Metrological remote identification of a human body by stereoscopic camera techniques
The stereoscopic and 3D triangulation techniques are widely used in close-range photogrammetry for metrological applications. Same techniques may also implement remote identification tasks of a human body for the security requirements using its intrinsic geometrical features remotely acquired by stereoscopic CCD camera pair. Within the initial technique, Human body and some other features at the background in motion are selected and separated from rest of background by the image differentiation algorithm which is already performed by video frame grabber card in real-time. The further differentiation between the human body and other features in motion (e.g. birds, clouds, trees, etc.) is made by stereoscopic measurement techniques in relation with a simple human body features data base. The data base may contain; possible ranges of height, width of the body, maximum velocity, etc. Since the distance between the camera and human body could be measured, its velocity may easily be calculated.
Unable to display preview. Download preview PDF.
- Akita K.. “Image Sequence Analysis of Real World Human Motion”, Pattern Recognition, Vol.17, No.1, pp.73–83, 1984.Google Scholar
- Chin R.T and C.R. Dyer. Model-based Recognition in Robot Vision, ACM Computing Surveys, 18(1), March 1986.Google Scholar
- Lee H.J. and Z. Chen. “Determination of 3D human body postures from a single view”, Comput. Vision Graphics Image Processing, 30, pp. 148–168, 1985.Google Scholar
- Maylor M.K and Y. Yang. “Human Body Motion Segmentation Complex Scene”, Pattern Recognition, Vol. 20, No.1, pp. 55–64, 1987.Google Scholar
- Moffit, F.H. and E.M Mikhail. Photogrammetry, Harper & Row publisher Inc., Newyork-USA, 1980.Google Scholar