Paper Title
Extraction Of Text From Video Clips
Abstract
Text in video is very compact and accurate clue for video indexing and summarization. I n this paper, an algorithm
is designed that the new Fourier-Statistical Features (FSF) in RGB space for detecting text in video frames of unconstrained
background, different fonts, different scripts and different font sizes. This work consists of two parts namely automatic
classification of text frames from a large database of text and non-text frames and FSF in RGB for text detection in the
classified text frames. For text frame classification, presents novel features based on three visual cues; Max-Min method and
sharpness in filter-edge maps to identify a true text frame. For text detection in video frames, presents the new Fourier
transform based features in RGB space with statistical features and the computed FSF features from RGB bands are subject
to Fuzzy C-means clustering to classify text pixels from the background of the frame. Text blocks of the classified text pixels
are determined by analyzing the projection profiles and extract the text part from the video frame.