Orientation Contrast Detection in Space-Variant Images

In order to appropriately act in a dynamic environment, any biological or artificial agent needs to be able to locate object boundaries and use them to segregate the objects from each other and from the background. Since contrasts in features such as luminance, color, texture, motion and stereo may signal object boundaries, locations of high feature contrast should summon an agent's attention. In this paper, we present an orientation contrast detection scheme, and show how it can be adapted to work on a cortical data format modeled after the retino-cortical remapping of the visual field in primates. Working on this cortical image is attractive because it yields a high resolution, wide field of view, and a significant data reduction, allowing real-time execution of image processing operations on standard PC hardware. We show how the disadvantages of the cortical image format, namely curvilinear coordinates and the hemispheric divide, can be dealt with by angle correction and filling-in of hemispheric borders.

[1]  H. C. Nothdurft,et al.  Texture segmentation and pop-out from orientation contrast , 1991, Vision Research.

[2]  A. H. Bunt,et al.  Foveal sparing. New anatomical evidence for bilateral representation of the central retina. , 1977, Archives of ophthalmology.

[3]  Eric L. Schwartz,et al.  Design considerations for a space-variant visual sensor with complex-logarithmic geometry , 1990, [1990] Proceedings. 10th International Conference on Pattern Recognition.

[4]  Giulio Sandini,et al.  An anthropomorphic retina-like structure for scene analysis , 1980 .

[5]  Bruce Fischl,et al.  The Local Structure of Space-variant Images , 1997, Neural Networks.

[6]  Martin D. Levine,et al.  A Review of Biologically Motivated Space-Variant Data Reduction Models for Robotic Vision , 1998, Comput. Vis. Image Underst..

[7]  Carl F. R. Weiman,et al.  Logarithmic spiral grids for image-processing and display , 1979 .

[8]  H. Nothdurft.  The role of features in preattentive vision: Comparison of orientation, motion and color cues , 1993, Vision Research.

[9]  Giulio Sandini,et al.  On the Advantages of Polar and Log-Polar Mapping for Direct Estimation of Time-To-Impact from Optical Flow , 1993, IEEE Trans. Pattern Anal. Mach. Intell..

[10]  B. Boycott,et al.  Retinal ganglion cell density and cortical magnification factor in the primate , 1990, Vision Research.

[11]  Ramesh C. Jain,et al.  Motion Stereo Using Ego-Motion Complex Logarithmic Mapping , 1987, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[12]  Eric L. Schwartz,et al.  Computational anatomy and functional architecture of striate cortex: A spatial mapping approach to perceptual coding , 1980, Vision Research.

[13]  Kostas Daniilidis.  Computation of 3-D-Motion Parameters Using the Log-Polar Transform , 1995, CAIP.