Radiant Vector Flow Method for Arbitrarily Oriented Scene Text Detection

Text detection and recognition is a hot topic for researchers in the field of image processing. It gives attention to Content based Image Retrieval community in order to fill the semantic gap between low level and high level features. Several methods have been developed for text detection and extraction that achieve reasonable accuracy for natural scene text as well as multi-oriented text. However, it is noted that most of the methods use classifier and large number of training samples to improve the text detection accuracy. The multi-orientation problem can be solved using the connected component analysis method. Since the images are high contrast images, the classifier with connected component analysis based feature training work well for achieving better accuracy. It cannot be used directly for text detection in video because of low contrast and complex background which causes problem such as disconnections and loss of shapes. The deciding classifier and geometrical features of the components is not that much easy in this case. To overcome from this problem our proposed research uses radiant Vector Flow and Grouping based Method for Arbitrarily Oriented Scene text Detection method. The GVF of edge pixels in the Sobel edge map of the input frame is explored to identify the dominant edge pixels which represent the text components. These extracts edge components method corresponding to dominant pixels in the Sobel edge map, which we call Text Candidates (TC) of the text lines. Experimental results on different datasets including arbitrarily oriented text data, horizontal and non-horizontal text data, Hua’s data and ICDAR-03 data.