REINFORCEMENT OF LINEAR STRUCTURE USING PARAMETRIZED RELAXATION LABELING

The problem of reinforcing local evidence of linear structure while suppressing unwanted information in noisy images is considered, using a new modified form of relaxation labeling. The methodology is based on parametrizing a continuous set of orientation labels via a single vector and using a sigmoidal thresholding function to bias neighborhood influence and ensure convergence to meaningful stable states. Label strength and label/no-label decisions are incorporated into a single functional. Optimal points of the functional represent the cases where as many pixels (objects) as possible have achieved the desirable linear-structure-reinforced and noise-suppressed labelings. Three different linear structure reinforcement tasks are considered within the general framework: edge reinforcement, edge reinforcement with thinning, and bar (line segment) reinforcement. Results are presented from several image data sets. The primary advantages to this approach are that it can directly handle continuous feature information (both magnitude and direction) from low-level image analysis operators (primarily edge and line detectors) and that the computational complexity of labeling is reduced due to the parametrization. Both of these improvements are of considerable practical utility.