Review on Hand Gesture Recognition using Artificial Intelligence based application

DOI

No DOI

Published In

Volume 3, Issue 4 (2022)

Published Online

July-August 2022

Pages

119-122

Article Type

Research Article

Check for updates

No DOI available for cross-reference

Abstract: The analysis is concentrated on style and development of an economical framework for hand gesture recognition used for the implementation of hand mouse and a sensible ward system in super speciality hospitals. within the field of laptop vision and pattern analysis the most important difficult drawback is that the vision-based hand gesture recognition. the issues like dynamic background, hand segmentation, camera activity, speed and want of external knowledge gloves has redoubled the analysis work on the vision – primarily based hand gesture recognition system. The hand gestures are obtained from the live video internet camera and kinect sensing element camera. The work is enforced on MatLab tool version thirteen. The analysis advises the segmentation of the hand gesture exploitation the skin color detection, motion detection and our projected rule that is that the integration of skin color detection and motion detection rule. actuality positive, False positive, True negative, False negative, sensitivity, property, preciseness and F1_Score values are determined for all 3 algorithms and compared. The Face detection rule is employed to delete the face detected thus on confine the detection solely to the hand. The higher than technique was sensitive to the lighting condition, human skin color, the background, the shadow effects, etc. to beat the higher than limitations segmentation supported depth data is adopted. The projected work is performed by detection and recognizing the gesture obtained through the Kinect camera. Here, the coaching includes knowledge assortment and have extraction. Secondly, the trained knowledge is assessed exploitation K Nearest Neighbors (KNN), Support Vector Machines (SVM) and Artificial Neural Networks (ANN) strategies. To adopt the most effective classifier this paper compares the accuracy of all the higher than techniques. Mode choice operation has been tested with 3 completely different classifiers and SVM is tried to be best out of them. The analysis is finished supported varied performances metrics like classification effectiveness, accuracy and recognition rate.