Download Full Text (1.8 MB)
Pipeline right-of-ways (ROWs) monitoring and safety pre-warning is a vital way to guarantee safe operation of the oil/gas transportation. Any construction equipment or heavy vehicle intrusion is a potential safety hazard to pipeline infrastructure. Since millions of miles of pipes buried along the length and breadth of the United States, monitoring is required to know if pipeline ROW is under threat or not. Taken into account of less population of the vast amount of area, high cost of labor and rapid advancements in sensor technology and automated techniques for image analysis, aerial monitoring is found to be the most viable option. The images captured by aerial data acquisition system, such as fixed-wing air-crafts or unmanned air vehicles are affected by a lot of factors including varying illumination conditions, environmental conditions, camera characteristics, etc. To deal with the above mentioned problems, an automatic intrusion detection system, which is capable of dealing with the constraints of the aerial imagery caused by low resolution, lower frame rate, large variations in illumination, motion blurs, etc., is being developed to assist the threat detection as part of the ROW automated monitoring program. The automated pipeline monitoring system is designed to be in three phases: background elimination, part-based object detection and risk assessment. In the first phase, a region of interest (ROI) detector is developed to extract potential regions that may contain objects by utilizing monogenic phase features into a cascade of pre-trained classifiers. In the second phase, a part-based object detection model is built for searching specific targets, which are considered as threat objects. In the third phase, a safe pre-warning system is built to access the severity of the threats to pipelines by computing the geolocation and temperature information of the threat objects. In addition, in order to assign more precise warning, the impacts caused by different types of vehicles will be taken into account by developing new feature extraction and classification algorithms.
Graduate Research - Graduate
Vijayan K Asari
Primary Advisor's Department
Electrical and Computer Engineering
Stander Symposium poster
"Automatic Intrusion Detection on Oil/Gas Pipeline Right-of-Ways" (2017). Stander Symposium Posters. 927.