Norshuhada Samudin

Video stabilization and tampering detection for surveillance systems using homography

Video stabilization is one of the key enhancements of video technology to compensate the variations in camera orientation and translation. In this paper, a method is proposed for surveillance systems to stabilize the camera vibrations and also to detect camera tampering. This method uses SURF feature detection and matching using RANSAC algorithm to achieve homogenous coordinates and then estimates the motion between consecutive frames. The motion is compensated in the region of interest using the calculated transformation matrix. The displacement of the region of interest in image sequence is also calculated to detect the attempts for camera tampering. The result shows the developed system has a promising performance in comparison with current methods in removing camera fluctuations from the input video and detecting camera tampering with high accuracy.

An Automatic Robust Image Registration Algorithm for Aerial Mapping

Aerial mapping is attracting more attention due to the development in unmanned aerial vehicles (UAVs) and their availability and also vast applications that require a wide aerial photograph of a region in a specific time. The cross-modality as well as translation, rotation, scale change and illumination are the main challenges in aerial image registration. This paper concentrates on an algorithm for aerial image registration to overcome the aforementioned issues. The proposed method is able to sample automatically and align the sensed images to form the final map. The results are compared with satellite images that shows a reasonable performance with geometrically correct registration.

Ghost removal for background subtraction using color similarity comparison

This paper aims to solve the problem of detecting ghost object; which is a common problem in background subtraction algorithm. Ghost object is the false object detected which is not corresponding to any actual object in current image. In this work, we proposed ghost detection and removal method using color similarity comparison. Proposed solution is designed based on the assumption that ghost problem occurs due to the existence of the object in background image instead of in the current image. We are using color similarity between detected foreground area and its surrounding area to first determine whether the object appear in background or current image, consequently identify whether the detected object is a ghost or an actual object. Proposed solution has been tested using various datasets including PETS2001 and own datasets and it is proved that the proposed method is able to solve ghost problem.

Video analytics algorithm for detecting objects crossing lines in specific direction using blob-based analysis

This paper presents a video analytics algorithm for detecting event of objects crossing predetermined line-of-interest in the scene in specific direction. A fast blob-based analysis is formulated to detect the event, combined with the object detection and tracking to detect and tracked the object as motion blobs. Proposed algorithm is tested in real outdoor surveillance environment for 24 hours in 3 days to evaluate the detection accuracies in different scenarios. For comparison, the testing is done against a commercial surveillance system. The results show that the proposed algorithm provides better accuracy in all scenarios, while maintaining real-time processing capacity.

Real-time estimation of point-of-interest in video surveillance based on visual fixation

Analyzing and interpreting the behavior of humans in a sequence of images from a video stream is an active and growing topic in computer vision and its applications. This paper focuses on real-time visual fixation analysis in which the study of humans gaze allows investigation and reasoning of a point-of-interest (POI) in the scene. The systems architecture comprises of 3 major components; i) head detection, ii) head direction detection and iii) POI analysis. We employ the 3D color space (RGB) model as proposed in the installation of “15 Seconds of Fame” [8] to detect skin color pixels for head detection. The head direction specified is based on estimating the orientation in tilt angle and pan angle using neural network. Finally, the POI is determined via the correlation between multiple head directions. Experimental results are given to demonstrate the reliability of our system in estimating POI in a sequence of low-resolution images.