Architectures for detecting and solving conflicts: two-stage classification and support vector classifiers

Vuurpijl, L.G.; Schomaker, Lambert; Erp, Merijn van

doi:10.1007/s10032-003-0104-1

Cited by 20 publications

(11 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The author try to get the best of both generative and discriminative classifiers. In (Vuurpijl et al, 2003), several ways to detect conflicts in any first stage classifier or ensemble of classifiers are described. Majority vote is explored, to activate pairwise SVM again.…”

Section: Two Stage Classificationmentioning

confidence: 99%

Hybrid generative/discriminative classifier for unconstrained character recognition

Prévost

Oudot

Moises

et al. 2005

Pattern Recognition Letters

View full text Add to dashboard Cite

Section: Two Stage Classificationmentioning

confidence: 99%

Hybrid generative/discriminative classifier for unconstrained character recognition

Prévost

Oudot

Moises

et al. 2005

Pattern Recognition Letters

View full text Add to dashboard Cite

“…Note that trying to cover all variations in handwriting is an ill-posed problem, since it has been shown that handwriting is individual 33 and so, each new writer adds new shapes. 42 We are currently pursuing a better coverage of character shapes by prototypes in two ways. The first elaborates on the experiments presented in this paper by using more data.…”

Section: Resultsmentioning

confidence: 99%

Automatic Allograph Matching in Forensic Writer Identification

Niels

Vuurpijl

Schomaker

2007

Int. J. Patt. Recogn. Artif. Intell.

Self Cite

View full text Add to dashboard Cite

A well-established task in forensic writer identification focuses on the comparison of prototypical character shapes (allographs) present in handwriting. In order for a computer to perform this task convincingly, it should yield results that are plausible and understandable to the human expert. Trajectory matching is a well-known method to compare two allographs. This paper assesses a promising technique for so-called humancongruous trajectory matching, called Dynamic Time Warping (DTW). In the first part of the paper, an experiment is described that shows that DTW yields results that correspond to the expectations of human users. Since DTW requires the dynamics of the handwritten trace, the "online" dynamic allograph trajectories need to be extracted from the "offline" scanned documents. In the second part of the paper, an automatic procedure to perform this task is described. Images were generated from a large online dataset that provides the true trajectories. This allows for a quantitative assessment of the trajectory extraction techniques rather than a qualitative discussion of a small number of examples. Our results show that DTW can significantly improve the results from trajectory extraction when compared to traditional techniques.

show abstract

“…Small style elements which are present within a character are the result of the writer's physiological make up as well as education and personal preference. Experiences on style variation in on-line handwriting recognition show evidence that the amount of shape information at the level of the characters is increasing as a function of the number of writers [16]. It should be noted that the essence of our method does not seem to be located in an exhaustive enumeration of all possible connectedcomponent allographic part shapes.…”

Section: Discussionmentioning

confidence: 96%

Automatic Writer Identification Using Fragmented Connected-Component Contours

Schomaker¹,

Bulacu²,

Franke

Ninth International Workshop on Frontiers in Handwriting Recognition

View full text Add to dashboard Cite

In this paper, a method for off-line writer identification is presented, using the contours of fragmented connectedcomponents in mixed-style handwritten samples of limited size. The writer is considered to be characterized by a stochastic pattern generator, producing a family of character fragments (fraglets). Using a codebook of such fraglets from an independent training set, the probability distribution of fraglet contours was computed for an independent test set. Results revealed a high sensitivity of the fraglet histogram in identifying individual writers on the basis of a paragraph of text. Large-scale experiments on the optimal size of Kohonen maps of fraglet contours were performed, showing usable classification rates within a non-critical range of Kohonen map dimensions.Further validation experiments on variable-sized random subsets from an independent set of 215 writers gives additional support for the proposed method. The proposed automatic approach bridges the gap between imagestatistics approaches and manual character-based methods.

show abstract

Architectures for detecting and solving conflicts: two-stage classification and support vector classifiers

Cited by 20 publications

References 23 publications

Hybrid generative/discriminative classifier for unconstrained character recognition

Hybrid generative/discriminative classifier for unconstrained character recognition

Automatic Allograph Matching in Forensic Writer Identification

Automatic Writer Identification Using Fragmented Connected-Component Contours

Contact Info

Product

Resources

About