N.A. Zainuddin

Autonomous Navigation of Mobile Robot Using Kinect Sensor

The problem of achieving real time process in depth camera application, in particular when used for indoor mobile robot localization and navigation is far from being solved. Thus, this paper presents autonomous navigation of the mobile robot by using Kinect sensor. By using Microsoft Kinect XBOX 360 as the main sensor, the robot is expected to navigate and avoid obstacles safely. By using depth data, 3D point clouds, filtering and clustering process, the Kinect sensor is expected to be able to differentiate the obstacles and the path in order to navigate safely. Therefore, this research requirement to propose a creation of low-cost autonomous mobile robot that can be navigated safely.

Detection of Different Classes Moving Object in Public Surveillance Using Artificial Neural Network (ANN)

Public surveillance monitoring is rapidly finding its way into Intelligent Surveillance Systems. Street crimes such as snatch theft is increasing drastically in recent years, cause a serious threat to human life worldwide. In this paper, a moving object detection and classification model was developed using novel Artificial Neural Network (ANN) simulation with the aim to identify its suitability for different classes of moving objects, particularly in public surveillance conditions. The result demonstrated that the proposed method consistently performs well with different classes of moving objects such as, motorcyclist, and pedestrian. Thus, it is reliable to detect different classes of moving object in public surveillance camera. It is also computationally fast and applicable for detecting moving objects in real-time.

Analysis of Artificial Neural Network and Viola-Jones Algorithm Based Moving Object Detection

In recent years, the worrying rate of street crime has demanded more reliable and efficient public surveillance system. Analysis of moving object detection methods is presented in this paper, includes Artificial Neural Network (ANN) and Viola-Jones algorithm. Both methods are compared based on their precision of correctly classify the moving objects. The emphasis is on two major issues involve in the analysis of moving object detection, and object classification to two groups, pedestrian and motorcycle. Experiments are conducted to quantitatively evaluate the performance of the algorithms by using two types of dataset, which are different in term of complexity of the background. The utilization of cascade architecture to the extracted features, benefits the algorithm. The algorithms have been tested on simulated events, and the more suitable algorithm with high detection rate is expected to be presented in this paper.

Adaptive Background Reconstruction for Street Surveillance

In recent years, adaptive background reconstruction works have found interest in many researchers. However, the existing algorithms that have been proposed by other researchers still in the early stage of development and many aspects need to be improved. In this paper, an adaptive background reconstruction is presented. Past pixel observation is used. The proposed algorithm also has eliminated the need of the pre-training of non-moving objects in the background. The proposed algorithm is capable of reconstructing the background with moving objects in video sequence. Experimental results show that the proposed algorithms are able to reconstruct the background correctly and handle illumination and adverse weather that modifies the background.

Analysis on Background Subtraction for Street Surveillance

Background subtraction is a well-known technique used in computer vision applications. However, in public surveillance system, the utilization of background subtraction still new and far from being solved. Insufficient analysis of the background subtraction algorithms made the situation getting worse. The analysis of the commonly-used algorithms is presented in this paper. Experiments are conducted to quantitatively evaluate the performance of the algorithms by using three video sequences. The more suitable algorithm for various conditions is expected to be presented as the results in this paper.