Fadhlan Hafizhelmi Kamaru Zaman
Universiti Teknologi MARA

Published : 5 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 5 Documents
Search

Gender classification using custom convolutional neural networks architecture Fadhlan Hafizhelmi Kamaru Zaman
International Journal of Electrical and Computer Engineering (IJECE) Vol 10, No 6: December 2020
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (941.246 KB) | DOI: 10.11591/ijece.v10i6.pp5758-5771

Abstract

Gender classification demonstrates high accuracy in many previous works. However, it does not generalize very well in unconstrained settings and environments. Furthermore, many proposed Convolutional Neural Network (CNN) based solutions vary significantly in their characteristics and architectures, which calls for optimal CNN architecture for this specific task. In this work, a hand-crafted, custom CNN architecture is proposed to distinguish between male and female facial images. This custom CNN requires smaller input image resolutions and significantly fewer trainable parameters than some popular state-of-the-arts such as GoogleNet and AlexNet. It also employs batch normalization layers which results in better computation efficiency. Based on experiments using publicly available datasets such as LFW, CelebA and IMDB-WIKI datasets, the proposed custom CNN delivered the fastest inference time in all tests, where it needs only 0.92ms to classify 1200 images on GPU, 1.79ms on CPU, and 2.51ms on VPU. The custom CNN also delivers performance on-par with state-of-the-arts and even surpassed these methods in CelebA gender classification where it delivered the best result at 96% accuracy. Moreover, in a more challenging cross-dataset inference, custom CNN trained using CelebA dataset gives the best gender classification accuracy for tests on IMDB and WIKI datasets at 97% and 96% accuracy respectively.
Improving steering convergence in autonomous vehicle steering control Amir Ashraf Mohamad; Fadhlan Hafizhelmi Kamaru Zaman; Fazlina Ahmat Ruslan
Indonesian Journal of Electrical Engineering and Computer Science Vol 13, No 1: January 2019
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v13.i1.pp279-285

Abstract

Steering control is a critical design element in autonomous vehicle development since it will determine whether the vehicle can navigate safely or not. For the prototype of UiTM Autonomous Vehicle 0 (UiTM AV0), Vexta motor is used to control the steering whereas Pulse Width Modulation (PWM) signal is responsible to drive the motor. However, by using PWM signal it is difficult to converge to the desired steering angle and furthermore time taken for steering angle to converge is much longer. Thus, Proportional Integral Derivative (PID) has been introduced in this autonomous vehicle steering controller to improve the convergence of the steering. Meanwhile a microcontroller was used to control the Vexta Motor direction and perform the calculation of the desired steering angle. Simulation results showed PID controller showed better time taken and preicison of successful convergence of the desired steering angle compared to the PWM controller.  Analysis results showed that PID controller significantly reduce the overshooting of steering angle and significantly improve the time taken for convergence by up to 37 seconds faster than PWM controller in UiTM AV0.
Prediction of energy consumption using recurrent neural networks (RNN) and nonlinear autoregressive neural network with external input (NARX) Wan Muhammad Zafri Wan Yahaya; Fadhlan Hafizhelmi Kamaru Zaman; Mohd Fuad Abdul Latip
Indonesian Journal of Electrical Engineering and Computer Science Vol 17, No 3: March 2020
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v17.i3.pp1215-1223

Abstract

Recurrent Neural Networks (RNN) and Nonlinear Autoregressive Neural Network with External Input (NARX) are recently applied in predicting energy consumption. Energy consumption prediction for depth analysis of how electrical energy consumption is managed on Tower 2 Engineering Building is critical in order to reduce the energy usage and the operational cost. Prediction of energy consumption in this building will bring great benefits to the Faculty of Electrical Engineering UiTM Shah Alam. In this work, we present the comparative study on the performance of prediction of energy consumption in Tower 2 Engineering Building using RNN and NARX method. The model of RNN and NARX are trained using data collected using smart meters installed inside the building. The results after training and testing using RNN and NARX show that by using the recorded data we can accurately predict the energy consumption in the building. We also show that RNN model trained with normalized data performs better than NARX model.
Learning face similarities for face verification using hybrid convolutional neural networks Fadhlan Hafizhelmi Kamaru Zaman; Juliana Johari; Ahmad Ihsan Mohd Yassin
Indonesian Journal of Electrical Engineering and Computer Science Vol 16, No 3: December 2019
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v16.i3.pp1333-1342

Abstract

Face verification focuses on the task of determining whether two face images belong to the same identity or not. For unrestricted faces in the wild, this is a very challenging task. Besides significant degradation due to images that have large variations in pose, illumination, expression, aging, and occlusions, it also suffers from large-scale ever-expanding data needed to perform one-to-many recognition task. In this paper, we propose a face verification method by learning face similarities using a Convolutional Neural Networks (ConvNet). Instead of extracting features from each face image separately, our ConvNet model jointly extracts relational visual features from two face images in comparison. We train four hybrid ConvNet models to learn how to distinguish similarities between the face pair of four different face portions and join them at top-layer classifier level. We use binary-class classifier at top-layer level to identify the similarity of face pairs which includes a conventional Multi-Layer Perceptron (MLP), Support Vector Machines (SVM), Native Bayes, and another ConvNet. There are 3 face pairing configurations discussed in this paper. Results from experiments using Labeled face in the Wild (LFW) and CelebA datasets indicate that our hybrid ConvNet increases the face verification accuracy by as much as 27% when compared to individual ConvNet approach. We also found that Lateral face pair configuration yields the best LFW test accuracy on a very strict test protocol without any face alignment using MLP as top-layer classifier at 87.89%, which on-par with the state-of-the-arts. We showed that our approach is more flexible in terms of inferencing the learned models on out-of-sample data by testing LFW and CelebA on either model.
Improving night driving behavior recognition with ResNet50 Muhammad Firdaus Ishak; Fadhlan Hafizhelmi Kamaru Zaman; Ng Kok Mun; Syahrul Afzal Che Abdullah; Ahmad Khushairy Makhtar
Indonesian Journal of Electrical Engineering and Computer Science Vol 33, No 3: March 2024
Publisher : Institute of Advanced Engineering and Science

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.11591/ijeecs.v33.i3.pp1974-1988

Abstract

The issue of driving behavior at night poses significant challenges due to reduced visibility and increased risk of accidents. Recent works have leveraged deep learning techniques to enhance night-time driving safety. However, the limited availability of high-quality training data and the lack of robustness in existing models present significant problems. In this work, we propose a novel approach to improve driving behavior recognition at night using ResNet50 with contrast limited adapted histogram equalization (CLAHE). We collected a new dataset and developed a more effective and robust model that can accurately recognize driving behaviors under low-illumination conditions, thereby reducing the likelihood of collisions and improving overall road safety. The experimental results demonstrate significant improvements in the deep learning model’s performance compared to conventional methods. Notably, the ResNet50 model delivers the best performance with accuracy rates of 90.73% using NIGHT-VIS-CLAHE data, demonstrating a 16% improvement in accuracy. For benchmark purposes, the InceptionV3, GoogleNet, and MobileNetV2 models also show enhanced accuracy through CLAHE implementation. Furthermore, NIGHT-VIS-CLAHE implementation in ResNet50 achieved 90.29% accuracy, surpassing the best NIGHT-IR InceptionV3 at 89.27%, highlighting the advantage of ResNet50 with CLAHE in low-light conditions even against infra-red sensor.