Aerial photograph of the Immaculate Conception Chapel

Threat Object Detection

1. Background

Rapid advances made in the area of camera and sensor technology has enabled the use of video acquisition systems to monitor the right-of-way. Huge amount of data is thus made available for analysis. However, it would be very expensive to employ analysts to scan through the data and identify threats to the pipeline right-of-way in the vast amount of low altitude aerial imagery. This warrants the development of an automated mechanism that is able to detect threats to the pipeline right-of-way and send out warnings in the event of detecting a threat.

2. Objective

The main goal of threat object detection program is to develop a real-time application system that has capability of detecting and localizing threat objects on pipeline to prevent human-caused damages to surface pipelines.

Threat Detection System

3. Approach

A novel algorithmic framework for the robust detection and classification of objects on right-of-way (ROW) is designed from there directions: visibility improvement, semantic segmentation, and part-based object recognition. The developed algorithm system is capable of detecting construction equipment along the pipeline right-of-way with a very high degree of accuracy in comparison with manual threat identification. Currently, the algorithm is being fine-tuned to be integrated with an image acquisition system to provide streamlined performance for accurate and efficient threat detection.

Threat Detection Scheme

4. Sample Results

1) Visibility Improvement

Visibility Improvement

2) Semantic Segmentation

Semantic Segmentation

3) Part-based Object Detection

Part-based Segment Detection

Related Publications:

  1. Paheding Sidike, Theus Aspiras, Vijayan K. Asari, and Mohammad S. Alam, "A rotation-invariant pattern recognition using spectral fringe-adjusted joint transform correlator and histogram representa-tion," SPIE Conference on Defense + Security: Optical Pattern Recognition XXV, Baltimore, MD, USA, 05-09 May 2014.
  2. Yakov Diskin and Vijayan K. Asari, "Illumination invariant 3D change detection," IS&T/SPIE In-ternational Conference on Electronic Imaging: Image Processing: Machine Vision Applications VII, vol. 9024, 90240D, February 2014.
  3. Paheding Sidike and Vijayan K. Asari, "Detection of machinery threat on pipeline right-of-way," (Poster) 2013 Pipeline Research Council International (PRCI) Research Exchange Meeting, Atlanta, Georgia, USA, 04-06 February 2014.
  4. Paheding Sidike, Vijayan K. Asari and Mohammad S. Alam, "Illumination-invariant pattern recogni-tion using fringe-adjusted joint transform correlator and monogenic signal," IS&T/SPIE International Conference on Electronic Imaging: Image Processing: Machine Vision Applications VII, vol 9024, 90240C, February 2014.
  5. Alex Mathew and Vijayan K. Asari, "Rotation-invariant histogram features for threat object detec-tion on pipeline right-of-way," IS&T/SPIE International Conference on Electronic Imaging: Video Surveillance and Transportation Imaging Applications 2014, San Francisco, CA, USA, 2 - 6 February 2014.
  6. Binu Nair, Varun Santhaseelan, Chen Cui, and Vijayan K. Asari, "Intrusion detection on oil pipeline right of way (ROW) using monogenic signal representation," Proceedings of SPIE Conference on De-fense, Security, and Sensing: Signal Processing, Sensor Fusion, and Target Recognition XXII, Balti-more, MD, USA, vol. 8745, 29 April - 03 May 2013.
  7. Saibabu Arigela and Vijayan K. Asari, "An automatic visibility improvement and object detection system for wide area surveillance applications," 2013 National Geospatial-Intelligence Agency (NGA) Academic Research Program (NARP) Symposium and Workshops, the National Academy of Sciences Keck Center, Washington DC, September 10-12, 2013.
  8. Varun Santhaseelan and Vijayan K. Asari, "Extracting context information from aerial imagery for aiding threat detection," Brother Joseph W Stander Symposium 2013, University of Dayton, Dayton, OH, USA, 16-17 April 2013.
  9. Vijayan K. Asari, Binu Nair, Varun Santhaseelan, Saibabu Arigela, Yakov Diskin, and Alex Mathew, "Automatic pipeline right-of-way threat detection by advanced image analysis," (Invited Talk), 2013 Pipeline Research Council International (PRCI) Research Exchange Meeting, Dallas, TX, USA, 05-07 February 2013.
  10. Saibabu Arigela and Vijayan K. Asari, "Image enhancement and super resolution on RAM aerial imagery," (Poster) 2013 Pipeline Research Council International (PRCI) Research Exchange Meeting, Dallas, TX, USA, 05-07 February 2013.
  11. Binu Nair and Vijayan K. Asari, "Detection of machinery threat on pipeline right-of-way ," (Poster) 2013 Pipeline Research Council International (PRCI) Research Exchange Meeting, Dallas, TX, USA, 05-07 February 2013.
  12. Varun Santhaseelan and Vijayan K. Asari, "Extracting context information from aerial imagery," (Poster) 2013 Pipeline Research Council International (PRCI) Research Exchange Meeting, Dallas, TX, USA, 05-07 February 2013.
  13. Yakov Diskin and Vijayan K. Asari, "Change detection using 3D scene representation models," (Poster) 2013 Pipeline Research Council International (PRCI) Research Exchange Meeting, Dallas, TX, USA, 05-07 February 2013.