA, the interframe transition relation calculation is realized on the DSP, the algorithm complexity and low-power requirement of an embedded board are considered, a lot of image data is processed in real time, and compared with the prior art, the algorithm processing speed is improved by an order of magnitude.

Description

technical field

[0001] The invention belongs to the field of digital image signal processing, and in particular relates to a ground target tracking device based on feature matching. Background technique

[0002] Target tracking in computer vision has a variety of research applications. In the embedded field, DSP is often used as the core device for tracking methods. However, in embedded DSP, it is difficult to realize complex tracking in DSP due to power consumption and real-time requirements. target tracking algorithm. Image feature points are invariant to rotation, scale, and illumination, making them widely used in image registration, object tracking, and other fields. However, the feature point algorithm itself is relatively complicated, which makes its application in the embedded field have many obstacles. For example, feature extraction algorithms like SIFT and SURF need to establish a scale space, and Gaussian filter images at multiple scales need to be obtained for...

Images