Suzaimah Ramli

Plastic bottle shape classification using partial erosion-based approach

In this paper, a work on representing plastic bottle shape using erosion based approach for an automated classification is reported. Morphological operations are used to describe the structure or form of an image. By using the two-dimensional description of plastic bottle silhouettes, edge detection of the object silhouette is performed followed by the erosion process. This work will compare two versions of erosion which are regular erosion, the Matlab function imerode and the improved version of erosion which is called partial erosion. Normalization procedure in which the sum pixel value after erosion is divided by the sum pixel of the whole silhouette is done. The normalized values are grouped into histograms of 9 bins of sum pixel value(9HbSPV), find its maximum number to form as a feature set and is then used as inputs to train a neural network for plastic bottle shape classification. Results obtained showed that the proposed feature extraction method can be applied to discriminate plastic bottles according to shape, either slim or broad bottles, efficiently.

Robust Identification of Polyethylene Terephthalate (PET) Plastics through Bayesian Decision

Recycling is one of the most efficient methods for environmental friendly waste management. Among municipal wastes, plastics are the most common material that can be easily recycled and polyethylene terephthalate (PET) is one of its major types. PET material is used in consumer goods packaging such as drinking bottles, toiletry containers, food packaging and many more. Usually, a recycling process is tailored to a specific material for optimal purification and decontamination to obtain high grade recyclable material. The quantity and quality of the sorting process are limited by the capacity of human workers that suffer from fatigue and boredom. Several automated sorting systems have been proposed in the literature that include using chemical, proximity and vision sensors. The main advantages of vision based sensors are its environmentally friendly approach, non-intrusive detection and capability of high throughput. However, the existing methods rely heavily on deterministic approaches that make them less accurate as the variations in PET plastic waste appearance are too high. We proposed a probabilistic approach of modeling the PET material by analyzing the reflection region and its surrounding. Three parameters are modeled by Gaussian and exponential distributions: color, size and distance of the reflection region. The final classification is made through a supervised training method of likelihood ratio test. The main novelty of the proposed method is the probabilistic approach in integrating various PET material signatures that are contaminated by stains under constant lighting changes. The system is evaluated by using four performance metrics: precision, recall, accuracy and error. Our system performed the best in all evaluation metrics compared to the benchmark methods. The system can be further improved by fusing all neighborhood information in decision making and by implementing the system in a graphics processing unit for faster processing speed.

Implementing Low Level Features for Human Aggressive Movement Detection

In this real world, being able to identify the signs of imminent abnormal behaviors such as aggression or violence and also fights, is of extreme importance in keeping safe those in harm’s way. This research propose an approach to figure out human aggressive movements using Horn-Schunck optical flow algorithm in order to find the flow vector for all video frames. The video frames are collected using digital camera. This research guides and discovers the patterns of body distracted movement so that suspect of aggression can be investigated without body contact. Using the vector of this method, the abnormal and normal video frames are then classified and utilized to define the aggressiveness of humans. Preliminary experiment result showed that the low level of feature extraction can classify human aggressive and non-aggressive movements.

Fear as Emotion Assessment in Information Security Using Kansei Engineering Methodology

Information security assessment needs to be further developed due to the increased need to improve information security. Human factors need to be considered in formulizing and establishing information security methodology. A preliminary study of fear as emotional assessment in information security using Kansei Engineering methodology was discussed in this paper. The concept of fear in information security was explored to better understand the emotion of fear, for the purpose of implementating it as user assessment. Based on the literature reviews, it shows that fear appeal does play a crucial role in changing an individual’s behaviour and attitude in the context of risk measures. However, too much induced fear could cause affect how much users comply to security policies. Therefore, it is vital to have the optimum level of fear appeal applied in information security. This study surveyed the relationship between fear level and users action in applying security measures and serve as preliminary study to use fear as emotion assessment in information security using Kansei Engineering.

Knowledge Driven Interface to Determine Degree of Exposure of Young Adult to Pedophile Online

In the era of Internet of Thing (IoT) which a lot of devices are connected to the internet, children are spending more hours online interacting in cyber space that increase exposure to cyber security including pedophile activity. Increase of time spend online could increase the potential of online sexual grooming behaviours of child molesters. Since that the behaviour are not easily identified prior to the abuse, this study gathers and collect information about child sexual abuse by pedophile and propose a comprehensive decision support system to educate children base on knowledge-driven method about online grooming by molesters. An interactive system is built to provide knowledge to children regarding child sexual abuse and pedophile in terms of definition and each characteristics of it. The main purpose of the system is compiling database about child sexual abuse and pedophiles in order to determine the level of child’s exposure to pedophile in term of five attributes which is selection of victims, gaining access, grooming, trust and approach.

A review on cyberbullying in Malaysia from digital forensic perspective

This paper is a review on cyberbullying generally on the issues around it and specifically in Malaysia. The topic that will be covered in this review begins with the definition of cyberbullying, what types of cyberbullying that occur nowadays, the roles of person involve and statistic of who being targeted, and what type of common data that being used in cyberbullying. This paper will give perspective towards digital forensic investigation such as what type of data and medium are being used by the cyberbullies.

Comparison between Horn-Schunck and Brox techniques for Motion Detection in Thermal Video

The Motion Detection From The Thermal Video Using Optical Flow Technique was created to study motion detection techniques from thermal video using optical flow. The purpose of this application is to identify which techniques is more effective in detecting motion in thermal videos and to some extent it can drive technological development in the context of image processing. This application is developed to detect motion of an object in the distance occurred in the video. The program of the development Graphical User Interface (GUI) for this application is to facilitate users in order to use this application and to understand how these applications work. GUI is an intermediate medium between the user and a system. The technique developed for this application is called the Horn-Schunck and Brox technique. The analysed scene involved a video recording of a human dynamic motion. The analysis was done through an algorithm to detect motion based on the differences of the flow direction of the magnitude that occurs between image frames. The result of the analysis showed which one of the techniques is more effective in detecting the motion of objects that occurs in the video.

Lecturer Promotion Assessment System based on Fuzzy Logic

Lecturer Promotion Assessment System for Department of Computer Science, Faculty of Science and Technology Defence was built to help the promotion of lecturers featured appraiser to conduct an initial assessment of the eligibility status of lecturers. The lecturer will be treated as eligible, in observation or tidal worth in value that has been set to proceed with the application promotion. It is because, in the application for promotion of lecturers it includes several procedures such as application delivery and assessment meetings between committee members. Thus the system helps the evaluator by using the Fuzzy Logic method consists of several phases which are fuzzification, fuzzy inference system and defuzzification. The system uses the engine type Mamdani fuzzy inference system and defuzzification through technique of centroid of Gravity (COG) as the output. Each of the estimated assessment will be compared with a database that has been set before produced as output. There are several ways to analyze the results of these feasibility of using graphs and message box. The construction of this system is expected to help ease the burden of the assessors and settle the problems that arise. In fact, this system may also be a drive to the use of more high-tech systems and artificial intelligence features along with the rapidly evolving technology now.

Facial expression change detection algorithm using optical flow technique

This study is aimed to investigate the use of optical flow techniques in detecting changes in facial expressions. A person face expression detection is very important as we can identify a persons emotional and mental state. When communicating with people, human can identify someones expression accurately compared to computer. Although many approaches are tackled on this topic, there are still several drawbacks and limitations. In order to yield better results, we have applied optical flow technique to detect facial expression. In addition, we have the technique with Horn-Schunck method to optimize the results. Based on the experiment conducted, the average value represented by every facial expression can be identified, and the values are significant for future research that is focused on facial expression classification.

Temperature determining method from motion detection using thermal images

This paper present a basic determining method to illustrate the optical flow estimation using motion information to detect the elevated temperature of febrile individual in thermal image sequences. Object detection on this paper focus on motion based rather than feature based. Target classification using physical based approach and optical flow algorithm to identifies the moving object. Proposed method in this paper will merge visual and thermal images to create “heat picture” of a object. Using thermal camera, a pointer automatically shows the hottest area in the picture, which is detect of object/human body temperature from motion recognition.