Siti Fatimah Abdul Razak

Interactive android-based indoor parking lot vehicle locator using QR-code

QR code has been applied in many ways from marketing products, locating promotional items on shelves, finding stores and etc. In this study, we report on an android based application development aimed to provide navigation services to locate parked vehicles in an indoor parking space of shopping malls. We utilise the motion sensor, bar code scanner function and camera function built in smartphones. This application is able to show the route from user current location to his parked vehicle based on an indoor map of the parking area stored in a database. In addition, it is also able to automatically detect users current movement based on steps calculation. A field test was conducted in a shopping mall indoor parking space to evaluate the performance of the application. In general, the application has shown promising results.

Real Time Driver Anger Detection

The field of artificial intelligence has seen an increasing number of researches being done related to facial expression recognition. Different methods have been proposed with some of them yielding good results and some performing poorly. Apart from that, anger plays a pivotal role in road accidents since road rage is stated to be one of the contributing factors to road accidents. In order to cater road rage and considering it being harmful to drivers and passengers, this paper proposes a real time driver anger detection. The project classifies human facial expressions, mainly anger expression in real time from a live video in order to warn the driver and eventually road accidents can be reduced.

A Study of Wireless Communication Technologies for Vehicular Communication

In recent years, vehicle communication is an advanced technology that has attain attention in both industries and academician all over the world. The initiation on vehicular communication is to improve road safety, efficiency and comfort. This paper studies the availability of the wireless communication technologies for vehicular communication and the possible implementation of the suitable wireless communication for vehicle communication in the context of Malaysia.

Flash it read it: A Chinese character flashcard recognizer

This paper proposes a Chinese character flashcard recognizer for kids learning purposes. Firstly, the system will capture an image of the flashcard shown to the webcam. Secondly, a preprocessing procedure based on thresholding is performed to remove noisy edges and convert the color image into binary image. Subsequently, a Principal Component Analysis based feature extraction is conducted to encode the image into a compact representation. Lastly, character recognition is performed using Euclidean distance. The kids are able to learn Chinese characters with their Pinyin and meaning using the system. From the experiments, promising recognition results are achieved.

Finger spelling recognition using neural network

Finger spelling is a way of communication by expressing words using hand signs in order to ensure deaf and dumb community can communicate with others effectively. Therefore, a system that can understand finger spelling is needed. As a result of that, this work is conducted to primarily develop a tutoring system for finger spelling. To develop a robust real-time finger spelling tutoring system, it is necessary to ensure the accuracy of the finger spelling recognition. Even though there are existing solutions available for a decade, but most of them are just focusing on improving accuracy rate without implementing their solutions as a complete tutoring system for finger spelling. Consequently, it inspires this research project to develop a tutoring system for finger spelling. Microsoft Kinect sensor is used to acquire color images and depth images of the finger spells. Depth images are used to perform segmentation on the color images. After that, the segmented images are used as input and pass into a two hidden layers backpropagation neural network for classification.

Developing a Rule Base for Recommending Insurance Products

This paper presents a fuzzy rule base for insurance prospects profiling based on two main sources, i.e. from dataset and experts. A historical dataset of Islamic insurance policyholders was used as a case study for this purpose. The Wang and Mendel algorithm (WMA) (Wang and Mendel, IEEE Trans Syst Man Cybern 22:1414–1472, [1]) was applied on the dataset to generate fuzzy rules which includes five types of policies with five pre-determined most influential variables, namely age, gender, marital status, monthly salary and job class. Fuzzy rules from the dataset and expert were compared based on similarity and accuracy measures. The rule base is proposed to be used by insurance advisors to make decisions and provide appropriate proposals according to the prospect’s profile.