The Red Light Camera Evaluation - Interim Report - July 1985

Red light cameras were introduced in August 1983 to deter run-the-red offences and therefore to reduce the incidence of right-angle accidents at signalised intersections in Melbourne. This report was prepared after two years of operation of the program. It provides a detailed account of the technical aspects of the program, but does not provide any detailed, evaluative analyses of accident data.

Investigating limitations of SURF approach for thermal imaging analysis and mapping

Wildlife conservation involves an understanding of a specific animal, its environment and the interaction within a local ecosystem. Unmanned Aerial Vehicles (UAVs) present cost effective, non-intrusive solution for detecting animals over large areas and the use thermal imaging cameras offer the ability detect animals that would otherwise be concealed to visible light cameras. This report examines some of limitations on using SURF for the development of large maps using multiple stills images extracted from the thermal imaging video camera which contain wildlife (eg. Koala in them).

Tracking people in 3D using position, size and shape

This paper presents a prototype tracking system for tracking people in enclosed indoor environments where there is a high rate of occlusions. The system uses a stereo camera for acquisition, and is capable of disambiguating occlusions using a combination of depth map analysis, a two step ellipse fitting people detection process, the use of motion models and Kalman filters and a novel fit metric, based on computationally simple object statistics. Testing shows that our fit metric outperforms commonly used position based metrics and histogram based metrics, resulting in more accurate tracking of people.

Mapping of indoor environments by robots using low-cost vision sensors

For robots to operate in human environments they must be able to make their own maps because it is unrealistic to expect a user to enter a map into the robot’s memory; existing floorplans are often incorrect; and human environments tend to change. Traditionally robots have used sonar, infra-red or laser range finders to perform the mapping task. Digital cameras have become very cheap in recent years and they have opened up new possibilities as a sensor for robot perception. Any robot that must interact with humans can reasonably be expected to have a camera for tasks such as face recognition, so it makes sense to also use the camera for navigation. Cameras have advantages over other sensors such as colour information (not available with any other sensor), better immunity to noise (compared to sonar), and not being restricted to operating in a plane (like laser range finders). However, there are disadvantages too, with the principal one being the effect of perspective. This research investigated ways to use a single colour camera as a range sensor to guide an autonomous robot and allow it to build a map of its environment, a process referred to as Simultaneous Localization and Mapping (SLAM). An experimental system was built using a robot controlled via a wireless network connection. Using the on-board camera as the only sensor, the robot successfully explored and mapped indoor office environments. The quality of the resulting maps is comparable to those that have been reported in the literature for sonar or infra-red sensors. Although the maps are not as accurate as ones created with a laser range finder, the solution using a camera is significantly cheaper and is more appropriate for toys and early domestic robots.

Adaptive unsupervised learning of human actions

Automatic detection of suspicious activities in CCTV camera feeds is crucial to the success of video surveillance systems. Such a capability can help transform the dumb CCTV cameras into smart surveillance tools for fighting crime and terror. Learning and classification of basic human actions is a precursor to detecting suspicious activities. Most of the current approaches rely on a non-realistic assumption that a complete dataset of normal human actions is available. This paper presents a different approach to deal with the problem of understanding human actions in video when no prior information is available. This is achieved by working with an incomplete dataset of basic actions which are continuously updated. Initially, all video segments are represented by Bags-Of-Words (BOW) method using only Term Frequency-Inverse Document Frequency (TF-IDF) features. Then, a data-stream clustering algorithm is applied for updating the system's knowledge from the incoming video feeds. Finally, all the actions are classified into different sets. Experiments and comparisons are conducted on the well known Weizmann and KTH datasets to show the efficacy of the proposed approach.

Crowd Counting Using Multiple Local Features

In public venues, crowd size is a key indicator of crowd safety and stability. Crowding levels can be detected using holistic image features, however this requires a large amount of training data to capture the wide variations in crowd distribution. If a crowd counting algorithm is to be deployed across a large number of cameras, such a large and burdensome training requirement is far from ideal. In this paper we propose an approach that uses local features to count the number of people in each foreground blob segment, so that the total crowd estimate is the sum of the group sizes. This results in an approach that is scalable to crowd volumes not seen in the training data, and can be trained on a very small data set. As a local approach is used, the proposed algorithm can easily be used to estimate crowd density throughout different regions of the scene and be used in a multi-camera environment. A unique localised approach to ground truth annotation reduces the required training data is also presented, as a localised approach to crowd counting has different training requirements to a holistic one. Testing on a large pedestrian database compares the proposed technique to existing holistic techniques and demonstrates improved accuracy, and superior performance when test conditions are unseen in the training set, or a minimal training set is used.

Improved detection and tracking of objects in surveillance video

Surveillance networks are typically monitored by a few people, viewing several monitors displaying the camera feeds. It is then very difficult for a human operator to effectively detect events as they happen. Recently, computer vision research has begun to address ways to automatically process some of this data, to assist human operators. Object tracking, event recognition, crowd analysis and human identification at a distance are being pursued as a means to aid human operators and improve the security of areas such as transport hubs. The task of object tracking is key to the effective use of more advanced technologies. To recognize an event people and objects must be tracked. Tracking also enhances the performance of tasks such as crowd analysis or human identification. Before an object can be tracked, it must be detected. Motion segmentation techniques, widely employed in tracking systems, produce a binary image in which objects can be located. However, these techniques are prone to errors caused by shadows and lighting changes. Detection routines often fail, either due to erroneous motion caused by noise and lighting effects, or due to the detection routines being unable to split occluded regions into their component objects. Particle filters can be used as a self contained tracking system, and make it unnecessary for the task of detection to be carried out separately except for an initial (often manual) detection to initialise the filter. Particle filters use one or more extracted features to evaluate the likelihood of an object existing at a given point each frame. Such systems however do not easily allow for multiple objects to be tracked robustly, and do not explicitly maintain the identity of tracked objects. This dissertation investigates improvements to the performance of object tracking algorithms through improved motion segmentation and the use of a particle filter. A novel hybrid motion segmentation / optical flow algorithm, capable of simultaneously extracting multiple layers of foreground and optical flow in surveillance video frames is proposed. The algorithm is shown to perform well in the presence of adverse lighting conditions, and the optical flow is capable of extracting a moving object. The proposed algorithm is integrated within a tracking system and evaluated using the ETISEO (Evaluation du Traitement et de lInterpretation de Sequences vidEO - Evaluation for video understanding) database, and significant improvement in detection and tracking performance is demonstrated when compared to a baseline system. A Scalable Condensation Filter (SCF), a particle filter designed to work within an existing tracking system, is also developed. The creation and deletion of modes and maintenance of identity is handled by the underlying tracking system; and the tracking system is able to benefit from the improved performance in uncertain conditions arising from occlusion and noise provided by a particle filter. The system is evaluated using the ETISEO database. The dissertation then investigates fusion schemes for multi-spectral tracking systems. Four fusion schemes for combining a thermal and visual colour modality are evaluated using the OTCBVS (Object Tracking and Classification in and Beyond the Visible Spectrum) database. It is shown that a middle fusion scheme yields the best results and demonstrates a significant improvement in performance when compared to a system using either mode individually. Findings from the thesis contribute to improve the performance of semi-automated video processing and therefore improve security in areas under surveillance.

Dense Correspondence Extraction in Difficult Uncalibrated Scenarios

The relationship between multiple cameras viewing the same scene may be discovered automatically by finding corresponding points in the two views and then solving for the camera geometry. In camera networks with sparsely placed cameras, low resolution cameras or in scenes with few distinguishable features it may be difficult to find a sufficient number of reliable correspondences from which to compute geometry. This paper presents a method for extracting a larger number of correspondences from an initial set of putative correspondences without any knowledge of the scene or camera geometry. The method may be used to increase the number of correspondences and make geometry computations possible in cases where existing methods have produced insufficient correspondences.

Diurnal variation of corneal shape and thickness

Purpose: To investigate associations between the diurnal variation in a range of corneal parameters, including anterior and posterior corneal topography, and regional corneal thickness. ----- Methods: Fifteen subjects had their corneas measured using a rotating Scheimpflug camera (Pentacam) every 3-7 hours over a 24-hour period. Anterior and posterior corneal axial curvature, pachymetry and anterior chamber depth were analysed. The best fitting corneal sphero-cylinder from the axial curvature, and the average corneal thickness for a series of different corneal regions were calculated. Intraocular pressure and axial length were also measured at each measurement session. Repeated measures ANOVA were used to investigate diurnal change in these parameters. Analysis of covariance was used to examine associations between the measured ocular parameters. ----- Results: Significant diurnal variation was found to occur in both the anterior and posterior corneal curvature and in the regional corneal thickness. Flattening of the anterior corneal best sphere was observed at the early morning measurement (p < 0.0001). The posterior cornea also underwent a significant steepening (p < 0.0001) and change in astigmatism 90/180° at this time. A significant swelling of the cornea (p < 0.0001) was also found to occur immediately after waking. Highly significant associations were found between the diurnal variation in corneal thickness and the changes in corneal curvature. ----- Conclusions: Significant diurnal variation occurs in the regional thickness and the shape of the anterior and posterior cornea. The largest changes in the cornea were typically evident upon waking. The observed non-uniform regional corneal thickness changes resulted in a steepening of the posterior cornea, and a flattening of the anterior cornea to occur at this time.

Corneal topography with Scheimpflug imaging and videokeratography : comparative study of normal eyes

PURPOSE: To compare the repeatability within anterior corneal topography measurements and agreement between measurements with the Pentacam HR rotating Scheimpflug camera and with a previously validated Placido disk–based videokeratoscope (Medmont E300). ------ SETTING: Contact Lens and Visual Optics Laboratory, School of Optometry, Queensland University of Technology, Brisbane, Queensland, Australia. ----- METHODS: Normal eyes in 101 young adult subjects had corneal topography measured using the Scheimpflug camera (6 repeated measurements) and videokeratoscope (4 repeated measurements). The best-fitting axial power corneal spherocylinder was calculated and converted into power vectors. Corneal higher-order aberrations (HOAs) (up to the 8th Zernike order) were calculated using the corneal elevation data from each instrument. ----- RESULTS: Both instruments showed excellent repeatability for axial power spherocylinder measurements (repeatability coefficients <0.25 diopter; intraclass correlation coefficients >0.9) and good agreement for all power vectors. Agreement between the 2 instruments was closest when the mean of multiple measurements was used in analysis. For corneal HOAs, both instruments showed reasonable repeatability for most aberration terms and good correlation and agreement for many aberrations (eg, spherical aberration, coma, higher-order root mean square). For other aberrations (eg, trefoil and tetrafoil), the 2 instruments showed relatively poor agreement. ----- CONCLUSIONS: For normal corneas, the Scheimpflug system showed excellent repeatability and reasonable agreement with a previously validated videokeratoscope for the anterior corneal axial curvature best-fitting spherocylinder and several corneal HOAs. However, for certain aberrations with higher azimuthal frequencies, the Scheimpflug system had poor agreement with the videokeratoscope; thus, caution should be used when interpreting these corneal aberrations with the Scheimpflug system.

Vision-based topological mapping and localisation

Competent navigation in an environment is a major requirement for an autonomous mobile robot to accomplish its mission. Nowadays, many successful systems for navigating a mobile robot use an internal map which represents the environment in a detailed geometric manner. However, building, maintaining and using such environment maps for navigation is difficult because of perceptual aliasing and measurement noise. Moreover, geometric maps require the processing of huge amounts of data which is computationally expensive. This thesis addresses the problem of vision-based topological mapping and localisation for mobile robot navigation. Topological maps are concise and graphical representations of environments that are scalable and amenable to symbolic manipulation. Thus, they are well-suited for basic robot navigation applications, and also provide a representational basis for the procedural and semantic information needed for higher-level robotic tasks. In order to make vision-based topological navigation suitable for inexpensive mobile robots for the mass market we propose to characterise key places of the environment based on their visual appearance through colour histograms. The approach for representing places using visual appearance is based on the fact that colour histograms change slowly as the field of vision sweeps the scene when a robot moves through an environment. Hence, a place represents a region of the environment rather than a single position. We demonstrate in experiments using an indoor data set, that a topological map in which places are characterised using visual appearance augmented with metric clues provides sufficient information to perform continuous metric localisation which is robust to the kidnapped robot problem. Many topological mapping methods build a topological map by clustering visual observations to places. However, due to perceptual aliasing observations from different places may be mapped to the same place representative in the topological map. A main contribution of this thesis is a novel approach for dealing with the perceptual aliasing problem in topological mapping. We propose to incorporate neighbourhood relations for disambiguating places which otherwise are indistinguishable. We present a constraint based stochastic local search method which integrates the approach for place disambiguation in order to induce a topological map. Experiments show that the proposed method is capable of mapping environments with a high degree of perceptual aliasing, and that a small map is found quickly. Moreover, the method of using neighbourhood information for place disambiguation is integrated into a framework for topological off-line simultaneous localisation and mapping which does not require an initial categorisation of visual observations. Experiments on an indoor data set demonstrate the suitability of our method to reliably localise the robot while building a topological map.

Camera Botanica 1

Camera Botanica 1 - testing a design process (unrealised buildings). ---------- Sited in a highly biodiverse and bushfire prone heathlands on the South-east coast of Western Australia, Camera Botanica 1 is a test of a new design methodology for achieving ecologically sustainable architecture in biodiverse, bushfire prone landscapes. ---------- The design methods were intensively site-based with the author-designer conducting his own site surveys using high-end professional grade surveying equipment such as: Real Time Kinematic GPS (landform survey); Terrestrial laser scanning (vegetation survey); laser levelling and Total Station surveys (erection of scaffolds and contour lines). ---------- This was the first time, internationally, that terrestrial laser scanning was used to measure vegetation. These precise surveys enabled the construction of highly detailed models and drawings - a facility that has not been available prior to this technology. ---------- Designed for a real client and a real site - Camera Botanica 1 is a hypothetical design outcome which demonstrates the efficacy of a new design methodology and thus expands on knowledge of the applicability of new surveying technologies to the design of ecologically sustainable architecture in biodiverse landscapes.

Camera Botanica 2

Camera Botanica 2 - testing a design process (unrealised building). Sited in a highly biodiverse and bushfire prone heathlands on the South-east coast of Western Australia, Camera Botanica 2 is a test of a new design methodology for achieving ecologically sustainable architecture in biodiverse, bushfire prone landscapes. ---------- The design method was intensively site-based with the author-designer conducting his own site surveys using high-end professional grade surveying equipment such as: Real Time Kinematic GPS (landform survey); Terrestrial laser scanning (vegetation survey); laser levelling and Total Station surveys (erection of scaffolds and contour lines). ---------- This was the first time, internationally, that terrestrial laser scanning was used to measure vegetation. These precise surveys enabled the construction of highly detailed models and drawings - a facility that has not been available prior to this technology. ---------- Designed for a real client and a real site - Camera Botanica 2 is a hypothetical design outcome which demonstrates the efficacy of a new design methodology and thus expands on knowledge of the applicability of new surveying technologies to the design of ecologically sustainable architecture in biodiverse landscapes.