Title :
Implementation of MD algorithm for text extraction from video
Author :
Sonam ; Kumar, Manoj
Author_Institution :
Dept. of Comput. Eng., Technol. Inst. of Textile & Sci., Bhiwani, India
Abstract :
With fast intensification of existing multimedia documents and mounting demand for information indexing and retrieval, much effort has been done on extracting the text from images and videos. Extracting the text from video is demanding due to complex background, varying font size, different style, high blurring, lower resolution, different position, viewing angle and so on. In this paper, implementation of text extraction from the video has been done by using the MD algorithm. In this paper, a region based approach is used to perform such kind of text extraction. The approach is divided into the three main layers: first, extraction of unique frames from video; second, identify the text frame from these unique frames; third, extraction of the text from these text frames. DWT performs the decomposition in the region and Morphological Operator performs the edge detection and the region classification.
Keywords :
discrete wavelet transforms; edge detection; feature extraction; information retrieval; mathematical morphology; text detection; DWT; MD algorithm; edge detection; images; information indexing; information retrieval; morphological operator; multimedia documents; region classification; text extraction; video; Algorithm design and analysis; Data mining; Discrete wavelet transforms; Image resolution; Morphology; Shape; DWT (Discrete wavelet transform); MD (Morphology-DWT); MIP (Mathematical image processing); TIE (Text information extraction system);
Conference_Titel :
Engineering (NUiCONE), 2013 Nirma University International Conference on
Conference_Location :
Ahmedabad
Print_ISBN :
978-1-4799-0726-7
DOI :
10.1109/NUiCONE.2013.6780069
