Please use this identifier to cite or link to this item:
http://148.72.244.84/xmlui/handle/xmlui/15839
Title: | Human Gait Recognition using an Enhanced Convolutional Neural Network |
Authors: | Esmail Sadeq, Fatima Tariq Mustafa Al-Ta'i, Ziyad |
Keywords: | Gait recognition, Soft Biometrics, MediaPipe, Enhanced Convolutional Neural Networks |
Issue Date: | 31-Jul-2024 |
Publisher: | University of Diyala |
Abstract: | Gait is a kind of behavioral biometric feature, which is defined as the way a person walks. Unlike other biometrics like face and iris which are limited by the distance. Soft biometric are features that can be extracted remotely and do not require human interaction. The force of gait, is that it does not require cooperative subjects and it is recognizable from low-resolution surveillance videos. This paper presents a proposed framework for gait recognition by building the required dataset. This work include two steps. First, nine gait attributes are extracted using MediaPipe and second, recognition is done using an Enhanced Convolutional Neural Network (ECNN). The proposed model achieved an accuracy of 89.583%.Although the accuracy is not high, yet the gait recognition is very important, especially in a remote viewing environment. |
URI: | http://148.72.244.84/xmlui/handle/xmlui/15839 |
ISSN: | 2958-4612 |
Appears in Collections: | مجلة ديالى للعلوم الاكاديمية / Academic Science Journal (Acad. Sci. J.) |
Files in This Item:
File | Description | Size | Format | |
---|---|---|---|---|
13-796.pdf | 1.03 MB | Adobe PDF | View/Open |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.