基于红外图像处理的埋地石油管道自动探测技术

Auto-Detection Technology of Underground Petroleum Pipeline Based on Infrared Image Processing

埋地石油管道的辅助巡线以及对偷油支管的快速定位与排查是目前油田亟待解决的问题。为此,根据输油管道加温后辐射红外能量的特点,提出了应用红外热成像技术探测埋地输油管道的方法．介绍了基于非制冷红外焦平面阵列的,具有便携、实时成像特点的红外管道探测仪的硬件设计;在对原始红外图像进行去噪和增强的基础上,采用基于最大类内距离比准则的递归带通分割法对输油管道的红外图像进行自动阈值分割,并结合Freeman直线链码方法对细化后的图像进行骨架跟踪处理,提取管道主干,标记管道走向并定位偷油支管分支．实验证明,本算法自适应性好．执行效率高．管道定位准确。实现了埋地输油管道的自动检测．

Assisting perambulation of underground petroleum pipeline and rapid detection of its stolen branches is an urgent problem in oilfield . According to the fact that the heated petroleum pipeline radiates infrared energy, the infrared imaging method is proposed. The hardware architecture uses uncooled infrared focal plane array as its sensor and has the capability of realtime imaging and portability. On the basis of image filtering and enhancement, the methods of ＂iterative segmentation based on the rule of maximal distance in class＂ and ＂Freeman chain-code skeleton track＂ are used to mark direction of pipeline and its branches. Experimental result shows that the proposed image processing algorithm can accomplish auto-detection of underground petroleum pipeline and has excellent performance in auto-adaption, speed and precision.