Yit Kwong Chin

Image segmentation via normalised cuts and clustering algorithm

Image segmentation has been widely applied in image analysis for various areas such as biomedical imaging, intelligent transportation systems and satellite imaging. The main goal of image segmentation is to simplify an image into segments that have a strong correlation with objects in the real world. Homogeneous regions of an image are regions containing common characteristics and are grouped as single segment. One of the graph partitioning methods in image segmentation, normalised cuts, has been recognised producing reliable segmentation result. To date, normalised cuts in image segmentation of various sized images is still lacking of analysis of its performance. In this paper, segmentation on synthetic images and natural images are covered to study the performance and effect of different image complexity towards segmentation process. This study gives some research findings for effective image segmentation using graph partitioning method with computation cost reduced. Because of its cost expensive and it becomes unfavourable in performing image segmentation on high resolution image especially in online image retrieval systems. Thus, a graph-based image segmentation method done in multistage approach is introduced here.

Multiple intersections traffic signal timing optimization with genetic algorithm

Traffic congestion in the urban area occurs more frequent than the past due to rapidly increasing on road vehicle usage rates. It could seriously hinder the development of urban area if a well management system has not being established. These scenarios necessitate the development of advance traffic management systems to increase the performance of signalized intersection. Traffic signal timing management (TSTM) system which comprise of genetic algorithm based optimization is proposed. Using a proper TSTM system, network traffic flow can be improved with considerably less cost than other infrastructural improvements. The proposed genetic algorithm based optimization approach allows signal timing parameters such as offset, cycle time, green split and phase sequence to be optimized with objective of minimum delay and better traffic fluency. The proposed GATSTM system has the ability to handle and manage the dynamic changes of the traffic networks condition by calibrating the system parameters accordingly.

Genetic algorithm based PID optimization in batch process control

The primary aim in batch process is to enhance the process operation in order to achieve high quality and purity product while minimising the production of undesired by-product. However, due to the difficulties to perform online measurement, batch process supervision is based on the direct measurable quantities, such as temperature. During the process, a large amount of exothermic heat is released when the reactants are mixed together. The exothermic behaviour causes the reaction to become unstable and consequently the quality and purity of the final product will be affected. Therefore, it is important to have a control scheme which is able to balance the needs of process safety with the product quality and purity. Since the chemical industries are still applying PI and PID to control the batch process, researchers are keen to optimize PID parameters using artificial intelligence (AI) techniques. However, most of these PID optimization techniques need online process model to predetermine the optimizer parameters. However in practice, the dynamic model of the batch process is poorly known. As a result, majority of the studies focused on acceptable performance instead of optimum performance of the batch process control. This paper proposes a new genetic algorithm (GA) optimizer which consists of additional information of the online estimated model parameters in addition to the PID parameters as the string of the GA. The simulation results show that the proposed GA auto-tuning method is a better candidate than the regular GA where the estimated model parameters in fitness function is capable to control the process temperature while avoiding model mismatch and disturbance condition.

Fuzzy Logic Based MPPT for PV Array under Partially Shaded Conditions

This paper presents the fuzzy logic based maximum power point tracking for the optimization of the solar photovoltaic (PV) array under partially shaded conditions. The PV system is modelled in MATLAB/SIMULINK where the PV array is formed by five PV modules connected in series. The PV characteristic of PV module and PV array under uniform solar irradiance are nonlinear but there are one maximum power point (MPP) can be identified. Nevertheless, the PV characteristic becomes more complex with multiple MPP when the PV array under partially shaded conditions (PSC). In this paper, maximum power point tracking (MPPT) approach based on perturb and observe algorithm has been investigated. Fuzzy logic is adopted into the conventional MPPT to enhance the overall performance of the PV system. The performances of MPPT and FMPPT are investigated particularly on the transient response and the steady state response when the PV array is exposed under different partially shaded conditions. The simulation results show that FMPPT has better performance where it can facilitate the PV array to reach the MPP faster and provide more stable output power.

Exploring Q-Learning Optimization in Traffic Signal Timing Plan Management

Traffic congestions often occur within the entire traffic network of the urban areas due to the increasing of traffic demands by the outnumbered vehicles on road. The problem may be solved by a good traffic signal timing plan, but unfortunately most of the timing plans available currently are not fully optimized based on the on spot traffic conditions. The incapability of the traffic intersections to learn from their past experiences has cost them the lack of ability to adapt into the dynamic changes of the traffic flow. The proposed Q-learning approach can manage the traffic signal timing plan more effectively via optimization of the traffic flows. Q-learning gains rewards from its past experiences including its future actions to learn from its experience and determine the best possible actions. The proposed learning algorithm shows a good valuable performance that able to improve the traffic signal timing plan for the dynamic traffic flows within a traffic network.

Enhancement of Particle Filter Resampling in Vehicle Tracking Via Genetic Algorithm

Vehicle tracking is an essential approach that can help to improve the traffic surveillance or assist the road traffic control. Recently, the development of video surveillance infrastructure has incited the researchers to focus on the vehicle tracking by using video sensors. However, the amount of the on-road vehicle has been increased dramatically and hence the congestion of the traffic has made the occlusion scene become a challenge task for video sensor based tracking. Conventional particle filter will encounter tracking error during and after occlusion. Besides that, it also required more iteration to continuously track the vehicle after occlusion. Thus, particle filter with genetic operator resampling has been proposed as the tracking algorithm to faster converge and keep track on the target vehicle under various occlusion incidents. The experimental results show that enhancement of the particle filter with genetic algorithm manage to reduce the particle sample size.

Performance Analysis of Intelligent Transport Systems (ITS) with Adaptive Transmission Scheme

Various means of modern transports have already taken the initiative to incorporate computing and communication technology. This cross field area is coined as intelligent transportation systems (ITS). Transport systems of the future require fast and precise monitoring to ensure safety of such systems is guaranteed. Thus, the communication aspect demands seamless and minimal error whilst delivering vital data. However, the dynamic surroundings always post a challenge to wireless communications, taking account into a myriad of interference such as multipath fading, shadowing, dispersing and path loss. Furthermore, there is also case of mobility. In this paper, the performance of wireless communications with adaptive modulation and coding in vehicular scenarios were compared against rigid transmission techniques. Simulations were conducted over different mobility channel models against Signal-to-Noise ratio (SNR) to provide thorough analysis. Performance measure of Bit Error Rate (BER) and Packet Error Rate (PER) were used to provide understanding of the overall picture.

Agent-Based Traffic Flow Optimization at Multiple Signalized Intersections

Relieving urban traffic congestion has always been an urgent call in a dynamic traffic network. The objective of this research is to control the traffic flow within a traffic network consists of multiple signalized intersections with traffic ramp. The massive traffic network problem is dealt through Q-learning actuated traffic signalization (QLTS), where the traffic phases will be monitored so that immediate actions can be taken when congestion is happening to minimize the number of vehicles in queue. QLTS has better performance than the existing common fixed-time traffic signalization (FTS) in dealing with the ramp flow due to its flexibility in changing the traffic signal with accordance to the traffic conditions and necessity.

Overlapping vehicle tracking via adaptive particle filter with multiple cues

Vehicle tracking is a vital approach to assist the on-road traffic surveillance system. Since the on-road vehicles is increasing, occlusion and overlapping of vehicles is often happen in the traffic surveillance scene. Therefore, segmentation and tracking of the occlusion or overlapped vehicle can be a challenging task in surveillance system via image processing. In this paper, a multiple cues overlapping vehicle tracking algorithm is proposed to continuously track the occluded vehicle effectively. The earlier vehicle tracking systems are normally based on colour feature which will leads to inaccurate results when the background colour is complex or too similar with the target vehicle. On the other hand, shape feature will increase the accuracy but consume more computation time in the resampling process during overlapping. The experimental results show that enhancement of the particle filter resampling process with multiple cues is capable to track the overlapped vehicle with higher accuracy and without compromising the processing time.

CUSUM-Variance Ratio based Markov Chain Monte Carlo algorithm in overlapped vehicle tracking

Markov Chain Monte Carlo (MCMC) is one of the algorithms that have been widely implemented in tracking vehicle for traffic surveillance purposes. The sampling efficiency of the algorithm is essential to determine the vehicle position accurately. However, the sample size of the algorithm is still remaining an issue as non-optimal sample size will defect the tracking accuracy, especially when the moving vehicle is overlapped. Adaptive sample size of MCMC has been implemented using CUSUM Path Plot and Variance Ratio algorithms to perform vehicle tracking. CUSUM Path Plot determines the samples convergence rate by calculating the hairiness of the sample size whereas Variance Ratio method computes two sets of MCMC to determine the samples steady state. This paper proposes the fusion of CUSUM-Variance ratio algorithm to enhance the tracking efficiency. Experimental results shows that the CUSUM-Variance Ratio method have a better performance in tracking the overlapping vehicle with higher accuracy and more optimal sample size compared to the standalone CUSUM Path Plot and Variance Ratio approaches.