摘要
常规方法计算匹配图像像素点三维坐标时,匹配特征点相似度较低,为此提出基于实物ID的智能变电站三维数字化设计方法。激光扫描智能变电站实景,卷积点云图像和尺度函数,提取点云图像中目标物体特征点,度量特征点相似程度,匹配梯度接近的像素点,计算像素点三维空间坐标,编码标识模型中的目标物体,生成三维数据解释说明文件,数字化显示变电站场景。实验结果表明,旋转变电站设备数字化显示的三维数据,减小了质心偏移距离,提高了数据精度,充分保证数字化设备与实际变电站设备的体积重合度。
When calculating the three-dimensional coordinates of matched image pixels by conventional methods,the similarity of matching feature points is low.Therefore,a three-dimensional digital design method of intelligent substation based on physical ID is proposed.Laser scans the real scene of intelligent substation,convolutes the point cloud image and scale function,extracts the feature points of the target object in the point cloud image,measures the similarity of the feature points,matches the pixel points close to the gradient,calculates the three-dimensional spatial coordinates of the pixel points,codes and identifies the target object in the model,generates the three-dimensional data interpretation document,and digitally displays the substation scene.The experimental results show that the three-dimensional data digitally displayed by the rotating substation equipment reduces the offset distance of the center of mass,improves the data accuracy,and fully guarantees the volume coincidence of the digital equipment and the actual substation equipment.
作者
陈晨
李巍
宋景博
周怡
李凯
CHEN Chen;LI Wei;SONG Jing-bo;ZHOU Yi;LI Kai(State Grid Henan Economic Research Institute,Zhengzhou 450000 China;State Grid Henan Electric Power Company,Zhengzhou 450000 China)
出处
《自动化技术与应用》
2023年第10期129-132,共4页
Techniques of Automation and Applications
关键词
实物ID
变电站三维数字化
点云数据
特征点
实景扫描
physical ID
substation three-dimensional digitization
point cloud data
feature points
real scene scanning
