Handwritten Character Recognition Using SIFT Algorithm

Image processing is the procedure which is used to process various images. In this work, text is extracted from the natural scene images. The natural scene images are those images which are seen daily. In the recent times various techniques has been proposed and among these character detection and make apply grouping on these detected characters for batter recognizing. The daily scene images can oriented on many angles due to which accuracy of text extraction can be reduced. In this work, improvement will be proposed using SIFT algorithm for batter reorganization. The simulation of the algorithm is performed in MATLAB and it is analyzed that proposed algorithm performs well in terms of PSNR, MSE and Accuracy.