• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.20 no.9
• /
• pp.1657-1665
• /
• 2016

Various fields are being researched through Deep Learning using CNN(Convolutional Neural Network) and these researches show excellent performance in the image recognition. In this paper, we provide streamlined GoogLeNet of CNN architecture that is capable of learning a large-scale Korean character database. The experimental data used in this paper is PHD08 that is the large-scale of Korean character database. PHD08 has 2,187 samples for each character and there are 2,350 Korean characters that make total 5,139,450 sample data. As a training result, streamlined GoogLeNet showed over 99% of test accuracy at PHD08. Also, we made additional Korean character data that have fonts that are not in the PHD08 in order to ensure objectivity and we compared the performance of classification between streamlined GoogLeNet and other OCR programs. While other OCR programs showed a classification success rate of 66.95% to 83.16%, streamlined GoogLeNet showed 89.14% of the classification success rate that is higher than other OCR program's rate.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.10 no.8
• /
• pp.1869-1876
• /
• 2009

Scanner which is a kind of optical devices is used to convert paper documents into document image files. The assessment of scanned document image is performed to check if there are any modification on document image files in scanning process. In assessment of scanned documents, user checks the degree of skew, noise, folded state and etc This paper proposed to how to measure geometrical modifications of document image in scanning process. In this study, we check the degree of modification in document image file by image processing and we compare the evaluation value which means the degree of modification in each items with OCR success ratio in a document image file. To analyse the correlation between OCR success ratio and the evaluation value which means the degree of modification in each items, we apply Pearson Correlation Coefficient and calculate weight value for each items to score total evaluation value of image modification degrees on a image file. The document image which has high rating score by proposed method also has high OCR success ratio.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.30B no.8
• /
• pp.33-40
• /
• 1993

This paper presents a hardware for image alignment based on proposed new algorithm which can align a small misaligned document image simply by one transformation with a parallel shifting of pixels. This hardware is simulated with VHDL and estimated to be about 65 ms to align an image made up of 380 by 480 pixels. Also, we will demonstrate the effectiveness of the proposed image alignment algorithm in OCR system by comparing its characteristics with those of the existing image rotation algorithms.

• KIEE International Transactions on Power Engineering
• /
• v.5A no.4
• /
• pp.357-365
• /
• 2005

This research investigates the influences of distributed generations (DG), which are interconnected to the bus by the dedicated lines, on the overcurrent relays (OCR) of the neighboring distribution feeders and also proposes a novel method to reduce the negative effects on the feeder protection. Due to the grid connected DG, the entire short-circuit capacity of the distribution networks increases, which may raise the current of the distribution feeder during normal operations as well as fault conditions. In particular, during the switching period for loop operation, the current level of the distribution feeder can be larger than the pickup value for the fault of the feeder's OCR, thereby causing the OCR to perform a mal-operation. This paper proposes the adaptive setting algorithm for the OCR of the distribution feeders having the neighboring dedicated feeders for the DG to prevent the mal-operations of the OCR under normal conditions. The proposed method changes the pickup value of the OCR by adapting the power output of the DG monitored at the relaying point in the distribution network. We tested the proposed method with the actual distribution network model of the Hoenggye substation at the Korea Electric Power Co., which is composed of five feeders supplying the power to network loads and two dedicated feeders for the wind turbine generators. The simulation results demonstrate that the proposed adaptive protection method could enhance the conventional OCR of the distribution feeders with the neighboring dedicated lines for the DG.

• 제어로봇시스템학회:학술대회논문집
• /
• 1997.10a
• /
• pp.1364-1366
• /
• 1997

To operate electric power systems economically and stably, and to supply the electric power of good quality, it is necessary that the measured information (current, voltage, and so on) be detected and transmutted with high reliability and high effieincy. For the reason, the optical magnetic field sensor is possible to rapidly detect to over current and recover when electric power line have fault. In addition, the optical sensor have no electro magnetic distortion and no electric insulation. In this study, we designed OCR(Over Current Relay) using optical sensor. The designed OCR using optical sensor was measured characteristic and compared contentional OCR. This system have highest optical advantages and reliability.

• The Journal of the Korea institute of electronic communication sciences
• /
• v.18 no.5
• /
• pp.989-998
• /
• 2023

As diseases are already increasing due to aging population and changes in eating habits, interest in digital healthcare is increasing. According to the 2020 Elderly Status Survey, 84% of South Koreans aged 65 and older suffer from at least one chronic disease, and the number and duration of drugs taken by the elderly are also increasing. Effective management that enhances compliance of chronic disease patients can prevent the onset of complications, thereby averting progression to severe illnesses. Thus, a proper medication-taking habit is crucial. This paper proposes a medication information retrieval system using OCR technology. By leveraging Google Cloud Vision API, the system detects and recognizes the names of medicines. Once recognized, the medication name is searched in a database to provide users with medication information and medication schedule management services. By providing accurate medication information through the search, it is possible to induce changes in medication methods and habits. By eliminating the inconvenience of direct input through OCR technology, we anticipate enhancing user convenience by promptly delivering information.

• Journal of IKEEE
• /
• v.24 no.4
• /
• pp.1086-1092
• /
• 2020

An autonomous agent for real world should be able to recognize text in scenes. With the advancement of deep learning, various DNN models have been utilized for transformation, feature extraction, and predictions. However, the existing state-of-the art STR (Scene Text Recognition) engines do not achieve the performance required for real world applications. In this paper, we introduce a performance-improvement method through an add-on composed of an OCR (Optical Character Recognition) engine and a classifier for STR engines. On instances from IC13 and IC15 datasets which a STR engine failed to recognize, our method recognizes 10.92% of unrecognized characters.

• The KIPS Transactions:PartD
• /
• v.18D no.1
• /
• pp.67-80
• /
• 2011

On an automated business document processing system maintaining financial data, errors on query based retrieval of numbers are critical to overall performance and usability of the system. Automatic spelling correction methods have been emerged and have played important role in development of information retrieval system. However scope of the methods was limited to the symbols, for example alphabetic letter strings, which can be reserved in the form of trainable templates or custom dictionary. On the other hand, numbers, a sequence of digits, are not the objects that can be reserved into a dictionary but a pure markov sequence. In this paper we proposed a new OCR model for spelling correction for numbers using the multiple streams and the context based correction on top of probabilistic information retrieval framework. We implemented the proposed error correction model as a sub-module and integrated into an existing automated invoice document processing system. We also presented the comparative test results that indicated significant enhancement of overall precision of the system by our model.

• 제어로봇시스템학회:학술대회논문집
• /
• 2002.10a
• /
• pp.42.4-42
• /
• 2002

With the research of simplified neural networks, we propose an AND/OR neural network; a kind of brief, fast network Then, we present an OCR solution that equip the network in one-chip FPGA and design it by using HDL. We selected the representative hexadecimal character as the recognition feature class and used a Feature Vector Recognition Method in the statistic pattern recognition. The result feature vector was encoded into a 7 bit array and inputted into the AND/OR network to finish learning.

• Journal of Environmental Science International
• /
• v.20 no.8
• /
• pp.943-951
• /
• 2011

We investigated the effects of temperature changes on the oxygen consumption rhythm in Japanese eels, Anguilla japonica, using an automatic intermittent flow respirometer (AIFR). The endogenous rhythm of the oxygen consumption rate (OCR) in the eels (n = 18; 44-74 cm, 145-690 g), freshly collected by bag net from estuaries, was nearly synchronous with the tidal pattern of the estuarine collection site. The magnitude of mean OCR (mOCR) of eels showed variable range of 82.2 - 116.5 ml $O_2\;kg^{-1}ww \;h^{-1}$ under constant conditions. In case of increasing temperature from 25 to $38^{\circ}C$, the OCR of eels exhibited a gradually increasing trend with a rhythmic pattern until $36^{\circ}C$. Above $36^{\circ}C$ the rhythms of the OCR dampened and the OCR decreased rapidly at around $36-37^{\circ}C$. The OCR of the eels exhibited the maximum value at $38^{\circ}C$, and then it sharply decreased. The results suggested that the critical thermal maximum (CTM) regarding the endogenous rhythms of the eels was at around $36-37^{\circ}C$ when water temperature increased at $0.5^{\circ}C$/14 h following the acclimation at $25^{\circ}C$. In case of decreasing temperature ($0.5^{\circ}C$/14 h) from 25 to $0^{\circ}C$, the OCR of the eels displayed a abrupt decrease up to $23^{\circ}C$, and between at 23 and $20^{\circ}C$, there was an agitation which showed a slight increase in the OCR with a duration of 1-2 days. Below $9^{\circ}C$, the OCR rhythm of the eels showed a constant state regardless of temperature decreasing. These results suggest that the Japanese eel has an upper incipient lethal temperature at $36^{\circ}C$, with a lower thermal limit at $9^{\circ}C$. The biochemical aspects of the eels influenced by water temperature need to be further studied.