A dynamic statistical nonparametric cleaning and enhancement system for highly degraded ancient handwritten modi lipi documents
Manisha S. Deshmukh, Manoj P. Patil, Satish Ramesh Kolhe · 2017
A nonparametric technique for the cleaning and enhancement of the chronic old age Modi document images is presented here. The technique is based on the dynamic thresholding and connected component analysis. The automatic fixation of the Niblack factor k is proposed as per the document degradation type. The dynamic thresholding algorithm is used to segment the noisy pixels from the text area and background of the document image. To eliminate some sort of noise like clutter, marginal and blobs, connected components are calculated and analyzed. To improve the quality of the document image filtering techniques, and morphological operations are employed. The performance of the approach is tested and verified on the variety of old age degraded noisy Modi script documents. To evaluate the performance of the proposed approach, it is compared with benchmarking techniques illustrated in literature.