• Shrineeta Kurud
• Shalaka Deore
• Pooja Kumbharkar
• Yashshree Shinde
• Ashwini Kolekar

Segmentation of cursive handwriting is the challenging step of Optical Character Recognition (OCR).The recognition accuracy will highly depend on the good segmentation. Segmentation of cursive handwriting is very tough task. The segmentation may be done on the idea of segmentation, a line phase of text, a word phase from line and character phase from word. This can be done by the use of horizontal, vertical methods. By reviewing many basic and advanced techniques of handwritten word segmentation, we try to find the most optimal segmentation bred out of them.The importance of segmentation is to find the accurate new segmented words in order to make a new data set. There are many segmentation algorithms available but each algorithm has different accuracy. The Purpose of this project is to combine different structural and stochastic segmentation algorithms. The proposed system will be segmenting cursive handwritten English text. For this free available dataset from IAM database is used. By comparing the accuracy of other segmentation techniques we have tried to implement edge and region segmentation from structural segmentation & ANN from stochastic segmentation as hybrid segmentation.

Deep Learning, Image Segmentation, Structural Segmentation, Stochastic Segmentation, ANN.

"To perform Hybrid Segmentation on Cursive Handwritten Text", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.9, Issue 2, page no.a171-a174, February-2022, Available :http://www.jetir.org/papers/JETIR2202022.pdf

"To perform Hybrid Segmentation on Cursive Handwritten Text", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.9, Issue 2, page no. ppa171-a174, February-2022, Available at : http://www.jetir.org/papers/JETIR2202022.pdf