Handwritten Character Recognition Using Bayesian Decision Theory

Handwritten part course credit Using Bayesian last TheoryAbstract calibre perception (CR) spate solve more complex problem in written shargon and make acknowledgement easier. Handwriting section fruition (HCR) has received extensive attention in academic and labor fields. The knowledge remains squirt be both on cables length or offline. Offline handwritten eccentric acknowledgment is the sub fields of optical feature intelligence (OCR). The offline handwritten typeface intelligence confronts be pre off bent grassing, class, feature extraction and credit rating. Our aim is to modify missing genius set of an offline character recognition apply Bayesian ending theory.Key quarrel Character recognition, Optical character recognition, Off-line Handwriting, breakdown, deliver extraction, Bayesian decision theory.IntroductionThe recognition frame erect be either on-line or off-line. On-line bridge player recognition involves the automatic innovation of school text edition as it is written on a special digitized or PDA, where a sensor picks up the pen-tip movements as well as pen-up/pen-down switching. That form of selective nurture is known as digital ink and can be regarded as a dynamic pattern of handwriting. Off-line handwriting recognition involves the automatic conversion of text in an catch into letter codes which argon usable within selective reading processor and text-processing applications. The selective information obtained by this form is regarded as a static pay offation of handwriting.The aim of character recognition is to transmute human readable character to machine readable character. Optical character recognition is a process of translation of human readable character to machine readable character in optic bothy scanned and digitized text. Handwritten character recognition (HCR) has received extensive attention in academic and production fields.Bayesian decision theory is a fundamental statistical memor y access that quantifies the tradeoffs between various decisions using probabilities and costs that accompany such decision.They divided the decision process into the following five moveIdentification of the problem.Obtaining inevitable info.Production of possible solution.Evaluation of such solution.Selection of a outline for performance.They as well include a sixth arcdegree implementation of the decision. In the equaling accession missing data cannot be recognition which is make social occasion ofable in recognition historical data. In our approach we atomic number 18 recognition the missing row using Bayesian classifier. It first classifier the missing words to obtain minimize flaw. It can recover as much flaw as possible.Related WorkThe history of CR can be traced as earlyish as 1900, when the Russian scientist Turing attempted to develop an c ar for the visually handicapped 1. The first character recognizers appe ard in the middle of the forties with the culture of digital computers. The early work on the automatic recognition of characters has been supportd either upon machine-printed text or upon a small scar of well-distinguished handwritten text or symbols. Machine-printed CR formations in this level in general use template twinned in which an come across is compared to a program library of views. For handwritten text, low-level image processing techniques have been used on the binary image to extract feature vectors, which are then cater to statistical classifiers. Successful, but constrained algorithms have been implemented mostly for Latin characters and numerals. However, some studies on Japanese, Chinese, Hebrew, Indian, Cyrillic, Greek, and Arabic characters and numerals in both machine-printed and handwritten cases were as well initiated 2.The commercial character recognizers were available in the 1950s, when electronic tablets capturing the x-y coordinate data of pen-tip movement was first introduced. This i nnovation enabled the researchers to work on the on-line handwriting recognition problem. A good source of references for on-line recognition until 1980 can be embed in 3.Studies up until 1980 suffered from the lack of powerful computer hardware and data acquisition devices. With the explosion of information technology, the previously veritable mannerologies found a very fertile environment for rapid yield addition to the statistical methods. The CR research was focused basically on the shape recognition techniques without using any semantic information. This led to an upper limit in the recognition rate, which was not satisfactory in umpteen practical applications. Historical review of CR research and development during this period can be found in 4 and 3 for off-line and on-line cases, reckonively.The solid progress on CR systems is achieved during this period, using the new development tools and methodologies, which are empowered by the continuously growing information technologies.In the early 1990s, image processing and pattern recognition techniques were efficiently combined with cardboard intelligence (AI) methodologies. Researchers developed complex CR algorithms, which receive high-resolution comment data and require extensive number crunching in the implementation phase. Nowadays, in addition to the more powerful computers and more accurate electronic equipments such as electronic scanners, cameras, and electronic tablets, we have efficient, modern use of methodologies such as neural networks (NNs), hidden Markov models (HMMs), fuzzy differentiate reasoning, and natural language processing. The recent systems for the machine-printed off-line 2 5 and limited vocabulary, user-dependent on-line handwritten characters 2 12 are quite tolerable for restricted applications. However, there is still a long way to go in order to r separately the ultimate goal of machine cloak of fluent human reading, especially for unconstrained on-line and of f-line handwriting.Bayesian decision Theory (BDT), single of the statistical techniques for pattern classification, to identify each of the large number of black-and- pureness rectangular pel displays as ane of the 26 capital letter in the English alphabet. The character images were base on 20 antithetical fonts and each letter within 20 fonts was randomly distorted to produce a file of 20,000 unique instances 6. exist constitutionIn this overview, character recognition (CR) is used as an comprehensive term, which covers all types of machine recognition of characters in various application domains. The overview serves as an update for the state-of-the-art in the CR field, emphasizing the methodologies required for the change magnitude needs in newly emerging areas, such as development of electronic libraries, multimedia databases, and systems which require handwriting data entry. The study investigates the attention of the CR research, analyzing the limitations of methodolo gies for the systems, which can be classified based upon two study criteria 1) the data acquisition process (on-line or off-line) and 2) the text type (machine-printed or handwritten). No matter in which class the problem belongs, in general, there are five major dos token1 in the CR problem1) Preprocessing2) naval division3) Feature Extraction4) Recognition5) Post processing3.1. PreprocessingThe raw data, depending on the data acquisition type, is subjected to a number of preliminary processing steps to make it usable in the descriptive formats of character analytic thinking. Preprocessing aims to produce data that are easy for the CR systems to operate accurately.The main tendencyives of preprocessing are1) hinderance decrement2) Normalization of the data3) Compression in the amount of information to be retained.In order to achieve the above objectives, the following techniques are used in the preprocessing stage.Preprocessing naval divisionSplits tidingssFeature Extract ionRecognitionPost processingFigure 1. Character recognition3.1.1 Noise diminutionThe kerfuffle, introduced by the optical scanning device or the writing instrument, causes confounded line segments, bumps and gaps in lines, filled loops, etc. The torturing, including local variations, rounding of corners, dilation, and erosion, is also a problem. Prior to the CR, it is prerequisite to eliminate these imperfections. Hundreds of available disagreement reduction techniques can be categorized in tether major groups 7 8a) Filteringb) morphological Operationsc) Noise Modeling3.1.2 NormalizationNormalization methods aim to transmit the variations of the writing and obtain standardized data. The following are the basic methods for normalization 4 1016.a) skew Normalization and Baseline Extractionb) Slant Normalizationc) surface Normalization3.1.3 CompressionIt is well known that classical image contraction techniques transform the image from the space domain to domains, which are not suited for recognition. Compression for CR requires space domain techniques for preserving the shape information.a) Threshold In order to reduce storage requirements and to increase processing speed, it is often sexually attractive to represent gray- master or color images as binary images by option a threshold lever. Two categories of threshold exist global and local. world- panoptic threshold picks one threshold place for the entire account image which is often based on an estimation of the background level from the zeal histogram of the image. Local (adaptive) threshold use different values for each pixel according to the local area information.b) slip While it provides a horrendous reduction in data size, thinning extracts the shape information of the characters. Thinning can be considered as conversion of off-line handwriting to intimately on-line like data, with spurious branches and artifacts. Two basic approaches for thinning are 1) pixel owlish and 2) apo theosis wise thinning 1. Pixel wise thinning methods locally and iteratively process the image until one pixel wide skeleton remains. They are very sensitive to noise and may strive the shape of the character. On the some other(prenominal) hand, the no pixel wise methods use some global information about the character during the thinning. They produce a real median or centerline of the pattern directly without examining all the individual pixels. In crowd-based thinning method defines the skeleton of character as the cluster centers. Some thinning algorithms identify the singular points of the characters, such as end points, cross points, and loops. These points are the source of problems. In a nonpareil wise thinning, they are handled with global approaches. A vision of pixel wise and nonpareil wise thinning approaches is available in 9.3.2. sectionThe preprocessing stage yields a clean document in the sense that a sufficient amount of shape information, high compression, and low noise on a normalized image is obtained. The next stage is segmenting the document into its subcomponents. Segmentation is an all important(predicate) stage because the extent one can reach in separation of words, lines, or characters directly affects the recognition rate of the script. in that location are two types of naval division external partition, which is the closing off of various writing units, such as paragraphs, sentences, or words, and internal naval division, which is the isolation of letters, especially in cursively written words.1) foreign Segmentation It is the most critical part of the document analysis, which is a necessary step prior to the off-line CR Although document analysis is a relatively different research area with its own methodologies and techniques, segmenting the document image into text and non text regions is an integral part of the OCR software. Therefore, one who works in the CR field should have a general overview for document analy sis techniques. Page layout analysis is accomplished in two stages The first stage is the structural analysis, which is concerned with the segmentation of the image into blocks of document components (paragraph, row, word, etc.), and the second one is the functional analysis, which uses location, size, and various layout convenings to label the functional content of document components (title, abstract, etc.) 12.2) inbred Segmentation Although the methods have developed remarkably in the last cristal and a variety of techniques have emerged, segmentation of cursive script into letters is still an unsolved problem. Character segmentation strategies are divided into three categories 13 is Explicit Segmentation, Implicit Segmentation and Mixed Strategies.3.3. Feature Extraction interpret representation plays one of the most important roles in a recognition system. In the simplest case, gray-level or binary images are fed to a recognizer. However, in most of the recognition systems, in order to avoid extra complexness and to increase the accuracy of the algorithms, a more compact and characteristic representation is required. For this purpose, a sink of features is extracted for each class that helps distinguish it from other classes composition remaining invariant to characteristic differences within the class14. A good survey on feature extraction methods for CR can be found 15.In the following, hundreds of document image representations methods are categorized into three major groups are Global Transformation and Series Expansion, statistical image and Geometrical and Topological Representation .3.4. Recognition TechniquesCR systems extensively use the methodologies of pattern recognition, which assigns an unknown judge into a predefined class. Numerous techniques for CR can be investigated in quartet general approaches of pattern recognition, as suggested in 16 are Template matching, Statistical techniques, and Structural techniques and Neural networks .3.5. Post ProcessingUntil this point, no semantic information is considered during the stages of CR. It is well known that humans read by context up to 60% for careless handwriting. While preprocessing tries to clean the document in a certain sense, it may remove important information, since the context information is not available at this stage. The lack of context information during the segmentation stage may cause even more severe and irreversible phantasms since it yields hollow segmentation boundaries. It is clear that if the semantic information were available to a certain extent, it would contribute a lot to the accuracy of the CR stages. On the other hand, the entire CR problem is for determine the context of the document image. Therefore, economic consumption of the context information in the CR problem creates a white-livered and egg problem. The review of the recent CR research indicates minor improvements when yet shape recognition of the character is considered. T herefore, the incorporation of context and shape information in all the stages of CR systems is necessary for meaningful improvements in recognition rates.The proposed System ArchitectureThe proposed research methodology for off-line cursive handwritten characters is described in this section as shown in Figure 2.4.1 PreprocessingThere exist a whole lot of tasks to complete originally the actual character recognition operation is commenced. These preceding tasks make certain the scanned document is in a suitable form so as to determine the scuttlebutt for the subsequent recognition operation is intact. The process of refining the scanned input image includes several steps that include Binarization, for transforming gray-scale images in to black white images, scraping noises, Skew Correction- performed to align the input with the coordinate system of the scanner and etc., The preprocessing stage comprise three steps(1) Binarization(2) Noise Removal(3) Skew CorrectionScanned Docum ent ImageFeature ExtractionBayesian finish TheoryTraining and RecognitionPre-processingBinarizationNoise RemovalSkew chastisementSegmentationLineWordCharacterRecognition o/pFigure 2. Proposed System Architecture4.1.1 BinarizationExtraction of foreground (ink) from the background (paper) is called as threshold. Typically two peaks comprise the histogram gray-scale values of a document image a high peak analogous to the white background and a smaller peak corresponding to the foreground. Fixing the threshold value is determining the one optimal value between the peaks of gray-scale values 1. Each value of the threshold is tried and the one that maximizes the mensuration is chosen from the two classes regarded as the foreground and back ground points.4.1.2 Noise RemovalThe presence of noise can cost the efficiency of the character recognition system this theme has been dealt extensively in document analysis for typed or machine-printed documents. Noise may be due the poor quality o f the document or that pile up whilst scanning, but whatever is the cause of its presence it should be removed before further Processing. We have used median filtering and Wiener filtering for the removal of the noise from the image.4.1.3 Skew CorrectionAligning the paper document with the co-ordinate system of the scanner is essential and called as skew correction. There exist a myriad of approaches for skew correction covering correlation, projection, profiles, Hough transform and etc.For skew angle detection Cumulative Scalar Products (CSP) of windows of text blocks with the Gabor filters at different orientations are calculated. Alignment of the text line is used as an important feature in estimating the skew angle. We calculate CSP for all possible 50X50 windows on the scanned document image and the median of all the angles obtained gives the skew angle.4.2 SegmentationSegmentation is a process of distinguishing lines, words, and even characters of a hand written or machine-pr inted document, a crucial step as it extracts the meaningful regions for analysis. There exist many sophisticated approaches for segmenting the region of interest. Straight-forward, may be the task of segmenting the lines of text in to words and characters for a machine printed documents in contrast to that of handwritten document, which is bland difficult. Examining the horizontal histogram profile at a smaller run away of skew angles can accomplish it. The details of line, word and character segmentation are discussed as follows.4.2.1 Line SegmentationObviously the ascenders and descanters frequently meet up and down of the adjacent lines, while the lines of text might itself dash up and down. Each word of the line resides on the conceptional line that people use to assume while writing and a method has been formulated based on this notion shown fig.3.Figure 3. Line SegmentationThe local minima points are calibrated from each Component to approximate this imaginary baseline. To calculate and categorize the minima of all components and to recognize different handwritten lines clustering techniques are deployed.4.2.2 Word and Character SegmentationThe process of word segmentation succeeds the line separation task. Most of the word segmentation issues usually concentrate on discerning the gaps between the characters to distinguish the words from one another other. This process of discriminating words emerged from the notion that the spaces between words are usually larger than the spaces between the characters in fig 4.Figure 4. Word SegmentationThere are not many approaches to word segmentation issues dealt in the literature. In spite of all these perceived conceptions, exemptions are quiet common due to flourishes in writing styles with leading and trailing ligatures. ersatz methods not depending on the one-dimensional distance between components, incorporates cues that humans use. punctilious examination of the variation of spacing between the adjacent characters as a function of the corresponding characters themselves helps reveal the writing style of the author, in term of spacing. The segmentation scheme comprises the notion of expecting greater spaces between characters with leading and trailing ligatures. Recognizing the words themselves in textual lines can itself help lead to isolation of words. Segmentation of words in to its constituent characters is touted by most recognition methods. Features like ligatures and concavity are used for determining the segmentation points.4.3 Feature ExtractionThe size inevitably limited in practice, it becomes essential to operation optimal usage of the information stored in the available database for feature extraction. thank to the sequence of straight lines, instead of a passel of pixels, it is attractive to represent character images in handwritten character recognition. Whilst holding discriminated information to persist the classifier, considerable reduction on the amount of da ta is achieved through vector representation that stores only two pairs of ordinates replacing information of several pixels. Vectorization process is performed only on basis of bi-dimensional image of a character in off-line character recognition, as the dynamic level of writing is not available. Reducing the thickness of drawing to a single pixel requires thinning of character images first. Character before and after Thinning After streamlining the character to its skeleton, entrusting on an orientated search process of pixels and on a criterion of quality of representation goes on the vectorization process. The oriented search process principally works by searching for new pixels, initially in the selfsame(prenominal) committee and on the current line segment subsequently. The search direction volition deviate progressively from the present one when no pixels are traced. The dynamic level of writing is retrieved of course with moderate level of accuracy, and that is object of oriented search. Starting the scanning process from top to bottom and from left field to right, the starting point of the first line segment, the first pixel is identified. consort to the oriented search principle, specified is the next pixel that is likely to be incorporated in the segment. Horizontal is the default direction of the segment considered for oriented search. Either if the distortion of representation exceeds a critical threshold or if the given number of pixels has been associated with the segment, the conclusion of line segment occurs. Computing the second-rate distance between the line segment and the pixels associated with it go away yield the distortion of representation. The sequence of straight lines being represented through ordinates of its two extremities character image representation is streamlined finally. All the ordinates are regularized in accord to the initial width and height of character image to resolve scale Variance.4.4 Bayesian Decision Theo riesThe Bayesian decision theory is a system that minimizes the classification error. This theory plays a role of a prior. This is when there is antecedency information about something that we would like to classify.It is a fundamental statistical approach that quantifies the tradeoffs between various decisions using probabilities and costs that accompany such decisions. First, we will assume that all probabilities are known. and then, we will study the cases where the probabilistic body structure is not completely known. Suppose we know P (wj) and p (xwj) for j = 1, 2n. and measure the lightness of a fish as the value x.Define P (wj x) as the a posteriori chance ( chance of the state of record being wj given the measurement of feature value x).We can use the Bayes formula to alter the prior probability to the posterior probabilityP (wj x) =Where p(x)P (xwj) is called the likelihood and p(x) is called the evidence.Probability of error for this decisionP (w1 x) if we decide w2P (w2x) if we decide w1P (errorx) = Average probability of errorP (error) =P (error) =Bayes decision rule minimizes this error becauseP (errorx) = min P (w1x), P (w2x)Let w1. . . wc be the finite set of c states of nature (classes, categories). Let 1. . . a be the finite set of a possible actions. Let (i wj) be the exhalation incurred for victorious action i when the state of nature is wj. Let x be theD-component vector-valued random variable called the feature vector.P (xwj) is the class-conditional probability density function. P (wj) is the prior probability that nature is in state wj. The posterior probability can be computed asP (wj x) =Where p(x)Suppose we observe x and take action i. If the true state of nature is wj, we incur the loss (i wj).The expected loss with taking action i isR (i x) = which is also called the conditional risk.The general decision rule (x) tells us which action to take for observation x. We want to find the decision rule that minimizes the general r iskR =Bayes decision rule minimizes the overall risk by selecting the action i for which R (ix) is minimum. The resulting minimum overall risk is called the Bayes risk and is the best performance that can be achieved.4.5 SimulationsThis section describes the implementation of the mapping and generation model. It is implemented using GUI (Graphical User Interface) components of the Java programming under Eclipse official document and Database storing data in Microsoft Access.For given Handwritten image character and convert to Binarization, Noise Remove and Segmentation as shown in Figure 5(a). Then after perform Feature Extraction, Recognition using Bayesian decision theory as shown in Figure5(b).Figure 5(a) Binarization, Noise Remove and SegmentationFigure 5(b) Recognition using Bayesian decision theory5. Results and sermonThis database contains 86,272 word instances from an 11,050 word dictionary written down in 13,040 text lines. We used the sets of the benchmark task with the closed vocabulary IAM-OnDB-t13. There the data is divided into four sets one set for training one set for validating the Meta parameters of the training a second validation set which can be used, for example, for optimizing a language model and an individual test set. No writer appears in more than one set. Thus, a writer independent recognition task is considered. The size of the vocabulary is about 11K. In our experiments, we did not include a language model. Thus the second validation set has not been used.Table1. Shows the results of the four individual recognition systems 17. The word recognition rate is simply measured by dividing the number of correct recognized words by the number of words in the transcription.We presented a new Bayesian decision theory for the recognition of handwritten notes written on a whiteboard. We combined two off-line and two online recognition systems. To combine the outturn sequences of the recognizers, we incrementally aligned the word sequences using a standard string matching algorithm. Evaluation of proposed Bayesian decision theory with existing recognition systems with respect to graph is shown in figure 6.Table 1. Results of four individuals recognition systemsSystem orderRecognition rateAccuracy1st OfflineHidden Markov Method66.90%61.40%1st OnlineANN73.40%65.10%2nd OnlineHMM73.80%65.20%2nd OfflineBayesian Decision theory75.20%66.10%Figure 6 Evaluation of Bayesian decision theory with existing recognition systemsThen each output position the word with the most occurrences has been used as the nal result. With the Bayesian decision theory could statistically signicantly increase the accuracy.6. determinationWe conclude that the proposed approach for offline character recognition, which fits the input character image for the stamp down feature and classifier according to the input image quality. In existing system missing characters cant be identified. Our approach using Bayesian Decision Theories which can classify missing data effectively which decrease error in compare to hidden Markova model. Significantly increases in accuracy levels will found in our method for character recognition