Description:
Reference #: 1700
The University of South Carolina is offering licensing opportunities for Novel filters for image gradient computation and edge detection.
Background:
Gradient computation and edge determination are challenging yet crucial steps in many computer vision tasks. They are the first in multi-step processes, serving as the foundation for all subsequent operations. Their accuracy directly affects the success of any future processing techniques and final detection. The challenge of edge detection is derived from a variety of factors, including noise, image sharpness, orientation, empirical parameters, and computational complexity. Many traditional kernel-based operators excel at tackling one of these problems, but trade off their ability to handle others. Traditional edge detectors commonly result in false detection, missed detection, and inaccurate localization. Because of these issues, a systematic approach that can directly operate on the original images and raw pixel intensities and address most of the issues with minimal manual manipulation is still strongly needed. The biggest issues traditional kernels face are signal noise and obstruction, which interfere with the ability to obtain the true derivative values at a pixel.
Invention Description:
The proposed inventions (the wide view filter and the line filter) have been developed with the purpose of improving the ability to detect gradients in the presence of extreme noise and discontinuity. The proposed filters can accurately produce smooth connected gradient values suitable for edge determination better than any traditional gradient-based operator on the market. When combined with a developed novel angle detection technique, these filters are also capable of calculating gradient orientations within one degree of their true value.
Potential Applications:
The potential applications include autonomous vehicle driving, medical image processing, smart manufacturing, consumer photography, and consumer photography pipelines
Advantages and Benefits:
These filters are better than traditional filters because they are able to overcome problems like image signal noise, arbitrary edge orientation, edge discontinuity/obstruction, image sharpness, and low contrast boundaries, which traditional filters have not previously been able to overcome.