Approach For Character Recognition Research Articles

Handwritten character recognition using low resolutions

This paper deals with an experimental study into the recognition of handwritten characters using representations of low resolution. For this experiment several methodologies were used, such as an attributed graph OCR method, a back-propagation neural network (BNN), a recurrent neural network (RNN), and the Castor neural network (CNN). The attributed graph methodology is based on the representation (mapping) of the text characters onto a small size two-dimensional array of 12×9 cells. For the recognition process each character is considered as a composition of “main” and “secondary” features. The main features are the important parts of a character for its successful recognition. The secondary (or artistic) features are the parts of a character that contribute to its various representations. The attribute graph methodology presented in this paper attempts to prove that the recognition of a reduced-size character provides a robust approach for the recognition of handwritten text. The BNN approach is used here as a comparative recognition method, although it has some serious weaknesses at low resolution. The RNN approach for handwritten character recognition is based upon recurrent neural networks, which have a feedback mechanism. The feedback mechanism acts to integrate new values of feature vectors with their predecessors. The output is supervised according to a target function. These networks can deal with inputs and outputs that are explicit functions of time. A new way of associating shape information was used, which gives very consistent results for handwritten character recognition. In this scheme the “shadow” of each character was considered, to find the distances between the margins of the character. The distances are normalized with respect to the maximum distance in the entire shape to minimize the effect of disproportionally formed characters. In addition, the performance of Castor’s neural net was evaluated for the recognition of handwritten text characters, by using character data sets with various resolutions. For this effort the three neural networks and attributed graph approaches used a set of 5000 handwritten characters, and their results are compared.

On-Line Handwritten Chinese Character Recognition Directed by Components with Dynamic Templates

Structural matching has been recognized as a promising approach for on-line Chinese character recognition. In order to reduce its great computational complexity and improve its performance, people have been seeking for ways to direct the matching of a whole character by the result of partial matching. In this paper, the authors proposed 45 basic components for 3,755 categories of daily-used Chinese characters to direct the stroke segment matching of whole characters. Since they are always located at either the beginning or the end of the stroke segment string of characters, these components are easy to be extracted and separated from other parts of a character. Besides, in our approach, the reference templates of these components are extracted dynamically from the corresponding segment string of characters when a specific matching is carried out. This strategy avoids building multiple templates for the components of the same kind but at different places of characters. The experiments show that the segment matching computation has been reduced greatly without reducing the correctness of matching.

Ring-projection-wavelet-fractal signatures: a novel approach to feature extraction

In this brief, we present a novel approach to optical character recognition that utilizes ring-projection-wavelet-fractal signatures (RP-WFS). In particular, the proposed approach reduces the dimensionality of a 2-D pattern by way of a ring-projection method and, thereafter, performs Daubechies' wavelet transformation on the derived 1-D pattern to generate a set of wavelet transformation subpatterns, namely, curves that are nonself-intersecting. Further, from the resulting nonself-intersecting curves, the divider dimensions are readily computed. These divider dimensions constitute a new feature vector for the original 2-D pattern, defined over the curves' fractal dimensions. We have conducted several experiments in which a set of printed alphanumeric symbols of varying fonts and orientation were classified, based on the formulation of our new feature vector. The results obtained from these experiments have consistently shown the character recognition approach with the proposed feature vector can yield an excellent classification rate of 100%.

Handwritten Chinese character recognition by metasynthetic approach

Enlightened by the idea of metasynthesis, two integration approaches for handwritten Chinese character recognition are proposed in this paper. The first one is Integration based on a Linear Model and the second one is Network Integration based on Supervised Learning. Compared with previous integration approaches, the proposed methods succeed in automatically acquiring the parameters of the integrated systems by supervised learning which is very important for the large number of classes of pattern recognition problems. The experimental results show that the performances of the synthesized systems are much better than any of the individual classifiers.

On-Line Chinese Character Recognition Via a Representation of Spatial Relationships Between Strokes

Chinese characters are constructed by basic strokes based on structural rules. In handwritten characters, the shapes of the strokes may vary to some extent, but the spatial relations and geometric configurations of the strokes are usually maintained. Therefore these spatial relations and configurations could be regarded as invariant features and could be used in the recognition of handwritten Chinese characters. In this paper, we investigate the structural knowledge in Chinese characters and propose the stroke spatial relationship representation (SSRR) to describe Chinese characters. An On-Line Chinese Character Recognition (OLCCR) method using the SSRR is also presented. With SSRR, each character is processed and is represented by an attribute graph. The process of character recognition is thereby transformed into a graph matching problem. After careful analysis, the basic spatial relationship between strokes can be characterized into five classes. A bitwise representation is adopted in the design of the data structure to reduce storage requirements and to speed up character matching. The strategy of hierarchical search in the preclassification improves the recognition speed. Basically, the attribute graph model is a generalized character representation that provides a useful and convenient representation for newly added characters in an OLCCR system with automatic learning capability. The significance of the structural approach of character recognition using spatial relationships is analyzed and is proved by experiments. Realistic testing is provided to show the effectiveness of the proposed method.

Skew angle detection of digitized Indian script documents

Skew angle detection of scanned documents containing most popular Indian scripts (Devnagari and Bangla) is considered. Most characters in these scripts have horizontal lines at the top, called head lines. The character head lines mostly join one another in a word and the word appears as a single component. In the proposed method the components are at first labeled. The upper envelope of a component is found by columnwise scanning from an imaginary line above the component. Portions of upper envelope satisfying the properties of digital straight line are detected. They are clustered as belonging to single text lines. Estimates from individual clusters are combined to get the skew angle. Apart from accuracy and efficiency, an advantage of the method is that character segmentation and zone detection can be readily done from head line information, which is useful in optical character recognition approaches of these scripts.

Handwritten digit recognition using combined ID3-derived fuzzy rules and Markov chains

In this paper, we present an approach for handwritten digit recognition based on combined ID3-derived fuzzy rules and Markov chains. Both techniques use statistical models on the structural representation of digit images. Skeleton images are adopted to produce fuzzy rules based on the ID3 approach, and contour images are used for the Markov chain based approach. Decision trees produced from the ID3 algorithm are converted to a set of simplified rules which are then fuzzified into a set of fuzzy rules. To retain the classification performance, a two-layer perceptron is applied to optimize defuzzification parameters. On the other hand, a digit contour is traversed in a well-defined order and the Markov chain is used to perform sequential analysis and to match with models. The two classifiers are then combined to complement each other using a three-layer perceptron. The combined classifier achieves a good classification performance and at the same time it overcomes the difficulties in a conventional syntactic approach for handwritten character recognition, such as the scaling problem and lack of machine learning ability. Experimental results on NIST Special Database 3 show that the combined classifier has a significantly improved performance in terms of substitution versus rejection rates. After about 15% of digits that cannot be classified with high confidence by the combined classifier are re-classified by a nearest neighbor classifier using optimized prototypes, the overall classification rate can be as high as 98.6% without rejection.

Japanese language model based on bigrams and its application to on-line character recognition

This paper deals with a postprocessing method based on the n-gram approach for Japanese character recognition. In Japanese a small number of phonetic characters (Kana) and thousands of Kanji characters, which are ideographs, are used for describing ordinary sentences. In other words, Japanese sentences not only have a large character set, but also include characters with different entropies. It is therefore difficult to apply conventional methodologies based on n-grams to postprocessing in Japanese character recognition. In order to resolve the above two difficulties, we propose a method that uses parts of speech in the following ways. One is to reduce the number of Kanji characters by clustering them according to the parts of speech that each Kanji character is used in. Another is to increase the entropy of a Kana character by classifying it into more detailed subcategories with part-of-speech attributes. We applied a bigram approach based on these two techniques to a Japanese language model. Experiments yielded the following two results: (1) our language model resolved the imbalance between Kana and Kanji characters, and reduced the perplexity of Japanese to less than 100, when Japanese newspaper texts (containing a total of approximately three million characters) were used for the learning of our model, and (2) the postprocessing using the model for on-line character recognition rectified about half of all substitution errors when the correct characters were among the candidates.

A ROBUST STROKE EXTRACTION METHOD FOR HANDWRITTEN CHINESE CHARACTERS

The stroke analysis method is an effective approach for handwritten Chinese character recognition. But as we know, it is very difficult to accurately extract the strokes. In this paper, a robust stroke extraction method is proposed. First, smoothing and thinning processes are applied to smooth the shape and to obtain the skeleton of the observed character. Then the end point, internal point and fork point are detected by calculating their own crossing numbers while the corner points are determined by a knowledge-based iterative method. Virtual-end-points are introduced for separating a stroke into a certain number of line segments without losing the connection relations among them. By representing each line segment as a vertex and the connection relation of two segments as an edge, the observed character can be represented by an attributed graph. Finally, a stroke extraction procedure is proposed to extract the strokes from the global structures of the character. After each stroke of a character is extracted, the cross points can also be determined. Experimental results have shown that the proposed method is more effective than the other methods.3,5−6

Attributed string matching by split-and-merge for on-line Chinese character recognition

Consecutive strokes of Chinese characters tend to be connected in fast writing, and this causes a problem for most stroke-based recognition approaches. A recognition scheme for recognizing cursive Chinese characters under the constraint of correct stroke writing orders is proposed. The recognition scheme consists of two phases: candidate character selection and detailed matching. In the former phase, an input script with N strokes is used to split the strokes of each reference character into N corresponding parts. In the latter phase, the connected input strokes are broken into multiple strokes under the guidance of candidate characters. In both phases, dynamic programming is employed for stroke or character matching. Good experimental results prove the feasibility of the proposed approach for cursive Chinese character recognition.< <ETX xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">&gt;</ETX>

Arabic character recognition system: A statistical approach for recognizing cursive typewritten text

Character recognition systems can contribute tremendously to the advancement of automation process, and can improve the interface between man and machine (computers) in many applications, including office automation and data entry. In this report we present a recognition system for typed Arabic text, which involves a statistical approach for character recognition. This approach uses “Accumulative Invariant Moments” as an identifier, which helped in the segmentation of connected and overlapping Arabic characters. However, Invariant Moments proved to be very sensitive to slight changes in a character shape. These changes are normally due to typing and the scanning process, and cannot be avoided. The recognition zone was defined based on the mean and standard deviation for the moments of a large sample of each character. However, this zone was increased, using an empirical multiplier, to improve recognition rate. The system was implemented on a mainframe in APL programming language for ease of experimentation, and then transported to a PC environment in BASIC for better portability. The recognition rate achieved was 94%, with a recognition speed of 10.6 characters/minute, running on a PC/AT with a math co-processor.

Psychophysical techniques for investigating the distinctive features of letters

In an earlier paper ( Blesser, Shillman, Kuklinski, Cox, Eden & Ventura, 1974 ) a theoretical approach for character recognition based on human perception was proposed. The approach focuses on the study of ambiguous characters rather than on archetypes. In this paper three experimental techniques are described for studying ambiguous characters and for investigating the relationship between physical attributes and functional attributes. In particular, the mapping from a physical LEG (formed by a line extension) to a functional LEG is examined. This attribute is important in distinguishing Y from V, F from C, H from U and a variety of other letter pairs. Within a neutral graphical context the Physical to Functional Rule (PFR) was found to be constant, independent of letter label and to be a function only of measurable physical parameters of the stimuli; in a biased context (i.e. when the stimulus set is significantly shifted toward one character) the mapping changes. These results are expected to aid in the development of automatic character recognition algorithms since they are based on human perception.

A theoretical approach for character recognition based on phenomenological attributes

The importance of using a theory of characters in the designing of automatic computer character recognition algorithms for hand-printed letters is demonstrated. Ambiguously shaped characters or “difficult cases” constitute the basis for the theory. It is argued that a theory based on ambiguities, rather than on the classical archetypal shape of letters, leads to algorithms which will perform more accurately. Letters are described in terms of an abstract set of functional attributes, each of which can be related to a type of ambiguity between two letters. The relations between the functional attributes, which specify the letter's identity, and the physical attributes, which are derived from the physical image, are called graphical context rules. These rules can be determined from psychological experimentation.