Tesis sobre el tema "Handwritten characters"
Siga este enlace para ver otros tipos de publicaciones sobre el tema: Handwritten characters.
Crea una cita precisa en los estilos APA, MLA, Chicago, Harvard y otros
Consulte los 50 mejores tesis para su investigación sobre el tema "Handwritten characters".
Junto a cada fuente en la lista de referencias hay un botón "Agregar a la bibliografía". Pulsa este botón, y generaremos automáticamente la referencia bibliográfica para la obra elegida en el estilo de cita que necesites: APA, MLA, Harvard, Vancouver, Chicago, etc.
También puede descargar el texto completo de la publicación académica en formato pdf y leer en línea su resumen siempre que esté disponible en los metadatos.
Explore tesis sobre una amplia variedad de disciplinas y organice su bibliografía correctamente.
1
Al-Emami, Samir Yaseen Safa. "Machine recognition of handwritten and typewritten Arabic characters". Thesis, University of Reading, 1988. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.359173.
Texto completo
2
Wang, Jianguo. "Off-line computer recognition of unconstrained handwritten characters". Thesis, The University of Sydney, 2001. https://hdl.handle.net/2123/27805.
Texto completoResumen
This thesis presents several techniques for improving the performance of off—line
Optical Character Recognition (OCR) systems: broken character mending and
recognition, feature extraction methods in OCR and hybrid methods for
handwritten numeral recognition. As an application, form document image
compression and indexing is also introduced.
Broken characters mending techniques are investigated first. A macrostrtrcture
analysis (MSA) mending method is proposed based on skeleton and boundary
information and macrostructure analysis that investigates the stroke tendency and
other properties of handwritten characters. A new skeleton end extension
algorithm is also introduced. The MSA mending method is combined with a
skeleton-based recognition algorithm to verify its efficiency. Experiment results
indicate that significant improvement has been achieved.
The feature extraction methods in OCR are analyzed by comparing their
effectiveness in different situations. Several factors and their relation with the
effectiveness of each feather extraction method are investigated. A dynamic
feature extraction method is developed to improve the performance of hybrid OCR
systems.
Hybrid methods for handwritten numeral recognition are then described, which
combine two compensatory recognisers by analyzing their performance for several
aspects. The different performances of the two algorithms for broken, connected or
slanted numerals. and the rneasurement—level decision provided by the neural
network algorithm are detected and combined to develop matching rules for each
recognition method. Five combination methods are developed to meet different
requirements. Experiments with a large number of testing data show satisfactory
results for the approach.
Finally, a generic method for compressing and indexing multi—copy form
documents is developed using template extraction and matching (TEM) strategies
and OCR. De—skewing, location and distortion adjusting of form images are
employed to realise the TEM method for practical applications. A statistical
template extraction algorithm is developed using greyscale images created by
overlapping a number of binary form images. The TEM method exploits the
cmnponent—Ievel redundancy found in multi—copy form documents and reaches a
high compression rate while keeping the original resolution and readability.
3
Ziogas, Georgios. "Classifying Handwritten Chinese Characters using Convolutional Neural Networks". Thesis, Uppsala universitet, Institutionen för informationsteknologi, 2018. http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-371526.
Texto completoResumen
Image recognition applications have been increasingly gaining popularity, as computer hardware was getting more powerful and cheaper. This increase in computational resources, led researchers even closer to their target on creating algorithms that could achieve high accuracy in image recognition tasks. These algorithms are applied in many different fields, such as in medical images analysis and object recognition in real-time applications.Previously studies have shown that among many image recognition algorithms, artificial neural networks and specifically deep neural networks, perform outstandingly due to their ability to recognize extremely accurate patterns, shapes and specific characteristics in an image.In this thesis project we are going to investigate a specialized type of Deep Neural Networks, called Convolutional Neural Networks or CNNs, which are designed specifically for image recognition tasks. Furthermore we will analyze their hyper parameters, as well as explore different architectures, in order to understand how these affect the accuracy and speed of the recognition. Finally we will present the results of the different tests, in terms of accuracy and validate them according to specific statistical metrics. For the purpose of our research, a data-set of handwritten Chinese characters was used.
4
Mandal, Rakesh Kumar. "Development of Neural Network techniques to recognize handwritten characters". Thesis, University of North Bengal, 2015. http://ir.nbu.ac.in/handle/123456789/1790.
Texto completo
5
Hosseini, Habib Mir Mohamad. "Analysis and recognition of Persian and Arabic handwritten characters /". Title page, contents and abstract only, 1997. http://web4.library.adelaide.edu.au/theses/09PH/09phh8288.pdf.
Texto completo
6
Mitchell, John. "Computer based analysis of handwritten characters for hand-eye coordination therapy". Thesis, University of Kent, 1993. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.358603.
Texto completo
7
Sae-Tang, Sutat. "A systematic study of offline recognition of Thai printed and handwritten characters". Thesis, University of Southampton, 2011. https://eprints.soton.ac.uk/206079/.
Texto completoResumen
Thai characters pose some unique problems, which differ from English and other oriental scripts. The structure of Thai characters consists of small loops combined with curves and there is an absence of spaces between each word and sentence. In each line, moreover, Thai characters can be composed on four levels, depending on the type of character being written. This research focuses on OCR for the Thai language: printed and offline handwritten character recognition. An attempt to overcome the problems by simple but effective methods is the main consideration. A printed OCR developed by the National Electronics and Computer Technology Center (NECTEC) uses Kohonen self- organising maps (SOMs) for rough classification and back-propagation neural networks for fine classification. An evaluation of the NECTEC OCR is performed on a printed dataset that contains over 0.6 million tokens. Comparisons of the classifier, with and without the aspect ratio, and with and without SOMs, yield small, but statistically significant differences in recognition rate. A very straightforward classifier, the nearest neighbour, was examined to evaluate overall recognition performance and to compare with the classifier. It shows a significant improvement in recognition rate (about 98%) over the NECTEC classifier (about 96%) on both the original and distorted data (rotated and noisy), but at the expense of longer recognition times. For offline handwritten character recognition, three different classifiers are evaluated on three different datasets that contain, on average, approximately 10,000 tokens each. The neural network and HMMs are more effective and give higher recognition rates than the nearest neighbour classifier on three datasets. The best result obtained from the HMMs is 91.1% on ThaiCAM dataset. However, when evaluated on a different dataset, the recognition rates drastically reduce, due to differences in many aspects of online and offline handwritten data. An improvement in classification rates was obtained by adjusting the stroke width of a character in the online handwritten dataset (12 percentage points) and combining the training sets from the three datasets (7.6 percentage points). A boosting algorithm called AdaBoost yields a slight improvement in recognition rate (1.2 percentage points) over the original classifiers (without applying the AdaBoost algorithm).
8
陳國評 y Kwok-ping Chan. "Fuzzy set theoretic approach to handwritten Chinese character recognition". Thesis, The University of Hong Kong (Pokfulam, Hong Kong), 1989. http://hub.hku.hk/bib/B30425876.
Texto completo
9
Wang, Yongqiang. "A study on structured covariance modeling approaches to designing compact recognizers of online handwritten Chinese characters". Click to view the E-thesis via HKUTO, 2009. http://sunzi.lib.hku.hk/hkuto/record/B42664305.
Texto completo
10
Wang, Yongqiang y 王永強. "A study on structured covariance modeling approaches to designing compact recognizers of online handwritten Chinese characters". Thesis, The University of Hong Kong (Pokfulam, Hong Kong), 2009. http://hub.hku.hk/bib/B42664305.
Texto completo
11
Chai, Sin-Kuo. "Multiclassifier neural networks for handwritten character recognition". Ohio : Ohio University, 1995. http://www.ohiolink.edu/etd/view.cgi?ohiou1174331633.
Texto completo
12
Clarke, Eddie. "A novel approach to handwritten character recognition". Thesis, University of Nottingham, 1995. http://eprints.nottingham.ac.uk/14035/.
Texto completoResumen
A number of new techniques and approaches for off-line handwritten character recognition are presented which individually make significant advancements in the field. First. an outline-based vectorization algorithm is described which gives improved accuracy in producing vector representations of the pen strokes used to draw characters. Later. Vectorization and other types of preprocessing are criticized and an approach to recognition is suggested which avoids separate preprocessing stages by incorporating them into later stages. Apart from the increased speed of this approach. it allows more effective alteration of the character images since more is known about them at the later stages. It also allows the possibility of alterations being corrected if they are initially detrimental to recognition. A new feature measurement. the Radial Distance/Sector Area feature. is presented which is highly robust. tolerant to noise. distortion and style variation. and gives high accuracy results when used for training and testing in a statistical or neural classifier. A very powerful classifier is therefore obtained for recognizing correctly segmented characters. The segmentation task is explored in a simple system of integrated over-segmentation. Character classification and approximate dictionary checking. This can be extended to a full system for handprinted word recognition. In addition to the advancements made by these methods. a powerful new approach to handwritten character recognition is proposed as a direction for future research. This proposal combines the ideas and techniques developed in this thesis in a hierarchical network of classifier modules to achieve context-sensitive. off-line recognition of handwritten text. A new type of "intelligent" feedback is used to direct the search to contextually sensible classifications. A powerful adaptive segmentation system is proposed which. when used as the bottom layer in the hierarchical network. allows initially incorrect segmentations to be adjusted according to the hypotheses of the higher level context modules.
13
Xu, Zhengyan y Yibing Zhou. "Specific Handwritten Chinese Character Recognition Based on Artificial Intelligence". Thesis, Högskolan i Gävle, Avdelningen för Industriell utveckling, IT och Samhällsbyggnad, 2013. http://urn.kb.se/resolve?urn=urn:nbn:se:hig:diva-14599.
Texto completoResumen
As internet techniques are developing more and more quickly, internet becomes the main way to communicate with the outside world. In this case, written information on paper needs to be converted to digital information urgently, increasing the need for handwritten character recognition. The aim of this work is to discuss methods that can be used to recognize handwritten Chinese characters. We study geometric features and clustering of handwritten Chinese characters from three aspects, which are handwritten character preprocessing, feature extraction and clustering. To test the correctness of our method, an application was built that could learn to recognize five medium-hard handwritten Chinese characters by using a neural network.
14
Sawhney, Sumeet S. "Distance measurements and their combination in handwritten character recognition". Thesis, National Library of Canada = Bibliothèque nationale du Canada, 2001. http://www.collectionscanada.ca/obj/s4/f2/dsk3/ftp04/MQ59339.pdf.
Texto completo
15
Ansari, Nasser. "Handwritten character recognition by using neural network based methods". Ohio : Ohio University, 1992. http://www.ohiolink.edu/etd/view.cgi?ohiou1172080742.
Texto completo
16
Sahai, Anant. "Handwritten character recognition using the minimum description length principle". Thesis, Massachusetts Institute of Technology, 1996. http://hdl.handle.net/1721.1/11015.
Texto completo
17
Shi, Daming. "An active radical approach to handwritten Chinese character recognition". Thesis, University of Southampton, 2002. https://eprints.soton.ac.uk/257379/.
Texto completo
18
Manley-Cooke, Peter. "Handwritten character recognition using a multi-classifier neuro-fuzzy framework". Thesis, University of East Anglia, 2006. http://ethos.bl.uk/OrderDetails.do?uin=uk.bl.ethos.433914.
Texto completo
19
Kassel, Robert H. "A comparison of approaches to on-line handwritten character recognition". Thesis, Massachusetts Institute of Technology, 1995. http://hdl.handle.net/1721.1/11407.
Texto completoResumen
Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Electrical Engineering and Computer Science, 1995.Includes bibliographical references (p. 155-163).
by Robert Howard Kassel.
Ph.D.
20
Kunwar, Rituraj. "Incremental / Online Learning and its Application to Handwritten Character Recognition". Thesis, Griffith University, 2017. http://hdl.handle.net/10072/366964.
Texto completoResumen
In real world scenarios where we use machine learning algorithms, we often have to deal with cases where input data changes its nature with time. In order to maintain the accuracy of the learning algorithm, we frequently have to retrain our learning system, thereby making the system inconvenient and unreliable. This problem can be solved by using learning algorithms which can learn continuously with time (incremental/ online learning). Another common problem of real-world learning scenarios that we often have to deal with is acquiring large amounts of data which is expensive and time consuming. Semi-supervised learning is the machine learning paradigm concerned with utilizing unlabeled data to improve the precision of classifier or regressor. Unlabeled data is a powerful and easily available resource and it should be utilized to build an accurate learning system. It has often been observed that there is a vast amount of redundancy in any huge, real-time database and it is not advisable to process every redundant sample to gain the same (already acquired) knowledge. Active learning is the learning setting which can handle this issue. Therefore in this research we propose an online semi-supervised learning framework which can learn actively. We have proposed an "online semi-supervised Random
Naive Bayes (RNB)" classifier and as the name implies it can learn in an online manner and make use of both labeled and unlabeled data to learn. In order to boost accuracy we improved the network structure of NB (using Bayes net) to propose an Augmented Naive Bayes (ANB) classifier and achieved a substantial jump in accuracy. In order to reduce the processing of redundant data and achieve faster convergence of learning, we proposed to conduct incremental semi-supervised learning in active manner. We applied the proposed methods on the "Tamil script handwritten character recognition"
problem and have obtained favorable results. Experimental results prove that our proposed online classifiers does as well as and sometimes better than its batch learning counterpart. And active learning helps to achieve learning convergence with much less number of samples.Thesis (PhD Doctorate)
Doctor of Philosophy (PhD)
School of Information and Communication Technology
Science, Environment, Engineering and Technology
Full Text
21
He, Tingting y 何婷婷. "A study on several problems in online handwritten Chinese character recognition". Thesis, The University of Hong Kong (Pokfulam, Hong Kong), 2008. http://hub.hku.hk/bib/B42182086.
Texto completo
22
Gong, Shyh-Jier y 龔世傑. "Recognition of handwritten digit characters". Thesis, 1993. http://ndltd.ncl.edu.tw/handle/19034124518507636417.
Texto completoResumen
碩士大同工學院
資訊工程研究所
81
This paper presents a methodology for classifying syntactic patterns is using a feature matching against a set of proto- otypes. The prototypes are first classified and arranged into a hierarchical structure that facilitates this matching. Image of characters are described by a sequence of features extracted from the chain codes of their contours. A rotatio- nally invariant string distance measure is defined that com- pared two feature strings. The methodology discussed in this paper is compared to a nearest neighbor classifier that use 2,010 prototypes. The proposed technique can get a recognit- ion rate of greater than 97 percent, and the recognition sp- eed is 0.5 sec/char.
23
"Video-based handwritten Chinese character recognition". 2003. http://library.cuhk.edu.hk/record=b6073522.
Texto completoResumen
by Lin Feng."June 2003."
Thesis (Ph.D.)--Chinese University of Hong Kong, 2003.
Includes bibliographical references (p. [114]-130).
Electronic reproduction. Hong Kong : Chinese University of Hong Kong, [2012] System requirements: Adobe Acrobat Reader. Available via World Wide Web.
Electronic reproduction. Ann Arbor, MI : ProQuest Information and Learning Company, [200-] System requirements: Adobe Acrobat Reader. Available via World Wide Web.
Mode of access: World Wide Web.
Abstracts in English and Chinese.
24
Chen, Yang-En y 陳仰恩. "Matching Topological Structures for Handwritten Characters Recognition". Thesis, 2017. http://ndltd.ncl.edu.tw/handle/szkjv3.
Texto completoResumen
碩士國立臺灣大學
資訊工程學研究所
105
This work presents a matching process to resolve the variable distortions for recognition of handwritten characters. It can tolerate difficult distortions by preserving the topological structure of the character. The unknown character and all templates are represented in advance by features that resemble the receptive fields of visual system. The templates are loaded with flexible and variable features along their skeletons. These flexible features can tolerate local distortions. The global operations, shift, rotation, and scale, are applied to revise the whole template according to certain highly matched features. After these operations, the matched feature in the revised template will overlap its corresponding feature in the unknown character as much as possible. New matching score can be calculated for each feature of the revised template. This kind revision can be operated iteratively. The recognition is accomplished for the template which received the highest whole score. This work illustrates this matching process and its operations. This process indirectly overcomes difficult distortion problems.
25
TSO, FU-NUNG y 左富農. "A Recognition System for 3D Handwritten Characters". Thesis, 2016. http://ndltd.ncl.edu.tw/handle/75968645387820459939.
Texto completoResumen
碩士世新大學
資訊管理學研究所(含碩專班)
104
The scope of applications for contactless handwriting is very broad. For example, due to frequent addition of new members and update of information, as well as hygiene considerations, it was most appropriate for IKEA to implement contactless handwriting devices for processing members’ basic information. However, contactless handwritings are not the same as those written on a 2D plane, because the written marks do not necessarily lie on the same plane. Continuing using 2D character input methods, with no way to judge displacement in the third vector Z axis (the depth vector), could result in errors. Therefore, the objective of this thesis was to research and develop a handwriting Chinese-character-recognition system which, without users porting along any devices, could carry out both input and recognition of characters. Firstly, via the system derived in this work, 3D written stroke information would be transferred to 2D written stroke data, then analyzed and resolved. Subsequently, an algorithm of Speeded Up Robust Features (SURF) is applied to select certain properties/features of the handwritten image. Finally, character recognitions can be satisfactorily carried out according to users’ need. The contactless handwriting system, researched and developed in this thesis, could ultimately recognize not only previously programmed Chinese characters, but also new characters and styles of calligraphy, based on user input. This results in a software program with high potential for significant expansion and improvement in terms of efficacy.
26
謝禎冏. "Model-guided recognition of handwritten chinese characters". Thesis, 1992. http://ndltd.ncl.edu.tw/handle/48911061822496300574.
Texto completo
27
Sharma, Anand. "Devanagari Online Handwritten Character Recognition". Thesis, 2019. https://etd.iisc.ac.in/handle/2005/4633.
Texto completoResumen
In this thesis, a classifier based on local sub-unit level and global character level representations
of a character, using stroke direction and order variations independent features, is developed
for recognition of Devanagari online handwritten characters.
It is shown that online character corresponding to Devanagari ideal character can be analyzed
and uniquely represented in terms of homogeneous sub-structures called the sub-units. These
sub-units can be extracted using direction property of online strokes in an ideal character. A
method for extraction of sub-units from a handwritten character is developed, such that the
extracted sub-units are similar to the sub-units of the corresponding ideal character.
Features are developed that are independent of variations in order and direction of strokes in
characters. The features are called histograms of points, orientations, and dynamics of orientations
(HPOD) features. The method for extraction of these features spatially maps co-ordinates
of points and orientations and dynamics of orientations of strokes at these points. Histograms
of these mapped features are computed in di erent regions into which the spatial map is divided.
HPOD features extracted from the sub-units represent the character locally; and those
extracted from the character as a whole represent it globally.
A classifier is developed that models handwritten characters in terms of the joint distribution
of the local and global HPOD features of the characters and the number of sub-units in the
characters. The classifier uses latent variables to model the structure of the the sub-units.
The parameters of the model are estimated using the maximum likelihood method. The use
of HPOD features and the assumption of independent generation of the sub-units given the
number of sub-units, make the classifier independent of variations in the direction and order of
strokes in characters. This sub-unit based classifier is called SUB classifier.
Datasets for training and testing the classifiers consist of handwritten samples of Devanagari
vowels, consonants, half consonants, nasalization sign, vowel omission sign, vowel signs,
consonant with vowel sign, conjuncts, consonant clusters, and three more short strokes with
di erent shapes. In all, there are 96 di erent characters or symbols that have been considered
for recognition. The average number of samples per character class in the training and the test
sets are, respectively, 133 and 29. The smallest and the largest dimensions of the extracted
feature vectors are, respectively, 258 and 786. Since the size of the training set per class is not
large compared to the dimension of the extracted feature vectors, the training set is small from
the perspective of training any classifier. classifiers that can be trained on a small data set are
considered for performance comparison with the developed classifier.
Second order statistics (SOS), sub-space (SS), Fisher discriminant (FD), feedforward neural
network (FNN), and support vector machines (SVM) are the other classifiers considered
that are trained with the other features like spatio-temporal (ST), discrete Fourier transform
(DFT), discrete cosine transform (DCT), discrete wavelet transform (DWT), spatial (SP), and
histograms of oriented gradients (HOG) features extracted from the samples of the training set.
These classifiers are tested with these features extracted from the samples of the test set. SVM
classifier trained with DFT features has the highest accuracy of 90.2% among the accuracies of
the other classifiers trained with the other features extracted from the test set. The accuracy
of SUB classifier trained with HPOD features is 92.9% on the test set which is the highest
among the accuracies of all the classifiers. The accuracies of the classifiers SOS, SS, FD, FNN,
and SVM increase when trained with HPOD features. The accuracy of SVM classifier trained
with HPOD features is 92.9%, which is the highest among the accuracies of the other classifiers
trained with HPOD features.
SUB classifier using HPOD features has the highest accuracy among the considered classifiers
trained with the considered features on the same training set and tested on the same test set.
The better character discriminative capability of the designed HPOD features is re
ected by
the increase in the accuracies of the other classifiers when trained with these features
28
HU, KAI-TING y 胡凱婷. "Electroencephalogram as Assistance for Recognizing Handwritten Chinese Characters". Thesis, 2015. http://ndltd.ncl.edu.tw/handle/a5aft8.
Texto completoResumen
碩士中華大學
工業管理學系碩士班
103
In recent years, sensor technology has developed rapidly, and man-machine interface has reached a mature stage. The tradition of a keyboard, mouse and other hardware devices. Mobile devices transcended the button already. More and more people choose to use handwriting input or handwriting recognition but simultaneously Faced with a variety of personal handwriting text and text variability. Although that can identify accurately, but lost time-consuming process of writing. When we write the word,it will let us produce brain waves. Electroencephalogram is the most primitive human ideation output signal will apply Brainwave to handwriting recognition. Handwriting recognition will expect to make more rapid identification. Electroencephalogram signal whether medical or academic research is always indicative research. The relationship between mind and brain wave is always in vague knowledge areas. To achieve two objectives of this study respectively as follows: First, we get Brainwave digital data through the Brainwave signal acquisition means, and establishment the research programs .Then sampling and analysis feature for ideas and Electroencephalogram brainwave signal. Second, use the effective information as input sample,and import neural network as a continuous dynamic data identification mechanism. We use a small number of samples as a precursor through the neural network training phase and recall stage. To indentify the similarity of the sample,and then determine the classification and brainwave behavior.
29
Tseng, Yi-Hong y 曾逸鴻. "Knowledge-based Radical Extraction for Handwritten Chinese Characters". Thesis, 1994. http://ndltd.ncl.edu.tw/handle/25802980577157459242.
Texto completoResumen
碩士國立交通大學
資訊工程研究所
82
The goal of this thesis is to speed up the execution and increase the efficiency of character recognition by using some knowledge relative to radicals. In this thesis, we assume that we have accepted an imperfect result of stroke extraction. If the correct radicals can be extracted successfully, the task of character recognition will be simplified greatly. First, we defined about 400 radicals that can compose more than 2000 Chinese characters. According to the experimental results and our observation, we summarize some knowledge to decribe the structural properties of those reference radical models. In the recognition process, 2-D Chinese characters would be first processed by radical separation and stroke extraction. Next, we use the possible positions at which radicals locate and the stable stroke types to pre-select suitable extracted strokes and reference radical models before the radical matching process that is a dynamic programming method. Then, there are three post-checking methods: the related stroke length checking, the convex hull checking, and the radical overlap checking are used to check the legality of all candidate radicals and remove the illegal ones. Finally, we can extract the correct radicals by finding the maximum clique of a undirected graph constructed by the legal candidate radicals. The input method utilizing a tablet is also applied in our system. Computers can generate each reference radical model automatically according to the on-line information extracted from a tablet, and insert it into the proper position of the knowledge base. The extension of radical database can be achieved easily. The testing Chinese characters are selected from the database CCL/HCCR1 and the experimental results show the feasibility of the radical extraction method proposed in this thesis.
30
趙宏宇. "A new stroke extraction method for handwritten characters". Thesis, 1993. http://ndltd.ncl.edu.tw/handle/68317869402951653535.
Texto completo
31
CHEN, JUN-XIAN y 陳俊賢. "The study of neocognitron for handwritten characters recognition". Thesis, 1991. http://ndltd.ncl.edu.tw/handle/30131981839063060038.
Texto completo
32
Chiu, Hung-Pin y 邱宏彬. "Processing and Invariant Recognition of Handwritten Chinese Characters". Thesis, 1997. http://ndltd.ncl.edu.tw/handle/96967638286401011586.
Texto completoResumen
博士國立中央大學
資訊工程研究所
85
An invariant pattern recognition system recognizes patterns despite arbitrary variations in location, scale, and orientation. In our study, an adequate norm alization process is first used to normalize characters such that they are inv ariant to translation and scale, then several rotation-invariant features and non-rotation-invariant structural features are respectively extracted for inva riant recognition of small and middle sets of handwritten Chinese characters. In off-line character recognition, a thinning algorithm is generally used to o btain the character skeletons to expedite the feature extraction. However, the overlap and rough borders of character strokes make the thinned character ske letons distorted; the distorted skeletons will degrade the performance of feat ure extraction and then that of recognition. Two feature-extraction approaches are presented by specially considering the skeleton-distortion problem to mor e perfectly extracted the desired features. The first approach employ a simple smoothing technique to remove boundary roughness from character patterns to a void the hairy problem. After thinning, an efficient fork-merging procedure is applied to correct the distorted skeletons. From the distortion-corrected ske letons, several invariant features including a simple rotation-invariant featu res called ring data are effectively extracted. In the second approach, we pro pose a stroke-based extraction method to effectively extract skeletons and str uctural features for handwritten Chinese characters. Based on run-length infor mation of character strokes, we split character strokes into the overlapped fo rk segments and the non-overlapped stroke segments to solve the stroke-interse ction problem. Based on the run lengths of all segments, we estimate the strok e width of a character by which hairy branches are effectively deleted. The pr oposed stroke-based extraction method produced fine skeletons in which the geo metrical structures of crossing lines, sharp corners, and straight lines are m ostly preserved. The fine structures facilitate the extraction of corner point s, line segments, straight strokes, and stroke direction maps. Based on the ex tracted features, two kinds of invariant recognition approaches are proposed t o recognize arbitrarily-rotated handwritten Chinese characters. In the invaria nt-feature-based approaches, several invariant features are used for preclassi fication and the ring-data feature is used for invariant recognition. Since th e class boundaries of handwritten Chinese characters are generally ill-defined , fuzzy-set classifiers and neural networks are proposed to classify the invar iant features to achieve the invariant recognition of handwritten Chinese char acters. Factors which influence the performance of the invariant-feature-based recognition approach are discussed based on experiments. These factors includ e the normalization methods, thinning algorithm, structure of ring data, class ification method, and the shape-variation degree of character set. The compari son of the proposed ring-data features to moment invariants and the comparison of the proposed classifiers to two traditional statistical classifiers: the n earest-neighbor and the minimum-mean-distance classifiers are also performed t o verify the power of the proposed approach. In the structural-feature-based a pproach, stroke direction maps are used for invariant recognition. Stroke dire ction maps only tolerate a small range of rotation; thus we use the direction maps at several representative degrees to represent one character and propose a simple classifier to recognize arbitrarily rotated characters. The invariant recognition results of all proposed classification approaches are compared an d analyzed in this dissertation.
33
Lee, Hsuan-Pei y 李宣霈. "Handwritten Chinese Characters Recognition Based on Deep Learning". Thesis, 2019. http://ndltd.ncl.edu.tw/handle/435deh.
Texto completoResumen
碩士國立中央大學
資訊工程學系
107
Character recognition has already been a popular research field even when machine learning and deep learning haven’t been discussed frequently. For example, the technique of OCR(Optical Character Recognition) has already been quite mature. Along with the development of machine learning and deep learning these years, the research of character recognition has also made a great leap by using deep learning. English characters and digit recognition has already been quite mature. However, Chinese characters recognition hasn’t been as mature as English characters and digit recognition even if many researches were based on deep learning since the structure of Chinese characters is more complexed. In addition that the Chinese characters recognition is more difficult than English characters and digit recognition, due to the variance of the style of handwritten characters from one person to another person, handwritten characters is even more difficult to be detected or recognized if there are more than one style of handwritten characters on a piece of paper. Therefore, the purpose of this research is to find out whether the multi-style handwritten Chinese characters dataset can do better job on character detection and recognition compared to one-style or few-style handwritten Chinese characters dataset.
34
Zhang, Wen-Hong y 張文鴻. "Interfered Seals and Handwritten Characters Removal for Prescription Images". Thesis, 2017. http://ndltd.ncl.edu.tw/handle/77kez5.
Texto completoResumen
碩士臺北市立大學
資訊科學系
105
In life, text in many documents have interfered by seals and handwritten characters. These documents include hospital prescription, school documents, bank documents, etc. If want to process these documents automatically, however, these interferences will affect the detection and recognition of the text. In this study, we focus on the prescription documents for hospital. Furthermore, to develop an automatic prescription processing system in the future. In order to complete this system, the text in the prescription must be detected and recognized firstly. However, many seals and handwritten characters are appeared in the prescription images. Thus, this study proposed an interference removal method to remove the interfered seals and handwritten characters. The proposed method is based on Niblack method. In this method, it has a fixed parameter, k. For abovementioned document images, this fixed parameter cannot solve abovementioned interference problems. Thus, in our method, an adaptive parameter, noise factor (NF), is proposed to produce different value for different document. This noise factor and the Niblack method are used to remove the interferences on the prescription documents. From the experimental results show that the proposed method can remove the prescription interference effectively. Furthermore, the results of the proposed method are compared with the results of the other binariaztion methods, the proposed method is superior to the other methods. For the local prescription document images, the recognition rate is 95%. For the whole prescription document images, the recognition rate is 94.7%. Key words: Interfered seals, Interfered handwritten characters, Prescription document images, Noise factor, Binarization method, Interference removal.
35
Liu, Hsin-Hung y 劉信宏. "Extraction of Strokes in Off-line Handwritten Chinese Characters". Thesis, 1994. http://ndltd.ncl.edu.tw/handle/88179145787442207078.
Texto completoResumen
碩士國立交通大學
資訊工程研究所
82
In this thesis, we present a system to extract handwritten Chinese character strokes by using the knowledge of strokes. After the preprocessing an image would be expressed as 1-D set of line segments. There are five strps in the preprocessing: thinning, detection of feature points, extraction of line segments, determination of line segment types, and deletion of short line segments. After preprocessing, we will extract strokes from line segments according to defined eight types of strokes. To extract a specific type of strokes, we first select a starting line segment. Next, we find the possible adjacent line segment of the starting line segment. According to the knowlwdge for the types of strokes, we find a feasible path that consists of continuous line segments. If some feasible paths exist simultaneously, we will apply heuristic knowledge to select the best feasible path. After a stroke is extracted, the line segments of the stroke are deleted. Then, the same procedure for stroke extraction is preformed recursively until no new strokes can be extracted. Experimental results show that we can extract strokes correctly within a short time and exist only a few of redundant strokes.
36
CHEN, BIN y 陳斌. "Recognition of handwritten Chinese characters via short line segments". Thesis, 1990. http://ndltd.ncl.edu.tw/handle/55548786621449892606.
Texto completo
37
Wu, Yi-Fang y 武宜芳. "Stroke Extraction of Chinese Handwritten Characters by Normalized Vectors". Thesis, 1997. http://ndltd.ncl.edu.tw/handle/04833111208102310768.
Texto completo
38
Mishra, Tusar Kanti. "Development of Features for Recognition of Handwritten Odia Characters". Thesis, 2015. http://ethesis.nitrkl.ac.in/6921/1/TusharKanti_511CS107-PhD_2015.pdf.
Texto completoResumen
In this thesis, we propose four different schemes for recognition of handwritten atomic Odia characters which includes forty seven alphabets and ten numerals. Odia is the mother tongue of the state of Odisha in the republic of India. Optical character recognition (OCR) for many languages is quite matured and OCR systems are already available in industry standard but, for the Odia language OCR is still a challenging task. Further, the features described for other languages can’t be directly utilized for Odia character recognition for both printed and handwritten text. Thus, the prime thrust has been made to propose features and utilize a classifier to derive a significant recognition accuracy. Due to the non-availability of a handwritten Odia database for validation of the proposed schemes, we have collected samples from individuals to generate a database of large size through a digital note maker. The database consists of a total samples of 17, 100 (150 × 2 × 57) collected from 150 individuals at two different times for 57 characters. This database has been named Odia handwritten character set version 1.0 (OHCS v1.0) and is made available in http://nitrkl.ac.in/Academic/Academic_Centers/Centre_For_Computer_Vision.aspx for the use of researchers. The first scheme divides the contour of each character into thirty segments. Taking the centroid of the character as base point, three primary features length, angle, and chord-to-arc-ratio are extracted from each segment. Thus, there are 30 feature values for each primary attribute and a total of 90 feature points. A back propagation neural network has been employed for the recognition and performance comparisons are made with competent schemes. The second contribution falls in the line of feature reduction of the primary features derived in the earlier contribution. A fuzzy inference system has been employed to generate an aggregated feature vector of size 30 from 90 feature points which represent the most significant features for each character. For recognition, a six-state hidden Markov model (HMM) is employed for each character and as a consequence we have fifty-seven ergodic HMMs with six-states each. An accuracy of 84.5% has been achieved on our dataset. The third contribution involves selection of evidence which are the most informative local shape contour features. A dedicated distance metric namely, far_count is used in computation of the information gain values for possible segments of different lengths that are extracted from whole shape contour of a character. The segment, with highest information gain value is treated as the evidence and mapped to the corresponding class. An evidence dictionary is developed out of these evidence from all classes of characters and is used for testing purpose. An overall testing accuracy rate of 88% is obtained.
The final contribution deals with the development of a hybrid feature derived from discrete wavelet transform (DWT) and discrete cosine transform (DCT). Experimentally it has been observed that a 3-level DWT decomposition with 72 DCT coefficients from each high-frequency components as features gives a testing accuracy of 86% in a neural classifier. The suggested features are studied in isolation and extensive simulations has been carried out along with other existing schemes using the same data set. Further, to study generalization behavior of proposed schemes, they are applied on English and Bangla handwritten datasets. The performance parameters like recognition rate and misclassification rate are computed and compared. Further, as we progress from one contribution to the other, the proposed scheme is compared with the earlier proposed schemes.
39
Niu, Xiaoxiao. "Fusions of CNN and SVM Classifiers for Recognizing Handwritten Characters". Thesis, 2011. http://spectrum.library.concordia.ca/7370/1/Niu_MCompSc_S2011.pdf.
Texto completo
40
Ming, Chen Jyh y 陳志明. "Recognition of similar handwritten Chinese characters by artificial neural networks". Thesis, 1995. http://ndltd.ncl.edu.tw/handle/08846493337872930669.
Texto completoResumen
碩士國立交通大學
資訊工程研究所
83
This thesis presents an application of neural networks on off- line similar handwritten Chinese character recognition. The proposed method consists of three components:(1)confusing character sets construction,(2)feature selection,(3)modular neural network recognition. In order to evaluate the proposed recognition system, we choose 5401 frequently used Chinese characters as our trainning and testing domain. The database of each testing and trainning sample character was created by the Computer and Communication Laboratory of Industrial Technology Research Institute. Because the samples in this database were collected by more than 2600 people, our recognition system could reach a high generality and user-independence. Experimental results show that, the method improves recognition rate from 86.01% to 90.12%.
41
Li, Li Mei y 李麗美. "Multi-stroke relaxation matching method for handwritten Chinese characters recognition". Thesis, 1995. http://ndltd.ncl.edu.tw/handle/35173093896778146495.
Texto completo
42
Lee, Wei y 李煒. "Using Fuzzy Logic For Offline Recognition Of Chinese Handwritten Characters". Thesis, 1996. http://ndltd.ncl.edu.tw/handle/90651248205623924786.
Texto completoResumen
碩士國立中正大學
資訊工程學系
84
In this thesis, we inherit Mr. Ling Shr Ming's preprocessing technique for Chinese O.C.R. stroke retrieval [1], and propose a new method for Chinese O.C.R characterrecognition. This method uses fuzzy logic based matching. The features which are used by the method include line position, slope and coordinates. The fuzzy matching has better recognition rates and is more flexible than traditional methods, and has a high tolerance for error in an input character. Besides, our fuzzy matching also solves the problem of permutations in Mr. Fan Chen Huang's fuzzy matching algorithm [2], and it also solves the cycle problem [3]in traditional algorithms.
43
ZHENG, FANG-XUAN y 鄭芳炫. "An approach to recognition of handwritten Chinese characters by stroke analysis". Thesis, 1988. http://ndltd.ncl.edu.tw/handle/22079771789453220936.
Texto completo
44
Kuo, Hung-Yie y 郭鴻憶. "On the Recognition of Handwritten Chinese Characters Using Fuzzy Neural Networks". Thesis, 1994. http://ndltd.ncl.edu.tw/handle/17766309908291846174.
Texto completoResumen
碩士國立成功大學
電機工程研究所
82
Handwritten Chinese character recognition is known to be a very difficult problem because of their large vocabulary set of complex structures and writing variations. In this thesis, a hierarchical architecture is constructed to recognize handwritten Chinese characters. First, the preprocessing operations including smoothing, stroke fitting, skeletonization, feature point detection are performed to obtain a suitable data format for further processes. Then a stroke extraction technique without any operations to correct the distorted skeleton before extracting process is proposed to extract all strokes represented by a sequence of line segments. Based on the fact that the cross point is formed by one stroke running through another, the cross point can be easily detected by checking four adjacency line segments that form exact two strokes. Because the number of Chinese characters is very large, preclassification stage is usually needed for this work to reduce the number of candidates in matching process. The characters with the same number of stroke and cross point are grouped together in preclassification stage. Finally, for each group, a fuzzy neural network is proposed to resolve individual identification. A fuzzy neural network which possesses the advantages of fuzzy concepts and neural network is also presented for handwritten Chinese character recognition based upon the feature of stroke position, direction and length. The experimental results show that the proposed approach is effective.
45
Lien, Chun-Yu y 連俊宇. "Feature extraction of 3D handwritten Chinese characters based on Leap Motion". Thesis, 2014. http://ndltd.ncl.edu.tw/handle/80166046784066969036.
Texto completoResumen
碩士國立中央大學
資訊工程學系在職專班
102
Recently, the demand of handwriting input becomes more and more important due to the popularity of smartphones and 3C products. Handwriting input makes 3C products operations becomes easy. However, most of the input devices nowadays are generally not contactless so that user has to write directly on the device with a pen or finger. Although there are some video based air-finger-writing systems using a camera to capture hand/ finger trajectory motions while handwriting. However, they are easily affected by illuminations. In this thesis, we propose a novel feature extraction method of 3D handwritten Chinese characters based on Leap Motion which can extract the features from air-finger-writing Chinese characters without carrying or contacting any device. In the proposed system, the moving fingers and hands can be captured clearly even in the dusky environments because Leap Motion uses infrared sensor in capturing moving trajectories. With the trajectories being captured, all embedded features including coordinate, velocity, and total writing time of each input character captured from Leap Motion are applied to extract the real and virtual strokes of the input Chinese character. The extracted real strokes, virtual strokes, and embedded features provide significant discriminative information for later 3D signature verification and 3D Chinese character recognition. Experimental results verify the validity and effectiveness of our proposed method in extracting features of 3D handwriting Chinese characters.
46
Li, Yu-An y 李育安. "Handwritten and Printed Chinese Character Recognition By Using Computer Font Type Chinese Characters into Convolutional Neural Network". Thesis, 2018. http://ndltd.ncl.edu.tw/handle/99ad3j.
Texto completoResumen
碩士國立臺灣大學
工程科學及海洋工程學研究所
106
The main purpose of this paper is to improve Handwritten Chinese Character Recognition and traditional, non-modern Printed Chinese Character Recognition problem. By using the existing different style of Chinese font resources in computer system and online sources, we take most commonly used 5000 and 10000 words, then do several data deformation and preprocessing by image processing skills to produce training data. Combined with the technology of Convolutional Neural Networks in machine learning, we trained a distinguished model which can be used to recognize handwritten and printed Chinese character both. The main goal of this paper is to find the valid training features, optimize parameters and fine tune our model to get a better performance. The results of this paper mainly include: (1) How to train a model which can recognize both the handwritten font and the printed font simultaneously on by existing computer word font. (2) For the printed Chinese character font, we mainly focus on early traditional printed fonts, and improves the recognition problems, such as rare Chinese characters recognition and characters easily damaged or blur in the original text. (3) We conduct our experiments with the Beijing Civil News, the Biansha Tibetan Buddhist Dharma and the 2013 CASIA handwritten Chinese character public test set. The results show that the model and method we proposed in this paper can reach the accuracy of 69.9% on News, 89.29% on Buddhist Dharma, and 58.27% on handwriting testing set. Compared with the existing common OCR recognition software, our model can improve the accuracy about 2~3%. Key Word : HCCR、PCCR、Image Processing、Machine Learning、Convolutional Neural Networks
47
Liao, Chong-Wen y 廖崇文. "A Study on Handwritten and Printed Characters Discrimination Using the Fuzzy System". Thesis, 1999. http://ndltd.ncl.edu.tw/handle/14496415984573861630.
Texto completoResumen
碩士國立交通大學
電機與控制工程系
87
The handwritten and printed character recognition plays an important rule in the area of the optical character recognition (OCR) and the document analysis. In order to promote the performance of the OCR system, the character style in handwritten or in printed need to be known in advance. In this thesis, a fuzzy classification system which can discriminate handwritten and printed characters is implemented. In order to promote the correct rate of the classification system, the most useful features of the classification system are extracted by using the sensitivity analysis. Then these features are combined by using the concept of the fuzzy logic. The proposed method can apply to the Chinese and the English document images. Moreover, the experimental results show that the classification system has a good performance even in the document which is mixed handwritten and printed characters.
48
Huang, Chiung-Wei y 黃瓊緯. "A Fuzzy Rule-based System for the Radical Extraction on Handwritten Chinese Characters". Thesis, 1994. http://ndltd.ncl.edu.tw/handle/44008925819285204256.
Texto completoResumen
碩士國立臺灣科技大學
工程技術研究所
82
In this thesis, a fuzzy rule-based system for extracting radicals as the preclassification information for handwritten Chinese character recognition (HCCR) is proposed. Since the writings of handwritten Chinese characters vary a lot, we adopt fuzzy set theory to deal with the recognition of these fuzzy in nature patterns. Based on the structure of Chinese characters, the primitive but important features, i.e., strokes, are treated as fuzzy objects in the proposed system. That is, candidates of strokes are provided to obtain more reliable and accurate results. Furthermore, fuzzy rules which represent the character structures are used to combine the extracted strokes into compound strokes or radicals. That is, the capability of the system can be enhanced by increasing fuzzy rules appropriately. Thus, the proposed method is not only a reliable but also a flexible approach. Besides, since the number of rules of a fuzzy system is much less than that of a general rule-based system, e.g., in our system 60 rules are used to deal with 20 radicals, the computation effort is not heavy. Moreover, except the poor thinning results and mis-extracted characters, an average of 95% recognition rate on the radical recognition of 542 test characters which are selected from the 100th samples of HCCRBASE (character image database provided by CCL, ITRI, Taiwan) is obtained. It not only confirms the feasibility of the proposed system, but also suggests that applying fuzzy set theory on HCCR is an efficient and promising approach.
49
Sheu, Chung-Chieh y 徐仲杰. "A handwritten chinese characters recognition method based on primitive and fuzzy features via SEART neural net model". Thesis, 1995. http://ndltd.ncl.edu.tw/handle/92946634800767476278.
Texto completoResumen
碩士國立臺灣科技大學
工程技術研究所
83
A handwritten Chinese characters recognition method using SEART neural network model with primitive and compound fuzzy features is proposed. The primitive features are extracted in local and global view. Also they have good stability. Since the writings of handwritten Chinese characters vary a lot, we adopt fuzzyract the compound features in structural view. These categories of features are extracted in one pass, so thel effort is not heavy. We combine the two categories of features and use a fast classifier, named supervised extended ART (SEART) neural network model, to recognize the handwritten Chinese characters. The SEART classifier has excellent performance, fast, good generalization and exceptions handling ability in complex problems. Using the fuzzy set theory in features extraction and the neural network as a classifier are helpful for tolerating distortions, noises and variations. In spite of the poor thinning, an average of 90.24% recognition rate on the 605 test characters is obtained. The database used is HCCRBASE (provided by CCL, ITRI, Taiwan). It not only confirms the feasibility of the proposed system, but also suggests that applying the fuzzy set theory and neural networks on HCCR is an efficient and promising approach.
50
Ms, Shalini. "Handwritten Hindi Character Recognition". Thesis, 2017. http://ethesis.nitrkl.ac.in/8879/1/2017_MT_Shalini.pdf.
Texto completoResumen
Devanagari, the most accepted script in India and Hindi is the only dialect which is widely spoken and written, so Handwritten Hindi character Recognition is done. Optical Character Recognition (OCR) is used for pattern recognition, it can be online or offline. Handwritten text is electronically converted into machine learning language. Handwritten character Recognition has many applications like cheque reader,passport reader,address reader,specific tasks readers. Devanagari is troublesome because the characters present in a words are somewhat similar to other character or connected words may have problem in recognition as number of modifiers are present. The major challenge faced was removal of header line as header line cannot be always straight as it varies from person to person. The characters which are handwritten will not always have sharp corners, the header lines present will not be perfectly straight and the curves which are present will not be so smooth. Handwritten character recognition undergo three major steps (i) Pre-Processing (ii) Feature Extraction(iii)Classification. Pre processing is the first step which deals with binarization, noise removal, morphological operations and segmentation. Segmentation is major part in character recognition. Words are segmented into single single characters and these segmented characters are used for feature extraction. In second step Histogram of Oriented Gradients (HOG) is used as extraction of feature in an image so as to obtain the feature vector .Object detection can be easily done by using HOG in image processing and computer vision. HOG has intensity values which is obtained by gradient computation and will give rough idea of shape or pattern of an image. Last step concludes with classification, for classifying the samples Support Vector machine is implemented. SVM is basically used as binary classifier but in this project it has been used as Multiclass Classifier (One Vs. All). SVM constructs a hyper plane as data points are mapped into higher D-dimensional space. Non Linear SVM includes various kernels like polynomial kernel, radial basis kernel for mapping the data into higher D-dimensional space. The performance analysis is efficient for the kernels which are used. Accuracy rate can be improved for segmentation by using various other methods for segmentation. It can be extended to work on degraded text or broken characters and conversion of text to speech. Online recognition of character can be done.