4.4 - Track Targets Identification in Store Separation Tests using Digital Image Processing Techniques
- Event
- etc2014 - 34. European Telemetry and Test Conference
2014-06-03 - 2014-06-05
Nürnberg - Chapter
- 4. Video
- Author(s)
- A. Kusumoto, L. Guarino de Vasconcelos, N. Leite - Instituto de Pesquisas e Ensaios em Voo (IPEV), São José dos Campos (Brazil), R. Pirk - Instituto Tecnológico de Aeronáutica (ITA), São José dos Campos (Brazil)
- Pages
- 72 - 76
- DOI
- 10.5162/etc2014/4.4
- ISBN
- 978-3-9813484-7-7
- Price
- free
Abstract
The digital image processing techniques, specifically the area of Computer Vision, has long been used lately in applications with human-machine interaction in surveillance and military applications, such as for the identification of people in an environment monitored by cameras or extraction of information from different positions in images of an aircraft Head-Up Display (HUD). The same techniques can be used in other situations such as in flight tests. In Store Separation Tests, the data analysis occurs post flight using the commercial off-the-shelf applications, but still depends of the user accuracy which indicates where the track targets are. This process can be time consuming and unproductive. The agility to achieve the analysis results can be improved with the use of computer vision techniques. The Instituto de Pesquisas e Ensaios em Voo (IPEV) together with the Instituto Tecnológico de Aeronáutica (ITA) is currently developing an application that identify and tracks in real-time several reference points fixed in the store surface. This paper presents the study and application of techniques of digital image processing for identifying position of reference points in sequential images, the software implements algorithms that uses techniques to extract their features and Haar-like cascade and Local Binary Patterns (LBP) classifiers. The preliminary results were considered satisfactory.