Convolutional Neural Network Based American Sign Language Static Hand Gesture Recognition

Convolutional Neural Network Based American Sign Language Static Hand Gesture Recognition

Ravinder Ahuja (Jaypee Institute of Information Technology Noida, Hansi, India), Daksh Jain (Jaypee Institute of Information Technology Noida, Delhi, India), Deepanshu Sachdeva (Jaypee Institute of Information Technology Noida, Delhi, India), Archit Garg (Jaypee Institute of Information Technology Noida, Ghaziabad, India) and Chirag Rajput (Jaypee Institute of Information Technology Noida, New Delhi, India)
Copyright: © 2019 |Pages: 14
DOI: 10.4018/IJACI.2019070104
OnDemand PDF Download:
No Current Special Offers


Communicating through hand gestures with each other is simply called the language of signs. It is an acceptable language for communication among deaf and dumb people in this society. The society of the deaf and dumb admits a lot of obstacles in day to day life in communicating with their acquaintances. The most recent study done by the World Health Organization reports that very large section (around 360 million folks) present in the world have hearing loss, i.e. 5.3% of the earth's total population. This gives us a need for the invention of an automated system which converts hand gestures into meaningful words and sentences. The Convolutional Neural Network (CNN) is used on 24 hand signals of American Sign Language in order to enhance the ease of communication. OpenCV was used in order to follow up on further execution techniques like image preprocessing. The results demonstrated that CNN has an accuracy of 99.7% utilizing the database found on
Article Preview


Sign language has always been the primary way of verbal communication among people who are both deaf and dumb. While communicating these people become very helpless and thus are only dependent on hand gestures. Visual gestures and signs which are a vital part of ASL that provide deaf and mute people an easy and reliable way of communication. It consists of the well-defined code gesture where each sign conveys the particular meaning in terms of communication. Inspire of 143 existing different sign languages all over the world, only some of them finds their position in the main list. American Sign Language, Bruisers Sign Language, Japanese Sign Language, French Sign Language, and Indian Sign Language are some of them (Parvini & Shahabi, 2007). These main languages are in use worldwide and differ according to the natives of the particular region (Wang & Popović, 2009). American Sign Language (ASL) is being utilized by the people in the entire world and considered as a standard sign language. There are many techniques for seeking gestural data. But restricting to only main types there are two important known types: Sensor-based method and Vision-based Method. The sensor-based method collects data from the glove generated by hand movement. In the vision-based method, the image is taken with the help of cameras (Kevin, Ranganath & Ghosh, 2004). This method incorporates the qualities of the image such as coloring and texture part that is compulsory for finding out the particular hand gesture (Yun & Peng, 2009; Huong, Huu & Le Xuan, 2015).

In this paper, sign gesture recognition of American Sign Language is proposed using CNN. Input images are taken from the user’s webcam. The input image is preprocessed by applying smoothening and then tested. At the completion, a new and efficient approach which will add a new dimension to the research work that is looking forward to the disabled society in a broader aspect. The rest of the paper is organized as follows: Section II contains Literature Review, section III contains Data Set collection, section IV proposed approach, section V contains results, and section VI contains a conclusion.

Complete Article List

Search this Journal:
Open Access Articles
Volume 13: 6 Issues (2022): Forthcoming, Available for Pre-Order
Volume 12: 4 Issues (2021)
Volume 11: 4 Issues (2020)
Volume 10: 4 Issues (2019)
Volume 9: 4 Issues (2018)
Volume 8: 4 Issues (2017)
Volume 7: 2 Issues (2016)
Volume 6: 2 Issues (2014)
Volume 5: 4 Issues (2013)
Volume 4: 4 Issues (2012)
Volume 3: 4 Issues (2011)
Volume 2: 4 Issues (2010)
Volume 1: 4 Issues (2009)
View Complete Journal Contents Listing