Novel Approach for Segmentation

Abstract

Written by hand text line segmentation is viewed as a noteworthy test in record picture examination. It needs to explain a few hindrances that are remarkable in present day printed text. Among the most are: skewed lines, curvilinear lines, contacting and covering parts, normally words or letters, among lines and anomaly in geometrical properties of line. Text line extraction from unconstrained written by hand reports is a test in light of the fact that the text lines are frequently skewed and the space between lines isn't self-evident. The intricacy engaged with the segmentation of the Handwritten Documents for Indian dialects like Telugu, Tamil and Malayalam. Bended and non-parallel text lines manually written archives likewise make the segmentation and acknowledgment testing. Text line segmentation of manually written records written in Kannada content by utilizing Independent Component Analysis (ICA) is thought about.