700 Journals and 15,000,000 Readers Each Journal is getting 25,000+ ReadersThis Readership is 10 times more when compared to other Subscription Journals (Source: Google Analytics)
Research Article Open Access
Text provides important information about images or video sequences in a documented image, but it always remains difficult to modify the static documented image. To carry out modification in any of the text matter the text must be segmented out from the documented image, which can be used for further analysis. Taking consideration to video image sequence the isolation of text data from the isolated frame becomes more difficult due to its variable nature. Various methods were proposed for the isolation of text data from the documented image. Among which Wavelet transforms have been widely used as effective tool in text segmentation. Document images usually contain three types of texture information. various wavelet transformation have been proposed for the decomposition of these images into their fundamentals feature. Onto these wavelet families, it is one of the difficult tasks in selecting a proper wavelet transformation with proper scale level for text isolation. This paper work implements an efficient text isolation algorithm for the extraction of text data from the documented video clips. The implemented system carries out a performance analysis on various wavelet transforms for the proper selection of wavelet transform with multi level decomposition. Of the selected wavelet transform the obtained wavelet a coefficient are applied with morphological operators for text isolation and evaluates the contribution of decomposition levels and wavelet functions to the segmentation result in documented video image. The proposed task implements neural network for the recognition of text characters from the isolated text image for making it.
decomposition, images, isolation Wavelet transforms, Segmentation, Aerospace Engineering,Applied Electronics,Applied Sciences,Biogenetic Engineering,Bio medical Engineering,Fluid Dynamics.