• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.9
• /
• pp.1070-1075
• /
• 2019

In this paper, I propose an algorithm to generate an input digit image for a digit recognition system by detecting a digit string in an image and segmenting the digits constituting the digit string. The proposed algorithm detects blobbed digit string through blob detection, designates a digit string area and corrects digit string skew using the detected blob information. And the proposed algorithm corrects the digit skew and determines the boundary points for the digit segmentation in the corrected digit sequence using three CPgraphs newly defined in this paper. In digit segmentation experiments using the image group including digit strings printed with a range of the font sizes and the image group including handwritten digit strings, the proposed algorithm successfully segments 100% and 90% of the digits in each image group.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.32B no.9
• /
• pp.1231-1238
• /
• 1995

In on-line character recognition, to segment input character is important. This paper proposes an internal character segmentation algorithm. The internal segmentation algorithm produces candidate words by considering possible combinations of Korean alphabets. In this process, we make use of projections of strokes onto the horizontal axis to remove ambiguities among candidate words. As a result of experiments, the internal segmentation algorithm shows better performance than external segmentation algorithm as the gap between sample characters becomes smaller.

• Korean Journal of Cognitive Science
• /
• v.3 no.1
• /
• pp.61-78
• /
• 1991

This paper descibes the study of application of a modified Neocognitron model with backward path for the recognition of Hangul(Korean) syllabic characters. In this original report, Fukushima demonstrated that Neocognitron can recognize hand written numerical characters of $19{\times}19$ size. This version accepts $61{\times}61$ images of handwritten Hangul syllabic characters or a part thereof with a mouse or with a scanner. It consists of an input layer and 3 pairs of Uc layers. The last Uc layer of this version, recognition layer, consists of 24 planes of $5{\times}5$ cells which tell us the identity of a grapheme receiving attention at one time and its relative position in the input layer respectively. It has been trained 10 simple vowel graphemes and 14 simple consonant graphemes and their spatial features. Some patterns which are not easily trained have been trained more extrensively. The trained nerwork which can classify indivisual graphemes with possible deformation, noise, size variance, transformation or retation wre then used to recongnize Korean syllabic characters using its selective attention mechanism for image segmentation task within a syllabic characters. On initial sample tests on input characters our model could recognize correctly up to 79%of the various test patterns of handwritten Korean syllabic charactes. The results of this study indeed show Neocognitron as a powerful model to reconginze deformed handwritten charavters with big size characters set via segmenting its input images as recognizable parts. The same approach may be applied to the recogition of chinese characters, which are much complex both in its structures and its graphemes. But processing time appears to be the bottleneck before it can be implemented. Special hardware such as neural chip appear to be an essestial prerquisite for the practical use of the model. Further work is required before enabling the model to recognize Korean syllabic characters consisting of complex vowels and complex consonants. Correct recognition of the neighboring area between two simple graphemes would become more critical for this task.

• Journal of The Korean Association of Information Education
• /
• v.1 no.1
• /
• pp.28-37
• /
• 1997

This paper presents a new neural network based on fuzzy set and its application to invariant character recognition. The fuzzy neural network consists of five layers. The results of simulation show that the network can recognize characters in the case of distortion, translation, rotation and different sizes of handwritten characters and even with noise(8${\sim}$30%)). Translation, distortion, different sizes and noise are achieved by layer L2 and rotation invariant by layer L5. The network can recognize 108 examples of training with 100% recognition rate when they are shifted in eight directions by 1 pixel and 2 pixels. Also, the network can recognize all the distorted characters with 100% recognition rate. The simulations show that the test patterns cover a ${\pm}20^{\circ}$ range of rotation correctly. The proposed network can also recall correctly all the learned characters with 100% recognition rate. The proposed network is simple and its learning and recall speeds are very fast. This network also works for the segmentation and recognition of handwritten characters.

• Annual Conference on Human and Language Technology
• /
• 2001.10d
• /
• pp.388-393
• /
• 2001

유사한 형태의 필기 한글 문자쌍은 한글 인식 시 발생하는 오류의 많은 부분을 차지한다. 이는 유사한 문자들의 작은 차이를 인식기가 충분히 반영하기 어렵기 때문이다. 본 논문에서는 최근 주목 받고 있는 Support Vector Machine을 이용해 유사한 문자쌍을 검증하는 한글 인식 후처리 방법을 제안한다. 제안하는 방법은, 대부분의 문자 유사쌍이 한 두개의 자모만이 상이한 점에 착안하여 자모 단위로 문자 유사쌍을 구분한다. 기존 랜덤그래프를 이용한 한글 인식기를 이용하여 자모 분할을 수행하고, Support Vector Machine을 이용하여 분할된 결과를 검증한다. 제안한 방법은 유사쌍 구분에 중요한 자모만을 선택적으로 고려하여, 기존 한글 인식기의 부족한 점을 보완한다. 실험 결과, 자주 혼동되는 문자쌍들의 인식 오류가 정정되는 것을 볼 수 있었으며 그에 따라 한글 인식의 전체 성능이 향상되었다.

• Journal of the Korea Society of Computer and Information
• /
• v.14 no.8
• /
• pp.11-18
• /
• 2009

This paper proposes the way of improving learning speed in Levenberg-Marquardt algorithm using the principal submatrix of Jacobian matrix. The Levenberg-Marquardt learning uses Jacobian matrix for Hessian matrix to get the second derivative of an error function. To make the Jacobian matrix an invertible matrix. the Levenberg-Marquardt learning must increase or decrease ${\mu}$ and recalculate the inverse matrix of the Jacobian matrix due to these changes of ${\mu}$. Therefore, to have the proper ${\mu}$, we create the principal submatrix of Jacobian matrix and set the ${\mu}$ as the eigenvalues sum of the principal submatrix. which can make learning speed improve without calculating an additional inverse matrix. We also showed that our method was able to improve learning speed in both a generalized XOR problem and a handwritten digit recognition problem.

• The Transactions of the Korea Information Processing Society
• /
• v.2 no.6
• /
• pp.995-1001
• /
• 1995

The structure of a typical pattern recognition consists a pre-processing, a feature extraction(algorithm) and classification or recognition. In classification, when widely varying patterns exist in same category, we need the clustering which organize the similar patterns. Clustering algorithm is two approaches. Firs, statistical approaches which are k-means, ISODATA algorithm. Second, neural network approach which is T. Kohonen's LVQ(Learning Vector Quantization). Nikhil R. Palet al proposed the GLVQ(Generalized LVQ, 1993). This paper suggest the efficient feature extraction methods of handwritten numerals in GLVQ clustering network. We use the handwritten numeral data from 21's authors(ie, 200 patterns) and compare the proportion of misclassified patterns for each feature extraction methods. As results, when we use the projection combination method, the classification ratio is 98.5%.

• Journal of the Korean Institute of Telematics and Electronics C
• /
• v.36C no.4
• /
• pp.87-94
• /
• 1999

This paper presents a study on the off-line handwritten Hangul (Korean) character recognition using the partially connected neural network (PCNN), which is based on partial connections between the input receptive fields and the hidden nodes. The hidden nodes of three PCNNs have ten receptive fields and different input feature sets. And we introduce modular partially connected neural network (MPCNN), The MPCNN combines three PCNNs with a merging network. The learning scheme of the proposed networks is composed of two steps: PCNN learning step and the merging step of combining three PCNN s. In the merging step, another merging PCNN network is introduced and trained by regarding the hidden output of each PCNN as a new input feature vector. The performance of the proposed classifier is verified on the recognition of 18 off-line handwritten Hangul characters widely used in business cards in Korea.

• Journal of the Korea Society of Computer and Information
• /
• v.10 no.4 s.36
• /
• pp.385-396
• /
• 2005

In this paper, a new structural approach to on-line signature verification is presented. A primitive pattern is defined as a part segmented by a local minimal position of speed. And a structural description of signature is composed of subpatterns which are defined as such forms as rotation shape, cusp shape and bell shape, acquired by composition of the primitives regarding the directional changes. As the matching method to find identical parts between two signatures, a modified DP(dynamic programming) matching algorithm is presented. And also, variation and complexity of local parts are computed from the training samples, and reference model and decision boundary are derived from these. Error rate, execution time and memory usage are compared among the functional approach, the parametric approach and the proposed structural approach. It is found that the average error rate can be reduced from 14.2% to 4.05% when the local parts of a signature are weighted and the complexity is used as a factor of decision threshold. Though the error rate is similar to that of functional approaches. time consumption and memory usage of the proposed structural approach are shown to be very effective.

• Applied Chemistry for Engineering
• /
• v.10 no.3
• /
• pp.367-373
• /
• 1999

Increasing consumer demand for recycle of most paper products is challenging manufacturers to use lower value raw materials while maintaining high quality. This study was carried out to investigate the composition of old corrugated containers (OCC) and to manufacture printing and writing papers of over 85% (ISO) brightness by ECF and TCF bleaching from OCC. Fiber length of American old corrugated container (AOCC) was longer than that of Korean old corrugated container (KOCC) in fiber length. The former was composed of unbleached kraft pulp (UKP) of about 80 percentage, the latter about 20 percentage. OCC is feasible for manufacturing of pulps i.e., AOCC is for printing and writing papers and KOCC is for whiteboards, in the aspects of brightness and mechanical properties on the other hand, the cost of manufacturing upgraded papers by OCC is high because of bleaching chemicals. Neverthless, considering various aspects, especially environment, there are many advantages in recycling. In case of the treated AOCC pulp, over 85% (ISO) brightness can be achieved through TCF or ECF bleaching step, while the treated KOCC pulp did not allow high brightness through TCF bleaching. The tensile index of the bleached KOCC pulp was lower than that of the bleached AOCC pulp, but the burst and tear index of KOCC was higher than that of AOCC.