The problem of visual inspection of outdoor environments (e.g., airports, railway stations, roads,
etc.) has received growing attention in recent times. The work presented in this thesis is a
component of a larger project to apply intelligent Closed-Circuit Television (CCTV) to enhance
the counter-terrorism capability for the protection of mass transport systems. The purpose of
intelligent surveillance systems is to automatically perform surveillance tasks by applying cameras
in place of human eyes. Recently, with the development of video hardware, the video surveillance
system is becoming more widely applied and is attracting more researchers to develop fast and
robust algorithms.
In this thesis, we describe the proposed pedestrian classification and tracking system that is
able to track and label multiple people in an outdoor environment such as a railway station. We
propose an approach that combines blob matching with particle filtering to track multiple people
in the scene, i.e., the proposed method selects the successful features of blob matching and particle
filtering for tracking. In our proposed method, the system can easily track persons even when they
are partially occluded by each other and can track them correctly after merging. In addition, a
novel appearance model derived from the colour information from both the moving regions and theoriginal input colour image is proposed to track people in the event of poor foreground extraction.
Additionally, the proposed appearance model also includes spatial information of the human
body in both vertical and horizontal directions, making location more accurate. In the object
classification stage, hierarchical chamfer matching combined with the particle filter is applied to
classify commuters in the railway station example into several classes. Based on single camera
tracking, we extend our work to multiple camera-based people tracking using a two-level tracking
scheme that includes image level tracking and particle filter-based ground level tracking.
In addition, a novel method to extract cars from moving regions including shadow area, based
on shape and colour information, is proposed. Chamfer template matching score, and non-shadow
region edge score, are applied as the shape information; while the shadow confidence score
(SCS) is used as the colour information. Experimental results show that the proposed method
is significantly better than the approaches where only the colour information is considered.