高级检索

大功率半导体线激光图像处理方法研究

Study on image analysis of high power semiconductor line-laser

  • 摘要: 为了研究适于大功率半导体线激光断面检测系统的图像分割方法,采用几种阈值分割算法对所采集的激光光刀图像进行了处理及分析。通过在室外晴好天气下搭建平台,模拟出线激光投射大视场物体的情形,并用CMOS相机采集一系列图片。针对采集图片用3种不同算法对光刀进行阈值分割,用灰度重心法提取光刀中心位置,并对各算法的优劣进行了比较分析。结果表明,局部最大类间方差法为最佳算法,且在综合考虑信噪比、待测物体特性和线激光自身特性等因素之后,提出该算法在大功率半导体线激光系统下具有一定适用条件。

     

    Abstract: In order to study the image analysis method suitable for the sectional detection system based on high power semiconductor line-laser, several threshold algorithms were proposed to process and analyze the light-stripe images. An outdoor platform was built to simulate large-visual-angle laser projection, then images were collected by CMOS camera. Three threshold algorithms were introduced to analyze the images,and the center position of the light stripe was extracted by barycenter method, then advantages and disadvantages of the algorithms were compared. Results showed that the improved Otsu method could obtain a good performance. Besides,after considering signal-to-noise ratio, the character of objects and line-laser itself, the common applicable conditions of the algorithm were proposed under the line-laser system.

     

/

返回文章
返回