My Publication: Face Anti-spoofing Countermeasure: Efficient 2D Materials Classification Using Polarization Imaging

Spoofing is an act to impersonate a valid user of any biometric systems in order to gain access. In a face biometric system, an imposter might use some fake masks that mimic the real user face. Existing countermeasures against spoofing adopt face texture analysis, motion detection and surface reflection analysis. For the purpose of face anti-spoofing analysis, skin structure is a key factor in achieving the target of our study. Skin consists of multiple layers structure which produces multiple reflections: surface and subsurface reflections. In this paper, we proposed a measure to discriminate between a genuine face and a printed paper photo based on physical properties of the materials which contribute to its distinctive reflection values. In order to differentiate the reflections, polarized light (light that vibrates in a single direction) can be used. The Stokes parameters are applied to generate the Stokes images which are then used to produce the final image known as Stokes degree of linear polarization (SDOLP) image. The intensity of the SDOLP image is investigated statistically which has shown promising results in the materials classification, between the skin and the paper mask. Furthermore, comparison between the experimental results from two skin color groups, black and others show that the SDOLP data distribution of black skin is similar to the printed paper photo of the same skin group.

My Publication: On Evaluation of Depth Accuracy in Consumer Depth Sensors

This paper presents an experimental study of different depth sensors. The aim is to answer the question, whether these sensors give accurate data for general depth image analysis. The study examines the depth accuracy between three popularly used depth sensors; ASUS Xtion Prolive, Kinect Xbox 360 and Kinect for Windows v2. The main attention is to study on the stability of pixels in the depth image captured at several different sensor-object distances by measuring the depth returned by the sensors within specified time intervals. The experimental results show that the fluctuation (mm) of the random selected pixels within the target area, increases with increasing distance to the sensor, especially on the Kinect for Xbox 360 and the Asus Xtion Prolive. Both of these sensors provide pixels fluctuation between 20mm and 30mm at a sensor-object distance beyond 1500mm. However, the pixel’s stability of the Kinect for Windows v2 not affected much with the distance between the sensor and the object. The maximum fluctuation for all the selected pixels of Kinect for Windows v2 is approximately 5mm at sensor-object distance of between 800mm and 3000mm. Therefore, in the optimal distance, the best stability achieved.

My Publication: Determination of Mango Fruit from Binary Image Using Randomized Hough Transform

A method of detecting mango fruit from RGB input image is proposed in this research. From the input image, the image is processed to obtain the binary image using the texture analysis and morphological operations (dilation and erosion). Later, the Randomized Hough Transform (RHT) method is used to find the best ellipse fits to each binary region. By using the texture analysis, the system can detect the mango fruit that is partially overlapped with each other and mango fruit that is partially occluded by the leaves. The combination of texture analysis and morphological operator can isolate the partially overlapped fruit and fruit that are partially occluded by leaves. The parameters derived from RHT method was used to calculate the center of the ellipse. The center of the ellipse acts as the gripping point for the fruit picking robot. As the results, the rate of detection was up to 95% for fruit that is partially overlapped and partially covered by leaves.

(e-mail me for full-text version)