Analysis on Hand Gesture Recognition Using Artificial Neural Network

EasyChair Preprint 3166

Abstract

Hand gesture recognition system can be used in different area, for example, HCI remote control, android regulator, computer generated truth and so forth. Needle gesture recognition system is for the most part the investigation of identification and acknowledgment of different arrow motions like American Sign Verbal hand gestures, Danish Sign Language hand motions and so on by a processer system. This work is centered on three fundamental issues in building up a motion acknowledgment framework. This work is centered on three fundamental issues in building up a motion acknowledgment framework. Human Computer Interaction requires using various modalities (for example body position, speech, hand motions, Lip development, Facial articulations, and so on.) and coordinating them together for an increasingly vivid client experience. Hand signals are a natural yet ground-breaking correspondence methodology which as not been completely investigated for Human Computer Interaction The most recent computer vision, image processing methods make vision based hand gesture recognition plausible for Human Computer Interaction (HCI).Training includes data collection and feature extraction. Second, the artificial neural network (ANN) method is used to classify the training data. To adopt the best classifier, this paper compares the accuracy of all of the above technologies, demonstrates that the efficiency and effectiveness of the proposed system result performance are compared to the existing work by 90% and analyzes the algorithm% Cases achieve the highest accuracy.

Keyphrases: ANN, Gesture Recognition System, Hand Gesture Recognition system, Human Computer Interaction, feature extraction

@booklet{EasyChair:3166,
  author    = {Charu Chauhan and Hemant Maurya and Himanshu Tiwary and Anusha Jain},
  title     = {Analysis on Hand Gesture Recognition Using Artificial Neural Network},
  howpublished = {EasyChair Preprint 3166},
  year      = {EasyChair, 2020}}