The fact that the signature is widely used as a means of personal verification emphasizes the need for an automatic verification system because of the unfortunate side-effect of being easily abused by those who would feign the identification or intent of an individual. A great deal of work has been done in the area of off-line signature verification over the past few decades. Verification can be performed either Offline or Online based on the application. Online systems use dynamic information of a signature captured at the time the signature is made. Offline systems work on the scanned image of a signature. In this paper, we present a method for Offline Verification of signatures using a set of simple shape based geometric features. The features that are used are Area, Center of gravity, Eccentricity, Kurtosis and Skewness. Before extracting the features, preprocessing of a scanned image is necessary to isolate the signature part and to remove any spurious noise present. The system is initially trained using a database of signatures obtained from those individuals whose signatures have to be authenticated by the system. The details of preprocessing as well as the features depicted above are described throughout the discussion. Then artificial neural network (ANN) was used to verify and classify the signatures: exact or forged, and a classification ratio of about 93% was obtained under a threshold of 90%.The implementation details and simulation results are discussed in the thesis.

One of the most important problems in robot kinematics

Forest fires are among the most dangerous natural threats that bring calamities to a community and can turn it totally upside down.In this paper, to enable a prevention mechanism, we rely on analytics to build a novel fire danger index model that predicts the risk of a developing fire in north Lebanon. We use correlation methods such as statistical regression, Pearson, Spearman and Kendall's Tau correlation to identify the most affecting parameters on fire ignition during the last six years in north Lebanon.The correlations of these attributes with fire occurrence are studied in order to develop the fire danger index. The strongly correlated attributes are then derived. We rely on linear regression to model the fire index as function of a reduced set of weather parameters that are easy to measure.This is critical as it facilitates the application of such prevention models in developing countries like Lebanon. The outcomes resulting from validation tests of the proposed index show high performance in the Lebanese regions. An assessment versus common widespread weather models is then made and has showed the significance the selected parameters.It is strongly believed that this index will help improve the ability of fire prevention measures in the Mediterranean basin area.

No abstract

Falling and its resulting injuries are an important public health problem for older adults. The National Safety Council estimates that persons over the age of 65 have the highest mortality rate (death rate) from injuries. The risk of falling increases with age; one of three adults 65 or older falls every year. Demographic predictions of population aged 65 and over suggest the need for telemedicine applications in the eldercare domain. This paper presents an integrated monitoring system for the detection of people falls in home environment. The system consist of combining low level features extracted from a video and heart rate tracking in order to classify the fall event. The extracted data will be processed by a neural network for classifying the events in two classes: fall and not fall. Reliable recognition rate of experimental results underlines satisfactory performance of our system