•  
  •  
 

Keywords

overftting, data pre-processing, classifcation, multi-layer perceptron, recognition, WEKA

Document Type

Research Article

Abstract

Motion analysis has been an active research area for the past decade. Several approaches had been proposed to detect and recognize motion activity for diferent applications such as motion estimation, modeling, and reconstruction. However, a suitable classifer is required to be embedded with the surveillance system to ensure accurate motion recognition. During these processes, the recognition system compares the captured motion with the motion database in order to recognize the motion activity. However, the classifer can only recognize the motion activities that are closely ft with the database, and overftting has been an issue in this process. Hence, this paper is aimed at resolving overftting problem by using Artifcial Neural Network (ANN) for motion classifcation. The motion data was transformed into numerical data with an aid of Kinovea. Data mining software called WEKA was used to perform motion classifcation. Multi-Layer Perceptron (MLP), which is known as ANN, was modifed to recognize diferent motion activities in the classifcation process. It was observed that MLP is able to yield classifcation accuracy of 97.62%. Overftting issues were also solved by manipulating learning rates in the ANN classifer. A reduced learning rate from 0.3 to 0.1 improved the classifcation accuracy of jumping motion by up to 12.04%

html

Publication Date

30-6-2021

First Page

29

Last Page

37

Download

Share

COinS