• Vrushali Ovhal
• Amruta patil
• Prajakta naidu
• Prof S.S.Raskar
• Prof S.P.Deore

Many studies concerning author identification are done on totally different written documents however less number of researches has been according on Marathi or syllabary documents. Here, during this technique, we have a tendency to settle for handwriting as texture image and have extraction done from preprocessed documents image. Considerably, the feature of conferred methodology uses feature extraction that is appropriate for structure of Devnagari written texts. And a brand new feature extraction technique is additionally conferred that is predicated on classifier style. Firstly, we have a tendency to study totally different ways for feature extraction from output of preprocessing. This enforced system represents performance of mimicree estimation techniques on syllabary written documents. Mimicree could be a distinctive feature; it acts as an important role in author identification, GAN. Further to enhance the performance of those techniques, the extracted pictures area unit reconstructed victimisation GAN algorithmic program for Devnagri datasets area unit used. The experimental result shows that mimicree calculable for pictures reconstructed victimisation GAN algorithmic program provides higher results.

GAN,BGR,Thresholding,HandWriting,Image Preprocessing.

"Handwriting Perusal Using Gan", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.7, Issue 5, page no.623-625, May-2020, Available :http://www.jetir.org/papers/JETIR2005228.pdf

"Handwriting Perusal Using Gan", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.7, Issue 5, page no. pp623-625, May-2020, Available at : http://www.jetir.org/papers/JETIR2005228.pdf