Handwriting Model Research Articles

The method of regularizing the sign space of the biometric model of the keyboard handwriting of users of military information systems on the basis of factor analysis

The article addresses the current scientific task of regularizing the feature space of the biometric model of the keyboard handwriting of users of military information systems in order to increase the effectiveness of the user authentication procedure by access control and demarcation systems. The setting of this scientific task is due to the presence of the following shortcomings of the existing biometric models of users' keyboard handwriting: when the number of users in the system increases, the complexity of their classification increases; the complexity of formalizing the uniqueness of users; the implementation of determining the fact of replacing an already authorized user is too complex; the feature space of the existing biometric models is insignificant due to the limitation of the set of properties of the modern keyboard, which negatively affects the accuracy of the authentication procedure. A biometric model of keyboard handwriting was chosen, the feature of which is the formalization of the uniqueness of the user of the information system on the basis of the detection of the regularities of the keyboard handwriting by fuzzy rules. An improved method of regularization of the feature space of the biometric model of keyboard handwriting of users of military information systems is proposed. The essence of the proposed method, which distinguishes it from the existing ones, is that the increase in the set of features of the biometric model is achieved by adding to it discovered hidden facts from the set of own features based on factor analysis with the highest rate of their variability. The application of the proposed method allows solving the problem of non-linear separability of the n-dimensional feature space of the biometric model of keyboard handwriting, which in turn allows to increase the accuracy and reliability of the authentication procedure by control systems and access demarcation of military information systems.

Using Structural Equation Modeling to analyze handwriting of children and youth with Autism Spectrum Disorder.

This study tests a handwriting model for children and youth with Autism Spectrum Disorder (ASD) that displays the relationships between handwriting process and product characteristics, and the predictors of these characteristics. Structural Equation Modelling was used to test the model for children and youth with ASD (n = 50) and typically developing peers (n = 50), ages 10 to 15, for a copying and freestyle handwriting task. Findings suggest a generic handwriting model applying to both groups and both handwriting tasks. Unique effects for children and youth with ASD were identified for: (1) high interdependence of handwriting process characteristics, (2) longer pen stroke in air leading to better legibility, and (3) lower scores for executive functions leading to lower scores for legibility.

Technology-assisted Dysgraphic Children Handwriting Model

Difficulties in handwriting or dysgraphia can have a negative effect on one’s self-esteem and academic achievement in school. This phenomenon is considered a detriment when students have difficulty expressing knowledge through writing; causing extra hardship in mastering all subjects they are learning. Past studies did not deliberate on specific components related to the intervention of dysgraphic students’ handwriting skills development comprehensively, as can be seen in today’s existing handwriting models. Therefore, this study focused on the underlying models of dysgraphic students’involving the main components in the writing process, and the components of ICT support. The principles of modeling with the potential impact of 1) visualization, 2) imagination, and 3) automation in writing need to be considered in addition to the mastery of writing skills. With such a specific model, solutions to these dysgraphic students’ handwriting skill problems can be generalized. Therefore, the main objective of this study was to propose a technology-assisted Dysgraphic Children Handwriting Model (DCHM) to address the problems encountered. This DCHM was validated by four experts – three special education teachers and a paediatric and adolescent psychiatrist. The resulting score was calculated as a percentage value based on the total score obtained for each question. The average percentage for all the criteria evaluated was 91.5%. Overall, all four experts agreed on the key components of the DCHM supported by the ICT component. This study is important for considering handwriting development more holistically and to identify the performance of students’ handwriting in the context of a robust handwriting model.

Work in Hand: Script, Print, and Writing, 1690–1840. Aileen Douglas. Oxford: Oxford University Press, 2017. Pp. x+229.

<i>Work in Hand: Script, Print, and Writing, 1690–1840</i>. Aileen Douglas. Oxford: Oxford University Press, 2017. Pp. x+229.

MIMO human handwriting model in stochastic environment

The present paper deals with a multi inputs multi outputs (MIMO) human handwriting model to characterise the pen-tip displacement on the plan from two forearm muscles activities, called electromyography signals (EMG). The proposed model takes into account perturbations and incertitude whish can affect the handwriting process (instability of the writing support, psychical state of the writer, brusque movement, etc.). In this sense, an experimental approach was presented to record the displacements of a pen-tip moving on the plane and two EMG signals during the handwriting motion. The velocity of the writing and ARMAX structure are used to characterise this biological act. Recursive extended least square (RELS) algorithm is used to estimate the parameters of the proposed handwriting model. The obtained structure shows good concordance with the experimental recorded data.

MIMO human handwriting model in stochastic environment

MIMO human handwriting model in stochastic environment

Manuscript shapes generated by novel bi-axis control algorithm based on a mathematical handwriting model

&lt;span lang="EN-US"&gt;Handwriting movement is one of the most complex activities of human motions. It’s a blend of kinesthetic, cognitive, perceptual and motor components. The study of this biological process shows that bell-shaped velocity profiles are generally observed in the handwriting motion. In this paper, an identification technique, based on Recursive Least Square algorithm (RLS), is proposed to identify the pen-tip movement in human handwriting process, by using input and output data which present EMG signals and velocities according to x and y coordinates. Using the estimated coordinates that have resulted from the velocity model; we propose a novel algorithm to generate handwritten graphic traces, which is inspired from the idea of tracing circles by Bresenham bi-axis control algorithm. The effectiveness of this approach should be observed on predicting cursive Arabic letters and Arabic word written on (&lt;/span&gt;&lt;em&gt;&lt;span lang="EN-US"&gt;x,y&lt;/span&gt;&lt;/em&gt;&lt;span lang="EN-US"&gt;) plane, these shapes constituting a recorded experimental basis.&lt;/span&gt;

Applying a Handwriting Measurement Model for Capturing Cognitive Load Implications Through Complex Figure Drawing

The aim of the study was to examine the application of a computerized handwriting model for characterizing complex figure-drawing performance. We posit that spatial, temporal, and pressure measures that reflect figure-drawing behavior will differ significantly under two mental workload conditions, and that both drawing and handwriting process measures will predict the quality of what is drawn and/or written. Thirty participants copied the Rey–Osterrieth Complex Figure Test (ROCFT). They then reproduced it from memory and finally copied a paragraph on a digitizer that is part of the Computerized Penmanship Evaluation Tool (ComPET) system. Results indicated that certain computerized measures of the ROCFT copying significantly correlated with those of the paragraph-copying behavior (r = .38–.75). Significant differences were found between the spatial and temporal computerized measures of performance in the ROCFT copying and drawing-from-memory tasks. Stepwise regressions indicated that mean pressure predicted 12 % of the variance of the ROCFT and paragraph-copying quality scores and 6 % of the ROCFT drawing-from-memory score. Furthermore, 52 % of the variance of the ROCFT drawing-from-memory score was predicted by the mean velocity. The benefits and significance of obtaining computerized measures of the drawing process for better insight about human performance characteristics are discussed, and applications are suggested.

On finding optimal parameters of an oscillatory model of handwriting

In this paper, we show how optimization methods can be used efficiently to determine the parameters of an oscillatory model of handwriting. Because these methods have to be used in real-time applications, this involves that the optimization problems must be rapidely solved. Hence, we developed an original heuristic algorithm, named FHA. This code was validated by comparing it (accuracy/CPU-times) with a multistart method based on Trust Region Reflective algorithm.

A parsimonious oscillatory model of handwriting

We propose an oscillatory model that is theoretically parsimonious, empirically efficient and biologically plausible. Building on Hollerbach's (Biol Cybern 39:139-156, 1981) model, our Parsimonious Oscillatory Model of Handwriting (POMH) overcomes the latter's main shortcomings by making it possible to extract its parameters from the trace itself and by reinstating symmetry between the [Formula: see text] and [Formula: see text] coordinates. The benefit is a capacity to autonomously generate a smooth continuous trace that reproduces the dynamics of the handwriting movements through an extremely sparse model, whose efficiency matches that of other, more computationally expensive optimizing methods. Moreover, the model applies to 2D trajectories, irrespective of their shape, size, orientation and length. It is also independent of the endeffectors mobilized and of the writing direction.

Cognitive Mechanism of Handwritten Production

A handwriting model proposed by van Galen (1991) conceives handwriting as product of a series of hierarchically organized modules which operates in a parallel way. Based on this model, three issues in this domain from a psychological viewpoint are summarized: first, whether or not the handwritten production is mediated by phonological information? Second, what is the relevant functional unit in handwriting process? Third, what is the relationship between cognitive encoding and motor-executive processing in handwriting? On the other hand, the issues on the relationships between speaking and writing, and reading and writing are addressed, because they are inter-related language processes. Furthermore, the neural correlates and related debates of handwriting are concluded. Finally, in order to address the unique contributions of Chinese to handwritten production theory, the unique properties of Chinese are described in detail, and the relevant implications are discussed.

Confidence- and margin-based MMI/MPE discriminative training for off-line handwriting recognition

We present a novel confidence- and margin-based discriminative training approach for model adaptation of a hidden Markov model (HMM)-based handwriting recognition system to handle different handwriting styles and their variations. Most current approaches are maximum-likelihood (ML) trained HMM systems and try to adapt their models to different writing styles using writer adaptive training, unsupervised clustering, or additional writer-specific data. Here, discriminative training based on the maximum mutual information (MMI) and minimum phone error (MPE) criteria are used to train writer-independent handwriting models. For model adaptation during decoding, an unsupervised confidence-based discriminative training on a word and frame level within a two-pass decoding process is proposed. The proposed methods are evaluated for closed-vocabulary isolated handwritten word recognition on the IFN/ENIT Arabic handwriting database, where the word error rate is decreased by 33% relative compared to a ML trained baseline system. On the large-vocabulary line recognition task of the IAM English handwriting database, the word error rate is decreased by 25% relative.

A statistical handwriting model for style-preserving and variable character synthesis

Synthesizing handwritten-style characters is an interesting issue in today’s handwriting analysis field. The purpose of this study is to artificially generate training data, foster a deep understanding of human handwriting, and promote the use of the handwritten-style computer fonts, in which the individuality or variety of the synthesized characters is considered important. Research considering such two properties together, however, is very rare. In this paper, a handwriting model is proposed to synthesize various handwritten characters while preserving the writer’s individuality from a limited number of training data, using a statistical approach. The proposed model is verified in single- and multiple-stroke characters, such as Arabic numbers, small English letters, and Japanese Kanji letters. Synthesized characters are evaluated in three ways. First, they are analyzed visually using the selected samples, and the relationship between the training and synthesized characters is explained. Second, the personalities and varieties of all the data are evaluated using a conventional writer verification method. Third, a questionnaire is developed and administered to evaluate the subjective responses of the users regarding the personal styles of the synthesized characters. The results prove that the proposed model stably synthesizes personalized characters by being invariant to the number of training data, whereas the variety increases gradually as the data increase.

A quantitative evaluation of the AVITEWRITE model of handwriting learning

Much sensory-motor behavior develops through imitation, as during the learning of handwriting by children. Such complex sequential acts are broken down into distinct motor control synergies, or muscle groups, whose activities overlap in time to generate continuous, curved movements that obey an inverse relation between curvature and speed. The adaptive vector integration to endpoint handwriting (AVITEWRITE) model of Grossberg and Paine (2000) [A neural model of corticocerebellar interactions during attentive imitation and predictive learning of sequential handwriting movements. Neural Networks, 13, 999–1046] addressed how such complex movements may be learned through attentive imitation. The model suggested how parietal and motor cortical mechanisms, such as difference vector encoding, interact with adaptively-timed, predictive cerebellar learning during movement imitation and predictive performance. Key psychophysical and neural data about learning to make curved movements were simulated, including a decrease in writing time as learning progresses; generation of unimodal, bell-shaped velocity profiles for each movement synergy; size scaling with isochrony, and speed scaling with preservation of the letter shape and the shapes of the velocity profiles; an inverse relation between curvature and tangential velocity; and a two-thirds power law relation between angular velocity and curvature. However, the model learned from letter trajectories of only one subject, and only qualitative kinematic comparisons were made with previously published human data. The present work describes a quantitative test of AVITEWRITE through direct comparison of a corpus of human handwriting data with the model’s performance when it learns by tracing the human trajectories. The results show that model performance was variable across the subjects, with an average correlation between the model and human data of 0.89±0.10. The present data from simulations using the AVITEWRITE model highlight some of its strengths while focusing attention on areas, such as novel shape learning in children, where all models of handwriting and the learning of other complex sensory-motor skills would benefit from further research.

From Parallel Sequence Representations to Calligraphic Control: A Conspiracy of Neural Circuits

Calligraphic writing presents many challenges for motor control, including: learning and recall of stroke sequences; critical timing of stroke onsets and durations; fine control of grip and contact forces; and letterform invariance under size scaling, which entails fine control of stroke directions and amplitudes during recruitment and derecruitment of musculoskeletal degrees of freedom. Experimental and computational studies in behavioral neuroscience have progressed toward explaining the learning, planning, and control exercised in tasks that share features with calligraphic writing and drawing. This article highlights component operations ranging from parallel sequence representations to fine force control. Treated in succession are: competitive queuing models of sequence representation, performance, learning, and recall; letter size scaling and motor equivalence; cursive handwriting models in which sensory-motor transformations are performed by circuits that learn inverse differential kinematic mappings; and fine-grained control of timing and transient forces by circuit models that learn to solve inverse dynamics problems.

Advances in the study of drawing and handwriting

Since the early 1980s the study of drawing and handwriting movements has come to be known as the field of graphonomics. The term graphonomics intends to capture the multi-disciplinary scientific effort involved in identifying lawful relationships between the planning and generation of drawing and handwriting movements, and defining the nature and limitations of the processes taking place at various levels of the neuromotor system as these movements evolve. In this guest editorial, an overview is presented of computational handwriting models. In addition, the research issues that are addressed in five neurocognitive studies of drawing and handwriting published in this special journal issue following the tenth Biannial Conference of the International Graphonomics Society, are introduced.

The stability of pen–joint and interjoint coordination in loop writing

This study is concerned with pen–joint and interjoint coordination in handwriting. In particular, it focuses on the stability of coordination as a means to find the locus of coordination control in the arm–pen effector system. Twelve subjects generated loop sequences of varying length at various positions on the baseline of writing. Joint excursions and pen-tip displacements were recorded by means of a 3D-motion tracking system. The coordination stability of 15 pairs of 6 mechanical degrees of freedom (d.f.s) of the arm–pen effector system was investigated by means of relative phase analyses. Pen–joint coordination between horizontal pen-tip displacements and wrist excursions was found to be most stable; that between vertical pen-tip displacements and finger excursions was considerably less stable. Interjoint coordination was generally less stable than pen–joint coordination, and most stable between the wrist and the elbow. Sequence length and its position on the line differentially affected the coordination stability of the d.f. combinations. The results are discussed in relation to assumptions about joint coordination in writing as expressed by computational handwriting models.

On-line character recognition using handwriting modelling

An oscillatory model of handwriting, previously used for synthesis, is modified. The modified model better suits the dynamics of handwriting. It also provides parameters which can be used for on-line handwriting recognition. A user-dependent system, operating on segmented handwritten lower case characters yields a recognition performance exceeding 90% for cooperative users.

Automatic reading of the literal amount of bank checks

The studies presented in this paper deal with the global recognition of a restricted variety of handwritten words comprising the vocabulary used to write French bank checks. Several tools have been developed within the constraints of this application, tools that relate to the general problem of off-line cursive handwriting recognition. The first difficulty when one wants to read a text is the location of words. This is done by using a top-down analysis that first locates lines of text before segmenting them into individual words. A structural model of cursive handwriting, built on the median axis of the word, is proposed. This constitutes an alternative to the use of the conventional, letter-by-letter analytical model and could be used for other problems involving cursive word recognition. According to this structural model of cursive handwriting, an edit distance is computed between the extracted structural description and the reference descriptions that are interpreted as grapheme strings. This provides an ordered list of candidates for each individual word. Feature extraction in the binary image of the word is performed using a specific line-following algorithm. Since it is possible to express the syntax of the sentences by a finite grammar, this information is used to discard the inconsistent sentences from the possible ones. These various algorithms have been tested on personal data, as well as on real check images.

Cluster network for recognition of handwritten, cursive script characters

A neural net structure was developed to recognize isolated cursive script characters. Characters were composed of features suggested by an established model of handwriting. The model assumes that handwritten characters are formed from a limited number of primitive hand motions characterized by the X-Y oscillations in the vertical ( Y) and horizontal ( X) directions. Consistent with this model, representation for handwritten characters contain features that are invariant under rotation, translation, changes in size and/or style. Besides their applicability to categorization of handwritten characters, these features are also assumed to control neural “motor” activities during writing. The initial application of a single backward error propagation network to handwritten script recognition resulted in slow learning and was limited to approximately 50% correct classification. Changing the architecture to a cluster of smaller networks and adding some simple heuristics improved performance dramatically. In its current form, the approach yields recognition up to 80% for our test data.