• Journal of Information Technology Services
• /
• v.21 no.6
• /
• pp.39-51
• /
• 2022

As the 4th Industrial Revolution emerged as a key to improving national competitiveness, OCR technology, one of the major technologies in the 4th industry is in the spotlight. Since characters in various images contain a lot of information, OCR technology for recognizing these characters has evolved into technology used in many industries. In this paper, trends in OCR technology were identified and predicted using thesis data published in 'RISS' and patent data by International patent classification (IPC) under the theme of Optical character recognition (OCR). For patent data 20,000 patents related to OCR technology from 2002 to 2020 were used as data, and 432 papers from 2012 to 2022 were used as data. Through time-series analysis, each patent data and thesis data were investigated since when OCR technology has developed, and various keyword analysis predicted which technology will be used in the future. Finally, the direction of future OCR technology development was presented through network association analysis with patent data and thesis data.

• The Journal of the Convergence on Culture Technology
• /
• v.10 no.2
• /
• pp.453-458
• /
• 2024

Optical Character Recognition is the technology that recognizes text in images and converts them into digital format. Deep learning-based OCR is being used in many industries with large quantities of recorded data due to its high recognition performance. To improve medical services, deep learning-based OCR was actively introduced by the medical industry. In this paper, we discussed trends in OCR engines and medical OCR and provided a roadmap for development of medical OCR. By using natural language processing on detected text data, current medical OCR has improved its recognition performance. However, there are limits to the recognition performance, especially for non-standard handwriting and modified text. To develop advanced medical OCR, databaseization of medical data, image pre-processing, and natural language processing are necessary.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.18 no.6
• /
• pp.269-276
• /
• 2018

OCR(Optical Character Recognition) using computers has been developed for 20 years and applied to various fields such as parking management based on the recognition of license plates of cars. This technology was also used in the development of our smart OCR-based household ledger. In order to improve filling the purchase history into a smartphone based household account book, we can take pictures of receipts with the smarphone camera and automatically organize the purchase list. In this process, the recognition rate of the characters of the receipt image is not high enough with OCR technology. We could improve the rate by applying the image processing technology and adjusting the contrast of the receipt image. The rate improved from 89% to 92.5%.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.17 no.4
• /
• pp.190-197
• /
• 2024

The O2O store model dissolves the boundaries between online and offline platforms, providing significant convenience to customers. To effectively operate such platforms, small business owners must provide necessary information in digital format. Specifically, the process of digitizing Korean menus manually can lead to multiple issues, and the use of OCR technology often results in high error rates due to the low accuracy in recognizing Korean. In response, this paper proposes an enhanced OCR model based on the popular EasyOCR framework, aimed at improving the recognition accuracy of Korean. The proposed model integrates the structural advantages of VGG and ResNet, and incorporates an attention mechanism to significantly improve the recognition performance of Korean. Moreover, experimental results indicate that the proposed model achieved approximately a 3.5% improvement in accuracy and around a 1% improvement in both confidence score and normalized edit distance compared to EasyOCR. Therefore, this demonstrates that the proposed method effectively addresses the existing challenges.

• Annual Conference on Human and Language Technology
• /
• 2019.10a
• /
• pp.470-474
• /
• 2019

한글 OCR 성능을 높이기 위해 딥러닝 모델을 활용하여 문자인식 부분을 개선하고자 하였다. 본 논문에서는 폰트와 사전데이터를 사용해 딥러닝 모델 학습을 위한 한글 문장 이미지 데이터를 직접 생성해보고 이를 활용해서 한글 문장의 OCR 성능을 높일 다양한 모델 조합들에 대한 실험을 진행했다. 딥러닝 모델은 STR(Scene Text Recognition) 구조를 사용해 변환, 추출, 시퀀스, 예측 모듈 각 24가지 모델 조합을 구성했다. 딥러닝 모델을 활용한 OCR 실험 결과 한글 문장에 적합한 모델조합은 변환 모듈을 사용하고 시퀀스와 예측 모듈에는 BiLSTM과 어텐션을 사용한 모델조합이 다른 모델 조합에 비해 높은 성능을 보였다. 해당 논문에서는 이전 한글 OCR 연구와 비교해 적용 범위를 글자 단위에서 문장 단위로 확장하였고 실제 문서 이미지에서 자주 발견되는 유형의 데이터를 사용해 애플리케이션 적용 가능성을 높이고자 한 부분에 의의가 있다.

• Annual Conference on Human and Language Technology
• /
• 2016.10a
• /
• pp.310-312
• /
• 2016

OCR(Optical Character Recognition)의 오류를 줄이기 위해 본 논문에서는 교정 어휘 쌍의 혼동 행렬(confusion matrix)과 나이브 베이즈 분류기($na{\ddot{i}}ve$ Bayes classifier)를 이용한 철자 교정 시스템을 제안한다. 본 시스템에서는 철자 오류 중 한글에 대한 철자 오류만을 교정하였다. 실험에 사용된 말뭉치는 한국어 원시 말뭉치와 OCR 출력 말뭉치, OCR 정답 말뭉치이다. 한국어 원시 말뭉치로부터 자소 단위의 언어 모델(language model)과 교정 후보 검색을 위한 접두사 말뭉치를 구축했고, OCR 출력 말뭉치와 OCR 정답 말뭉치로부터 교정 어휘 쌍을 추출하고, 자소 단위로 분해하여 혼동 행렬을 만들고, 이를 이용하여 오류 모델(error model)을 구축했다. 접두사 말뭉치를 이용해서 교정 후보를 찾고 나이브 베이즈 분류기를 통해 확률이 높은 교정 후보 n개를 제시하였다. 후보 n개 내에 정답 어절이 있다면 교정을 성공하였다고 판단했고, 그 결과 약 97.73%의 인식률을 가지는 OCR에서, 3개의 교정 후보를 제시하였을 때, 약 0.28% 향상된 98.01%의 인식률을 보였다. 이는 한글에 대한 오류를 교정했을 때이며, 향후 특수 문자와 숫자 등을 복합적으로 처리하여 교정을 시도한다면 더 나은 결과를 보여줄 것이라 기대한다.

• Journal of Korea Society of Industrial Information Systems
• /
• v.12 no.2
• /
• pp.37-48
• /
• 2007

In order to become a mega hub port, major ports all over the world are making every effort to enhance their productivity through efficiency of internal operation and introduction of the state-of-the-art technologies. They are not only installing various kinds of high-technology equipments but also introducing advanced technologies for the development of an effective gate system. Recently thanks to the appearance of RFID (radio frequency identification) and OCR (optical character recognition) technology, major container terminals are stewing up the automation of truck and container identification at the container luminal gate. This study aim to develop an automated gate system for identification task based on RFID and OCR technology. It will make mn effective gate operations in a container terminal.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2014.01a
• /
• pp.283-286
• /
• 2014

본 논문에서는 카메라형 OCR (Optical Character Reader) 기술을 이용해 오프라인 중고서점의 효율적 장서관리 시스템을 구축하기 위한 디지털 데이터화 관리시스템 방안을 제안한다. OCR은 광학적으로 인식할 수 있는 문자를 컴퓨터가 읽을 수 있도록 하는 기술이다. 원리적으로 문자 한 개를 수십 개의 모눈으로 분할해 특정한 모눈의 흑백 또는 자획형상 특징에 의해 문자를 판독한다. 이 논문에서는 OCR 기술을 활용함으로써 디지털 데이터화의 효과는 물론 적용 환경의 개선효과를 기대해 볼 수 있는 오프라인 중고서점 시장을 목표로 했다. 오프라인 중고서점에서 보유하고 있는 장서의 디지털 데이터화는 기업형 중고서점과의 경쟁에 있어서도 생존을 위해 필요한 요소이다. 카메라형 OCR 기술을 활용한 장서 디지털 데이터화는 오프라인 중고서점 판매자가 도서재고 검색 및 판매 관리 효율을 높이도록 도와줄 뿐 아니라, 도서판매 유형, 소비자 분석과 수요 예측을 가능하게 한다. 또한 소비자에게 오프라인 중고서점에서 보유하고 있는 희귀 장서와 중고서적들을 검색해 구입할 수 있는 편의를 제공할 것이다. 오프라인 중고서점 판매를 촉진하고 활성화시킨다면 출판의 선순환적 구조를 만드는 데 기여할 것으로 예상된다.

• Geomechanics and Engineering
• /
• v.15 no.4
• /
• pp.987-995
• /
• 2018

In this study, a relationship between small-strain shear modulus ($G_{max}$) and overconsolidation ratio (OCR) based on shear wave velocity ($V_S$) measurement was established to identify the stress history of centrifuge model ground. A centrifuge test was conducted in various centrifugal acceleration levels including loading and unloading sequences to cause various stress histories on centrifuge model ground. The $V_S$ and vertical effective stress were measured at each level of acceleration. Then, a sensitivity analysis was conducted using testing data to ensure the suitability of OCR function for the tested cohesionless soils and found that OCR can be estimated based on $V_S$ measurements irrespective of normally-consolidated or overconsolidated loading conditions. Finally, the developed $G_{max}$-OCR relationship was applied to centrifuge models constructed and tested under various induced stress-history conditions. Through a series of tests, it was concluded that the induced stress history on centrifuge model by compaction, g-level variation, and past overburden load can be analysed quantitatively, and it is convinced that the OCR evaluation technique will contribute to better interpret the centrifuge test results.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.15 no.2
• /
• pp.51-60
• /
• 2020

Optical character recognition (OCR) has been studied for decades because it is very useful in a variety of places. Nowadays, OCR's performance has improved significantly due to outstanding deep learning technology. Thus, there is an increasing demand for commercial-grade but affordable OCR systems. We have developed a low-cost, high-performance OCR system for the industry with the cheapest embedded developer kit that supports GPU acceleration. To achieve high accuracy for industrial use on limited computing resources, we chose a state-of-the-art text recognition algorithm that uses an end-to-end deep learning network as a baseline model. The model was then improved by replacing the feature extraction network with the best one suited to our conditions. Among the various candidate networks, EfficientNet-B3 has shown the best performance: excellent recognition accuracy with relatively low memory consumption. Besides, we have optimized the model written in TensorFlow's Python API using TensorFlow-TensorRT integration and TensorFlow's C++ API, respectively.