- Project management
- Managing project progress and reporting
- Laboratory management
- Student and staff management
- Machine Learning for Real-time Applications
- Deep Learning
Monocular Camera Based System to Detect Potholes, Uneven Surfaces and Non-protruding Hazards
To assist disabled people with everyday living, this project contributes to developing automated computer vision algorithm to detect impediments using OpenCV. Two different types of laser patterns (mesh and cross-hair) are projected on to surfaces (such as roads, stairs, pedestrian pathways) and using a GoPro camera, video of the projected laser patterns are recorded. Patterns are detected based on number of laser grid (intersections) and strength of the laser pattern intensities.
More information is available from in the form of publications. It was covered by IEEE Spectrum, Engineers Australia, The University of Melbourne’s Pursuit. The work also received national media attention (News.com.au). It was also featured in the Australian Unity’s Flourish magazine.
IoT-based Real-time Urban Micro-climate Monitoring
Urban Heat Island (UHI) is an increasingly visible phenomenon. UHI is mainly attributed to urbanization with increasing buildings and reduction of urban trees. With increasing buildings, night temperatures at the center of the cities remain high compared to suburban areas. This causes the temperature at Central Business Districts (CBDs) to have higher temperatures.
The above figure shows the two locations where sensors were deployed to understand the micro-climate. I worked on designing IoT network architecture, programming and deploying of systems to enable real-time monitoring of tree micro-climate. This helps us to understand the effect of environmental parameters with respect to different tree canopy and tree sizes.
Crowd Event Detection
Detecting events of groups of people in crowded scenarios is an important application in video surveillance and computer vision. Crowd events such as walking, running, merging, separating into groups (“splitting”), dispersing, and evacuating are critical to understanding crowd behavior. However, video data lie in a high-dimensional space, whereas events lie in a low-dimensional space. Difficulty lies in detecting people and inferring their movements in the scene. This work introduces a new framework based on optical flow manifolds (OFM) to detect crowd events. Essentially, the events are recognized from optical flow vector on a manifold. Experiment results suggest that the proposed semi-supervised approach performs best in detecting merging, separating into group (“splitting”), and dispersion events compared with existing methods. The advantages of the semi-supervised approach are the requirement of a single parameter to detect crowd events, and results that are provided on a frame-by-frame basis.
Detecting Loitering (Suspicious/Anomalous) Behavior in Crowded Scenes
From a crowd management/surveillance viewpoint, it is important to have automated tools to detect loitering people. To this end, this work provides a framework to detect and track patterns of suspicious people. The framework uses spatial and spatio-temporal features. Specifically, Gray Level Co-occurrence Matrix (GLCM)-based texture features, such as contrast, correlation, energy, and homogeneity, are used. These features are encoded to represent each video frame as well as spatio-temporal relations. Later, Hyperspherical clustering is used to extract trajectories of suspicious people, resulting in loitering behavior detection.
Low-cost, Real-time Water Quality Monitoring
Good water quality is essential for the health of our aquatic ecosystems. Continuous water quality monitoring is an important tool for catchment management authorities, providing real-time data for environmental protection and tracking pollution sources; however, continuous water quality monitoring at high temporal and spatial resolution remains prohibitively expensive. In this work, a low-cost wireless water physiochemistry sensing system is presented.
The results indicate that with appropriate calibration, a reliable monitoring system can be established. This will allow catchment managers to continuously monitor at higher temporal resolutions and also understand the behavior of aquatic animals relative to water pollution using data analysis.
Crowd Density Estimation in Crowded Scenes
Crowd density estimation is to estimate number of people per given area in a 2D scene. Accurate density estimation assists security personnel and crowd managers to have a knowledge of the ground situation in tine event of untoward incidents. Density estimation also helps to provide likelihood of stampede and tailor consequent plans for emergency plannings.
However, automated density estimation often faces difficulties in detecting motion from the scene due to varying environmental conditions, occlusion and crowded scenes. Instead of detecting and tracking individual person, density estimation is an approximate method to count people. This work presents a semi-supervised approach using optical flow features and clustering those features to represent density in a 2D area.
Real-time Tracking of Assets Using GPS, Short Message Service (SMS) and Google Maps
For asset management. fleet management and logistics, it is necessary to track assets, including trucks, cars, in real time. The project involved developing hardware, firmware and embedded system to acquire, process and send GPS locations to a centralized server through SMS. The platform developed is a scalable to generic sensor networking platform having numerous applications ranging from medical to military and from aerospace to underwater
Figure (Courtesy: Google Maps) shows the path traced by a vehicle in real time from the developed system.