Optimization of retina-like sensor parameters based on visual task requirements

Abstract. Retina-like sensors maximize both field of view and resolution in addition to economizing on pixel count, so they play an important role in both biological and machine vision. A new retina-like sensor model for compensating motion blur introduced by forward motion imaging is proposed. Next, the determination of pixel arrangement of a retina-like sensor according to visual task requirements is formulated into a multiobjective optimization problem. Then, three retina-like sensors are designed to meet different visual task requirements using the particle swarm optimization algorithm. The results are robust and approximate to design criteria.

[1]  Paolo Gamba,et al.  Log-map analysis , 2008, Parallel Comput..

[2]  Christopher M. Brown,et al.  Real-time binocular smooth pursuit , 1993, International Journal of Computer Vision.

[3]  C. Koch,et al.  A saliency-based search mechanism for overt and covert shifts of visual attention , 2000, Vision Research.

[4]  Alexandre Bernardino,et al.  A review of log-polar imaging for visual perception in robotics , 2010, Robotics and Autonomous Systems.

[5]  Xuejie Zhang,et al.  A spatial variant approach for vergence control in complex scenes , 2011, Image Vis. Comput..

[6]  Frédéric Jurie,et al.  A new log-polar mapping for space variant imaging.: Application to face detection and tracking , 1999, Pattern Recognit..

[7]  Narendra Ahuja,et al.  Detecting Faces in Images: A Survey , 2002, IEEE Trans. Pattern Anal. Mach. Intell..

[8]  Wei Zhao,et al.  Wide field-of-view foveated imaging system , 2008 .

[9]  Ioan Cristian Trelea,et al.  The particle swarm optimization algorithm: convergence analysis and parameter selection , 2003, Inf. Process. Lett..

[10]  Alexandre Bernardino,et al.  Binocular tracking: integrating perception and control , 1999, IEEE Trans. Robotics Autom..

[11]  Jean Taboury,et al.  Design strategies to simplify and miniaturize imaging systems. , 2011, Applied optics.

[12]  T. Pajdla,et al.  OMNIVIEWS: direct omnidirectional imaging based on a retina-like sensor , 2002, Proceedings of IEEE Sensors.

[13]  Filiberto Pla,et al.  Similarity motion estimation and active tracking through spatial-domain projections on log-polar images , 2005, Comput. Vis. Image Underst..

[14]  Giulio Sandini,et al.  Retinalike space variant CCD sensor , 1990, Other Conferences.

[15]  Giulio Sandini,et al.  Direction Control for an Active Docking Behaviour Based on the Rotational Component of Log-Polar Optic Flow , 2000, ECCV.

[16]  Giulio Sandini,et al.  Anthropomorphic Visual Sensors , 2005 .

[17]  Filiberto Pla,et al.  Log-polar mapping template design: From task-level requirements to geometry parameters , 2008, Image Vis. Comput..

[18]  Robert B. Fisher,et al.  Object-based visual attention for computer vision , 2003, Artif. Intell..

[19]  Bart Dierickx,et al.  Space-variant nonorthogonal structure CMOS image sensor design , 1998, IEEE J. Solid State Circuits.

[20]  Zhang Lei.  Imaging Superiority of Non-Rectangular Detector Array in Forward Motion , 2011 .

[21]  Osamu Tsuzuki,et al.  Practical design and evaluation methods of omnidirectional vision sensors , 2012 .

[22]  Juan Domingo,et al.  On the advantages of combining differential algorithms and log-polar vision for detection of self-motion from a mobile robot , 2001, Robotics Auton. Syst..

[23]  Giulio Sandini,et al.  A retina-like CMOS sensor and its applications , 2000, Proceedings of the 2000 IEEE Sensor Array and Multichannel Signal Processing Workshop. SAM 2000 (Cat. No.00EX410).

[24]  Riccardo Poli,et al.  Particle swarm optimization , 1995, Swarm Intelligence.