Virtual 3D Models of Insects for Accelerated Quarantine Control

Chuong Nguyen, David Lovell, Rolf Oberprieler, Debbie Jennings, Matt Adcock, Eleanor Gates-Stuart, John La Salle; Proceedings of the IEEE International Conference on Computer Vision (ICCV) Workshops, 2013, pp. 161-167

We learn from the past that invasive species have caused tremendous damage to native species and serious disruption to agricultural industries. It is crucial for us to prevent this in the future. The first step of this process is to identify correctly an invasive species from native ones. Current identification methods, relying on mainly 2D images, can result in low accuracy and be time consuming. Such methods provide little help to a quarantine officer who has time constraints to response when on duty. To deal with this problem, we propose new solutions using 3D virtual models of insects. We explain how working with insects in the 3D domain can be much better than the 2D domain. We also describe how to create true-color 3D models of insects using an image-based 3D reconstruction method. This method is ideal for quarantine control and inspection tasks that involve the verification of a physical specimen against known invasive species. Finally we show that these insect models provide valuable material for other applications such as research, education, arts and entertainment.