Preliminary study of recognizing alphabet letter via hand gesture

Abstract

Communication is one of human basic necessity. People with disabilities usually use sign language to communicate with others. The study identifies the alphabet letters from the finger spelling based on American Sign Language (ASL) standard. The bending sensors were attached to three fingers of spelling glove where the bending movement of each finger contributes different resistance value of the bending sensors. The combinations of resistance value from each finger were detected from microcontroller's library and alphabet letters will be displayed. This study assists normal person who are new to sign language to understand the spelling letters when communicate with the disabilities.

A Review on Systems-Based Sensory Gloves for Sign Language Recognition State of the Art between 2007 and 2017

“…The change in resistance of flex sensors was fed into the Arduino Nano to implement the template matching algorithm in order to recognize the specific gesture and prerecorded audio command that was stored in the Secure Digital (SD) memory card; this was then transmitted through a speaker and display the text through a 16 × 2 LCD display [ 68 ]. The hardware structure of [ 63 ] consists of three flex sensors, an INA 126 instrumentation amplifier, an analog-to-digital converter (ADC), a 16F877A microcontroller, and an LCD. The sensors were placed on three fingers, namely, the ring, middle, and index fingers, to identify the letters of the alphabet from the finger spelling of ASL.…”

“…Thus, the availability of reliable and accessible data has an important impact on helping researchers conduct hassle-free data collection. The benefit of using pre-existing data is to save time and effort and to be accurate when relying on reliable data [ 63 ]. Two-way communication: At present, the electronics market is provided with various SLR devices to support the abled individuals.…”

“…The main device adopted by researchers as the output was the screen of computer [ 3 , 4 , 35 , 40 , 48 , 60 , 61 , 62 ]. Other devices attracting the attention of researchers are the liquid-crystal display (LCD) [ 12 , 35 , 57 , 63 ], speaker [ 38 , 46 , 54 , 64 ], or both [ 6 , 51 , 55 , 65 , 66 , 67 , 68 ]. Ultimately, the smartphone is another alternative chosen for system output [ 9 , 42 , 48 , 59 , 69 ].…”

“…Hidden Markov Models (HMMs) is a popular technique that has shown its potential in numerous applications such as computer vision, speech recognition, molecular biology, and SLR [ 53 , 73 , 74 , 75 , 76 ]. Besides the HMM, the KNN is also used to classify hand gestures [ 63 ], and the KNN classifier with support vector machines (SVM) has been applied in posture classification in [ 77 ]. The KNN is applied in work on the recognition of ASL signs [ 40 ].…”