Support Vector Machine Based Handwritten Hindi Character Recognition and Summarization

摘要

In today’s digital era, the text may be in form of images. This researchaims to deal with the problem by recognizing such text and utilizing the supportvector machine (SVM). A lot of work has been done on the English language forhandwritten character recognition but very less work on the under-resourced Hindilanguage. A method is developed for identifying Hindi language characters thatuse morphology, edge detection, histograms of oriented gradients (HOG), andSVM classes for summary creation. SVM rank employs the summary to extractessential phrases based on paragraph position, phrase position, numerical data,inverted comma, sentence length, and keywords features. The primary goal ofthe SVM optimization function is to reduce the number of features by eliminatingunnecessary and redundant features. The second goal is to maintain or improvethe classification system’s performance. The experiment included news articlesfrom various genres, such as Bollywood, politics, and sports. The proposed method’saccuracy for Hindi character recognition is 96.97%, which is good comparedwith baseline approaches, and system-generated summaries are compared tohuman summaries. The evaluated results show a precision of 72% at a compressionratio of 50% and a precision of 60% at a compression ratio of 25%, in comparisonto state-of-the-art methods, this is a decent result.