Marsyita Hanafi

Face Liveness Detection Using Dynamic Local Ternary Pattern (DLTP)

Face spoofing is considered to be one of the prominent threats to face recognition systems. However, in order to improve the security measures of such biometric systems against deliberate spoof attacks, liveness detection has received significant recent attention from researchers. For this purpose, analysis of facial skin texture properties becomes more popular because of its limited resource requirement and lower processing cost. The traditional method of skin analysis for liveness detection was to use Local Binary Pattern (LBP) and its variants. LBP descriptors are effective, but they may exhibit certain limitations in near uniform patterns. Thus, in this paper, we demonstrate the effectiveness of Local Ternary Pattern (LTP) as an alternative to LBP. In addition, we adopted Dynamic Local Ternary Pattern (DLTP), which eliminates the manual threshold setting in LTP by using Weber’s law. The proposed method was tested rigorously on four facial spoof databases: three are public domain databases and the other is the Universiti Putra Malaysia (UPM) face spoof database, which was compiled through this study. The results obtained from the proposed DLTP texture descriptor attained optimum accuracy and clearly outperformed the reported LBP and LTP texture descriptors.

The Design and Compilation of a Facial Spoof Database on Various Textures

Face biometrics plays an important role in various authentication applications. However, one of the main design challenges for an accurate face biometrics is detecting spoof artificial artifacts. The development of robust anti-spoofing algorithm requires for rigorous system training and testing on facial database that includes various possible spoof specimen which reflects different variations of spoofing attacks. Currently, the databases which are used to test face anti-spoofing algorithms are limited in terms of texture variations. Therefore, in this paper, our focus is to introduce a face spoofing database which is compiled from various types of spoofing textures. Our database was collected from 30 different subjects and fake faces were recaptured on various forms which include four different types of paper-based textures and three different digital display devices. All images were collected using a high precision camera device. This database would provide a more realistic and challenging platform for facial anti-spoofing research.

Wavelet-based medical image fusion via a non-linear operator

Medical image fusion has been extensively used to aid medical diagnosis by combining images of various modalities such as Computed Tomography (CT) and Magnetic Resonance Image (MRI) into a single output image that contains salient features from both inputs. This paper proposes a novel fusion algorithm through the use of a non-linear fusion operator, based on the low sub-band coefficients of the Discrete Wavelet Transform (DWT). Rather than employing the conventional mean rule for approximation sub-bands, a modified approach is taken by the introduction of a non-linear fusion rule that exploits the multimodal nature of the image inputs by prioritizing the stronger coefficients. Performance evaluation of CT-MRI image fusion datasets based on a range of wavelet filter banks shows that the algorithm boasts improved scores of up to 92% as compared to established methods. Overall, the non-linear fusion rule holds strong potential to help improve image fusion applications in medicine and indeed other fields.

Automated road marking detection system for autonomous car

In recent years, road markings detection has received great attention and has been widely explored due to the aim of producing a system that is able to detect various shape of road markings on the images that are captured under various imaging conditions. Generally, the road images are captured using a camera, which has been placed inside the vehicle at a fixed position. However, the quality of the resulting images decreases if the camera position has been changed accidentally, due to the movement of the car. Hence, in this paper, a road markings detection system that tackle the problems of detecting road markings on the images captured under various camera positions and illumination conditions is proposed. The system consists of a graph cut segmentation method, which is used to detect the road, an inverse perspective transform method, which is used to convert the image into a birds-eye view image, an image normalization method, which is CLAHE and a connected component analysis that is used to remove the background. We demonstrate the usefulness of the constructed algorithm by performing experiments on a database that consists of 400 road images.

Eye and mouth localization for various imaging conditions images

This paper presents an algorithm to detect the eyes and mouth of the faces under various imaging conditions such as with different poses, dimensions, illuminations, resolutions, wearing glasses or not and having different expressions is proposed. Initially, the algorithm converted the face region detected by the Viola-Jones algorithm into gray-scale and reduced the resolution by decimation to achieve low resolution region. The pixels that are darker than their surroundings are later located by 8 individual filters. Finally, the center of eyes and mouth are searched using a triangle model. Experiment show that, the algorithm produced 98.7% accuracy in detecting the eyes and mouth.

Palm Vein Pattern Visual Interpretation Using Laplacian and Frangi-Based Filter

Zarina Mohd Noh*, Abdul Rahman Ramli, Marsyita Hanafi, M Iqbal Saripan, Ridza Azri Ramlee 1,2,3,4 Department of Computer and Communication Systems Engineering, Faculty of Engineering, Universiti Putra Malaysia, 43400 UPM Serdang, Selangor, Malaysia 1,5 Fakulti Kejuruteraan Elektronik dan Kejuruteraan Komputer, Universiti Teknikal Malaysia Melaka (UTeM), Hang Tuah Jaya, 75450 Durian Tunggal, Melaka, Malaysia

A Real Time Road Marking Detection System on Large Variability Road Images Database

For no less than two decades, the development of autonomous systems has led to the development of embedded applications permitting to enhance the driving comfort and limit the hazard level of dangerous zones. One of the first embedded system is a lane detection system, which was implemented using road marking detection algorithms with the aim to produce a system that is able to detect various shapes of road markings on the images that are captured under various imaging conditions. Generally, the road images were captured using a camera, which has been placed inside a vehicle at a fixed position. In this paper, a road markings detection system that tackles the problems of detecting road markings on the images captured under various weather and illumination conditions is proposed. The proposed system consists of inverse perspective transform method, which is used to convert an image into a bird’s-eye view image, an image normalization method, namely CLAHE that tackle various illumination conditions and Sobel edge detection method for identifying the road marker. We demonstrate the usefulness of the constructed algorithm by performing experiments on our Large Variability Road Images database (LVRI) that consists of 22,500 road images with the accuracy of 96.53%.

Overview and challenges of palm vein biometric system

Palm vein biometric system is one of the biometric technologies that has grabbed the attention of scholarly researchers and industrial alike, due to its distinctive properties and hidden nature. Constant effort had been done in improving the palm vein biometric system performance through the design of its vein acquisition system and vein image analysis. This paper provides an overview of the underlying elements of a palm vein biometric system that summarises the works done, and predicts the upcoming research focus in this area.

Robotic cans surface inspection system based on shape features

Computer vision systems are one of the most widely used techniques in Automation and have been extensively used for industry automation. Industrial automation deals mainly with the automation of production, quality control and materials management processes. One trend is the increasing use of Machine vision to offer automatic inspection and robot guidance functions, while the other is a continued increase in the use of robots. The aim of this paper is to provide a robotic cans surface inspection system based on the shape. The proposed system is simple and user friendly yet accurate, uses Hu moment as a feature of detected shape in the image and compared to the range of acceptable Hu moment gained from training. It is composed of a camera attached to a PC with TCP/IP, image acquisition, analysis, and inspection implemented by Open CV Library for image processing. The method described in this paper checks on the statistical-based approaches for feature extraction such as moment feature as part of the final inspection system. Robotic arm is programed as a client server method to receive action and position from the PC, which carries out the image processing as well.

Assessment of near infrared LED radiation pattern using Otsu thresholding

This paper describes the use of Otsu thresholding method in assessing the radiation pattern emitted by near infrared (NIR) LED. The NIR LED configured in this paper is intended to be used as illumination source for the development of a NIR palm vein image acquisition device. The experiment is conducted using a single board computer (SBC) to promote a real-time embedded system development that can be readily integrated as a vein viewing device. Based on the Otsu thresholded image obtained, it is observed that the NIR LED radiation pattern can be accessed subjectively through the thresholding process. The resulted thresholded image can be used as preliminary assessment of the radiation pattern in developing a NIR image acquisition system that fully utilizes the NIR LED properties.