Adam Nowosielski

SmartMonitor: an approach to simple, intelligent and affordable visual surveillance system

The paper provides fundamental information about the SmartMonitor --- an innovative surveillance system based on video content analysis. We present a short introduction to the characteristics of the developed system and a brief review of methods commonly applied in surveillance systems nowadays. The main goal of the paper is to describe planned basic system parameters as well as to explain the reason for creating it. SmartMonitor is being currently developed but some experiments have already been performed and their results are provided as well.

Extraction of the Foreground Regions by Means of the Adaptive Background Modelling Based on Various Colour Components for a Visual Surveillance System

Intelligent monitoring systems based on visual content analysis are often composed of three main modules — background modelling, object extraction and object tracking. This paper describes a method for adaptive background modelling utilizing Gaussian Mixture Models (GMM) and various colour components. The description is based on the experimental results obtained during the development of the SmartMonitor — an innovative security system based on video content analysis. In this paper the main characteristics of the system are introduced. An explanation of GMM algorithm and a presentation of its main advantages and drawbacks is provided. Finally, some experimentally obtained images containing foreground regions extracted with the use of various background models are presented.

The Removal of False Detections from Foreground Regions Extracted Using Adaptive Background Modelling for a Visual Surveillance System

For recent surveillance systems, the false detection removal process is an important step which succeeds the extraction of foreground regions and precedes the classification of object silhouettes. This paper describes the false object removal process when applied to the ’SmartMonitor’ system — i.e. an innovative monitoring system based on video content analysis that is currently being developed to ensure the safety of people and assets within small areas. This paper firstly briefly describes the basic characteristics and advantages of the system. A description of the methods used for background modelling and foreground extraction is also given. The paper then goes on to explain the artefacts removal process using various background models. Finally the paper presents some experimental results alongside a concise explanation of them.

Automatic Analysis of Vehicle Trajectory Applied to Visual Surveillance

In this paper we discuss a problem of automatic analysis of vehicles trajectories in the context of illegal movements. It is crucial to detect restricted or security critical behaviour on roads, especially for safety protection and fluent traffic. Here, we propose an vision-based algorithm for vehicle detection and tracking, which is later employed to recognize patterns in resultant trajectories. Experiments were performed on real video streams. They gave encouraging results.

Touchless Input Interface for Disabled

New interfaces for human-computer interaction are based on sophisticated methods of image processing and pattern recognition. They are based on some kind of visual system and offer touchless and convenient medium for communication for the user. On the other hand, there is a great number of people who suffer from some form of disability. They are unable to use standard input devices (mouse and keyboard) or take advantage from new sensors (e.g. Kinect). For people unable to control hand or body movements special assistive devices and interfaces are proposed. An interface for controlling an operating system is proposed in the article. It is based on processing video stream from a simple webcam. The solution is open and extensible. It can be used by disabled users or in hands-busy applications.

Minimal Interaction Touchless Text Input with Head Movements and Stereo Vision

Many modern interfaces for human-computer interaction are based on some kind of visual system and offer touchless and convenient medium for communication for the user. These non-contact interfaces predominantly limit their operations to gesture recognition or a pointer manipulation in a graphical user interface. The problem of typing in most cases is omitted since touchless interfaces are not designed for text entry purposes. There is, however, a great number of people who suffer from some form of disability. They are unable to use standard input devices (mouse and keyboard) or modern touchless interfaces. For those physically challenged people a minimal interaction interface for text entry operated with head movements is proposed in the paper. It is based on processing video streams from a stereo webcam, uses only directional head movements and a dictionary support.

Applying Image Features and AdaBoost Classification for Vehicle Detection in the ‘SM4Public’ System

The main goal of works described in the paper is to test and select algorithms to be implemented in the ‘SM4Public’ security system for public spaces. The paper describes the use of cascading approaches in the scenario concerning the detection of vehicles in static images. Three feature extractors were used along with benchmark datasets in order to prepare eight various cascades of classifiers. The algorithms selected for feature extraction are Histogram of Oriented Gradients, Local Binary Patterns and Haar-like features. AdaBoost was used as a classifier. The paper briefly introduces the ‘SM4Public’ system characteristics, characterizes the employed algorithms and presents sample experimental results.

QWERTY- and 8pen- Based Touchless Text Input with Hand Movement

Natural user interface offer touchless and convenient medium for communication. Hands-free control of electronic devices is largely based on sophisticated methods of image processing and pattern recognition. While mouse alternative may be successfully achieved with the use of camera and body movements, the main problem is the text entry. New touchless interfaces are generally introduced for convenient manipulation in Graphical User Interface but casual text entry using the same medium of communication is still necessary. The problem of non-contact onscreen keyboard interface is addressed in the article. Two interfaces for touchless text input are proposed and compared. The first one uses standard onscreen QWERTY keyboard operated with hand movement in front of the camera. The second solution is the adaptation of a stroke-based text entry technique 8pen [1] to touchless interface.

Pedestrian Traffic Distribution Analysis Using Face Recognition Technology

Pedestrian traffic distribution analysis in big public transport interchanges is aimed at improving transfer conditions, schedule optimization and route planning. Traditional survey of people flow is based on counting and interviewing of travellers. In the article, a new method of transport interchange analysis using image processing method is presented. In November 2009, Bemowo-Ratusz interchange in Warsaw has been subjected to detailed examination. The study was associated with the planned redevelopment. In parallel with the survey based on image processing methods, the traditional interview-based survey has been conducted in order to compare the results. Passenger transfer matrices obtained from optical analysis and traditional survey show good correlation. This proves the effectiveness and shows potential of image analysis methods for this kind of application.

3-Steps Keyboard: Reduced Interaction Interface for Touchless Typing with Head Movements

This paper introduces a novel technique for touchless typing with head movements allowing to reach any alphabet character in only three steps. Head movements are frequently used for human-computer interaction by users with motor impairments unable to operate standard computer input devices. In such interfaces great difficulty is typing. Many directional head movements are required to reach subsequent characters using the on-screen keyboard and additional mechanism (like eye blink or mouth open) supplements the selection process. In this paper, a reduced interaction keyboard for touchless typing with head movements is proposed. The solution is based on recognition of head movements in four main directions.