• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.16 no.5
• /
• pp.892-898
• /
• 2012

Rough classification methods are used to improving the recognition speed in many character recognition problems. In this case, some irreversible result can occur by an error in rough classification. Methods for duplicating each model in several classes are used in order to reduce this risk. But the errors by rough classfication can not be completely ruled out by these methods. In this paper, an recognition method is proposed to increase speed that matches models selectively without any increase in error. This method constructs a model graph using similarity between models. Then a search process begins from a particular point in the model graph. In this process, matching of unnecessary models are reduced that are not similar to the input pattern. In this paper, the proposed method is applied to the recognition problem of handwriting numbers and upper/lower cases of English alphabets. In the experiments, the proposed method was compared with the basic method that matches all models with input pattern. As a result, the same recognition rate, which has shown as the basic method, was obtained by controlling the out-degree of the model graph and the number of maintaining candidates during the search process thereby being increased the recognition speed to 2.45 times.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.10 no.1
• /
• pp.63-68
• /
• 2010

Recently, to accelerate the Ubiquitous generation, the input interface of the mobile machinery and tools are actively being researched. In addition with the existing interfaces such as the keyboard and curser (mouse), other subdivisions including the handwriting, voice, vision, and touch are under research for new interfaces. Especially in the case of small-sized mobile machinery and tools, there is a increasing need for an efficient input interface despite the small screens. This is because, additional installment of other devices are strictly limited due to its size. Previous studies on handwriting recognition have generally been based on either two-dimensional images or algorithms which identify handwritten data inserted through vectors. Futhermore, previous studies have only focused on how to enhance the accuracy of the handwriting recognition algorithms. However, a problem arisen is that when an actual handwriting is inserted, the user must select the classification of their characters (e.g Upper or lower case English, Hangul - Korean alphabet, numbers). To solve the given problem, the current study presents a system which distinguishes different languages by analyzing the form/shape of inserted handwritten characters. The proposed technique has treated the handwritten data as sets of vector units. By analyzing the correlation and directivity of each vector units, a more efficient language distinguishing system has been made possible.

• The Journal of the Korea Contents Association
• /
• v.18 no.7
• /
• pp.495-502
• /
• 2018

The advent of deep learning technology has made rapid progress in handwritten letter recognition in many languages. Handwritten Chinese recognition has improved to 97.2% accuracy while handwritten Japanese recognition approached 99.53% percent accuracy. Hanguel handwritten letters have many similar characters due to the characteristics of Hangeul, so it was difficult to recognize the letters because the number of data was small. In the handwritten Hanguel recognition using Hybrid Learning, it used a low layer model based on lenet and showed 96.34% accuracy in handwritten Hanguel database PE92. In this paper, 98.64% accuracy was obtained by organizing deep CNN (Convolution Neural Network) in handwritten Hangeul recognition. We designed a new network for handwritten Hangeul data based on GoogLenet without using the data augmentation or the multitasking techniques used in Hybrid learning.

• Journal of Korea Society of Industrial Information Systems
• /
• v.6 no.1
• /
• pp.82-89
• /
• 2001

This paper presents recognition of unconstrained handwritten numeral strings using decision value generator, which is combined with both isolated digit identifier and recognizer designed with structural characteristics of digits. Numerical string recognition system is composed of three modules, which are pre-segmentation, segmentation and recognition. Pre-segmentation module classifies a numeral string into sub-images, which are isolated digit, touched digits or broken digit, using confidence value of decision value generator. Segmentation module segments touched digits using reliability value of decision value generator that will separate the leftmost digit from touched string of digits. Segmentation-based and segmentation-free methods have used for classification and segmentation, respectively. To evaluate proposed method, experiments have carried out with handwritten numeral strings of NIST SD19 and higher recognition performance than previous works has obtained with 96.7%.

• Journal of KIISE:Software and Applications
• /
• v.31 no.1
• /
• pp.8-19
• /
• 2004

In this paper, we propose an efficient method for locating destination address block on both of machine-Printed and handwritten Korean mail piece images. The proposed method extracts connected components from the binary mail piece image, generates text lines by merging them, and then groups the text fines into nine clusters. The destination address block is determined by selecting some clusters. Considering the geometric characteristics of address information on Korean mail piece, we split a mail piece image into nine areas with an equal size. The nine clusters are initialized with the center coordinate of each area. A modified Manhattan distance function is used to compute the distance between text lines and clusters. We modified the distance function on which the aspect ratio of mail piece could be reflected. The experiment done with live Korean mail piece images has demonstrated the superiority of the Proposed method. The success rate for 1, 988 testing images was about 93.56%.

• Journal of Korea Society of Industrial Information Systems
• /
• v.29 no.3
• /
• pp.13-22
• /
• 2024

Currently, with the proliferation of digital devices, the significance of handwritten texts in daily lives is gradually diminishing. As the use of keyboards and touch screens increase, a decline in Korean handwriting quality is being observed across a broad spectrum of Korean documents, from young students to adults. However, Korean handwriting still remains necessary for many documentations, as it retains individual unique features while ensuring readability. To this end, this paper aims to implement an application designed to improve and correct the quality of handwritten Korean script The implemented application utilizes the CRAFT (Character-Region Awareness For Text Detection) model for handwriting area detection and employs the VGG-Feature-Extraction as a deep learning model for learning features of the handwritten script. Simultaneously, the application presents the user's handwritten Korean script's reliability on a syllable-by-syllable basis as a recognition rate and also suggests the most similar fonts among candidate fonts. Furthermore, through various experiments, it can be confirmed that the proposed application provides an excellent recognition rate comparable to conventional commercial character recognition OCR systems.

• Proceedings of the Korea Contents Association Conference
• /
• 2005.11a
• /
• pp.384-388
• /
• 2005

Currently most of the online handwriting recongition system are using free style input method. However, it has disadvantages of ill-recongition. In this paper, we present a new online Korean HAndwriting recongition system(KOHA) which give a slice restriction and remove the ill-recongition. KOHA uses boundary lines of input window and the stenography is possible with KOHA. Also, KOHA has the advantage of Unistroke.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.04b
• /
• pp.565-567
• /
• 2000

필기 한글 인식에서 원래 패턴의 모양을 유지하는 골격선 추출은 중요하다. 세선화에 의존하는 기존 방법은 작은 잡음에 민감하다는 단점을 안고 있다. 본 논문은 필기 한글 패턴에 적합한 새로운 골격선 추출 방법을 제안한다. 먼저 한글 패턴은 T-접점과 B-접점이라는 두가지 모양 특징을 중심으로 분할할 수 있다는 관찰에 근거하여 유사블록으로 이루어진 부품 집합으로 분할한다. 또한 세 개 이상의 획이 복잡한 형태로 만나는 지점을 결합 부품으로 분할한다. 그런 다음, 각 부품에서 접점의 형태에 따라 결합 부품을 추가 탐지한다. 결합 부품과 인접한 부품들의 연관 관계에 따라 골격선을 구하고, 골격선의 연결성을 보장하기 위해서 선분 연장을 수행한다. 본 논문에서 기존의 방법과의 비교를 위해 다섯 가지 비교 기준을 설정하고, 이를 기반으로 비교 분석하였다. 본 논문에서 제안한 방법이 여러 기준에서 세선화-기반 방법보다 우수함을 보였다.

• Annual Conference on Human and Language Technology
• /
• 1993.10a
• /
• pp.295-301
• /
• 1993

제약이 없이 자유롭게 쓴 오프라인 필기체 한글을 인식하는 문제는 응용분야에 따른 도메인의 정보를 이용함으로써 보다 쉽게 접근할 수 있다. 본 연구는 오프라인 필기체 한글 인식을 위한 한 도메인으로 우편봉투를 대상으로 하였을 때, 우편번호가 할당된 지명과 건물명을 대상으로 글자의 종류와 빈도수를 통계 분석하였다. 분석 결과 가능한 한글 조합 11,172자중 403자만이 쓰이고 있음을 알았다. 이러한 정보는 자소 분할이 어려운 오프라인 필기체 한글 인식에 있어, 문자 단위 정합을 사용했을 때 인식속도 및 인식률 향상에 기여 할 것으로 생각된다.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.10b
• /
• pp.440-442
• /
• 2000

필기 한글 문자 인식을 위해서는 패턴을 구성하는 획 성분을 분석하는 작업이 필수적이다. 획 성분 추출을 위해 사용한 세선화 방법은 입력 영상을 왜곡하는 단점을 가지고 있다. 이를 극복하기 위하여 본 논문은 입력 영상을 왜곡하지 않고 의미 있는 부품 단위로 분할하는 방법을 제안한다. 의미 있는 부품이란 유사 볼록하게 분할된 영역을 의미한다. 분할 방법은 먼저 입력 영상에 볼록 헐 연산을 적용하여 오목 영역을 생성한다. 이 오목 영역에서 분할 기준(anchor point)점을 탐지하고 획의 반대편 외곽선 상에서 분할 끝(terminal point)점을 찾아 분할 경로를 구성하여 획을 분할한다. 모든 부품이 유사 볼록 조건을 만족할 때까지 위 과정을 반복 수행한다. 제안한 방법은 두 개의 파라미터만을 가지며 간단한 프로시져로 구성되어 있다. 또한 필기 한글 패턴뿐 아니라 여러 언어에 적용 가능하다는 장점을 갖는다.