On-line computer recognition of proposed standard ANSI(USASI) handprinted characters

A robust real-time system for recognition of handprinted characters of the upper case English alphabet is described. The basic system is suited to implementation on small computers and has been designed to accept characters conforming to the stroke types and sequences suggested by a proposed ANSI(USASI) standard. Experiments with 2340 samples from 10 untrained subjects yielded an overall character recognition accuracy of 98.3%. The system is quite robust with respect to size and stylistic variations. The robustness and real-time operation of the system are largely attributed to the preprocessing and stroke identification techniques developed, which include a new two-stage syntactic classifier for the identification of curvilinear strokes.     

On-line recognition of handprinted Kanji characters

Difficulties in Kanji (Chinese character) recognition stem from its large character set (about 5000 characters) and the large number of strokes (up to about sixty) in each character.

The paper describes a preliminary approach to this Kanji recognition problem. In the present method, a handprinted Kanji character is coded into a symbol string using the binary relation between stroke and reference zone. Two symbol string recognition methods are proposed and investigated; the direct matching recognition (DMR) method and the unit structure recognition (USR) method.

The DMR method worked efficiently for characters which have up to five strokes. The USR method represents Kanji characters with a structural unit combination. This method worked efficiently for multi-stroke characters and greatly reduced dictionary update labor.     

Recognition of handprinted Hebrew characters using features selected in the Hough transform space

Automatic identification of handprinted Hebrew characters is described in this paper. The recognition model devised constitutes a multi-stage system. In the first stage a coarse classifier allocates the input patterns into one of 17 categories, based on the number and the location of end points within predetermined regions in the characters matrix. The second stage uses features extracted in the Hough transform space to classify characters assigned to each of 16 categories. The remaining one category, composed of similar, square-like (rotated L shape) classes, is recognized by structural analysis and a statistical classifier. An additional step of postprocessing is added to compensate for the sensitivity of the Hough transform to the existence of similar classes within some of the categories. Experiments were conducted with a multi-author (40 writers) data base. An average recognition rate of 86.9% was observed for the system. This compared favorably with the results of two other recognition methods.     

Preprocessing for Chinese character recognition and global classification of handwritten Chinese characters

Techniques for calculating the stroke directions of thinned binary characters and for detecting the intersections and end points of strokes by means of pattern matching and weighting method are proposed as a preprocessing of handwritten Chinese character recognition. We also propose a method for global classification of handwritten Chinese characters by means of projection profiles of strokes and show that the method is available for the Chinese characters written in the square style.     

Classification of handprinted Kanji characters by the structured segment matching method

A method for handprinted Kanji character classificatin is proposed in this paper. After extraction of directional line segments and partitioning of the character frame area, a feature vector that represents the distribution of strokes is generated and is matched with average vectors in a dictionary.     

Classification of characters by man and by machine

In this paper the performance of a nearly optimal system for character recognition is compared to human performance on the same data set. The recognition system uses a linear feature extraction method which is superior to discrete Karhunen-Loeve expansion. The experiments consider binary and multiple classification of handprinted characters, binary classification of similar characters of one font corrupted by additive white normal noise, and multiple classification of truncated handprinted characters. It turns out that the human visual system is superior in recognizing handprinted characters and inferior in the case of single font characters with additive noise.     

CHITRA: Cognitive handprinted input-trained recursively analyzing system for recognition of alphanumeric characters

A novel system for recognition of handprinted alphanumeric characters has been developed and tested. The system can be employed for recognition of either the alphabet or the numeral by contextually switching on to the corresponding branch of the recognition algorithm. The two major components of the system are the multistage feature extractor and the decision logic tree-type catagorizer. The importance of good features over sophistication in the classification procedures was recognized, and the feature extractor is designed to extract features based on a variety of topological, morphological and similar properties. An information feedback path is provided between the decision logic and the feature extractor units to facilitate an interleaved or recursive mode of operation. This ensures that only those features essential to the recognition of a particular sample are extracted each time. Test implementation has demonstrated the reliability of the system in recognizing a variety of handprinted alphanumeric characters with close to 100% accuracy.Most of this work was carried out at the School of Automation, Indian Institute of Science, Bangalore, India.     

Position-width-pulse algorithm for recognition of handwritten characters

An algorithm for the recognition of handwritten characters based on the position-width-pulse method of recognition of curves is presented in the article. An algorithm for transformation of characters into curves is presented and a recognition procedure described. Numerical estimators of the proximity S _w of curves that graphically map the image of characters to be recognized relative to printed characters used as reference characters are calculated. A conclusion that assigns a recognized character to a corresponding reference character is arrived at on the basis of the minimum value of S _w with specified reliability.     

An application of the hough transform to the recognition of printed hebrew characters

This paper describes computer recognition of printed Hebrew characters, in which the Hough Transform space is used as the feature domain. The sequentially built recognition scheme has a first stage coarse classification by end points of strokes and a second stage classifier divided into two parts: in one part the decision is based on matching features in the transform space and in the other part a structural classification is applied. The method was tested on a sample set of eighteen print-simulated alphabets and a recognition rate of 99.6% was achieved. A performance comparison with two other recognition methods is also given.     

Modified region decomposition method and optimal depth decision tree in the recognition of non-uniform sized characters – An experimentation with Kannada characters

In contrast to English alphabets, some characters in Indian languages such as Kannada, Hindi, Telugu may have either horizontal or vertical or both the extensions making it difficult to enclose every such character in a standard rectangular grid as done quite often in character recognition research. In this work, an improved method is proposed for the recognition of such characters (especially Kannada characters), which can have spread in vertical and horizontal directions. The method uses a standard sized rectangle which can circumscribe standard sized characters. This rectangle can be interpreted as a two-dimensional, 3×3 structure of nine parts which we define as bricks. This structure is also interpreted as consecutively placed three row structures of three bricks each or adjacently placed three column structures of three bricks each.

It is obvious that non-uniform sized characters cannot be contained within the standard rectangle of nine bricks. The work presented here proposes to take up such cases. If the character has horizontal extension, then the rectangle is extended horizontally by adding one column structure of three bricks at a time, until the character is encapsulated. Likewise, for vertically extended characters, one row structure is added at a time. For the characters which are smaller than the standard rectangle, one column structure is removed at a time till the character fits in the shrunk rectangle. Thus, the character is enclosed in a rectangular structure of m×n bricks where m3 and n1. The recognition is carried out intelligently by examining certain selected bricks only instead of all mn bricks. The recognition is done based on an optimal depth logical decision tree developed during the Learning phase and does not require any mathematical computation.     

Evaluation of incremental learning algorithms for HMM in the recognition of alphanumeric characters

We present an evaluation of incremental learning algorithms for the estimation of hidden Markov model (HMM) parameters. The main goal is to investigate incremental learning algorithms that can provide as good performances as traditional batch learning techniques, but incorporating the advantages of incremental learning for designing complex pattern recognition systems. Experiments on handwritten characters have shown that a proposed variant of the ensemble training algorithm, employing ensembles of HMMs, can lead to very promising performances. Furthermore, the use of a validation dataset demonstrated that it is possible to reach better performances than the ones presented by batch learning.