Boundary Detection Using Double-Opponency and Spatial Sparseness Constraint
Brightness and color are two basic visual features integrated by the human visual system (HVS) to gain a better understanding of color natural scenes. Aiming to combine these two cues to maximize the reliability of boundary detection in natural scenes, we propose a new framework based on the color-opponent mechanisms of a certain type of color-sensitive double-opponent (DO) cells in the primary visual cortex (V1) of HVS. This type of DO cells has oriented receptive field with both chromatically and spatially opponent structure. The proposed framework is a feed forward hierarchical model, which has direct counterpart to the color-opponent mechanisms involved in from the retina to V1. In addition, we employ the spatial sparseness constraint (SSC) of neural responses to further suppress the unwanted edges of texture elements. Experimental results show that the DO cells we modeled can flexibly capture both the structured chromatic and achromatic boundaries of salient objects in complex scenes when the cone inputs to DO cells are unbalanced. Meanwhile, the SSC operator further improves the performance by suppressing redundant texture edges. With competitive contour detection accuracy, the proposed model has the additional advantage of quite simple implementation with low computational cost.
Object boundaries represent important cues for visual perception such as scene understanding and object recognition. Boundary detection is also a fundamental building block for a large variety of computer vision applications, such as image segmentation and object detection.
Among numerous computational boundary detection, typical methods include zero crossing , phase congruency, Canny detector. To detect boundaries from color images, many early studies focused on extending the standard edge detectors, such as Canny to color space.
DISADVANTAGES OF EXISTING SYSTEM:
These methods are inherently difficult to discriminate salient object boundaries and texture edges due that they respond to all the discontinuities in image intensity, color or texture.
Our new boundary detection system is based on the double-opponent (DO) mechanism and has the amazing property of jointly extracting color- and luminance-defined edges, which is really different from the two-step way of some existing models that explicitly extract the color and luminance edges in separate channels and then combine them, e.g., with a supervised learning.
A new strategy of spatial sparseness constraint (SSC) was introduced to weight the edge responses of the proposed CO system, which provides a simple while efficient way for texture suppression.
ADVANTAGES OF PROPOSED SYSTEM:
Competitive performance for edge detection and texture suppression with only low-level local information.
Flexibility in responding to color- and luminance-defined boundaries.
As few as only one free parameter (i.e., cone weight) (the model is almost insensitive to another parameter, Gaussian scale σ).
Quite low computational cost.
System : Pentium IV 2.4 GHz.
Hard Disk : 40 GB.
Floppy Drive : 44 Mb.
Monitor : 15 VGA Colour.
Ram : 512 Mb.
Operating system : Windows XP/7.
Coding Language : MATLAB
Tool : MATLAB R2013A
Kai-Fu Yang, Shao-Bing Gao, Ce-Feng Guo, Chao-Yi Li, and Yong-Jie Li, Member, IEEE, “Boundary Detection Using Double-Opponency and Spatial Sparseness Constraint”, IEEE TRANSACTIONS ON IMAGE PROCESSING, VOL. 24, NO. 8, AUGUST 2015.