• Journal of Intelligence and Information Systems
• /
• v.26 no.2
• /
• pp.1-25
• /
• 2020

In this paper, we suggest an application system architecture which provides accurate, fast and efficient automatic gasometer reading function. The system captures gasometer image using mobile device camera, transmits the image to a cloud server on top of private LTE network, and analyzes the image to extract character information of device ID and gas usage amount by selective optical character recognition based on deep learning technology. In general, there are many types of character in an image and optical character recognition technology extracts all character information in an image. But some applications need to ignore non-of-interest types of character and only have to focus on some specific types of characters. For an example of the application, automatic gasometer reading system only need to extract device ID and gas usage amount character information from gasometer images to send bill to users. Non-of-interest character strings, such as device type, manufacturer, manufacturing date, specification and etc., are not valuable information to the application. Thus, the application have to analyze point of interest region and specific types of characters to extract valuable information only. We adopted CNN (Convolutional Neural Network) based object detection and CRNN (Convolutional Recurrent Neural Network) technology for selective optical character recognition which only analyze point of interest region for selective character information extraction. We build up 3 neural networks for the application system. The first is a convolutional neural network which detects point of interest region of gas usage amount and device ID information character strings, the second is another convolutional neural network which transforms spatial information of point of interest region to spatial sequential feature vectors, and the third is bi-directional long short term memory network which converts spatial sequential information to character strings using time-series analysis mapping from feature vectors to character strings. In this research, point of interest character strings are device ID and gas usage amount. Device ID consists of 12 arabic character strings and gas usage amount consists of 4 ~ 5 arabic character strings. All system components are implemented in Amazon Web Service Cloud with Intel Zeon E5-2686 v4 CPU and NVidia TESLA V100 GPU. The system architecture adopts master-lave processing structure for efficient and fast parallel processing coping with about 700,000 requests per day. Mobile device captures gasometer image and transmits to master process in AWS cloud. Master process runs on Intel Zeon CPU and pushes reading request from mobile device to an input queue with FIFO (First In First Out) structure. Slave process consists of 3 types of deep neural networks which conduct character recognition process and runs on NVidia GPU module. Slave process is always polling the input queue to get recognition request. If there are some requests from master process in the input queue, slave process converts the image in the input queue to device ID character string, gas usage amount character string and position information of the strings, returns the information to output queue, and switch to idle mode to poll the input queue. Master process gets final information form the output queue and delivers the information to the mobile device. We used total 27,120 gasometer images for training, validation and testing of 3 types of deep neural network. 22,985 images were used for training and validation, 4,135 images were used for testing. We randomly splitted 22,985 images with 8:2 ratio for training and validation respectively for each training epoch. 4,135 test image were categorized into 5 types (Normal, noise, reflex, scale and slant). Normal data is clean image data, noise means image with noise signal, relfex means image with light reflection in gasometer region, scale means images with small object size due to long-distance capturing and slant means images which is not horizontally flat. Final character string recognition accuracies for device ID and gas usage amount of normal data are 0.960 and 0.864 respectively.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.8 no.2
• /
• pp.207-211
• /
• 2007

It is a quite quality concerning to control the temperature of single crystalline growth as it does when we get most of heat treating products. It is also important factor to control the temperature when we make the $Al_2O_3$(single crystalline) used to artificial jewels, glass of watches, and heat resistant transparent glasses. Thus, it is a major interest to get the proper temperature in accordance with the time process while we are making mixture of oxygen and hydrogen to have the right temperature. In this paper, we will study of electrical valve positioning system with DC-Motor for the gas mixture to improve the quality of products.

• Journal of the Korean Society for information Management
• /
• v.29 no.2
• /
• pp.155-171
• /
• 2012

In text categorization, core terms of an input document are hardly selected as classification features if they do not occur in a training document set. Besides, synonymous terms with the same concept are usually treated as different features. This study aims to improve text categorization performance by integrating synonyms into a single feature and by replacing input terms not in the training document set with the most similar term occurring in training documents using Wikipedia. For the selection of classification features, experiments were performed in various settings composed of three different conditions: the use of category information of non-training terms, the part of Wikipedia used for measuring term-term similarity, and the type of similarity measures. The categorization performance of a kNN classifier was improved by 0.35~1.85% in $F_1$ value in all the experimental settings when non-learning terms were replaced by the learning term with the highest similarity above the threshold value. Although the improvement ratio is not as high as expected, several semantic as well as structural devices of Wikipedia could be used for selecting more effective classification features.

• Journal of KIISE:Computing Practices and Letters
• /
• v.11 no.1
• /
• pp.26-35
• /
• 2005

Generally, we employ FSMs for the design of controllers in digital systems. FSMs are Implemented with state diagrams generated from control flow. With HDL, we design and verify FSMs based on state diagrams. As the number of states in the system increases, the verification or modification processes become complicated, error prone and time consuming. In this paper, we propose a control flow oriented hardware description language at the register transfer level called Cycle-C. Cycle-C describes FSMs with timing information and control How intensive algorithms. The Cycle-C description is automatically converted into FSMs in the form of synthesizable RTL VHDL. In experiments, we design FSMs for control intensive interface circuits. There is little area difference between Cycle-C design and manual design. In addition, Cycle-C design needs only 10~50% of the number lines of manual RTL VHDL designs.

• Journal of KIISE:Computer Systems and Theory
• /
• v.35 no.7
• /
• pp.340-353
• /
• 2008

A computational color palette extraction model is introduced to describe paint brush objectively and efficiently. In this model, a color palette is defined as a minimum set of colors in which a painting can be displayed within error allowance and extracted by the two step processing of color grouping and major color extraction. The color grouping controls the resolution of colors adaptively and produces a basic color set of given painting images. The final palette is obtained from the basic color set by applying weighted k-means clustering algorithm. The extracted palettes from several famous painters are displayed in a 3-D color space to show the distinctive palette styles using RGB and CIE LAB color models individually. And the two experiments of painter classification and color transform of photographic image has been done to check the performance of the proposed method. The results shows the possibility that the proposed palette model can be a computational color analysis metric to describe the paint brush, and can be a color transform tool for computer graphics.

• Journal of Broadcast Engineering
• /
• v.8 no.1
• /
• pp.54-62
• /
• 2003

This paper presents a Soccer Game 3D Reencatment System which reenact the Important scene like getting a goal with image processing and computer graphics technologies. KBS Research Institute of Technology has developed the 3D Reenactment System of Soccer Game called ‘VPlay' to provide TV viewers with fresh images in soccer games. Vplay generates the reenactment of exciting and important soccer scenes by using computer graphics. Vplay extracts legion of players from video with color information, and then computes precise positions of players on the ground by using global motion estimation model and playground axis transformation model. The results are applied to locomotion generation module that generates the locomotion of virtual characters automatically. Using predefined motion and model library, Vplay reenacts the important scene in a quick and convenient manner Vplay was developed for live broadcasting of soccer games that demands rapid producing time and was used efficiently during past WorldCup and Asian Game.

• Journal of Digital Contents Society
• /
• v.19 no.2
• /
• pp.237-243
• /
• 2018

Most of public underground facilities are installed under the ground, thus it is difficult to recognize the accurate location even with the drawings. Studies are conducted to understand exact position of underground facilities using augmented reality. However, in those studies, establishing of additional 3D object model database is needed when AR system is used at field. Because most of public underground facility information are established as 2 dimensional. In this study, AR system is developed as mobile application which can use original 2D underground facility data to transfer 3D AR data automatically without additional 3D database establishment.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2010.05a
• /
• pp.1916-1920
• /
• 2010

영상수위계는 카메라에 의해서 수위표를 촬영하여 촬영된 영상을 처리하여 수위값으로 변환하여 자동적으로 수위를 측정하는 장비이다. 이 수위계는 기존 수위측정 장비인 부자식, 압력식, 기포식, 초음파식, 레이다식과는 달리 수위표를 촬영한 영상으로부터 수위를 직접 눈으로 확인할 수 있는 장점이 있다. 이로 인해 영상자료로부터 측정된 수위를 검증할 수 있어 수위측정의 정확도를 향상시킬 수 있다. 그리고 수위표 영상과 더불어 관측지점 주변의 전체 영상을 동시에 촬영하여 실시간으로 전송하기 때문에 홍수시 하천 상황에 대한 모니터링 목적으로 사용될 수 있다. 영상수위계 시스템은 크게 메인제어기, 전원부, 서버부 및 카메라부로 구성되어 있다. 현재 운영되고 있는 시스템은 전원장치 리셋시 전 시스템을 리셋해야하고, 상전 단전시 전 시스템이 off된다. 그리고 별도의 통신모듈을 사용하여 장비간 통신이 이루어지고 있다. 또한 카메라부에는 렌즈와 팬/틸트를 제어하기 위한 별도의 장비가 포함되어 있고, 백색 LED 조명이 사용되어 야간에 수위인식주기마다 조명이 on/off 되고 있다. 위와 같은 전원장치의 운영으로 시스템을 안정적으로 운영할 수 없다. 그리고 수위 오인식을 최소화하기 위해서는 연속적인 수위 인식이 필요하지만, 백색 LED조명과 1초에 2프레임을 캡쳐하는 비디오 캡쳐방식에 의해 시스템을 상시로 운영하는 것이 곤란하다. 현재 운영 중인 영상수위계를 안정적으로 운영할 수 있도록 장비별로 제어가 가능하도록 전원 제어장치를 개발하였고, 상전 단전시 최소 30분 정도 전원을 유지할 수 있도록 무정전전원장치(UPS)를 설치하였으며, 측정자료의 저장장치를 하드디스크 타입에서 Flash SSD 메모리 타입으로 교체하였다. 또한 영상수위계 시스템을 상시 운영할 수 있도록 백색 LED조명을 적외선 LED조명으로 교체하였고, 1초에 1회 수위를 인식하도록 수위인식주기와 1초에 25프레임 캡쳐할 수 있도록 비디오 캡쳐방식을 개선하였다. 위와 같은 시스템의 개선으로 시스템을 안정적으로 운영할 수 있게 되어 시스템 고장에 의해 발생하는 수위 결측을 감소시킬 수 있고, 시스템의 상시 운영으로 수 위 오인식을 최소화시킬 수 있을 것으로 판단된다.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.27 no.2
• /
• pp.195-202
• /
• 2009

As the generation and application of spatial information is gradually expanded not only in traditional surveying fields but also a CNS and an ITS recently. The Accuracy and the newest of data grow to be an important element. But digital map is updated with system based tile. So, it is hard to get the newest of data and to be satisfied with user requirements. In this study, management system is developed to manage feature change efficiently using bid informations from NaraJangter which service the bid informations. A construction works with change possibility of feature from bid informations are classified and are made DB. And the DB is used as the feature change forecast informations. Also, It is converted from bid information of text form to positioning informations connected to spatial information data. If this system is made successfully, this system contributes to reduce the cost for the update of digital map and to take the newest date of spatial informations.

• Journal of KIISE:Software and Applications
• /
• v.32 no.3
• /
• pp.202-214
• /
• 2005

The semantic web is the web paradigm that represents not general link of documents but semantics and relation of document. In addition it enables software agents to understand semantics of documents. We propose a semantic search based on inference with ontologies, which has the following characteristics. First, our search engine enables retrieval using explicit ontologies to reason though a search keyword is different from that of documents. Second, although the concept of two ontologies does not match exactly, can be found out similar results from a rule based translator and ontological reasoning. Third, our approach enables search engine to increase accuracy and precision by using explicit ontologies to reason about meanings of documents rather than guessing meanings of documents just by keyword. Fourth, domain ontology enables users to use more detailed queries based on ontology-based automated query generator that has search area and accuracy similar to NLP. Fifth, it enables agents to do automated search not only documents with keyword but also user-preferable information and knowledge from ontologies. It can perform search more accurately than current retrieval systems which use query to databases or keyword matching. We demonstrate our system, which use ontologies and inference based on explicit ontologies, can perform better than keyword matching approach .