Papers by MOHD RAZIF SHAMSUDDIN
Sentiment Analysis on Online Learning for Higher Education During Covid-19
2022 3rd International Conference on Artificial Intelligence and Data Sciences (AiDAS)
PCA Method in Facial Features Extraction for Attendance Tracking Prototype
2022 3rd International Conference on Artificial Intelligence and Data Sciences (AiDAS)
Flower Recognition using Deep Convolutional Neural Networks
IOP Conference Series: Earth and Environmental Science
This study investigates the suitable model for flower recognition based on deep Convolutional Neu... more This study investigates the suitable model for flower recognition based on deep Convolutional Neural Networks (CNN) with transfer learning approach. The dataset used in the study is a benchmark dataset from Kaggle. The performance of CNN for plant identification using images of flower are investigated using two popular image classification models: AlexNet and VGG16. Results show that CNN is proven to produce outstanding results for object recognition, but its achievement can still be influenced by the type of images and the number of layers of the CNN architecture. The models produced adequate performance rates, with the VGG16 model achieving the best results. AlexNet and VGG16 models achieved the accuracy of 85.69% and 95.02% respectively. This model can be replicated for flower recognition in other areas, especially in our national heritage, Taman Negara which is among the richest flora ecosystem in the world. The significant feature extraction processes were discussed as well, an...
Herbal Plant Analysis Based on Leaf Features using K-Means Clustering
IOP Conference Series: Earth and Environmental Science
Plants are essential in the Earth, as it supplies the oxygen needed by human beings and animals, ... more Plants are essential in the Earth, as it supplies the oxygen needed by human beings and animals, and becomes the source of foods and medical treatments. Many medicinal plants can treat diseases and it is also called herbal plants. Traditionally, these plants are processed and transformed as traditional medicines to cure any diseases. Nowadays, there are still practices that use medicinal plants. However, it is quite challenging to find herbal plants and these herbal plants come with different features such as size, shape, and colour. Therefore, this paper presents a machine learning approach, namely clustering, to classify the herbal plant species through images. We focused on six herbal plants in Malaysia which are Peacock Fern, Misai Adam, Mempisang, Tapak Sulaiman, Pandan Serapat and Kacip Fatimah. These species were collected from Taman Negara Pahang, Kuala Keniam, Malaysia. The k-means algorithm was employed by experimenting with several numbers of clusters in the range of two,...
Eye Detection for Drowsy Driver Using Artificial Neural Network
Communications in Computer and Information Science, 2017
Driving is one of the common activities in people’s everyday life and therefore improving driving... more Driving is one of the common activities in people’s everyday life and therefore improving driving skill to reduce car crashes is an important issue. Even though a lot of studies and work has been done on road and vehicle designs to improve driver’s safety yet the total number of car crashes is increasing day by day. Therefore, the most factors that cause an accident is fatigue driver rather than other factors which are distraction, speeding, drinking driver, drugs and depression. To prevent car crashes that occur due to drowsy driver, it is essential to have an assistive system that monitors the vigilance level of driver and alert the driver in case of drowsy detection. This system presents a drowsy detection system based on eye detection of the driver. Vision-based approach is adopted to detect drowsy eye because other developed approaches are either intrusive (physical approach) that makes the driver uncomfortable or less sensitive (vehicle based approach). The data collected from 26 volunteers will have four (4) different type of image. Thus, the total input will be 10,800 nodes. This thesis will be classified into two (2) outputs which are drowsy eye and non-drowsy eye. The algorithm that will be used is Back-propagation Neural Network (BPNN) and will be applied in MATLAB software. The experimental result shows that this system could achieve 98.1% accuracy.
Recent research in Deep Neural Networks (DNN) shows promising potential of applications in machin... more Recent research in Deep Neural Networks (DNN) shows promising potential of applications in machine learning. Generally, the goal of DNN is to model complex, hierarchical features in data. Classifying complex features of an input in Neural Network remains as a challenge despite the rapid development in Deep Learning research. DNN is not classified by its algorithm, but the depth of the networks. Usually any Neural Architecture that has a depth of more than three layers are considered deep. Although many report suggest DNN performs better than the shallow networks, it is prone to suffer from the vanishing gradient problem and overfitting. This paper proposes a method to study the problems in DNN and suggests a method to enhance the capability of the network. The study aims to create a new DNN architecture that can perform well with complex input patterns. Key-Words: Deep Learning, Deep Neural Network, Neural Network Architecture, Classification, Machine Learning
Everyone would require to have a signature for authorization and other important tasks that needs... more Everyone would require to have a signature for authorization and other important tasks that needs identification. Thus, signature has become one of a method to represent its writer uniquely. Signature has many hidden features that are difficult to extract. Some of the identified features that a signature should have are slanting, baseline, proportion and size. In this proposed study, slanting is chosen to be identified in a signature. Signatures are captured using a tablet and saved in a digitized format of x and y values. A slant algorithm is created and coded into a functional system. An experiment consisting of 50 signatures are tested and the finding shows the angle and degree of the slant in every signature. The creation of this algorithm would be able to give some degree of contribution in the area of signature recognition. Key-Words: Slant, Slant Recognition, Online Signature, Curved Stroke, Curved Slant
IOP Conference Series: Materials Science and Engineering, 2020
This paper explores the capabilities of a graph optimization-based Simultaneous Localization and ... more This paper explores the capabilities of a graph optimization-based Simultaneous Localization and Mapping (SLAM) algorithm known as Cartographer in a simulated environment. SLAM refers to the problem in which an agent attempts to determine its location in the immediate environment as well as constructing the map(s) of its environment. SLAM is one of the most important aspects in the implementation of autonomous vehicle. In this paper, we explore the capabilites of the Cartographer algorithm which is based on the newer graph optimization approach in improving SLAM problems. A series of experiments were tested in order to discover its Cartographer capabilities in tackling SLAM problems. Then, we compare the results of Cartographer with Hector SLAM, another graph-based SLAM algorithm. We present the results from the experiments which show some promising findings based on the amount of computer resources used and the quality of the map(s) produced.
IOP Conference Series: Materials Science and Engineering, 2020
This research is a part of a major research on automation of malware identification using Deep De... more This research is a part of a major research on automation of malware identification using Deep Denoising Autoencoders. Malicious software, or in short called malware refers to any software designed to cause damage to a single computer, server, or computer network. This malware term includes all kind of malicious software such as computer virus and spyware. All these malicious malware behaviour is monitored, logged and recorded using a cuckoo sandbox with the help of an x86 hosted supervisor software. The intent of recording the malware behaviour is to understand the pattern of behaviour of each known malware family. This collected data will be further trained to a Deep Denoising Autoencoders to automate the identification process of new malware within the identified malware families. However, the raw behaviour data is not suitable for an optimum training process. This paper will discuss the process of transforming the text based behavioural dataset to a more suitable dataset for dee...
Exploratory Analysis of MNIST Handwritten Digit for Machine Learning Modelling
Communications in Computer and Information Science, 2018
This paper is an investigation about the MNIST dataset, which is a subset of the NIST data pool. ... more This paper is an investigation about the MNIST dataset, which is a subset of the NIST data pool. The MNIST dataset contains handwritten digit images that is derived from a larger collection of NIST data which contains handwritten digits. All the images are formatted in 28 × 28 pixels value with grayscale format. MNIST is a handwritten digit images that has often been cited in many leading research and thus has become a benchmark for image recognition and machine learning studies. There have been many attempts by researchers in trying to identify the appropriate models and pre-processing methods to classify the MNIST dataset. However, very little attention has been given to compare binary and normalized pre-processed datasets and its effects on the performance of a model. Pre-processing results are then presented as input datasets for machine learning modelling. The trained models are validated with 4200 random test samples over four different models. Results have shown that the normalized image performed the best with Convolution Neural Network model at 99.4% accuracy.
Shallow Network Performance in an Increasing Image Dimension
Communications in Computer and Information Science, 2016
This paper describes the performance of a shallow network towards increasing complexity of dimens... more This paper describes the performance of a shallow network towards increasing complexity of dimension in an image input representation. This paper will highlight the generalization problem in Shallow Neural Network despite its extensive usage. In this experiment, a backpropagation algorithm is chosen to test the network as it is widely used in many classification problems. A set of three different size of binary images are used in this experiment. The idea is to assess how the network performs as the scale of the input dimension increases. In addition, a benchmark MNIST handwritten digit sampling is also used to test the performance of the shallow network. The result of the experiment shows the network performance as the scale of input increases. The result is then discussed and explained. From the conducted experiments it is believed that the complexity of the input size and breadth of the network affects the performance of the Neural Network. Such results can be as a reference and guidance to people that is interested in doing research using backpropagation algorithm.
Personality identification based on signature features recognition using neuro-fuzzy / Mohd Razif Shamsuddin
Personality analysis is time consuming and requires complex tasks. Personality experts are burden... more Personality analysis is time consuming and requires complex tasks. Personality experts are burdened with piles of questionnaire tools that sometimes required high attention to be evaluated. Moreover, some of the available personality tools cannot be reused because of copyright reasons. Thus, this has cause personality identification tool to become very expensive when the sample of respondent population is high. This research is conducted to model an integration of graphology and Cattell's 16 personality traits theory.
Controlling activated sludge system using genetic algorithms / Mohd Razif Shamsuddin
Heuristic knowledge is an instinct naturally embedded in animals for navigational purposes. Ant C... more Heuristic knowledge is an instinct naturally embedded in animals for navigational purposes. Ant Colony Optimization (ACO) has captured this mechanism into a well defined technique to solve TSP and routing problems. Now, ACO gives a general framework to solve other problems in similar nature. This paper tends to explain how knowledge discovery is done in ACO and tested the modified ACO in solving robot's path planning problem. The results have been encouraging for ACO to produce a good navigational path for a robot to follow but the performance has been inferior to fuzzy approach used in the same problem domain. The results are discussed in detail and the comparative findings are justified clearly focusing on the unique criteria of the chosen ACO technique.
Graphology and Cattell's 16PF Traits Matrix (HoloCatT Matrix)
2008 Third International Conference on Convergence and Hybrid Information Technology, 2008
... Mohd Razif Shamsuddin1, Ku Shairah Jazahanim1, Zaidah Ibrahim1, Rahmattullah Khan Abdul Wahab... more ... Mohd Razif Shamsuddin1, Ku Shairah Jazahanim1, Zaidah Ibrahim1, Rahmattullah Khan Abdul Wahab Khan2, Azlinah Mohamed1 1Faculty of Information Technology & Quantitative Sciences Universiti Teknologi MARA MALAYSIA 2 Applied Psychology Center International ...
Slant Classification Using FuzzySIS
2008 Third International Conference on Convergence and Hybrid Information Technology, 2008
Everyone would require to have a signature for authorization and other important tasks that needs... more Everyone would require to have a signature for authorization and other important tasks that needs identification. Thus, signature has become one of a method to represent its writer uniquely. Signature has many hidden features that are difficult to extract. Some of the identified features that a signature should have are slanting, baseline, proportion and size. In this proposed study, slanting is chosen to be identified in a signature. Signatures are captured using a tablet and saved in a digitized format of x and y values. A slant algorithm is created and coded into a functional system. An experiment consisting of 50 signatures are tested and the finding shows the angle and degree of the slant in every signature. The creation of this algorithm would be able to give some degree of contribution in the area of signature recognition.
According to the American National Science and Technology Council (NSTC), the first signature rec... more According to the American National Science and Technology Council (NSTC), the first signature recognition system was developed in 1965. Then the research continued in 1970 focusing on the potential of geometric characteristic of a signature rather than dynamic characteristic. Nowadays, signature is a commonly used identification procedure. Everyone would be required having a signature for authorization and other important tasks that needs identification. Thus, signature has become one of a method to represent its writer uniquely. Signature has many hidden features that are difficult to extract. Some of the identified features that a signature should have are slanting, baseline, proportion and size. This paper covers the area of signature slant identification. Signatures are captured using a tablet and saved in a digitized format of x and y values. Then it is filtered and calculated for its angle and degree. In the end the signature will be classified to its slant category. A slant algorithm is created and coded into a functional system. An experiment consisting of 50 signatures are tested and the finding shows the angle and degree of the slant in every signature. The result is then tested for its accuracy with an available 10 sample of created proofed signatures. The result shows a favorable accuracy of 80% correct slant identification. The creation of this algorithm would be able to give some degree of contribution in the area of signature recognition.
Baseline Image Classification Approach Using Local Minima Selection
Lecture Notes in Computer Science, 2009
... Minima Selection Mohd. Razif Shamsuddin and Azlinah Mohamed ... 737 [5] Mohd Razif, S., Azlin... more ... Minima Selection Mohd. Razif Shamsuddin and Azlinah Mohamed ... 737 [5] Mohd Razif, S., Azlinah, M.: Online Slant Identification Algorithm for Curved Stroke. In: 7th International Conference Advances on Software Engineering, Parallel and Distrib-uted System, pp. 1318. ...
Uploads
Papers by MOHD RAZIF SHAMSUDDIN