Lambert Schomaker

UNIPEN project of on-line data exchange and recognizer benchmarks

We report the status of the UNIPEN project of data exchange and recognizer benchmarks started two years ago at the initiative of the International Association of Pattern Recognition (Technical Committee 11). The purpose of the project is to propose and implement solutions to the growing need of handwriting samples for online handwriting recognizers used by pen-based computers. Researchers from several companies and universities have agreed on a data format, a platform of data exchange and a protocol for recognizer benchmarks. The online handwriting data of concern may include handprint and cursive from various alphabets (including Latin and Chinese), signatures and pen gestures. These data will be compiled and distributed by the Linguistic Data Consortium. The benchmarks will be arbitrated the US National Institute of Standards and Technologies. We give a brief introduction to the UNIPEN format. We explain the protocol of data exchange and benchmarks.

Text-Independent Writer Identification and Verification Using Textural and Allographic Features

The identification of a person on the basis of scanned images of handwriting is a useful biometric modality with application in forensic and historic document analysis and constitutes an exemplary study area within the research field of behavioral biometrics. We developed new and very effective techniques for automatic writer identification and verification that use probability distribution functions (PDFs) extracted from the handwriting images to characterize writer individuality. A defining property of our methods is that they are designed to be independent of the textual content of the handwritten samples. Our methods operate at two levels of analysis: the texture level and the character-shape (allograph) level. At the texture level, we use contour-based joint directional PDFs that encode orientation and curvature information to give an intimate characterization of individual handwriting style. In our analysis at the allograph level, the writer is considered to be characterized by a stochastic pattern generator of ink-trace fragments, or graphemes. The PDF of these simple shapes in a given handwriting sample is characteristic for the writer and is computed using a common shape codebook obtained by grapheme clustering. Combining multiple features (directional, grapheme, and run-length PDFs) yields increased writer identification and verification performance. The proposed methods are applicable to free-style handwriting (both cursive and isolated) and have practical feasibility, under the assumption that a few text lines of handwritten material are available in order to obtain reliable probability estimates

Text detection from natural scene images: towards a system for visually impaired persons

We propose a system that reads the text encountered in natural scenes with the aim to provide assistance to the visually impaired persons. This paper describes the system design and evaluates several character extraction methods. Automatic text recognition from natural images receives a growing attention because of potential applications in image retrieval, robotics and intelligent transport system. Camera-based document analysis becomes a real possibility with the increasing resolution and availability of digital cameras. However, in the case of a blind person, finding the text region is the first important problem that must be addressed, because it cannot be assumed that the acquired image contains only characters. At first, our system tries to find in the image areas with small characters. Then it zooms into the found areas to retake higher resolution images necessary for character recognition. In the present paper, we propose four character-extraction methods based on connected components. We tested the effectiveness of our methods on the ICDAR 2003 Robust Reading Competition data. The performance of the different methods depends on character size. In the data, bigger characters are more prevalent and the most effective extraction method proves to be the sequence: Sobel edge detection, Otsu binarization, connected component extraction and rule-based connected component filtering.

Automatic writer identification using connected-component contours and edge-based features of uppercase Western script

In this paper, a new technique for offline writer identification is presented, using connected-component contours (COCOCOs or CO/sup 3/s) in uppercase handwritten samples. In our model, the writer is considered to be characterized by a stochastic pattern generator, producing a family of connected components for the uppercase character set. Using a codebook of CO/sup 3/s from an independent training set of 100 writers, the probability-density function (PDF) of CCs was computed for an independent test set containing 150 unseen writers. Results revealed a high-sensitivity of the CO/sup 3/ PDF for identifying individual writers on the basis of a single sentence of uppercase characters. The proposed automatic approach bridges the gap between image-statistics approaches on one end and manually measured allograph features of individual characters on the other end. Combining the CO/sup 3/ PDF with an independent edge-based orientation and curvature PDF yielded very high correct identification rates.

An overview and comparison of voting methods for pattern recognition

In pattern recognition, there is a growing use of multiple classifier combinations with the goal to increase recognition performance. In many cases, plurality voting is a part of the combination process. In this article, we discuss and test several well known voting methods from politics and economics on classifier combination in order to see if an alternative to the simple plurality vote exists. We found that, assuming a number of prerequisites, better methods are available, that are comparatively simple and fast.

Using Pen-Based Outlines for Object-Based Annotation and Image-Based Queries

A method for image-based queries and search is proposed which is based on the generation of object outlines in images by using the pen, e.g., on color pen computers. The rationale of the approach is based on a survey on user needs, as well as on considerations from the point of view of pattern recognition and machine learning. By exploiting the actual presence of the human users with their perceptual-motor abilities and by storing textually annotated queries, an incrementally learning image retrieval system can be developed. As an initial test domain, sets of photographs of motor bicycles were used. Classification performances are given for outline and bitmap-derived feature sets, based on nearest-neighbour matching, with promising results. The benefit of the approach will be a user-based multimodal annotation of an image database, yielding a gradual improvement in precision and recall over time.

Finding structure in diversity: a hierarchical clustering method for the categorization of allographs in handwriting

The paper introduces a variant of agglomerative hierarchical clustering techniques. The new technique is used for categorizing character shapes (allographs) in large data sets of handwriting into a hierarchical structure. Such a technique may be used as the basis for a systematic naming scheme of character shapes. Problems with existing methods are described and the proposed method is explained. After application of the method to a very large set of characters, separately for all the letters of the alphabet, relevant clusters are identified and given a unique name. Each cluster represents an allograph prototype.

Writer identification using directional ink-trace width measurements

As suggested by modern paleography, the width of ink traces is a powerful source of information for off-line writer identification, particularly if combined with its direction. Such measurements can be computed using simple, fast and accurate methods based on pixel contours, the combination of which forms a powerful feature for writer identification: the Quill feature. It is a probability distribution of the relation between the ink direction and the ink width. It was tested in writer identification experiments on two datasets of challenging medieval handwriting and two datasets of modern handwriting. The feature achieved a nearest-neighbor accuracy in the range of 63-95%, which even approaches the performance of two state-of-the-art features in contemporary-writer identification (Hinge and Fraglets). The feature is intuitive and explainable and its principle is supported by a model of trace production by a quill. It illustrates that ink width patterns are valuable. A slightly more complex variant of Quill, QuillHinge, scored 70-97% writer identification accuracy. The features are already being used by domain experts using a graphical interface.

Using stroke- or character-based self-organizing maps in the recognition of on-line, connected cursive script

Abstract Comparisons are made between a number of stroke-based and character-based recognizers of connected cursive script. In both approaches a Kohonen self-organizing neural network is used as a feature-vector quantizer. It is found that a “best match only” character-based recognizer performs better than a “best match only” stroke-based recognizer at the cost of a substantial increase in computation. However, allowing up to three multiple stroke interpretations yielded a much larger improvement on the performance of the stroke-based recognizer. Within the character-based architecture, a comparison is made between temporal and spatial resampling of characters. No significant differences between these resampling methods were found. Geometrical normalization (orientation, slant) did not significantly improve the recognition. Training sets of more than 500 cursive words appeared to be necessary to yield acceptable performance.

Effects of motor programming on the power spectral density function of finger and wrist movements

Power spectral density analysis was applied to the frequency content of the acceleration signal of pen movements in line drawing. The relative power in frequency bands between 1 and 32 Hz was measured as a function of motoric and anatomic task demands. Results showed a decrease of power at the lower frequencies (1-4 Hz) of the spectrum and an increase in the middle (9-12 Hz), with increasing motor demands. These findings evidence the inhibition of visual control and the disinhibition of physiological tremor under conditions of increased programming demands. Adductive movements displayed less power than abductive movements in the lower end of the spectrum, with a simultaneous increase at the higher frequencies. The relevance of the method for the measurement of neuromotor noise as a possible origin of delays in motor behavior is discussed.