A Mechatronics Approach for Design of Multiview Image Acquisition Setup for Scene Reconstruction with Single Camera

Conventional design approach is being actively replaced by concurrent design approach in the context of interdisciplinary systems. The proposed research work intends to develop a single moving camera based stereo vision system for scene reconstruction with the intrinsic advantage of multi-directional fields of view. The conventional stereo vision setup uses two stationary passive cameras to capture images of a scene from different vantage points. The proposed system imparts varying mechanical degrees of freedom motion for both the object and the camera which aids in acquiring sequence of images which covers all the visible regions of the object of interest. This gives better detail of the scene under consideration as compared to the conventional two images based stereopsis. A mechatronics design approach has been presented which carefully integrates various elements of the system such as the mechanisms, actuators, sensors and the electronic controller. The paper clearly pin points the cues for the design of the mechanical system which are obtained from the requirements of the computer vision system. The relative pose between the camera and the scene is governed by three independent degrees of freedom namely rotation angle for the object, tilt and working distance for the camera. The selection of the aforementioned parameters is decided by the specifications such as field of view, size of the object and sensor and spatial resolution. The proposed design predicts the system to enjoy benefits of reduced cost and improved flexibility in general.

[1]  Ramakant Nevatia,et al.  Segment-based stereo matching , 1985, Comput. Vis. Graph. Image Process..

[2]  Ruzena Bajcsy,et al.  3D reconstruction of environments for virtual collaboration , 1998, Proceedings Fourth IEEE Workshop on Applications of Computer Vision. WACV'98 (Cat. No.98EX201).

[3]  Jonathan M. Garibaldi,et al.  Real-Time Correlation-Based Stereo Vision with Reduced Border Errors , 2002, International Journal of Computer Vision.

[4]  Miska M. Hannuksela,et al.  Transform domain similarity measures in stereo matching , 2012, 2012 3DTV-Conference: The True Vision - Capture, Transmission and Display of 3D Video (3DTV-CON).

[5]  R. A. Hamzah,et al.  Sum of Absolute Differences algorithm in stereo correspondence problem for stereo matching in computer vision application , 2010, 2010 3rd International Conference on Computer Science and Information Technology.

[6]  Alan C. Bovik,et al.  FOVEA: a foveated vergent active stereo vision system for dynamic three-dimensional scene recovery , 1998, IEEE Trans. Robotics Autom..

[7]  D Marr,et al.  A computational theory of human stereo vision. , 1979, Proceedings of the Royal Society of London. Series B, Biological sciences.

[8]  R. A. Hamzah,et al.  A pixel matching process and multiple ROI for stereo images in stereo vision application , 2011, 2011 IEEE 3rd International Conference on Communication Software and Networks.

[9]  Jiri Matas,et al.  Robust wide-baseline stereo from maximally stable extremal regions , 2004, Image Vis. Comput..

[10]  G. G. Stokes "J." , 1890, The New Yale Book of Quotations.

[11]  Yanjun Li,et al.  Design and calibration of a single-camera-based stereo vision sensor , 2006 .

[12]  Bärbel Mertsching,et al.  Pre-attentive detection of depth saliency using stereo vision , 2010, 2010 IEEE 39th Applied Imagery Pattern Recognition Workshop (AIPR).

[13]  Fuqiang Zhou,et al.  A novel way of understanding for calibrating stereo vision sensor constructed by a single camera and mirrors , 2013 .