A novel ring radius transform for video character reconstruction

Character recognition in video is a challenging task because low resolution and complex background of video cause disconnections, loss of information, loss of shapes of the characters etc. In this paper, we introduce a novel ring radius transform (RRT) and the concept of medial pixels on characters with broken contours in the edge domain for reconstruction. For each pixel, the RRT assigns a value which is the distance to the nearest edge pixel. The medial pixels are those which have the maximum radius values in their neighborhood. We demonstrate the application of these concepts in the problem of character reconstruction to improve the character recognition rate in video images. With ring radius transform and medial pixels, our approach exploits the symmetry information between the inner and outer contours of a broken character to reconstruct the gaps. Experimental results and comparison with two existing methods show that the proposed method outperforms the existing methods in terms of measures such as relative error and character recognition rate.

► The Gradient Vector Flow and the least cost path finding for character segmentation.
► A novel Ring Radius Transform with medial axis for character reconstruction.
► Use of the medial axis values in novel way to fill the gap in character contours.
► Iterative filling algorithms to fill the large gaps.
► The mutual nearest neighbour with end points to fill the remaining gaps.