Online handwriting recognition with support vector machines - a kernel approach

Abstract. -The problem of automatic signature recognition has received little attention in comparison with the problem of signature verification despite its potential applications for accessing security-sensitive facilities and for processing certain legal and historical documents. This paper presents an efficient off-line human signature recognition system based on Support Vector Machines (SVM) and compares its performance with a traditional classification technique, Multi-Layer Perceptrons (MLP). In both cases we propose two approaches to the problem: (1) construct each feature vector using a set of global geometric and moment-based characteristics from each signature and (2) construct the feature vector using the bitmap of the corresponding signature. We also present a mechanism to capture the intrapersonal variability of each user using just one original signature. Our results empirically show that SVM, which achieves up to 71% correct recognition rate, outperforms MLP.

show abstract

“…Ref. [3] uses SVM for on-line handwriting recognition by designing a kernel able for sequential and non-fixed dimension data. Ref.…”

Section: Svm and Mlp For Automatic Off-linementioning

confidence: 99%

Support vector machines versus multi-layer perceptrons for efficient off-line signature recognition

Frı́as-Martı́nez

Sánchez

Vélez

2006

Engineering Applications of Artificial Intelligence

109

View full text Add to dashboard Cite

show abstract

“…It is well-known that the DTW distance is not a distance in a strict sense as it does not satisfy the triangle inequality and, therefore, it can not be used to define a positive definite kernel [3]. Despite this disadvantage, many variants of DTW and definitions of kernels based on DTW have been recently proposed in the literature.…”

Section: Related Workmentioning

confidence: 99%

“…Unfortunately, computing the DTW distance is prohibitively costly for many practical applications [33]. Moreover, it cannot be used to define a positive definite kernel since it violates the triangle inequality [3].…”

Section: Introductionmentioning

confidence: 99%

A multi-scale smoothing kernel for measuring time-series similarity

2015

View full text Add to dashboard Cite

In this paper a kernel for time-series data is introduced so that it can be used for any data mining task that relies on a similarity or distance metric. The main idea of our kernel is that it should recognize as highly similar time-series that are essentially the same but may be slightly perturbed from each other: for example, if one series is shifted with respect to the other or if it slightly misaligned. Namely, our kernel tries to focus on the shape of the time-series and ignores small perturbations such as misalignments or shifts. First, a recursive formulation of the kernel directly based on its definition is proposed. Then it is shown how to efficiently compute the kernel using an equivalent matrix-based formulation. To validate the proposed kernel three experiments have been carried out. As an initial step, several synthetic datasets have been generated from UCR time-series repository and the KDD challenge of 2007 with the purpose of validating the kernel-derived distance over shifted time-series. Also, the kernel has been applied to the original UCR time-series to analyze its potential in time-series classification in conjunction with Support Vector Machines. Finally, two real-world applications related to ozone concentration in atmosphere and electricity demand have been considered.

show abstract

“…SVM was developed firstly to solve the classification problem, but it is also applied to the domain of regression problems. It becomes popular because of its success in many applications, such as handwriting recognition [7], image clustering [44], text categorization [40], gene classification [34], protein structure prediction [33], etc. hyperplanes, see [27].…”

Section: Multi-category Classification Svmmentioning

confidence: 99%

Filter-based multiscale entropy analysis of complex physiological time series

Zhao

2013

Phys. Rev. E

View full text Add to dashboard Cite

Zhao, Liang, "Filter-based multiscale entropy analysis of complex physiological time series" (2013). Dissertations -ALL. Paper 3. ABSTRACTThe multiscale entropy (MSE) has been widely and successfully used in analyzing the complexity of physiologic time series. In this thesis, we re-interpret the averaging process in MSE as filtering a time series by a filter of a piecewise constant type. From this viewpoint, we introduce the filter-based multiscale entropy (FME) which filters a time series by filters to generate its multiple frequency components and then compute the blockwise entropy of the resulting components. By choosing filters adapted to the feature of a given time series, FME is able to better capture its multiscale information and to provide more flexibility for studying its complexity. Motivated by the heart rate turbulence theory which suggests that the human heartbeat interval time series (HHITS) can be described in piecewise linear patterns, we propose the piecewise linear filter multiscale entropy (PLFME) for the complexity analysis of the time series. Numerical results from PLFME are more robust to data of various lengths than those from MSE. We then propose wavelet packet transform entropy (WPTE) analysis. We apply WPTE analysis to HHITS using lower and higher piecewise linear filters. Numerical results show that WPTE using piecewise linear filters gives us the highest classification rates discriminating different cardiac systems among other multiscale entropy analysis. At the end, we discuss the application of FME on discrete time series. We introduce an 'eliminating' algorithm to examine and compare the complexity of coding and noncoding DNA sequences.

show abstract

Online handwriting recognition with support vector machines - a kernel approach

Cited by 229 publications

References 8 publications

Support vector machines versus multi-layer perceptrons for efficient off-line signature recognition

Support vector machines versus multi-layer perceptrons for efficient off-line signature recognition

A multi-scale smoothing kernel for measuring time-series similarity

Filter-based multiscale entropy analysis of complex physiological time series

Contact Info

Product

Resources

About