A Novel Vision-Based Finger-Writing Character Recognition System

Abstract

A novel video-based finger writing virtual character recognition system (FVCRS) is described in this paper. With this FVCR system, one can enter characters into a computer by just using the movement of fingertip, without any additional device such as a keyboard or a digital pen. This provides a new wireless character-inputting method. A simple but effective background model is built for segmenting human-finger movements from cluttered background. A robust fingertip detection algorithm based on feature matching is given, and recognition of the finger-writing character is by a DTWbased classifier. Experiments show that the FVCRS can successfully recognize fingerwriting uppercase and lowercase English alphabet with the accuracy of 95.3% and 98.7%, respectively.

Seeing double

“…Within the computer vision and human computer interaction communities, work on finger tracking for gesture recognition [8,32], virtual input systems [40,48], virtual object manipulation [25,26], and hand writing recognition [19,46] all share similarities to our application of finger motion analysis. Probably the most germane of these is the work of Iturbe et al [18] which uses finger movement to control a virtual input system in which a user's finger is modeled as a straight line and its movement is used to determine the button being pointed to by the user.…”

“…Unfortunately, their approach quickly fails for small mobile devices where the fingers need to bend in order to reach the keys. Similarly, Jin et al [19] utilized finger input for character writing recognition. In their approach, the user's finger is isolated using a background modeling technique.…”

“…Next, the path taken by the finger is tracked as the user writes individual letters, effectively recognizing the letter through the trajectory of the finger. Sadly, the approach of Jin et al [19] can not be directly applied to mobile devices as users do not spell words by forming one character at a time, but instead interact with a keyboard via a series of touch events.…”

Camera-Based Motion Recognition for Mobile Interaction

“…The other solution studied in this paper, vision-based finger tracking, is well studied problem on desktop computers with numerous applications [16,17]. On mobile devices, Henrysson et al [18] considered how a front-facing camera on the phone can be used for 3-D augmented reality interaction.…”

Gesture Recognition for Chinese Traffic Police

“…Kang et al used upper-body gestures as the interface between a video game and its player and achieved an average success rate of 93.36% for the recognition of 10 gesture commands [3]. Zhen and Huang et al developed a video-based system for recognizing characters written with a finger [4]. It allows one to enter characters into the computer program by using the movement of a fingertip.…”