Ahmad Kadri Junoh

K-means clustering and neural network for evaluating sound level vibration in vehicle cabin

One of the characteristics that may influence customers in vehicle purchasing is the level of comfort of the vehicle’s sound vibration in the vehicle cabin. The basic principle suggests that the sound vibration discomfort level is affected by a few factors which are mainly based on magnitudes, frequencies, directions and also the exposed periods. Normally, the phenomenon of sound vibration disrupts the performance of the driver by affecting the driver’s vision and also inducing a certain degree of stress due to the sound and vibration to which the driver and his or her passengers are exposed. The sound vibration is generally contributed by a few sources originated from the transmission of the vehicle’s engine, tire interactions with the road surface and also the exposure of vehicle’s body vibration during the movement. The objective of this study is to propose an approach that clusters the level of sound and vibration into a few categories and classifies them into those categories without implementing the subjective test that normally involves human assessment. The study has observed the changes of the sound quality and the level of vibration at particular points in the vehicle cabin over the changes of engine speeds. In reference to the results, the study has successfully provided a technical procedure in order to cluster, and also to classify, the level of sound vibration by taking into account the correlation between experienced noise and exposed vibration in the vehicle cabin.

Noise Annoyance Fuzzy Index in Passenger Car Cabin

Vehicle acoustical comfort and vibration in passenger car cabin are the factors which attract buyers to a vehicle, in order to have a comfortable driving environment. The comfort of the driving will affect the driver by influencing their driving performance. Vehicle acoustical comfort index (VACI) has been introduced to represent the level of annoyance in passenger car cabin. This index has been used in the computation related to the acoustics aspect in the car cabin. Noise annoyance is categorised into five states: most annoying, medium annoying, marginal, medium pleasant and most pleasant. In order to improve the VACI index, this study carries out an approach to classify and categorise the aforementioned five states using the fuzzy set theory approach. At the end of several studies, the noise annoyance fuzzy index (NAFI) has been introduced. By using NAFI, automotive researchers will be able to classify and make judgement about the state of annoyance from the exposed noise in the car cabin numerically, where the use of this index is regarded to be more precise and has practically been used in engineering fields especially where the fuzzy logic system is concerned.

Fast infant pain detection method

within this paper, pain detection is exposed and reviewed for detecting facial changes of patient in a hospital in Neonatal Intensive Care Unit (NICU). The system propesed three stage. The first stage implements Haar Cascade detection to detect the infant face. Secondly, PCA was employed for feature extraction. The third module extracts the PCA features of faces by measuring certain dimensions of pain and no pain regions with Support Vector Machine classifier. From 300 samples of face images, it is found that the identification rate of reaches 93.18%.

A novel hybrid fuzzy nonlinear weighted goal programming for optimising interior acoustics level in car cabin

The factors that are normally considered by customers for comfortable driving upon purchasing a car are acoustical comfort and exposed vibration. Keeping this in mind, the level of noise annoyance in reference to the vehicle acoustical comfort index is proposed. This index can be categorized by five states which are generally used in the computation that involve acoustical levels in the car cabin. This study is carried out in order to improve this index by classifying and categorizing these five states using the fuzzy set theory approach. Besides, identified sources of vibration are usually linked with engine transmission and tire-road interaction. This study includes the observation on the effects of the vibration due to tire interaction with the road, as well as the pattern of the trends toward the experienced noise against the engine speed [rpm] at both stationary and non-stationary positions. A combination of fuzzy set theory and nonlinear programming has been employed to develop a multi-objective model of hybrid fuzzy nonlinear weighted goal programming. This developed model adopts the results that have been provided in order to optimize the acoustics level by referring to the vibration level which is required at a certain value of the engine speed [rpm]. The proposed model has a significant impact towards a better environment, in the perspective of acoustics – a key factor in vehicle manufacturing process.

Application of Feed-Forward Neural Networks for Classifying Acoustics Levels in Vehicle Cabin

Vehicle acoustical comfort and vibration in a passenger car cabin are the main factors that attract a buyer in car purchase. Numerous studies have been carried out by automotive researchers to identify and classify the acoustics level in the vehicle cabin. The objective is to form a special benchmark for acoustics level that may be referred for any acoustics improvement purpose. This study is focused on the sound quality change over the engine speed [rp to recognize the noise pattern experienced in the vehicle cabin. Since it is difficult for a passenger to express, and to evaluate the noise experienced or heard in a numerical scale, a neural network optimization approach is used to classify the acoustics levels into groups of noise annoyance levels. A feed forward neural network technique is applied for classification algorithm, where it can be divided into two phases: Learning Phase and Classification Phase. The developed model is able to classify the acoustics level into numerical scales which are meaningful for evaluation purposes.

Computational method for predicting the effects of vibrations to acoustical comfort in vehicle cabin

One of the main features that attract the customer to purchase a vehicle is the vehicle acoustical and vibration comfort in the vehicle cabin. The exposed noise and vibration will affect the driver’s performance in a way that the noise and vibration will distract their vision, which altogether will be stressful to the driver and passenger. Based on previous studies, vibrations are contributed by two main sources, namely the engine transmission and interaction between the tyre and road surface, while the vehicle is moving. Through this study, an approach has been done to estimate numerically the amount of noise influenced by the vibration caused by the interaction between the tyre and road surface. The studies have focused on the observation of the trends of sound quality changes over the changes of engine speeds. At the end of this study, a technical method is provided to show the correlation between the acoustical comfort exposed with the vibration caused by the interaction between tyres and road surface.

Optimisation of acoustical comfort in vehicle cabin using goal programming

Car cabin acoustical comfort is one of the main features that may attract customers to purchase a new vehicle. The noise in a passenger car cabin is closely related to the vibration generated in the vehicle system. In this study, the effects of vibration on noise in the passenger car cabin have been studied. The vehicle acoustical comfort index (VACI) was used to evaluate the noise annoyance level and the vibration dose value (VDV) was used to evaluate the vibration level. Because engine noise is one of the dominant sources of interior noise, an approach has been used to study the correlation between engine speed and the level of generated vibration. According to the changes in the noise and the vibration level as a function of engine speed, a goal programming model was used to optimise the noise annoyance level in the passenger car cabin. At the end of the study, a multi-objective model was successfully built to optimise the noise annoyance levels by looking at the required vibration dose value at certain engine speeds.

Crime Detection with DCT and Artificial Intelligent Approach

Crime rate in Malaysia is almost in awareness stage. The centre for Public Policy Studies Malaysia reports that the ratio of police to population is 3.6 officers to 1,000 citizens in Malaysia. This lack of manpower sources ratios alone are not a comprehensive afford of crime fighting capabilities. Thus, dealing with these circumstances, we present a comprehensive study to determine bandit behavior with Discrete Cosine Transform (DCT), Support vector machine (SVM) and k Nearest Neighbor (k-NN) Classifier. This system provided a good justification as a monitoring supplementary tool for the Malaysian police arm forced.

Safety system based on Linear Discriminant Analysis

A Linear Discriminant Analysis Classifier for home security system we describe in this paper. Images were taken in uncontrolled indoor environment using video cameras of various qualities. Database contains 4,005 static images (in visible and infrared spectrum) of 267 subjects. Images from different quality cameras should mimic real-world conditions and enable robust face recognition algorithms testing, emphasizing different law enforcement and surveillance use case scenarios. In addition to database description, this paper also elaborates on possible uses of the database and proposes a testing protocol. A baseline Haar Cascade Method for face recognition algorithm was tested following the proposed protocol based on LDA Classifier. Other researchers can use these test results as a control algorithm performance score when testing their own algorithms on this dataset. Database is available to research community through the procedure described at http://www.lrv.fri.uni-lj.si/facedb.html.

Home security system based on Fuzzy k-NN Classifier

A Fuzzy k-nn Classifier for home security system we describe in this paper. Images were taken in uncontrolled indoor environment using video cameras of various qualities. Database contains 4,005 static images (in visible and infrared spectrum) of 267 subjects. Images from different quality cameras should mimic real-world conditions and enable robust face recognition algorithms testing, emphasizing different law enforcement and surveillance use case scenarios. In addition to database description, this paper also elaborates on possible uses of the database and proposes a testing protocol. A baseline Principal Component Analysis (PCA) face recognition algorithm was tested following the proposed protocol based on k-nn Classifier. Other researchers can use these test results as a control algorithm performance score when testing their own algorithms on this dataset. Database is available to research community through the procedure described at http://www.lrv.fri.uni-lj.si/facedb.html.