DESEO: an active vision system for detection, tracking and recognition

[EN]In this paper, a basic conceptual architecture aimed at the design of Computer Vision System is qualitatively described. The proposed architecture addresses the design of vision systems in a modular fashion using modules with three distinct units or components: a processing network or diagnostics unit, a control unit and a communications unit. The control of the system at the modules level is designed based on a Discrete Events Model. This basic methodology has been used to design a realtime active vision system for detection, tracking and recognition of people. It is made up of three functional modules aimed at the detection, tracking, recognition of moving individuals plus a supervision module.

Exploring the cardiovascular disease continuum: blood pressure and target organ damage

Introduction The objectives of this thesis are to: (1) examine how ambulatory blood pressure monitoring (ABPM) refines office blood pressure (BP) measurement; (2) determine if absolute ambulatory BP or dipping status is better associated with target organ damage (TOD); (3) explore the association of isolated nocturnal hypertension (INH) with TOD; and (4) investigate the association of night-time BP with ultrasound markers of cardiovascular damage. Methods Data from the Mitchelstown Cohort Study was analysed to deliver objectives 1 and 2. Objective 3 was addressed by a systematic review and analysis of data from the Mitchelstown Study. A sample of participants from the Mitchelstown Study underwent an echocardiogram for speckle tracking analysis and carotid ultrasound to achieve objective 4. Results ABPM reclassifies hypertension status in approximately a quarter of individuals, with white coat and masked hypertension prevalence rates of 11% and 13% respectively. Night-time systolic BP is better associated with TOD than daytime systolic BP and dipping level. In multi-variable models the odds ratio (OR) for LVH was 1.4 (95% CI 1.1 -1.8) and for albumin:creatinine ratio ≥ 1.1 mg/mmol was 1.5 (95% CI 1.2 – 1.8) for each 10 mmHg rise in night-time systolic BP. The evidence for the association of INH with TOD is inconclusive. Night-time systolic BP is significantly associated with global longitudinal strain (GLS) (beta coefficient 0.85 for every 10 mmHg rise, 95% CI 0.3 – 1.4) and carotid plaques (OR 1.9 for every 10 mmHg rise, 95% CI 1.1 – 3.2) in univariable analysis. The findings persist for GLS in sex and age adjusted models but not in multivariable models. Discussion Hypertension cannot be effectively managed without using ABPM. Night-time systolic BP is better associated with TOD than daytime systolic BP and dipping level, and therefore, may be a better therapeutic target in future studies.

Diversity of LFPs Activated in Different Target Regions by a Common CA3 Input

Identifying the pathways contributing to local field potential (LFP) events and oscillations is essential to determine whether synchronous interregional patterns indicate functional connectivity. Here, we studied experimentally and numerically how different target structures receiving input from a common population shape their LFPs. We focused on the bilateral CA3 that sends gamma-paced excitatory packages to the bilateral CA1, the lateral septum, and itself (recurrent input). The CA3-specific contribution was isolated from multisite LFPs in target regions using spatial discrimination techniques. We found strong modulation of LFPs by target-specific features, including the morphology and population arrangement of cells, the timing of CA3 inputs, volume conduction from nearby targets, and co-activated inhibition. Jointly they greatly affect the LFP amplitude, profile, and frequency characteristics. For instance, ipsilateral (Schaffer) LFPs occluded contralateral ones, and septal LFPs arise mostly from remote sources while local contribution from CA3 input was minor. In the CA3 itself, gamma waves have dual origin from local networks: in-phase excitatory and nearly antiphase inhibitory. Also, waves may have different duration and varying phase in different targets. These results indicate that to explore the cellular basis of LFPs and the functional connectivity between structures, besides identifying the origin population/s, target modifiers should be considered.

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.

Person tracking using motion detection and optical flow

Person tracking systems to date have either relied on motion detection or optical flow as a basis for person detection and tracking. As yet, systems have not been developed that utilise both these techniques. We propose a person tracking system that uses both, made possible by a novel hybrid optical flow-motion detection technique that we have developed. This provides the system with two methods of person detection, helping to avoid missed detections and the need to predict position, which can lead to errors in tracking and mistakes when handling occlusion situations. Our results show that our system is able to track people accurately, with an average error less than four pixels, and that our system outperforms the current CAVIAR benchmark system.

Adaptive optical flow for person tracking

Person tracking systems are dependent on being able to locate a person accurately across a series of frames. Optical flow can be used to segment a moving object from a scene, provided the expected velocity of the moving object is known; but successful detection also relies on being able segment the background. A problem with existing optical flow techniques is that they don’t discriminate the foreground from the background, and so often detect motion (and thus the object) in the background. To overcome this problem, we propose a new optical flow technique, that is based upon an adaptive background segmentation technique, which only determines optical flow in regions of motion. This technique has been developed with a view to being used in surveillance systems, and our testing shows that for this application it is more effective than other standard optical flow techniques.

Improved maritime target tracker using colour fusion

Searching for humans lost in vast stretches of ocean has always been a difficult task. This paper investigates a machine vision system that addresses this problem by exploiting the useful properties of alternate colour spaces. In particular, the paper investigates the fusion of colour information from the HSV, RGB, YCbCr and YIQ colour spaces within the emission matrix of a Hidden Markov Model tracker to enhance video based maritime target detection. The system has shown promising results. The paper also identifies challenges still needing to be met.