基于FPGA的补偿滤波与焊缝识别方法

发布时间：2018-04-15 10:18

本文选题：FPGA + 图像滤波　；参考：《热加工工艺》2017年23期

【摘要】：在视觉检测中,对焊缝图像的识别通常需要使用激光作为辅助工具,一束线型的激光中往往包含着大量的焊缝特征信息。由于CMOS图像传感器采集到的图像数字信息存在较大的噪声,提出一种基于线性滤波器的补偿滤波方法,利用图像传感器的红色通道对激光的强敏感性,用红色像素R的像素值去补偿其周边的像素值,以加强激光在图像中的特征。利用现场可编程逻辑门阵列(Field programmable gate array,FPGA)并行处理的优点,设计了一种FPGA驱动CMOS图像传感器,并采集其输出的图像数字信息的方法。图像数字信息以流水线的形式被FPGA处理,并通过焊缝特征图像计算出偏移信息。该方法减少了数据存取操作,从而减小了处理延时,具有很高的实时性。
[Abstract]:In visual detection, laser is usually used as auxiliary tool to identify weld image. A beam of linear laser usually contains a lot of weld characteristic information.Because of the large noise in the digital image information collected by CMOS image sensor, a compensation filtering method based on linear filter is proposed, which makes use of the red channel of image sensor to be sensitive to laser.The pixel value of the red pixel R is used to compensate the pixel value around it to enhance the characteristics of the laser in the image.Taking advantage of the parallel processing advantage of Field programmable gate FPGA (Field Programmable Gate Array), a method of FPGA driving CMOS image sensor is designed, and the output image digital information is collected.The image digital information is processed by FPGA in the form of pipeline, and the offset information is calculated by the weld seam feature image.This method reduces the data access operation, thus reduces the processing delay, and has high real-time performance.
【作者单位】：国家冷轧板带装备及工艺工程技术研究中心;燕山大学工业计算机控制工程河北省重点实验室;浙江富日材料科技有限公司;
【基金】：河北省高等学校创新团队领军人才培训计划资助项目(LJRC013) 河北省自然科学基金面上项目(F2016 203249) 国家冷轧板带装备及工艺工程技术研究中心开放课题资助项目(2012006)
【分类号】：TP212;TP391.41

【相似文献】