摘要
在井下作业过程中往往因为操作不当导致工具坠井、落鱼失踪,这将严重影响井下作业的进度。为了提高落鱼打捞的成功率,采用双目视觉技术对落鱼进行精确定位测量。基于井下特殊的成像环境,运用改进的SGBM算法,对校正后的双目图像进行立体匹配处理,从而获取落鱼的视差图,然后基于双目视觉三维测量原理以及落鱼视差信息进行定位测量。为验证算法的有效性,进行地面模拟实验验证,结果表明,该方法容易实现,计算速度快,而且在一定范围内的精度基本满足实际应用的需要,同时验证了双目视觉在落鱼检测中的可行性,也为可视化定量测量技术提供了参考。
In order to improve the success rate of fish fishing,binocular vision technology is used to accurately locate and measure fish falling.Based on the special imaging environment in the borehole,the improved SGBM algorithm is applied to the corrected binocular image for stereo matching processing,so as to obtain the parallax map of the falling fish,and then the positioning measurement is carried out based on the binocular vision 3D measurement principle and the parallax information of the falling fish.In order to verify the effectiveness of the algorithm,ground simulation experiments are carried out,and the results showed that the proposed method is easy to implement,the calculation speed is fast,and the accuracy in a certain range basically met the needs of practical applications.
出处
《工业控制计算机》
2024年第12期63-65,共3页
Industrial Control Computer
关键词
双目视觉
立体匹配
落鱼
binocular vision
stereo matching
falling fish