Online Handwriting Recognition

Plamondon, Réjean; Lopresti, Daniel; Schomaker, Lambert; Srihari, Rohini K.

doi:10.1002/047134608x.w5508

Cited by 34 publications

(20 citation statements)

References 96 publications

(33 reference statements)

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Our re-sampling approach is straightforward and closely related to widely accepted techniques [29]. The results presented in Section V indicate that the forging accuracy for inter-tablet experiments versus intra-tablet experiments were statistically insignificant.…”

Section: B Accounting For Hardware Variabilitymentioning

confidence: 96%

Forgery Quality and Its Implications for Behavioral Biometric Security

Ballard

Lopresti

Monrose

2007

IEEE Trans. Syst., Man, Cybern. B

View full text Add to dashboard Cite

Abstract-Biometric security is a topic of rapidly growing importance in the areas of user authentication and cryptographic key generation. In this paper, we describe our steps toward developing evaluation methodologies for behavioral biometrics that take into account threat models which have been largely ignored. We argue that the pervasive assumption that forgers are minimally motivated (or, even worse, naïve) is too optimistic and even dangerous. Taking handwriting as a case in point, we show through a series of experiments that some users are significantly better forgers than others, that such forgers can be trained in a relatively straightforward fashion to pose an even greater threat, that certain users are easy targets for forgers, and that most humans are a relatively poor judge of handwriting authenticity and hence their unaided instincts cannot be trusted. Additionally, to overcome current labor-intensive hurdles in performing more accurate assessments of system security, we present a generative attack model based on concatenative synthesis that can provide a rapid indication of the security afforded by the system. We show that our generative attacks match or exceed the effectiveness of forgeries rendered by the skilled humans we have encountered.

show abstract

Section: B Accounting For Hardware Variabilitymentioning

confidence: 96%

Forgery Quality and Its Implications for Behavioral Biometric Security

Ballard

Lopresti

Monrose

2007

IEEE Trans. Syst., Man, Cybern. B

View full text Add to dashboard Cite

show abstract

“…In case of illegible script, human readers activate a number of cognitive facilities. If superficial association fails, the reader may solve the underlying reading puzzle by analytically combining elements from (a) the linguistic context, (b) the shape context [19,31] as well as (c) the pragmatic context. At the very moment of being handed a handwritten shopping list, the human reader is precued to a very limited subset of the total lexicon, even before having attempted to read the first character.…”

Section: (H) Processing Typementioning

confidence: 99%

Reading Systems: An Introduction to Digital Document Processing

Schomaker

2007

Digital Document Processing

View full text Add to dashboard Cite

show abstract

“…; N. Here, the random vector s represents the style whose identity is entirely determined by its class-conditional means. 3 We make the following assumptions on the feature distributions: 3. A style is generated by selecting the class means.…”

Section: Model For Continuous Stylesmentioning

confidence: 99%

“…Human readers achieve higher accuracy on word groups from single writers than on word groups from different writers [3]. Hong and Hull use visual relations between images, i.e., similarities between words or word-parts, to increase word recognition accuracy [4].…”

Section: Introductionmentioning

confidence: 99%

Style context with second-order statistics

Veeramachaneni

Nagy²

2005

IEEE Trans. Pattern Anal. Machine Intell.

View full text Add to dashboard Cite

Abstract-Patterns often occur as homogeneous groups or fields generated by the same source. In multisource recognition problems, such isogeny induces statistical dependencies between patterns (termed style context). We model these dependencies by secondorder statistics and formulate the optimal classifier for normally distributed styles. We show that model parameters estimated only from pairs of classes suffice to train classifiers for any test field length. Although computationally expensive, the style-conscious classifier reduces the field error rate by up to 20 percent on quadruples of handwritten digits from standard NIST data sets.

show abstract

Online Handwriting Recognition

Abstract: The sections in this article are Handwriting Data Acquisition and Preprocessing Recognition Postprocessing Conclusion Acknowledgments

Cited by 34 publications

References 96 publications

Forgery Quality and Its Implications for Behavioral Biometric Security

Forgery Quality and Its Implications for Behavioral Biometric Security

Reading Systems: An Introduction to Digital Document Processing

Style context with second-order statistics

Contact Info

Product

Resources

About