Damian Chu - Academia.edu (original) (raw)

Damian Chu

Uploads

Papers by Damian Chu

Research paper thumbnail of I C Prolog II a Language for Implementing Multi Agent Systems

This paper examines how we may prototype Multi Agent Systems We informally enu merate the low lev... more This paper examines how we may prototype Multi Agent Systems We informally enu merate the low level technical support needed for such systems and show how IC Prolog II is a good candidate language IC Prolog II is a new implementation of Prolog that is particu larly suited to distributed applications It features multiple threads high level communication primitives and an object oriented extension A fully worked example of specifying an agent is given to illustrate use of the language This shows that it is possible to give a high level description of an agent and that this description can be executed directly making fast proto typing of agents a reality With this new tool researchers in Multi Agent Systems may gain practical experience in exploring ideas on a real implementation Familiarity with the Prolog programming language is assumed

Research paper thumbnail of Automatic detection of targets using fractal dimension

Proceedings. 2005 IEEE International Geoscience and Remote Sensing Symposium, 2005. IGARSS '05.

Automatic detection of targets in a radar image is an important problem with many applications su... more Automatic detection of targets in a radar image is an important problem with many applications such as assessing the battlefield situation over large areas to targeting individual targets on land, sea, or air. To detect a target in a clutter, the radar echoes from a target have to compete with the echoes from the clutter. A clutter signal can be modeled by a nonlinear deterministic dynamical system. As a result, clutter modeling becomes a problem of chaotic system reconstruction from time series measurements. On the other hand, for the case of an application without time series of radar images, the clutter can be treated as a spatial chaotic system because of the chaotic scattering and nonlinear wave phenomena. A dynamic learning neural network (DLNN) has shown to be able to reconstruct the clutter signal. In this paper, fractal dimension is used to quantify the chaotic behavior of the clutter on the geometric aspects. The proposed approach is applied to a SAR image from the MSTAR public release data set for target detection as an example. The target detection results of using the conventional CFAR detection technique, the DLNN detection approach, and the fractal detection method are compared. The experimental results suggest that the use of fractal dimension results in better detection evaluation measures, including detection rate, false detection rate, and loss detection rate than those of other detectors.

Research paper thumbnail of I.C. Prolog II: a Multi-threaded Prolog System

Implementations of Logic Programming Systems, 1994

This paper introduces IC-Prolog II { a new implementation of Prolog that is particularly suited t... more This paper introduces IC-Prolog II { a new implementation of Prolog that is particularly suited to distributed applications. Unlike o t h e r w ork on distributed logic programming, we do not aim to improve t h e r a w performance of a logic program nor do we require multi-processor machines or specialised hardware. Instead, our aim is to widen the applicability of logic programming to encompass new classes of practical applications which require the coordination of concurrently executing programs on separate workstations to communicate over the network. IC-Prolog II features multiple threads, a Parlog subsystem and high-level communication primitives. Multiple threads enables the concurrent execution of independent goals. The Parlog subsystem allows local ne-grained parallelism to be speci ed. In IC-Prolog II, Prolog can call Parlog and vice-versa. The combination of the two logic languages o ers greater expressive p o wer than simply the sum of the two since di erent components of the same application may use either Parlog or Prolog or both. The high-level communication primitives provide the means for independent IC-Prolog II processes on di erent m a c hines on a network to communciate. The result is a language well-suited for writing network-friendly applications.

Research paper thumbnail of Data Fusion Of Remote Sensing Images For Terrain Classification With A Variance Reduction Technique

ABSTRACT First Page of the Article

Research paper thumbnail of Correction to 'On the numerical computation in systems and control

IEEE Transactions on Automatic Control, 2003

Research paper thumbnail of I C Prolog II a Language for Implementing Multi Agent Systems

This paper examines how we may prototype Multi Agent Systems We informally enu merate the low lev... more This paper examines how we may prototype Multi Agent Systems We informally enu merate the low level technical support needed for such systems and show how IC Prolog II is a good candidate language IC Prolog II is a new implementation of Prolog that is particu larly suited to distributed applications It features multiple threads high level communication primitives and an object oriented extension A fully worked example of specifying an agent is given to illustrate use of the language This shows that it is possible to give a high level description of an agent and that this description can be executed directly making fast proto typing of agents a reality With this new tool researchers in Multi Agent Systems may gain practical experience in exploring ideas on a real implementation Familiarity with the Prolog programming language is assumed

Research paper thumbnail of Automatic detection of targets using fractal dimension

Proceedings. 2005 IEEE International Geoscience and Remote Sensing Symposium, 2005. IGARSS '05.

Automatic detection of targets in a radar image is an important problem with many applications su... more Automatic detection of targets in a radar image is an important problem with many applications such as assessing the battlefield situation over large areas to targeting individual targets on land, sea, or air. To detect a target in a clutter, the radar echoes from a target have to compete with the echoes from the clutter. A clutter signal can be modeled by a nonlinear deterministic dynamical system. As a result, clutter modeling becomes a problem of chaotic system reconstruction from time series measurements. On the other hand, for the case of an application without time series of radar images, the clutter can be treated as a spatial chaotic system because of the chaotic scattering and nonlinear wave phenomena. A dynamic learning neural network (DLNN) has shown to be able to reconstruct the clutter signal. In this paper, fractal dimension is used to quantify the chaotic behavior of the clutter on the geometric aspects. The proposed approach is applied to a SAR image from the MSTAR public release data set for target detection as an example. The target detection results of using the conventional CFAR detection technique, the DLNN detection approach, and the fractal detection method are compared. The experimental results suggest that the use of fractal dimension results in better detection evaluation measures, including detection rate, false detection rate, and loss detection rate than those of other detectors.

Research paper thumbnail of I.C. Prolog II: a Multi-threaded Prolog System

Implementations of Logic Programming Systems, 1994

This paper introduces IC-Prolog II { a new implementation of Prolog that is particularly suited t... more This paper introduces IC-Prolog II { a new implementation of Prolog that is particularly suited to distributed applications. Unlike o t h e r w ork on distributed logic programming, we do not aim to improve t h e r a w performance of a logic program nor do we require multi-processor machines or specialised hardware. Instead, our aim is to widen the applicability of logic programming to encompass new classes of practical applications which require the coordination of concurrently executing programs on separate workstations to communicate over the network. IC-Prolog II features multiple threads, a Parlog subsystem and high-level communication primitives. Multiple threads enables the concurrent execution of independent goals. The Parlog subsystem allows local ne-grained parallelism to be speci ed. In IC-Prolog II, Prolog can call Parlog and vice-versa. The combination of the two logic languages o ers greater expressive p o wer than simply the sum of the two since di erent components of the same application may use either Parlog or Prolog or both. The high-level communication primitives provide the means for independent IC-Prolog II processes on di erent m a c hines on a network to communciate. The result is a language well-suited for writing network-friendly applications.

Research paper thumbnail of Data Fusion Of Remote Sensing Images For Terrain Classification With A Variance Reduction Technique

ABSTRACT First Page of the Article

Research paper thumbnail of Correction to 'On the numerical computation in systems and control

IEEE Transactions on Automatic Control, 2003

Log In