Scene Text Extraction from Videos Using Hybrid Approach

With fast intensification of existing multimedia documents and mounting demand for information indexing and retrieval, much endeavor has been done on extracting the text from images and videos. The prime intention of the projected system is to spot and haul out the scene text from video. Extracting the scene text from video is demanding due to complex background, varying font size, different style, lower resolution and blurring, position, viewing angle and so on. In this paper we put forward a hybrid method where the two most well-liked text extraction techniques i.e. region based method and connected component (CC) based method comes together. Initially the video is split into frames and key frames obtained. Text region indicator (TRI) is being developed to compute the text prevailing confidence and candidate region by performing binarization. Artificial Neural network (ANN) is used as the classifier and Optical Character Recognition (OCR) is used for character verification. Text is grouped by constructing the minimum spanning tree with the use of bounding box distance.