Research Output
A comparative study on interior acoustic comfort level of compact cars using data mining approach
Supervised segmentation on fusarium macroconidia spore in microscopic images via analytical approaches
1 - 2 of 2
Publication
Vehicle acoustic comfort is one of the ergonomic measurement criteria that are essential for car occupants. Furthermore, interior cabin noise of a car may affect the driver's concentration when driving. This study is to investigate the noise comfort level of car interior on several compact cars. The objective is to measure interior cabin noise for all three cars and then to compare their acoustic comfort level using subfield data mining approach. A deduction was made to rate the best car among the three in term of acoustic comfort. The interior cabin noise was obtained for the cases where engine speed is varied while the cars are in stationary and moving condition. The noise was assessed according to pre-determined subjective and objective criteria. The sound quality parameters was assessed by regression analysis. In subjective assessment, the recorded noise was evaluated based on jury assessment. Then, the data mining approach was implemented to illustrate the noise level. The collected noise data were divided into five clusters through hierarchical clustering method. To assess the accuracy of noise data clusters, the method of k-nearest neighbours was performed and the results show a high accuracy rate (> 95%). Finally, the interior noise of the three cars was compared by using the analysis of variation. The vehicle acoustic comfort index was produced for the three cars tested in this study. In addition, the acoustic quality among the three cars is presented using anova. Annoyance index of the three cars was generated using data mining method. From the results, Axia car model has the best acoustic comfort compared to the other two cars by objective evaluation. By subjective evaluation, Axia car model recorded the lowest level of annoyance.
Publication
Fungi are one of the major causes that contributed to plant diseases. There are lots of fungi species but it is estimated that only 10% have been described. There are two major approaches to identifying fungi species, morphological identification, and molecular test which need cautious clarification to make good interpretations and are time-consuming. In this paper, we propose a Machine Learning approach that involves the use of the K-Means clustering technique, and Decision Tree to highlight the observed fungi spore images taken under the microscopic view and discard background pixels to produce digital images database which later can be used for Deep Learning.