TEXT EXTRACTION FROM IMAGES/VIDEOS USING MORPHOLOGICAL OPERATION

Abstract

Popularity of digital image & video is increasing rapidly, Text characters embedded in images and video sequences represents a rich source of information for content-based indexing and retrieval applications However, these text characters are difficult to be detected, recognized, extracted, due to their various sizes, gray scale values and complex backgrounds. This thesis investigates methods for building an efficient application system for text extraction of any gray scale values embedded in images and video sequences. Text embedded in multimedia data, as a well-defined model of concepts for humans‘ communication, contains much semantic information related to the content. This text information can provide a much truer form of content–based access to the image and video documents if it can be extracted efficiently. My work in this thesis is not focused on Text Recognition rather than on Text Extraction parts(Text Detection). Text Extraction plays a major role in finding vital and valuable information. These text characters are difficult to be detected and recognized due to their deviation of size, font, style, orientation, alignment, contrast, complex colored, textured background. Due to rapid growth of available multimedia documents and growing requirement for information, identification, indexing and retrieval, many researches have been done on text extraction in images & videos. Several techniques have been developed for extracting the text from an image & video. The proposed methods were based on morphological operators, wavelet transform, artificial neural network, skeletonization operation, edge detection algorithm, histogram technique etc. All these techniques have their benefits and restrictions. This thesis discusses morphological operation based scheme (using extraction of region containing Text) for extracting the text from an image & video. Text extraction is a challenging problem due to the presence of complex backgrounds and the variable length of the text Strings appearing on screen with variable heights. My work would follow in the sequence like, I converted frames into gray scale ,then localization of text frame, followed by ROI(Region of Extraction) steps &then binary conversion(Binarization) in themeanwhile morphological operation has been done on binary images, in last execution of text extraction/detection step taken place.