Signature Pattern Recognition Research Articles

Application of The Speed-Up Robust Features Method To Identify Signature Image Patterns On Single Board Computer

Through the development of a signature pattern recognition program on SBC Beagle-bone Black, this research seeks to determine how to differentiate between real and false signatures. Three techniques of gathering data were employed in this study: interviews, observations, and a review of the literature. The quick application development method is the approach that is applied. The rapid, efficient, and brief development cycle (RAD) is emphasized. This study uses a use-case diagram to illustrate the application's logic and data flow. In this study, OpenCV is used as a digital image processing library along with the C++ programming language and QT creator as an integrated development environment (IDE). This application was subjected to both accuracy and functional testing. The following conclusions are drawn from the findings of the investigation and testing that was done: Using the fast library approach for approximate nearest neighbors (FLANN) and the speeded-up robust features (SURF) feature extraction method, the signature pattern recognition program on the Beagle-bone black SBC can differentiate between real and fraudulent signatures. Through the processes of generating image scale space, feature localization, and feature description, the SURF approach extracts feature from signature images. This signature pattern recognition application is one of the digital image processing apps that can be run on the Beagle-bone Black single board computer. This indicates that the specifications of the SBC Beagle-bone Black for digital image processing are good.

Signature Identification Menggunakan Metode Template Matching dan Fuzzy K-Nearest Neighbor

Abstract — Signature is the result of the process of writing a person of a particular nature as a symbolic substance, which means a symbol or mark. Signature is usually used as an identifying mark of a person, each person must have his own signature in a different pattern. Because it's used as a person's identifying badge, Signatures now become particularly susceptible to counterfeiting and abuse that require check with a signature pattern recognition. This research has created a signature pattern recognition system using methods Template Matching and Fuzzy K-Nearest Neighbor to help recognize a person's signature pattern. The number of signatures used is 110 in two categories: the original signature with 100 data and the false signature with 10 data, and there were 10 classes taken using smartphone cameras. From this research, it was found that the best value from the image size of 200x200 pixels was 92% of the class that owned the signature legible, Positive Predictive Value (PPV) 88% and False Rejection Rate (FRR) 12%, with a k=3 on the original signature, and 90% of the class that owned the signature legible, Negative Predictive Value (NPV) 90% dan False Acceptance Rate (FAR) 10% with a k=9 on the false signature. From these results, it could be concluded that methods Template Matching and Fuzzy K-Nearest Neighbor could be used for signature pattern recognition.

Signature Pattern Recognition using Kohonen Network

A signature is a special form of handwriting that used for human identification process. The current identification process is extremely ineffective. People have to manually compare signatures with the previously stored data. This study proposed SOM Kohonen algorithm as the method of signature pattern recognition. This method has able to visualize high-dimensional data. The image processing method is used in this study in pre-processing data phase. The accuracy of SOM Kohonen was 70 %, indicated the method used was good enough for pattern recognition.

N-Latticed Bounding Box-Based Signature Pattern Recognition by Accumulated ASCII Difference

<i>N</i>-Latticed Bounding Box-Based Signature Pattern Recognition by Accumulated ASCII Difference

JARINGAN SARAF TIRUAN BACKPROPAGATION UNTUK APLIKASI PENGENALAN TANDA TANGAN

Back propagation neural network is part of a multilayered feedforward neural network (MFN) which has been developed and reliable enough to solve the problem of approximation and pattern classification. Application of artificial neural network (ANN) in pattern recognition is one of the signature pattern recognition. Signature of each person are generally identical but not the same. This means that often a person's signature changes every time. This change concerns the position, size and pressure factors signature. Signature is the most widely used form of identification of a person. In general, to identify the signature is still done manually, by matching signatures at the time of the transaction with a valid signature. Therefore, we need a system that can analyze the characteristic signature making it easier to identify the person's signature. The research methodology used in the development of the system is a method Rappid Guidelines for Application Engineering (GRAPPLE), which only covers the design stage needs (Requirement Gathering), analysis (Analysis), the design (Design), and development (Development). This signature recognition process through several stages. First image through image processing stages, where the image will be used as the image of the gray / grayscaling. Once the image is converted into binary data by using thresholding. After going through the binary image processing, the data obtained will be the input value to the training process by using the backpropagation method. The results of the training will be used for the process of signature recognition. Image signatures used in this study were 80 image signatures from 10 respondents. The ratio between training data and testing data is 5:3. The test results show that the signature is able to recognize applications built with precision signature 84% of the tested signatures. Errors in the identification of signatures occur for several reasons, namely: the position of the signature, the image file is damaged, and the learning process is not maximized.

Gesture recognition for smart home applications using portable radar sensors.

In this article, we consider the design of a human gesture recognition system based on pattern recognition of signatures from a portable smart radar sensor. Powered by AAA batteries, the smart radar sensor operates in the 2.4 GHz industrial, scientific and medical (ISM) band. We analyzed the feature space using principle components and application-specific time and frequency domain features extracted from radar signals for two different sets of gestures. We illustrate that a nearest neighbor based classifier can achieve greater than 95% accuracy for multi class classification using 10 fold cross validation when features are extracted based on magnitude differences and Doppler shifts as compared to features extracted through orthogonal transformations. The reported results illustrate the potential of intelligent radars integrated with a pattern recognition system for high accuracy smart home and health monitoring purposes.

Non-parametricdamage detection and characterization using smart piezoceramicmaterial

The detection of damages by modal analysis and similarvibration techniques depends upon the knowledge and estimation ofvarious modal parameters. In addition to the associated difficulties,such low-frequency dynamic response based techniques fail to detectincipient damages. Smart piezoelectric ceramic (PZT) transducers,which act both as actuators and sensors in a self-analyzing manner,are emerging to be effective in non-parametric health monitoring ofstructural systems. In this paper we present the results of anexperimental study for the detection and characterization of damagesusing PZT transducers on aluminum specimens. The method of extractingthe impedance characteristics of the PZT transducer, which iselectromechanically coupled to the host structure, is adopted fordamage detection. Three types of damage are simulated and assessed bythe bonded PZT transducers for characterization. We present theeffectiveness of PZT transducers in the detection andcharacterization of incipient damages without the need to know themodal parameters. The PZT transducers are found to have asignificantly large sensing area for detecting even small incipientdamages. The possibility of replicating the pristine state signaturesof different transducers under similar conditions of bonding andgeometrical location is also explored. For appropriatecharacterization of damages, a few statistical signature patternrecognition techniques are evaluated.