Showing papers on "Intelligent word recognition published in 1992"

Computer recognition of unconstrained handwritten numerals

Abstract: Four independently, developed expert algorithms for recognizing unconstrained handwritten numerals are presented. All have high recognition rates. Different experimental approaches for incorporating these recognition methods into a more powerful system are also presented. The resulting multiple-expert system proves that the consensus of these methods tends to compensate for individual weaknesses, while preserving individual strengths. It is shown that it is possible to reduce the substitution rate to a desired level while maintaining a fairly high recognition rate in the classification of totally unconstrained handwritten ZIP code numerals. If reliability is of the utmost importance, substitutions can be avoided completely (reliability=100%) while retaining a recognition rate above 90%. Results are compared with those for some of the most effective numeral recognition systems found in the literature. >

Speech recognition system having word-based and phoneme-based recognition means

Abstract: A speech recognition system includes a parameter extracting section for extracting a speech parameter of input speech, a first recognizing section for performing recognition processing by word-based matching, and a second recognizing section for performing word recognition by matching in units of word constituent elements. The first word recognizing section segments the speech parameter in units of words to extract a word speech pattern and performs word recognition by matching the word speech pattern with a predetermined word reference pattern. The second word recognizing section performs recognition in units of word constituent elements by using the extracted speech parameter and performs word recognition on the basis of candidates of an obtained word constituent element series. The speech recognition system further includes a recognition result output section for obtaining a recognition result on the basis of the word recognition results obtained by the first and second recognizing sections and outputting the obtained recognition result. The speech recognition system further includes a word reference pattern learning section for performing learning of a word reference pattern on the basis of the recognition result obtained by the recognizing result output section and the word speech pattern.

Online recognition of freely handwritten Japanese characters using directional feature densities

Abstract: The authors propose an online handwritten Japanese character recognition method permitting both stroke number and stroke order variations. The method is based on the pattern matching technique. Matching is done by the multiple similarity method using directional feature densities, which are independent of both stroke number and stroke order. This method has achieved a good recognition rate, 91%, for 2965 freely written Japanese kanji characters. >

Hidden Markov models for character recognition

Abstract: A hierarchical system for character recognition with hidden Markov model knowledge sources which solve both the context sensitivity problem and the character instantiation problem is presented. The system achieves 97-99% accuracy using a two-level architecture and has been implemented using a systolic array, thus permitting real-time (1 ms per character) multifont and multisize printed character recognition as well as handwriting recognition. >

Character segmentation techniques for handwritten text-a survey

Abstract: The paper is a survey of techniques for segmenting images of handwritten text into individual characters. The topic is broken into two categories: segmentation and segmentation-recognition techniques. Several approaches to each are outlined, and each is analyzed for its relevance to printed, cursive, on-line and off-line input data. >

Analysis and recognition of alphanumeric handprints by parts

Abstract: An advanced hierarchical model has been proposed to produce a more effective character recognizer based on the probability of occurrence of the patterns. New definitions such as crucial parts, efficiency ratios, degree of confusion, similar character pairs, etc. have also been given to facilitate pattern analysis and character recognition. Using these definitions, computer algorithms have been developed to recognize the characters by parts, including halves, quarters, and sixths. The recognition rates have been analyzed and compared with those obtained from subjective experiments. Based on the results of both computer and human experiments, a detailed analysis of the crucial parts and the Canadian standard alphanumeric character set has been made revealing some interesting fundamental characteristics of these handprint models. The results should be useful for pattern analysis and recognition, character understanding, handwriting education, and human-computer communication. >

SARAT-a system for the recognition of Arabic printed text

Abstract: Presents a new system for the automatic recognition of grabic printed text. The system is still under development. Here the concept of the so called SARAT-system is presented together with some very promising first results. The main features of the system are the segmentation into single characters through recognition, contour based features, statistical distance classification, and a word module. >

Online handwritten character recognition

Abstract: An online handwritten character recognition system which performs the narrowing of candidates for handwritten character recognition quickly and very accurately by simple processing of a small amount of operations. A predetermined characteristic amount is extracted for each stroke, a characteristic amount word is created having a binary value of 1 only in one or more bit positions corresponding to selected values of the characteristic amount, an AND operation is performed bit-by-bit between the reference word of the corresponding stroke of the character of interest, and it is determined if all the bits of the results of the AND operation are zero. If the number of binary values of the results of the zero-determining operation for all the strokes of the character of interest exceeds a threshold, it is judged to be a candidate.

Off-line handwritten word recognition (HWR) using a single contextual hidden Markov model

Abstract: A complete scheme for totally unconstrained handwritten word recognition based on a single contextual hidden Markov model (HMM) is proposed. The scheme includes a morphology- and heuristics-based segmentation algorithm and a modified Viterbi algorithm that searches the (l+1)st globally best path based on the previous l best paths. The results of detailed experiments for which the overall recognition rate is up to 89.4% are reported. >

DRS: a workstation-based document recognition system for text entry

Abstract: Document recognition system (DRS), a workstation-based prototype document analysis system that uses optical character recognition (OCR), is described. The system provides functions for image capture, block segmentation, page structure analysis, and character recognition with contextual postprocessing, as well as a user interface for error correction. All the functions except image capture and character recognition have been implemented by means of software for the Japanese edition of OS/2. >

Handwritten word recognition using HMM with adaptive length Viterbi algorithm

Abstract: The authors have developed a handwritten word recognition scheme based on a single contextual, discrete symbol probability hidden Markov model (HMM) incorporated with an adaptive length Viterbi algorithm. This work attempts to extend the earlier HMM scheme for naturally segmented word recognition to cursive and nonsegmented word recognition. The algorithm presegments the script into characters and/or fractions of characters, dynamically selects the correct segmentation points, determines the word length, and recognizes the word according to the maximum path probability. The HMM is on top of, but independent of, script segmentation and character recognition techniques, and therefore leaves room for further improvement. The experiments have shown promising results and directions for further improvement. >

Shape description by UNL Fourier features-an application to handwritten character recognition

Abstract: Handwritten characters are forms. Thus they are accessible to form description methods. The authors present a universal shape characterization method applied to the domain of off-line handwritten characters. The test universe is restricted to the ten Arabic numerals but can be extended to any other characters. The authors point out previous work in the area of 2D shape classification and handwriting recognition. The theoretical concept of the authors' shape descriptor is introduced together with a recipe to apply it in the real case of discrete binary images. Results of classification experiments are presented. >

Massively parallel implementation of character recognition systems

Abstract: A massively parallel character recognition system has been implemented. The system is designed to study the feasibility of the recognition of handprinted text in a loosely constrained environment. The NIST handprint database, NIST Special Database 1, is used to provide test data for the recognition system. The system consists of eight functional components. The loading of the image into the system and storing the recognition results from the system are I/O components. In between are components responsible for image processing and recognition. The first image processing component is responsible for image correction for scale and rotation, data field isolation, and character data location within each field; the second performs character segmentation; and the third does character normalization. Three recognition components are responsible for feature extraction and character reconstruction, neural network-based character recognition, and low-confidence classification rejection. The image processing to load and isolate 34 fields on a scientific workstation takes 900 seconds. The same processing takes only 11 seconds using a massively parallel array processor. The image processing components, including the time to load the image data, use 94 of the system time. The segmentation time is 15 ms/character and segmentation accuracy is 89 for handprinted digits and alphas. Character recognition accuracy for medium quality machine print is 99.8. On handprinted digits, the recognition accuracy is 96 and recognition speeds of 10,100 characters/second can be realized. The limiting factor in the recognition portion of the system is feature extraction, which occurs at 806 characters/second. Through the use of a massively parallel machine and neural recognition algorithms, significant improvements in both accuracy and speed have been achieved, making this technology effective as a replacement for key data entry in existing data capture systems.© (1992) COPYRIGHT SPIE--The International Society for Optical Engineering. Downloading of the abstract is permitted for personal use only.

World image matching as a technique for degraded text recognition

Abstract: A technique is presented that determines equivalences between word images in a passage of text. A clustering procedure is applied to group visually similar words. Initial hypotheses for the identities of words are then generated by matching the word groups to language statistics that predict the frequency at which certain words will occur. This is followed by a recognition step that assigns identifications to the images in the clusters. This paper concentrates on the clustering algorithm. A clustering technique is presented and its performance on a running text of 1062 word images is determined. It is shown that the clustering algorithm can correctly locate groups of short function words with better than a 95 percent correct rate. >

Offline recognition of handwritten cursive words

Abstract: A robust algorithm for offline cursive script recognition is described. The algorithm uses a generate-and-test paradigm to analyze cursive word images. The generate phase of the algorithm intelligently segments the word after analyzing certain structural features present in the word. The test phase determines the most likely character candidates among the segmentation points by using a recognition algorithm trained on generalized cursive letter shapes. In a sense, word recognition is done by sliding a variable sized window across the word looking for recognizable characters and strokes. The output of this system is a list of all plausible interpretations of the word. This list is then analyzed by a two-step contextual post- processor which first matches all of the interpretations to a supplied dictionary using a string matching algorithm. This eliminates the least likely interpretations. The remaining candidates are then analyzed for certain character spatial relationships (local reference line finder) to finally rank the dictionary. The system has the advantage of not requiring explicit word training yet is able to recognize many handwriting styles. This system is being successfully tested on a database of handwritten words extracted from live mail with dictionary sizes of up to 300 words. Planned extensions include developing a multilevel generate-and-test paradigm which can handle any type of handwritten word.

Recognition of handwritten digits by image processing and neural network

Abstract: Recognition of handwritten digits has been one of the first applications of neural networks. The authors propose an intermediate approach between classical methods, which are based on extraction of a small set of parameters, and pure neural methods, in which the neural network is fed with raw image data. Complexity and learning time are reduced with still good performance. Experimental results and comparisons of various parameters and classifiers, for a database of 2589 digits obtained from 30 persons are provided. >

An A* algorithm for very large vocabulary continuous speech recognition

Abstract: We present a new search algorithm for very large vocabulary continuous speech recognition. Continuous speech recognition with this algorithm is only about 10 times more computationally expensive than isolated word recognition. We report preliminary recognition results obtained by testing our recognizer on "books on tape" using a 60,000 word dictionary.

Cursive script online character recognition with a recurrent neural network model

Abstract: A study was conducted to assess the performance of a discrete-time recurrent neural network in cursive script character recognition. The pen coordinates were sampled at discrete times and sequentially entered on two separate channels to a bank of neural-network-based recognizers, each trained to recognize one specific character. The recognizers' outputs were collected and reconverted into a string of characters, with associated probabilities. This method was tried on a restricted alphabet of six letters. The results of the study are presented, and its extension to more complex situations is discussed. >

Temporal clues in handwriting

Abstract: Handwritten character recognition is typically classified as online or offline depending on the nature of the input data. Online data consists of a temporal sequence of instrument positions while offline data is in the form of a 2D image of the writing sample. Online recognition techniques have been relatively successful but have the disadvantage of requiring the data to be gathered during the writing process. This paper presents work on the extraction of temporal information from static images of handwriting and its implications for character recognition. >

A handwritten Arabic character recognition technique for machine reader

Abstract: A technique for real-time recognition of unconstrained Arabic characters is presented. The proposed technique does not require any constraints of the character forms other than limiting them to a reasonable size and orientation. Structural features, which are more suitable for handwritten character recognition, are selected. Structural features that are independent of the writer style, which are called stable features, use a list of integer values (vector) to describe the character. On the other hand CHAIN CODE is used for other structural features (decisive) that are suitable for more variation of the writer style. A suitable clustering technique is chosen to accomplish the classifier procedure. The algorithm can be extended to cursive words after introducing the additional segmentation stage. >

The efficiency of holistic template matching in the recognition of unconstrained handwritten digits

Abstract: Psychological evidence suggests that simple visual patterns can be recognized by the use of internal representations as holistic templates, but the efficiency of holistic template matching in the recognition of real-life patterns, such as handwritten characters, has been doubted. To clarify this issue, we measured the efficiency of holistic template matching in machine recognition of totally unconstrained handwritten digits. Our learning and recognition algorithm was simple; no previous knowledge of handwritten digits was presupposed, and preprocessing was limited to Gaussian smoothing and normalization with respect to position, size, and orientation. For patterns presented in a known orientation, recognition rates were .69, .77, and .88, respectively, when about 5, 10, or 50 templates had been learned for each type of digit. For patterns presented in unknown orientations, recognition rates were slightly lower. High levels of reliability could be attained by the discounting of classifications based on weak evidence. Apparently, in high reliability recognition, holistic template matching can be used as a first operation by which recognition is achieved for most of the handwritten digits that are seen in real life.

A segmentation method for handwritten Japanese character lines based on transitional information

Abstract: The authors propose a method to segment a character line image into individual character images and recognize them. To segment Japanese handwriting accurately, it is necessary to use character recognition results and contexts. However, recognition results might be wrong, or recognition confidence scores might be inaccurate. Dictionary consulting is not sufficient to deal with such ambiguous character recognition results. The paper reports on a method which hierarchically uses transitional information and a word dictionary for recognition results for all possible characters. Experimental results show that, for character line samples written roughly, 91.7% recognition rate is achieved while the recognition rate for a method without transitional information is 78.3%. >

A hypothesis testing approach to word recognition using dynamic feature selection

Abstract: A top-down approach to word recognition is proposed. Discussions are presented on dynamically selecting the most effective feature combinations, which are applied to discriminate between a limited set of word hypotheses. >

Integration of a structural features-based preclassifier and a man-machine interactive classifier for a fast multi-stroke character recognition

Abstract: A transputer-based parallel machine for handwritten character recognition is proposed. An algorithm based on structural features and on a tree classifier was used to accomplish the pre-classification of the unknown sample in order to speed up the recognition process. The algorithm for the final classification is based on the description of the strokes through Fourier descriptors. The learning phase is accomplished through a man-machine interactive process. The proposed system can expand its knowledge base. A special representation of this knowledge base is proposed in order to record a great amount of data in a suitable way. A fast multistroke handwritten isolated character recognition system is presented. The test of this system was performed on a PC based prototype while the realization of a parallel transputer based working machine is in progress. Experimental results obtained applying these machines to handwritten numerals recognition are reported. >

A segmentation-free approach to OCR

Abstract: When confronted with degraded images of text, however, the performance of many commercial OCR systems deteriorates severely. This occurs because all these systems rely on a segmentation step that is prone to error in the presence of image noise and printing artifacts. The authors present a novel OCR approach that overcomes this problem by eliminating the segmentation step altogether. This approach is based on the concept and techniques of occluded object recognition. To achieve high efficiency as well as robustness, they incorporate the notions of indexing and voting, and tailor them to the problem of OCR. Preliminary experimental results are given. >