Extraction journals sub titles via wavelet threshold

摘要

Text extraction is one of most interest for researchers. In this paper, we propose a new robust and efficient algorithm to become such a software extracts text of colored book and journal cover sheets is. First, we perform wavelet decomposition. Next for edge detecting from detail coefficients, we use a dynamic threshold. By blurring approximate coefficients with alternative heuristic thresholding, considering horizontal coefficient, achieve effective edge. Afterward, ROI technique get binary image. Finally, text boxes would be extracted with new projection profile.