桥梁识别在军事上和民用上都有重要的意义,国内外已有的一些研究虽然取得了一定效果,但大多数都对目标图像有些限制要求,使用范围不广.而且都是基于PC机的仿真算法,不能用于实时处理的场合.本文提出了一种基于FPGA和DSP的知识基桥梁目标自动识别算法,本算法对桥梁形状没有特别要求,能够对航拍或遥感图像中的桥梁进行实时识别.不仅能够对近景图像进行识别,也能对远景图像进行识别,而且能识别一幅图像中的多个桥梁.

Recognition of bridges plays important roles in both military and civil fields.The existing researches have got some achievements,but most of them can only be applied in some limited images.What’s more,all of them are simulation algorithms based on PC,so they cannot be used in real-time applications.A real-time knowledge-based automatic recognition algorithm of bridges was introduced.This algorithm is based on FPGA and DSP,which can be widely used in aerial images or remote sensing images.It can be applied to not only close shot images,but also distant shot ones.At the same time,it can recognize several bridges in one picture.