Weng Kin Lai

Human posture classification using hybrid Particle Swarm Optimization

Accurate identification and categorization of different types of human postures is a vital element of a real time video-based surveillance system. In this paper, we present an approach known as the hybrid Particle Swarm Optimization (PSO+K) to classify human postures into their respective clusters. The PSO algorithm is used to search for possible optimal solution from the solution space. Then the results of the PSO are used as initial cluster centroids of the K-Means for further refinement to find the final optimal solution. Experimental results from the algorithm are compared with the K-Means and the conventional PSO algorithm using our posture dataset and the result shows that PSO+K produces better accuracies compared to other algorithms.

Fuzzy qualitative complex actions recognition

Understanding actions is a complex issue in many aspects. However, most of the literature on action recognition deals with only simple actions. In this paper, we proposed the fuzzy qualitative robot kinematics framework to complex actions over time, e.g. walk then run and over the body, walk while wave hand etc. The human limbs is modelled as articulated rigid bodies and its motion is represented by a series of such models in terms of time. With this, we eventually converted the human motion analysis into a conventional robotic problem which has been well studied. Experimental results has shown that the action model built in this manifold offers few advantages. e.g. handles the tradeoffs in the off-the-shelf tracking algorithm and avoid using generative model where the size of the training data typically goes as the square of the number of states.

A Comparison Of Posture Recognition Using Supervised And Unsupervised Learning Algorithms.

Recognition of human posture is one step in the process of analyzing human behaviour. However, it is an ill-defined problem due to the high degree of freedom exhibited by the human body. In this paper, we study both supervised and unsupervised learning algorithms to recognise human posture in image sequences. In particular, we are interested in a specific set of postures which are representative of typical applications found in video analytics. The algorithms chosen for this paper are Kmeans, artificial neural network, self organizing maps and particle swarm optimization. Experimental results have shown that the supervised learning algorithms outperform the unsupervised learning algorithms in terms of the number of correctly classified postures. Our future work will focus on detecting abnormal behaviour based on these recognised static postures.

Investigations into Particle Swarm Optimization for Multi-class Shape Recognition

There has been a significant drop in the cost as well as an increase in the quality of imaging sensors due to stiff competition as well as production improvements. Consequently, real-time surveillance of private or public spaces which relies on such equipment is gaining wider acceptance. While the human brain is very good at image analysis, fatigue and boredom may contribute to a less-than-optimum level of monitoring performance. Clearly, it would be good if highly accurate vision systems could complement the role of humans in round-the-clock video surveillance. This paper addresses an image analysis problem for video surveillance based on the particle swarm computing paradigm. In this study three separate datasets were used. The overall finding of the paper suggests that clustering using Particle Swarm Optimization leads to better and more consistent results, in terms of both cluster characteristics and subsequent recognition, as compared to traditional techniques such as K-Means.

Improving Neural Network Generalization by Combining Parallel Circuits with Dropout

In an attempt to solve the lengthy training times of neural networks, we proposed Parallel Circuits (PCs), a biologically inspired architecture. Previous work has shown that this approach fails to maintain generalization performance in spite of achieving sharp speed gains. To address this issue, and motivated by the way Dropout prevents node co-adaption, in this paper, we suggest an improvement by extending Dropout to the PC architecture. The paper provides multiple insights into this combination, including a variety of fusion approaches. Experiments show promising results in which improved error rates are achieved in most cases, whilst maintaining the speed advantage of the PC approach.

DropCircuit : A Modular Regularizer for Parallel Circuit Networks

How to design and train increasingly large neural network models is a topic that has been actively researched for several years. However, while there exists a large number of studies on training deeper and/or wider models, there is relatively little systematic research particularly on the effective usage of wide modular neural networks. Addressing this gap, and in an attempt to solve the problem of lengthy training times, we proposed Parallel Circuits (PCs), a biologically inspired architecture based on the design of the retina. In previous work we showed that this approach fails to maintain generalization performance in spite of achieving sharp speed gains. To address this issue, and motivated by the way dropout prevents node co-adaptation, in this paper, we suggest an improvement by extending dropout to the parallel-circuit architecture. The paper provides empirical proof and multiple insights into this combination. Experiments show promising results in which improved error rates are achieved in most cases, whilst maintaining the speed advantage of the PC approach.

Hybrid particle swarm optimisation for data clustering

Finding a best clustering algorithm to tackle the problem of finding the optimal partition of a data set is always an NP-hard problem. In general, solutions to the NP-hard problems involve searches through vast spaces of possible solutions and evolutionary algorithms have been a success. In this paper, we explore one such approach which is hardly known outside the search heuristic field - the Particle Swarm Optimisation+k-means (PSOk) for this purpose. The proposed hybrid algorithm consists of two modules, the PSO module and the k-means module. For the initial stage, the PSO module is executed for a short period to search for the clusters centroid locations. Succeeding to the PSO module is the refining stage where the detected locations are transferred to the k-means module for refinement and generation of the final optimal clustering solution. Experimental results on two challenging datasets and a comparison with other hybrid PSO methods has demonstrated and validated the effectiveness of the proposed solution in terms of precision and computational complexity.

Analysis of Gait Features between Loaded and Normal Gait

Ever since the beginning of research on gait recognition, the main focus has been on investigating gait as a biometric. For security surveillance systems, the detection of suspicious behavior is considered to be more relevant than the recognition of identity, which applies more to security authentication systems. Hence, this research uses gait features as cues to detect suspicious behavior. In this work, the effect of load bearing on gait is investigated by analyzing the kinematics parameters between normal and loaded gaits to find out the ground truth between them. The focus is on two features computed throughout a video sequence: silhouette attributes attraction and limbs angular displacements attraction. The silhouette attributes use the area and center of mass of objects. The volunteers were carrying 5kg, 10kg, 15kg and 20kg weights in a bag pack attached either to the back or to the front of their bodies. The results show that silhouette area can be a useful descriptor for discriminating between loaded and normal gait. The second feature (limbs angular displacements attraction) also gives a positive result (92.2%) for both loads starting from 10kg attached at the back and front of subjects.