Driving Active Contours to Concave Regions

Broken characters restoration represents the major challenge of optical character recognition (OCR). Active contours, which have been used successfully to restore ancient documents with high degradations drawback in restoring deep concavity boundaries. Deep problem main obstacle, has prevented Gradient Vector Flow active contour converge objects complex In this paper, we proposed a technique enhance (GVF) using particle swarm optimization (PSO) through directing snake points (snaxels) toward correct positions into boundaries broken by comparing genetic algorithms as an method. Our experimental results showed that outperform on algorithm capturing converged areas and save spent time process.