Tri-level handwritten text segmentation techniques for Gujarati language

Objectives: To improve the efficiency of tri-level segmentation tasks for handwritten Gujarati text. Methods: Using hybrid methods segmentation, we have used line, word and character from image. This study presents a paradigm that works with touching characters, slop line written on page, overlapping, etc. It evaluated dataset 500+ images created by us different writing sentences people. We Horizontal projection technique Scale-space Vertical segmentation. Findings: The experimental results show proposed method is more efficient text diacritics. obtained accuracy level 82%, word-level 90% line-level 87%. Novelty: designed methodology to segment diacritics at all three levels including words lines. Applications: which pre-processing task can be in any recognition systems i.e. OCR. Keywords: Deep learning; trilevel segmentation;