Binocular video camera and image processing method thereof

Abstract

The invention provides a binocular video camera and an image processing method thereof. The video camera obtains image data of a target scene through two image collection sensors. The video camera comprises a shell and a lens; a front board, a rigid flexible board, a main board, a power board and a storage board are arranged in the shell. A sensor A and a sensor B are arranged on the front board. An A / D chip is welded to the front board. An FPGA and a DSP are welded to the main board. According to the binocular video camera, the multi-core framework is adopted, and the image sensors, the FPGA and the high-performance DSP are involved; the intelligent analysis algorithm and image processing algorithms for automatic exposure, automatic white balance adjustment and the like are embedded into the binocular video camera, meanwhile, the dual-sensor synchronous imaging technology is adopted, and the functions of moving object detection and snapshooting, object feature recognition and the like are achieved; on the basis of the target scene simultaneous imaging technology, the processes of image feature extracting, recognizing, matching and rebuilding are completed, a target object can be more efficiently and reliably recognized and positioned, and the binocular video camera has the advantages of being high in frame rate, low in power consumption and noise and the like.

Description

technical field [0001] The invention belongs to the field of machine vision, in particular to a binocular camera and an image processing method thereof. Background technique [0002] With the rapid development of computer vision technology and image acquisition technology, many scenes not only need to collect the large-scale motion trajectory of moving objects, but also need to see the characteristic details of moving objects. [0003] However, the traditional image acquisition system has only one image sensor. Otherwise, if you use a wide-angle lens, you can only capture large scenes and cannot see details; otherwise, you can use a close-up lens to see details, but you cannot capture large motion trajectories. Therefore, how to realize the acquisition of the motion trajectory of the large scene and the ability to see the details has always been the goal pursued by technicians in the field of computer vision. Contents of the invention [0004] In order to realize the acqu...

AI Technical Summary

Problems solved by technology

[0003] However, the traditional image acquisition system has only one image sensor. Otherwise, if you use a wide-angle lens, you can only capture large scenes and cannot see details; otherwise, you can use a close-up lens to see details, but you cannot capture large motion trajectories.

Images