Fumitaka Kimura

Mie University, Tsu-shi, Mie-ken, Japan

Are you Fumitaka Kimura?

Claim your profile

Publications (134)5.9 Total impact

  • Source
  • Source
  • N. Bhattacharya, U. Pal, F. Kimura
    [show abstract] [hide abstract]
    ABSTRACT: Recognition of Bangla compound characters has rarely got attention from researchers. This paper deals with segmentation and recognition of online handwritten Bangla cursive text containing basic and compound characters and all types of modifiers. Here, at first, we segment cursive words into primitives. Next primitives are recognized. A primitive may represent a character/compound character or a part of a character/compound character having meaningful structural information or a part incurred while joining two characters. We manually analyzed all the input texts written by different groups of people to create a ground truth set of distinct classes of primitives for result verification and we obtained 251 valid primitive classes. For automatic segmentation of text into primitives, we discovered some rules analyzing different joining patterns of Bangla characters. Applying these rules and using combination of online and offline information the segmentation technique was proposed. We achieved correct primitive segmentation rate of 97.89% from the 4984 online words. Directional features were used in SVM for recognition and we achieved average primitive recognition rate of 97.45%.
    Document Analysis and Recognition (ICDAR), 2013 12th International Conference on; 01/2013
  • [show abstract] [hide abstract]
    ABSTRACT: This paper proposes a new combined signature verification technique called segmentation-verification based on three different off-line feature vectors extracted from full name Japanese signature image and from the sub-images of the first name and the last name. The Mahalanobis distance for each feature vector is calculated and the final decision (verification) is performed by SVM based on the three Mahalanobis distance. In the evaluation test the proposed technique achieved 94.30% verification accuracy, which is 1.03% higher than the best accuracy obtained from the full name signature image. This result shows that the proposed segmentation-verification approach improves Japanese signature verification accuracy significantly.
    Document Analysis and Recognition (ICDAR), 2013 12th International Conference on; 01/2013
  • R. Malik, P.P. Roy, U. Pal, F. Kimura
    [show abstract] [hide abstract]
    ABSTRACT: In this paper, we present a novel approach for retrieval of handwritten musical documents using a query sequence/word of musical scores. In our algorithm, the musical score-words are described as sequences of symbols generated from a universal codebook vocabulary of musical scores. Staff lines are removed first from musical documents using structural analysis of staff lines and symbol codebook vocabulary is created in offline. Next, using this symbol codebook the music symbol information in each document image is encoded. Given a query sequence of musical symbols in a musical score-line, the symbols in the query are searched in each of these encoded documents. Finally, a sub-string matching algorithm is applied to find query words. For codebook, two different feature extraction methods namely: Zernike Moments and 400 dimensional gradient features are tested and two unsupervised classifiers using SOM and K-Mean are evaluated. The results are compared with a baseline approach of DTW. The performance is measured on a collection of handwritten musical documents and results are promising.
    Document Analysis and Recognition (ICDAR), 2013 12th International Conference on; 01/2013
  • [show abstract] [hide abstract]
    ABSTRACT: In this paper, we study on Chinese text classification using character-based approach (N-gram) and word-based approach and propose the use of uni-gram, bi-gram and word features of length greater than or equal to three. A weight coefficient which can be used to give higher weights to word features is also introduced. We further investigate a serial approach based on feature transformation and dimension reduction techniques to improve the performance. Experimental results show that our proposed approach is efficient and effective for improving the performance of Chinese text classification.
    Document Analysis and Recognition (ICDAR), 2013 12th International Conference on; 01/2013
  • [show abstract] [hide abstract]
    ABSTRACT: Although there are some reports on offline Tamil isolated handwritten character recognition, to our knowledge there is only two reports on Tamil off-line handwritten word recognition. Also no city name dataset is available for Tamil script. In this paper we present a Tamil offline city name dataset, we developed, and propose a scheme for recognition. Because of the different writing style of various individuals, some of the characters in a Tamil city name may touch and accurate segmentation of such touching into individual characters is a difficult task. Avoiding proper segmentation here, we consider a city name string as a word and the recognition problem is treated as lexicon driven word recognition. In the proposed method, binarized city names are pre-segmented into primitives (individual character or its parts). Primitive components of each city name are then merged into possible characters to get the best city name using dynamic programming. For merging, total likelihood of characters is used as the objective function and character likelihood is computed based on Modified Quadratic Discriminant Function (MQDF), where direction features are applied. A dataset of 265 Tamil city names is developed. and the database will be available freely to the researchers. From the experiment of the proposed scheme 96.89% city name accuracy is obtained from this dataset.
    Document Analysis and Recognition (ICDAR), 2013 12th International Conference on; 01/2013
  • [show abstract] [hide abstract]
    ABSTRACT: We propose a new methodology for motion tracking of local myocardial tissue on m-mode echocardiograms. this methodology is applicable to the quantitative assessment of myocardial performance in clinics. The M-mode echocardiogram is widely used in clinics to measure diagnostic indexes like thickening and thinning of myocardial muscle layers. To measure such indexes, doctors are required to track myocardial motion manually, however the tracking of myocardial motion by hand is tedious and time-consuming process. Our proposed method is able to tracking the myocardial motion on mmode echocardiograms automatically by employing DPbased optimization. In this report we present the proposed method.
    Proceedings of the IASTED International Conference Signal and Image Processing ( IP 2012); 08/2012
  • U. Pal, R.K. Roy, F. Kimura
    [show abstract] [hide abstract]
    ABSTRACT: Under three-language formula, the destination address block of postal document of an Indian state is generally written in three languages: English, Hindi and the State official language. From the statistical analysis we found that 12.37%, 76.32% and 10.21% postal documents are written in Bangla, English and Devanagari script, respectively. Because of inter-mixing of these scripts in postal address writings, it is very difficult to identify the script by which a city name is written. To avoid such script identification difficulties, in this paper we proposed a lexicon-driven method for multi-lingual (English, Hindi and Bangla) city name recognition for Indian postal automation. In the proposed scheme, at first, to take care of slanted handwriting of different individuals a slant correction technique is performed. Next, a water reservoir concept is applied to pre-segment the slant corrected city names into possible primitive components (characters or its parts). Pre-segmented components of a city name are then merged into possible characters to get the best city name using the lexicon information. In order to merge these primitive components into characters and to find optimum character segmentation, dynamic programming (DP) is applied using total likelihood of the characters of a city name as an objective function. We tested our system on 16132 Indian trilingual city names and 92.25% overall recognition accuracy was obtained.
    Frontiers in Handwriting Recognition (ICFHR), 2012 International Conference on; 01/2012
  • [show abstract] [hide abstract]
    ABSTRACT: In this paper, several sets of experiments were carried out to study the impact of word segmentation errors on automatic Chinese text classification. Comparison experiment of four word-based approaches was first carried out and the results show that the performance was significantly reduced when using automatic word segmentation instead of manual word segmentation which means errors caused by automatic word segmentation have an obvious impact on classification performance. We further conducted the experiment using character-based approach (N-gram). Although N-gram approach produces a large number of ambiguous words, the results show that it performed better than automatic word segmentation.
    01/2012;
  • [show abstract] [hide abstract]
    ABSTRACT: This paper addresses the problem of recognizing rotated characters in scene and estimating the rotation angle using Weighted Direction Code Histogram (WDCH) and Modified Quadratic Discriminant Function (MQDF). In our previous paper [1], we proposed a rotation-free character recognition method and confirmed the feasibility of real world application. We also applied our method to scene analysis to detect surface normal from written characters by estimating rotation angle of the characters. In this research, we evaluate how precisely our proposed method can estimate rotation angle by comparing obtained result by 3D digitizer.
    Pattern Recognition (ICPR), 2012 21st International Conference on; 01/2012
  • [show abstract] [hide abstract]
    ABSTRACT: Automatic text classification (ATC) is important in applications such as indexing and organizing electronic documents in databases leading to enhancement of information access and retrieval. We propose a method which employs various types of feature sets and learning algorithms to improve classification effectiveness. Unlike the conventional methods of multi-classifier combination, the proposed method considers the contributions of various types of feature sets and classifiers. It can therefore be known as multiple feature-classifier combination (MFC) method. In this paper we present empirical evaluation of MFC using two benchmarks of text collections to determine its effectiveness. Empirical evaluation show that MFC consistently outperformed all compared methods.
    01/2012;
  • [show abstract] [hide abstract]
    ABSTRACT: In order to achieve high accuracy of face recognition, detection of facial parts such as eyes, nose, and mouth is essentially important. In this paper, we propose a method to detect eyes from frontal face images. The proposed method consists of two major steps. The first is two dimensional Hough transformation for detecting circle of unknown radius. The circular Hough transform first generates two dimensional parameter space (xc, yc) using the gradient of grayscale. The radius of circle r is determined for each local maximum in the (xc, yc) space. The second step of the proposed method is evaluation of likelihood of eye using histogram of gradient and Support Vector Machine (SVM). The eye detection step of proposed method firstly detects possible eye center by the circular Hough transform. Then it extracts histogram of gradient from rectangular window centered at each eye center. Likelihood of eye of the extracted feature vector is evaluated by SVM, and pairs of eyes satisfying predefined conditions are generated and ordered by sum of the likelihood of both eyes. Evaluation experiment is conducted using 1,409 images of the FERET database of frontal face image. The experimental result shows that the proposed method achieves 98.65% detection rate of both eyes.
    Pattern Recognition (ICPR), 2012 21st International Conference on; 01/2012
  • [show abstract] [hide abstract]
    ABSTRACT: This paper proposes a new SVM based technique for combining signature verification techniques using off-line features and on-line features. The off-line feature based technique employs gradient feature vector representing the shape of signature image, and the on-line feature based technique employs dynamic programming (DP) matching technique for time series data of the signatures. The final decision (verification) is performed by SVM based on output from those off-line and online techniques. In the evaluation test the proposed technique achieved 92.96% verification accuracy, which is 1.4% higher than the better accuracy obtained by the individual techniques. This result shows that combining multiple techniques by SVM improves signature verification accuracy significantly.
    Frontiers in Handwriting Recognition (ICFHR), 2012 International Conference on; 01/2012
  • Source
    [show abstract] [hide abstract]
    ABSTRACT: In this paper, we perform Chinese text classification using N-gram (uni-gram, bi-gram and mixed uni-gram/bi-gram) frequency feature instead of word frequency feature to represent documents and propose the use of mixed uni-gram/bi-gram after feature transformation. We further propose a serial approach based on feature transformation and dimension reduction techniques to improve the performance. Experimental results show that our proposed approach is efficient and effective for improving the performance of Chinese text classification. Furthermore, we present several experiments evaluating the selection of features based on part-of-speech analysis and the results show that suitable combination of part-of-speech can lead to better classification performance.
    Document Analysis and Recognition (ICDAR), 2011 International Conference on; 10/2011
  • Source
    [show abstract] [hide abstract]
    ABSTRACT: The technology of Optical Character Recognition (OCR) is used to generate texts in the process of digitizing print documents. Usually these texts need to be indexed and organized to simplify their access and retrieval. One of the powerful approaches in accomplishing this task is the use of Automated Text Classification. However, it is currently impossible for OCR technology to recognize all characters with an accuracy of 100%. We therefore propose the use of combined linguistic features in automated classification of OCR texts to formulate an informative feature set. The proposed method was experimentally evaluated using Japanese OCR texts. Empirical results indicate that the combination of linguistic features improved classification performance of OCR texts.
    2011 International Conference on Document Analysis and Recognition, ICDAR 2011, Beijing, China, September 18-21, 2011; 01/2011
  • Source
    Umapada Pal, Rami Kumar Roy, Fumitaka Kimura
    [show abstract] [hide abstract]
    ABSTRACT: Although for postal automation there are many pieces of work towards street name recognition on non-Indian languages, to the best of our knowledge there is no work on street name recognition on Indian languages. In this paper we proposed a scheme for recognition of Indian street name written in Bangla script. Because of the writing style of different individuals some of the characters in a street name may touch with its neighboring characters. Accurate segmentation of such touching into individual characters is a difficult task. To avoid such segmentation, here we consider a street name string as word and the street name recognition problem is treated as lexicon driven word recognition. Some of the street names may contain two or more words and we have concatenated these words to have a single word. In the proposed method, at first, street names are binarized and pre-segmented into possible primitive components (individual characters or its parts) analyzing their cavity portions. Pre-segmented components of a street name are then merged into possible characters to get the best street name. Dynamic programming (DP) is applied for the merging using total likelihood of characters as the objective function. To compute the likelihood of a character, modified quadratic discriminant function (MQDF) is used. Our proposed system shows 99.03% reliability with 18.80% rejection, and 0.79% error rates when tested on 4450 handwritten Bangla street name samples.
    2011 International Conference on Document Analysis and Recognition, ICDAR 2011, Beijing, China, September 18-21, 2011; 01/2011
  • Source
    [show abstract] [hide abstract]
    ABSTRACT: In this paper, we propose an efficient skew estimation technique based on Piece-wise Painting Algorithm (PPA) for scanned documents. Here we, at first, employ the PPA on the document image horizontally and vertically. Applying the PPA on both the directions, two painted images (one for horizontally painted and other for vertically painted) are obtained. Next, based on statistical analysis some regions with specific height (width) from horizontally (vertically) painted images are selected and top (left), middle (middle) and bottom (right) points of such selected regions are categorized in 6 separate lists. Utilizing linear regression, a few lines are drawn using the lists of points. A new majority voting approach is also proposed to find the best-fit line amongst all the lines. The skew angle of the document image is estimated from the slope of the best-fit line. The proposed technique was tested extensively on a dataset containing various categories of documents. Experimental results showed that the proposed technique achieved more accurate results than the state-of-the-art methodologies.
    2011 International Conference on Document Analysis and Recognition, ICDAR 2011, Beijing, China, September 18-21, 2011; 01/2011
  • Source
    [show abstract] [hide abstract]
    ABSTRACT: In this paper, we propose a new method for three dimensional rotation-free recognition of characters in scene. In the proposed method, we employ the Modified Quadratic Discriminant Function (MQDF) classifier trained with samples generated by three-dimensional rotation process in a computer. We assume that when recognizing individual characters, considering three-dimensional rotation can approximately handle the recognition of perspectively distorted characters. The results of the evaluation experiments using printed alphanumeric characters as an evaluation data set, consisting of approximately 600 samples/class for 62 character classes, show that the recognition rate is 99.34% for rotated characters while it is 99.59% for non rotated characters. We have empirically confirmed that the rotated characters given as the training data set do not negatively affect significantly to recognition of non rotated characters. Moreover, 437 characters extracted from 50 camera-captured scenes were correctly recognized and the feasibility of real world application of our method has been confirmed. Finally we describe on three dimensional rotation angle estimation of characters for detecting local normal of the surface on which the characters are printed aiming to scene analysis by shape from characters.
    2011 International Conference on Document Analysis and Recognition, ICDAR 2011, Beijing, China, September 18-21, 2011; 01/2011
  • [show abstract] [hide abstract]
    ABSTRACT: A new motion tracking method for local myocardial tissue using M-mode ultrasound signals is proposed. This method is applicable to the quantitative assessment of myocardial performance in clinics. M-mode echocardiograms are widely used in clinics for myocardial diagnosis. However, physicians are required to employ tedious and time-consuming diagnostic protocols. In the pro-posed method, the myocardial motion in M-mode echocardiograms is automatically tracked based on dynamic programming (DP) optimization. The method consists of two main stages. In the first stage, the correlation weighted mean algorithm estimates the velocity field in the myocardial wall. In the second stage, the instantaneous displacement of a set of tracking points is calculated from the estimated velocity field, and the DP tracking method tracks the motion of the tracking points. The experimental results obtained using clinical ultrasound data demonstrate that the proposed DP tracking method can provide higher performance in myocardial motion tracking as compared to other methods.

Publication Stats

933 Citations
5.90 Total Impact Points

Institutions

  • 1987–2011
    • Mie University
      • • Graduate School of Engineering
      • • Department of Electrical and Electronic Engineering
      • • Faculty of Engineering
      Tsu-shi, Mie-ken, Japan
  • 1991–2009
    • University of Michigan-Dearborn
      • Department of Electrical and Computer Engineering
      Dearborn, Michigan, United States
  • 2008
    • Autonomous University of Barcelona
      • Computer Vision Center
      Cerdanyola del Vall├Ęs, Catalonia, Spain
  • 2007
    • Indian Statistical Institute
      • Computer Vision and Pattern Recognition Unit (CVPR)
      Baranagore, Bengal, India