3D Text Segmentation and Recognition using Leap Motion Sensor

Author(s):

Prajakta Umeshchandra Patale-Patil , SBPCOE,Indapur; Suvidha Vijay Jagadale, S.B.P COE Indapur; Supriya Dilip Shinde, S.B.P COE Indapur; Priyanka Sanjay Tengale, S.B.P COE Indapur

Keywords:

Text segmentation, Feature point Extraction, Cosine Similarity algorithm, leap motion sensor, Air writing

Abstract:

3D texts drawn by fingers using Leap motion sensor can be provocative for existing text recognition framework. We present a method of Human-Computer-Interaction (HCI) through 3D air-writing. Previous system worked on pen-paper based system. The 3D gestures are recorded in air they produce often non-uniform text style and jitter-effect while writing. Air writing, characters, words, and lines are usually connected by continuous stroke that makes it difficult to recognize. In this, we will present 3D text segmentation and recognition using leap motion sensor. In this system we will use Feature Point Extraction Algorithm and Cosine Similarity Algorithm. It can be differentiated from it's neighbouring points. Feature with similar co-ordinate in both text will be compared. Cosine similarity in measure of similarity between two non-zero vectors of an inner product space that measure cosine of angle between them. It recognizes words and characters and 3D sequences of segmented words. For this we are creating a dataset. Using 3D text segmentation and recognition it reduce pen-paper based system.

Other Details:

Manuscript Id	:	IJSTEV4I5042
Published in	:	Volume : 4, Issue : 5
Publication Date	:	01/12/2017
Page(s)	:	142-143

Download Article