Proper maintenance of sewer pipes is vital for the healthy functioning of a city. Due to the difficulty of reach for sewage pipes, automating pipe inspection has high potential in providing an efficient and objective identification of defects which could lead to damaging the pipe system. A popular approach has been to send remote controlled robots to photograph the pipes and process the images to identify possible defects. However majority of the images contain regular pipe features such as the flow line, pipe joints and pipe connections. Regular features pose a challenge for automated defect detection algorithms which require high processing time. This paper proposes a self organising map based approach to leverage the regularity of image features to isolate regions of interest which could contain defects. As a result, the search space is narrowed down for the defect detection algorithms, decreasing the overall processing time. Novelty of the work lies in the feature extraction and the gradual isolation of the potential defective image features to a manageable size. Therefore, this technique is suitable for large scale real applications. We demonstrate the effectiveness of the proposed approach for a real pipe image data set.

We propose an algorithm for creating superpixels. The major step in our algorithm is simply minimizing two pseudo-Boolean functions. The processing time of our algorithm on images of moderate size is only half a second. Experiments on a benchmark dataset show that our method produces superpixels of comparable quality with existing algorithms. Last but not least, the speed of our algorithm is independent of the number of superpixels, which is usually the bottle-neck for the traditional algorithms of superpixel creation.

This paper presents a method of mining the data obtained by a collection of pressure sensors monitoring a pipe network to obtain information about the location and size of leaks in the network. This inverse engineering problem is effected by support vector machines (SVMs) which act as pattern recognisers. In this study the SVMs are trained and tested on data obtained from the EPANET hydraulic modelling system. Performance assessment of the SVM showed that leak size and location are both predicted with a reasonable degree of accuracy. The information obtained from this SVM analysis would be invaluable to water authorities in overcoming the ongoing problem of leak detection.

A multi-agent architecture for scene understanding systems is described. Such systems consist of a collection of coupled vision subsystems, each one dedicated to a particular sensor. Interaction between the vision subsystems is controlled by an organising agent. Each vision subsystem is implemented on the basis of a partition tree image labelling representation and an associated multi-agent hierarchy of classifiers controlled by an organising agent. Each classifier in the multi-agent hierarchy of classifiers may be implemented as a decision tree of neural network classifiers acting on feature vectors derived from regions in the partition tree. Therefore, the image labelling and image classification subsystems can be described in terms of hierarchical networks of neural networks from which an interpretation in terms of multi-agents follows very naturally. The vision subsystems have a multi-resolution image representation and concurrency between and within the image labelling and image classification vision subsystem components.