Showing papers on "Handwriting recognition published in 1990"

The state of the art in online handwriting recognition

Abstract: This survey describes the state of the art of online handwriting recognition during a period of renewed activity in the field. It is based on an extensive review of the literature, including journal articles, conference proceedings, and patents. Online versus offline recognition, digitizer technology, and handwriting properties and recognition problems are discussed. Shape recognition algorithms, preprocessing and postprocessing techniques, experimental systems, and commercial products are examined. >

Handwritten zip code recognition with multilayer networks

Abstract: An application of back-propagation networks to handwritten zip code recognition is presented. Minimal preprocessing of the data is required, but the architecture of the network is highly constrained and specifically designed for the task. The input of the network consists of size-normalized images of isolated digits. The performance on zip code digits provided by the US Postal Service is 92% recognition, 1% substitution, and 7% rejects. Structured neural networks can be viewed as statistical methods with structure which bridge the gap between purely statistical and purely structural methods. >

Complementary algorithms for the recognition of totally unconstrained handwritten numerals

Abstract: Two novel methods for recognizing totally unconstrained handwritten numerals are presented. One classifies samples based on structural features extracted from their skeletons; the other makes use of their contours. Both methods achieve high recognition rates (86.05%, 93.90%) and low substitution rates (2.25%, 1.60%). To take advantage of the inherent complementarity of the two methods, different ways of combining them are studied. It is shown that it is possible to reduce the substitution rate to 0.70%, while the recognition rate remains as high as 92.00% . Furthermore, if reliability is of utmost importance, one can avoid substitutions completely (reliability 100%) and still retain a fairly high recognition rate (84.85%). >

Handwriting recognition employing pairwise discriminant measures

Abstract: In a handwriting recognition system a method for differentiating a first character from a second character having similar characteristics. The method includes the steps of determining during a training session at least one pairwise discriminant measure associated with the first character and with the second character. The step of determining includes a step of evaluating with a potential discriminant measure a plurality of prototype first characters and a plurality of prototype second characters. Subsequently, during a handwriting recognition session, the method evaluates with at least one of the previously determined discriminant measures selected from a set of same an input character identified as being either the first prototype character or the second prototype character. The step of evaluating includes a step of analyzing a stroke or strokes associated with the input character in accordance with one or more of the pairwise discriminant measures selected from the set, summing the results of each discriminant measure analysis to obtain a result for each of the pair of characters associated with the pairwise discriminate measure, and selecting a character from the pair that has a maximum value result.

Multi-phase recognition of multifont photoscript Arabic text

Abstract: A new system for the recognition of a multifont photoscript Arabic text is introduced. The distinguishing feature of such text is that it is written cursively. This imposes an additional requirement of isolating each character or set of overlapping characters before recognition. The proposed system is composed of three interleaved phases. The segmentation phase attempts to produce an initial set of characters from the connected text according to a set of predefined rules. The output is then passed to a preliminary classification phase that attempts to label the unknown characters into one of ten possible classes according to a set of rules that acquire their parameter values through learning. The last phase contains a more elaborate set of rules that recognize characters within each class. This recognition phase is designed to allow errors during segmentation and/or classification to be rectified through an adaptive recognition technique. The system has been implemented and tested on several fonts with a recognition rate of 130 words/min and an error rate of less than 6%. >

Recognition and verification of postcodes in handwritten and hand-printed addresses

Abstract: An algorithmic architecture for a high-performance OCR system for hand-printed and handwritten addresses is proposed. The architecture integrates syntactic and contextual postprocessing with character recognition to optimize postcode recognition performance and verifies the postcode against simple features extracted from the remainder of the address to ensure a low error rate. An initial implementation of all parts of the proposed system is reported, showing an overall postcode recognition rate of 44% and correct extraction of verification information for 24% of upper-case addresses and 27% of mixed-case addresses. >

Online recognizer for runon handprinted characters

Abstract: A recognize-then-segment recognizer of unconstrained handprinting is described with uses a unified tablet-display to provide a paperlike computer interface. Whereas most handwriting recognition systems segment and then recognize, this one recognizes and then finds the best segmentation. It classifies strokes, generates character hypotheses, and verifies hypotheses to estimate the optimal character sequence for each word of runon handwritten characters. Linguistic constraints can limit the choices. The system is implemented on an IBM workstation, accepts runon characters written on a tablet, and performs recognition in real time. >

A Handwriting Recognition System Based on Properties of the Human Motor System

Abstract: The human reader of handwriting is unaware of the amount of back-ground knowledge that is constantly being used by a massive parallel computer, his brain, to decipher cursive script. Artificial cursive script recognizers do not have access to a comparable source of knowledge or of comparable computational power to perform top-down processing. Therefore, in an artificial script recognizer, there is a strong demand for reliable bottom-up processing. For the recognition of unrestricted script consisting of arbitrary character sequences, on-line recorded handwriting signals offer a more solid basis than the optically obtained grey-scale image of a written pen trace, because of the temporal information and the inherent vectorial description of shape. The enhanced bottom-up processing is based on implementing knowledge of the motor system in the handwriting recognition system. The bottom-up information will already be sufficient to recognize clearly written and unambiguous input. However, ambiguous shape sequences, such as m vs n.. or d vs cl, and sloppy stroke patterns still require top-down processing. The present paper discusses the handwriting recognition system as being developed at the NICI. The system contains six major modules: (1) On-line digitizing, pre-processing of the movements and segmentation into strokes. (2) Normalization of global handwriting parameters. (3) Extraction of motorically invariant, real-valued, feature values per stroke to form a multidimensional feature vector and subsequent feature vector quantization by a self-organizing two-dimensional Kohonen network. (4) Allograph construction, using a second network of transition probabilities between cell activation patterns of the Kohonen network. (5) Optional word hypothesization. (6) The system has to be trained by supervised learning, the user indicating prototypical stroke sequences and their symbolic interpretation (letter or N-gram naming).

An interactive system for the selection of handwritten numeral classes

Abstract: A method for recognizing handwritten Arabic numerals using Fourier descriptors is presented. The classification phase is based on a new similarity definition introduced from one of the Banach algebras of continuous and bounded plane curves. The learning phase is developed using a man-machine interactive system in which the role of man as character-source and the role of the machine as recognizer are partially interchangeable. >

A new distance measure for binary images

Abstract: A distance measure, called the generalized Euclidean distance, is developed for binary images to take into account perceptual distortions. Based on this distance measure, a type of transformation is devised to ensure that the generalized Euclidean distance of two images is the same as the Euclidean distance of two transformed images. A set of transformed images is then used to train and test a feed-forward neural network for handwritten numeral recognition. It is shown that the recognition rate is significantly improved by incorporating human perception into the neural network, and that the transformation step can be merged into the trained neural network so that no transformation is required during the recognition stage. >

Understanding Handwriting Recognition from the User's Perspective

Abstract: As an input technique, handwriting recognition offers benefits in ease of use, but poses special problems for the user when a recognition error occurs. When a recognition error occurs, the user is often surprised since the misrecognized character often looks acceptable to him/her. In contrast, when a typing error occurs with a keyboard interface, the user immediately understands what has happened. The purpose of this study was: 1. to gain insight into what people think when a recognition error occurs, and 2. to discover whether a simple monochrome display of a user's handwriting prototypes would provide information which could be used to improve recognition accuracy. Such a display might serve as a point of reference for understanding and avoiding recognition errors. The results of the study suggested that a display of handwriting prototypes can be used by people to improve recognition accuracy. The study also found that in a large percentage of instances, people do not have any insight into the cause of ...

An on-line handwriting-recognition system based on unreliable modules.

Abstract: In automatic recognition of unrestricted handwriting the ambiguities can be solved by top-down processing. However, automatic systems never have access to the extended background knowledge available to human readers. In order to replace this higher-level information we need to improve the reliability of the bottom-up processing. A handwriting-recognition system can be split up into six discrete blocks: (1) digitizing, word segmentation, pre-processing, and segmentation into strokes, (2) normalization of global handwriting parameters, (3) extraction of features per stroke, (4) allograph recognition, (5) optional word hypothesization, and, in order to allow recognition (6) a learning phase. The present paper discusses the design of three of these processing blocks: normalization, allograph recognition, and learning and brieey speciies feature extraction. Normalization concerns orientation, size, and slant. However, various alternative algorithms can be chosen and some algorithms yield more reliable results than others. A mechanism is proposed that will, sooner or later, nd the most appropriate normalization algorithms. Consequently, the features extracted from each stroke in the handwriting pattern will be more uniform within a writer and even between writers. In the recognition phase, handwriting patterns are segmented into allographs using an algorithm that can handle allographs with various numbers of strokes and with optional connection strokes between them. In order to teach the recognizer the allographs a method has been designed that builts non-interactively a lexicon of allographs by automatically discovering the allographs in a large corpus of cursive script.

On-line recognition of handwritten Chinese characters and alphabets

Abstract: An online handwritten-character recognition system is proposed based on stroke-sequence feature extraction and on a string-matching method called the dynamic neighboring matching method. The patterns to be recognized, including Chinese characters and alphabets, are allowed to be stroke-order- and stroke-number-free and tolerant of combined strokes and wrong strokes. Fewer learning samples are needed, but they must be within the constraint of normal handwriting. Reference patterns have been generated from 4000 categories of characters with stroke numbers ranging from 1 to 31. The recognition results are based upon the 3500 Chinese characters written by 10 people. The recognition rate is 93.5%, and the cumulative classification rate of choosing four most-similar characters is up to 98.5%. >

Optical character recognition: a technology driver for neural networks

Abstract: It is shown that a neural net can perform handwritten digit recognition with state-of-the-art accuracy. The solution required automatic learning and generalization from thousands of training examples and also required designing into the system considerable knowledge about the task-neither engineering nor learning from examples alone would have sufficed. The resulting network is well suited for implementation on workstations or PCs and can take advantage of digital signal processors (DSPs) or custom VLSI. >

Computer Processing of Handwriting

Abstract: Let's read! We will often find out this sentence everywhere. When still being a kid, mom used to order us to always read, so did the teacher. Some books are fully read in a week and we need the obligation to support reading. What about now? Do you still love reading? Is reading only for you who have obligation? Absolutely not! We here offer you a new book enPDFd computer processing of handwriting to read.

Handwritten character recognition with the Athena model

Abstract: The results of using the Athena neural network model in automatic recognition of handwritten English letters are presented. This model has several layers. The training patterns need to be presented once to each layer of neurons, in parallel. The model forms a binary tree of neurons, and the training is done from the root towards the leaves, processing each level in parallel. The learning time is shortened significantly compared to the network of D.E. Rumelhart et al. (1986). The net structure is dynamically decided during the learning process. The goal is to enhance the adaptive learning capability to recognize more and more patterns. In the approach presented, an incremental learning scheme based on the Athena neural net is introduced. The objective is to learn misclassified patterns adaptively by dynamically correcting the neural net topological structure and the individual neuron's weights and threshold. Incremental learning is performed without losing information about previously learned patterns. The experiments conducted on the recognition of handwritten English letters utilizing Athena and the incremental learning scheme show a significant recognition success rate. >

A mathematical modeling approach to size invariant Greek character recognition

Abstract: A new approach to the automatic machine recognition of Greek characters of any size and font is presented. Unlike most of the traditional methods, which are based on syntactic schemes, this method utilizes the topological characteristics of the characters to achieve recognition. Each character is individually modeled to a standard geometrical shape, and the least mean square method is used to form the best fit. Sample characters like alpha, beta, gamma, omega, theta, omicron, and rho have been successfully modeled with standard geometrical shapes, such as circles, lemniscates, cardioids, ellipses, and lines. An alternate approach to the automatic and reliable recognition of handwritten and printed Greek characters is also discussed. >