• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.1
• /
• pp.13-19
• /
• 2019

The accuracy of machine learning is greatly affected by amount of learning data and quality of data. Collecting existing Web-based learning data has danger that data unrelated to actual learning can be collected, and it is impossible to secure data transparency. In this paper, we propose a method for collecting data directly in parallel by blocks in a block - chain structure, and comparing the data collected by each block with data in other blocks to select only good data. In the proposed system, each block shares data with each other through a chain of blocks, utilizes the All-reduce structure of Parallel-SGD to select only good quality data through comparison with other block data to construct a learning data set. Also, in order to verify the performance of the proposed architecture, we verify that the original image is only good data among the modulated images using the existing benchmark data set.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2014.10a
• /
• pp.725-728
• /
• 2014

Continue to have an increasingly violent crimes against children every year, and as you know all seriousness is classified as a felony. However, efforts to reduce the underlying crime is low. Therefore, it is necessary to solve this problem, the security system. Is to protect the children and adults that exist that can pose a threat to children to identify and monitor tracking method in this paper. Was based on a Korean standard body size of a person, such as keys, arm length, leg length, head vertical length, head width proposed method. Also, separate the adults and children through the comparison of the reference value, the ratio and the ratio of the number of pixels of the detected object, the proposed method. Processing speed is fast because it detects only a specific object region in the entire image in the handling method in the proposed method the five nine minutes. The advantage is to enable comparison of the specific object, through which there is.

• Design & Manufacturing
• /
• v.15 no.1
• /
• pp.26-31
• /
• 2021

As the aging ages, the disease also increases, and the development of AI technology and X-ray equipment used to treat patients' diseases is also progressing a lot. X-ray tube converts only 1% of electron energy into X-ray and 99% into thermal energy. Therefore, when the cooling time of the anode and the X-ray tube are frequently used in large hospitals, the amount of X-ray emission increases due to temperature rise, the image quality deteriorates due to the difference in X-ray dose, and the lifespan of the overheated X-ray tube may be shortened. Therefore, in this study, temperature rise and cooling time of 60kW, 75kW, and 90kW of X-ray tube anode input power were studied. In the X-ray Tube One shot 0.1s, the section where the temperature rises fastest is 0.03s from 0s, and it is judged that the temperature has risen by more than 50%. The section in which the temperature drop changes most rapidly at 20 seconds of cooling time for the X-ray tube is 0.1 seconds to 0.2 seconds, and it is judged that a high temperature drop of about 65% or more has occurred. After 20 seconds of cooling time from 0 seconds to 0.1 seconds of the X-ray tube, the temperature is expected to rise by more than 3.7% from the beginning. In particular, since 90kW can be damaged by thermal shock at high temperatures, it is necessary to increase the surface area of the anode or to require an efficient cooling system.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.3
• /
• pp.958-979
• /
• 2023

Due to laws, regulations, privacy, etc., between 70-90 percent of providers do not share medical data, forming a "data island". It is essential to collaborate across multiple institutions without sharing patient data. Most existing methods adopt distributed learning and centralized federal architecture to solve this problem, but there are problems of resource heterogeneity and data heterogeneity in the practical application process. This paper proposes a collaborative deep learning modelling method based on the blockchain network. The training process uses encryption parameters to replace the original remote source data transmission to protect privacy. Hyperledger Fabric blockchain is adopted to realize that the parties are not restricted by the third-party authoritative verification end. To a certain extent, the distrust and single point of failure caused by the centralized system are avoided. The aggregation algorithm uses the FedProx algorithm to solve the problem of device heterogeneity and data heterogeneity. The experiments show that the maximum improvement of segmentation accuracy in the collaborative training mode proposed in this paper is 11.179% compared to local training. In the sequential training mode, the average accuracy improvement is greater than 7%. In the parallel training mode, the average accuracy improvement is greater than 8%. The experimental results show that the model proposed in this paper can solve the current problem of centralized modelling of multicenter data. In particular, it provides ideas to solve privacy protection and break "data silos", and protects all data.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.15 no.4
• /
• pp.257-270
• /
• 2022

In this paper, we design a smart factory risk improvement system and risk improvement method using AI edge devices. The smart factory risk improvement system collects, analyzes, prevents, and promptly responds to the worker's work performance process in the smart factory using AI edge devices, and can reduce the risk that may occur during work with improving the defect rate when workers perfom jobs. In particular, based on worker image information, worker biometric information, equipment operation information, and quality information of manufactured products, it is possible to set an abnormal risk condition, and it is possible to improve the risk so that the work is efficient and for the accurate performance. In addition, all data collected from cameras and IoT sensors inside the smart factory are processed by the AI edge device instead of all data being sent to the cloud, and only necessary data can be transmitted to the cloud, so the processing speed is fast and it has the advantage that security problems are low. Additionally, the use of AI edge devices has the advantage of reducing of data communication costs and the costs of data transmission bandwidth acquisition due to decrease of the amount of data transmission to the cloud.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.46 no.1
• /
• pp.23-31
• /
• 2009

Recently, biometric techniques such as face recognition, finger-print recognition and iris recognition have been widely applied for various applications including door access control, finance security and electric passport. This paper presents the method of using finger-vein pattern for the personal identification. In general, when the finger-vein image is acquired from the camera, various conditions such as the penetrating amount of the infrared light and the camera noise make the segmentation of the vein from the background difficult. This in turn affects the system performance of personal identification. To solve this problem, we propose the novel and fast method for extracting the finger-vein region. The proposed method has two advantages compared to the previous methods. One is that we adopt a locally adaptive thresholding method for the binarization of acquired finger-vein image. Another advantage is that the simple morphological opening and closing are used to remove the segmentation noise to finally obtain the finger-vein region from the skeletonization. Experimental results showed that our proposed method could quickly and exactly extract the finger-vein region without using various kinds of time-consuming filters for preprocessing.

• Applied Chemistry for Engineering
• /
• v.8 no.6
• /
• pp.1014-1021
• /
• 1997

Studies were carried out on the preparation of several kinds of pulps from Sponge gourd fiber by KP, ASP, SP PAP, AP and addition of AQ pulping process. These unbeaten and beaten pulping fibers were observed their characteristics and fiber structure by SEM, FQA, Image analyzer and Micro projector. The results were summarized as follows; 1) The cooking condition which is the possible defibrilation of Shives are KP base($160^{\circ}C$, 2hr.), ASP base($155^{\circ}C$, 4hr.), PAP base($160^{\circ}C$, 1hr.). From the results, the kappa no. had the range of 12, 25, 10 each other. 2) The pulp yields of sponge gourd fiber obtained the range of KP 50~55%, ASP&60~70% and PAP 45~50%. SP base have the highest and contnets of KP&PAP base are much the same as woods. 3) Increasing amount of NaOH on Pulping was accelerated the defibrilation of Shives and was changed a morphology of pulping fiber quality such as fiber length, curl and kink index. 4) Addition of AQ on pulping process of sponge gourd fiber had a affect to raise the rate of delignification while protecting cellullosic components against degradation, especially defibrilation was very excellent, beated pulp much more easily and increased the fibrilation. 5) ASP system have higher bulk density, fiber bonding and protecting cellullosic components against degradation than KP or PAP. 6) The color reactions of the "C" stain solution showed blue or blue-gray with clean and transparency thin cell wall.

• Journal of Cadastre & Land InformatiX
• /
• v.52 no.2
• /
• pp.53-65
• /
• 2022

According to the advancement of the GNSS satellite positioning system, the module of hardware and operation software reflecting accuracy and economical efficiency is implemented in the user sector including the multi-channel GNSS receiver, the multi-frequency external antenna and the mobile app (App) base public positioning analysis software etc., and the multichannel GNSS RTK positioning of the active configuration method (DIY, Do it yourself) is possible according to the purpose of user. Especially, as the infrastructure of multi-GNSS satellite is expanded and the potential of expansion of utilization according to various modules is highlighted, interest in the utilization of multi-channel low-cost GNSS receiver module is gradually increasing. The purpose of this study is to review the multi-channel low-cost GNSS receivers that are appearing in the mass market in various forms and to analyze the utilization plan of the "address information facility investigation project" of the Ministry of Public Administration and Security by constructing the multi-channel low-cost GNSS positioning module based RTK survey system (hereinafter referred to as "multi-channel GNSS RTK module positioning system"). For this purpose, we constructed a low-cost "multi-channel GNSS RTK module positioning system" by combining related modules such as U-blox's F9P chipset, antenna, Ntrip transmission of GNSS observation data and RTK positioning analysis app through smartphone. Kinematic positioning was performed for circular trajectories, and static positioning was performed for address information facilities. The results of comparative analysis with the Static positioning performance of the geodetic receivers were obtained with 5 fixed points in the experimental site, and the good static surveying performance was obtained with the standard deviation of average ±1.2cm. In addition, the results of the test point for the outline of the circular structure in the orthogonal image composed of the drone image analysis and the Kinematic positioning trajectory of the low cost RTK GNSS receiver showed that the trajectory was very close to the standard deviation of average ±2.5cm. Especially, as a result of applying it to address information facilities, it was possible to verify the utility of spatial information construction at low cost compared to expensive commercial geodetic receivers, so it is expected that various utilization of "multi-channel GNSS RTK module positioning system"

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.2
• /
• pp.69-80
• /
• 1999

Recently, social requirements of personal identification techniques are rapidly expanding in a number of new application ares. Especially fingerprint recognition is the most important technology. Fingerprint recognition technologies are well established, proven, cost and legally accepted. Therefore, it has more spot lighted among the any other biometrics technologies. In this paper we propose a new on-line fingerprint recognition algorithm for non-inked type live scanner to fit their increasing of security level under the computing environment. Fingerprint recognition system consists of two distinct structural blocks: feature extraction and feature matching. The main topic in this paper focuses on the feature matching using the fingerprint minutiae (ridge ending and bifurcation). Minutiae matching is composed in the alignment stage and matching stage. Success of optimizing the alignment stage is the key of real-time (on-line) fingerprint recognition. Proposed alignment algorithm using clique shows the strength in the search space optimization and partially incomplete image. We make our own database to get the generality. Using the traditional statistical discriminant analysis, 0.05% false acceptance rate (FAR) at 8.83% false rejection rate (FRR) in 1.55 second average matching speed on a Pentium system have been achieved. This makes it possible to construct high performance fingerprint recognition system.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.12C
• /
• pp.1288-1298
• /
• 2002

To implement elliptic curve cryptosystem in GF(2$\^$m/) at high speed, a fast divider is required. Although bit-parallel architecture is well suited for high speed division operations, elliptic curve cryptosystem requires large m(at least 163) to support a sufficient security. In other words, since the bit-parallel architecture has an area complexity of 0(m$\^$m/), it is not suited for this application. In this paper, we propose a new serial-in serial-out systolic array for computing division operations in GF(2$\^$m/) using the standard basis representation. Based on a modified version of tile binary extended greatest common divisor algorithm, we obtain a new data dependence graph and design an efficient bit-serial systolic divider. The proposed divider has 0(m) time complexity and 0(m) area complexity. If input data come in continuously, the proposed divider can produce division results at a rate of one per m clock cycles, after an initial delay of 5m-2 cycles. Analysis shows that the proposed divider provides a significant reduction in both chip area and computational delay time compared to previously proposed systolic dividers with the same I/O format. Since the proposed divider can perform division operations at high speed with the reduced chip area, it is well suited for division circuit of elliptic curve cryptosystem. Furthermore, since the proposed architecture does not restrict the choice of irreducible polynomial, and has a unidirectional data flow and regularity, it provides a high flexibility and scalability with respect to the field size m.