摘要
针对单一图像识别系统和单一声纳识别系统对事物特征值数据量采集的不足造成识别误差这一缺陷,从仿生学中蝙蝠眼睛依旧存在的角度考虑设计了一种基于FPGA+DSP的声纳图像传感融合系统。该系统采用FPGA发射声纳信号与采集回波,并将预处理的回波数据传给DSP做进一步处理;采用CCD摄像头仿蝙蝠的眼睛采集图像数据,最后将采集到的声纳数据与图像数据在DSP中进行进一步算法处理,最终输出人耳能够识别的信号,借助于人脑参与分析识别,并对前方探测物做出规避动作。
Aimed at the deficiency that the single image recognition system and the single sonar recognition system cannot recognize the object precisely with insufficient characteristic information,based on the bionics principle,a sonar and image dual-sensor fusion system based on FPGA and DSP was designed.The system used the FPGA to transmit the sonar signal and acquire echo signal,and pretreated the echo data,and sent the echo to DSP for further processing.The system used the CCD camera imitating the bat's eyes to capture the image data.Both the echo data and the image data were further treated by DSP.Finally,signal would be sent out in the form that the human ears can recognize.With the aid of the brain involved in analysis of recognition,the bland man can detect the objects out ahead and take evasive action.
出处
《机械工程与自动化》
2015年第5期4-6,共3页
Mechanical Engineering & Automation
基金
国家自然科学基金资助项目(60772004)
关键词
声纳
图像
传感融合
设计
sonar
image
sensor fusion
design