• Journal of the Korean Institute of Intelligent Systems
• /
• v.26 no.1
• /
• pp.56-63
• /
• 2016

In this study, we introduce a design of Fuzzy RBFNNs-based digit recognition system using the incremental-PCA in order to recognize the handwritten digits. The Principal Component Analysis (PCA) is a widely-adopted dimensional reduction algorithm, but it needs high computing overhead for feature extraction in case of using high dimensional images or a large amount of training data. To alleviate such problem, the incremental-PCA is proposed for the computationally efficient processing as well as the incremental learning of high dimensional data in the feature extraction stage. The architecture of Fuzzy Radial Basis Function Neural Networks (RBFNN) consists of three functional modules such as condition, conclusion, and inference part. In the condition part, the input space is partitioned with the use of fuzzy clustering realized by means of the Fuzzy C-Means (FCM) algorithm. Also, it is used instead of gaussian function to consider the characteristic of input data. In the conclusion part, connection weights are used as the extended diverse types in polynomial expression such as constant, linear, quadratic and modified quadratic. Experimental results conducted on the benchmarking MNIST handwritten digit database demonstrate the effectiveness and efficiency of the proposed digit recognition system when compared with other studies.

• Economic and Environmental Geology
• /
• v.41 no.6
• /
• pp.747-761
• /
• 2008

In order to experiment the acoustic remote classification of seabed sediment, we achieved ground-truth data(i.e. video and grab samples, etc.) and developed post-processing for automatic classification procedure on the basis of 300 kHz MultiBeam Echo Sounder(MBES) backscattering data, which was acquired using KONGBERG Simrad EM3000 at Sock-Cho Port, East Sea of South Korea. Sonar signal and its classification performance were identified with geo-referenced video imagery with the aid of GIS (Geographic Information System). The depth range of research site was from 5 m to 22.7 m, and the backscattering amplitude showed from -36dB to -15dB. The mean grain sizes of sediment from equi-distanced sampling site(50 m interval) varied from 2.86$(\phi)$ to 0.88(\phi). To acquire the main feature for the seabed classification from backscattering amplitude of MBES, we evaluated the correlation factors between the backscattering amplitude and properties of sediment samples. The performance of seabed remote classification proposed was evaluated with comparing the correlation of human expert segmentation to automatic algorithm results. The cross-model perception error ratio on automatic classification algorithm shows 8.95% at rocky bottoms, and 2.06% at the area representing low mean grain size.

• The KIPS Transactions:PartC
• /
• v.14C no.5
• /
• pp.425-430
• /
• 2007

Without the aid of stationary infrastructure, maintaining routing information for all nodes is inefficient in the Mobile Ad hoc Networks(MANET). It is more efficient when every time routing information is necessary that the source node broadcasts a query message to neighbour nodes. The source node using Ad hoc On-Demand distance Vector(AODV), which is one of the routing protocols of MANET, uses the Expanding Ring Search(ERS) algorithm which finds a destination node efficiently. In order to reduce the congestion of the network, ERS algorithm does not broadcast Route REQuest(RREQ) messages in the whole network. When the timer expires, if source node does not receive Route REPly(RREP) messages from the destination node, it gradually increases TTL value and broadcasts RREQ messages. Existing AODV cost a great deal to find a destination node because it uses a fixed NODE_TRAVERSAL_TIME value. Without the message which is added in existing AODV protocols, this paper measures delay time among the neighbours' nodes by making use of HELLO messages. We propose Adaptive ERS(AERS) algorithm that makes NET_TRAVERSAL_TIME optimum which apply to the measured delay time to NODE_TRAVERSAL_TIME. AERS suppresses the unnecessary messages, making NET_TRAVERSAL_TIME optimum in this paper. So we will be able to improve a network performance. We prove the effectiveness of the proposed method through simulation.

• Korean Journal of Agricultural and Forest Meteorology
• /
• v.19 no.4
• /
• pp.270-279
• /
• 2017

Regional assessment of crop productivity using a gridded simulation approach could aid policy making and crop management. Still, little effort has been made to develop the systems that allows gridded simulations of crop growth using ORYZA 2000 model, which has been used for predicting rice yield in Korea. The objectives of this study were to develop a series of data processing modules for creating input data files, running the crop model, and aggregating output files in a region of interest using gridded data files. These modules were implemented using C++ and R to make the best use of the features provided by these programming languages. In a case study, 13000 input files in a plain text format were prepared using daily gridded weather data that had spatial resolution of 1km and 12.5 km for the period of 2001-2010. Using the text files as inputs to ORYZA2000 model, crop yield simulations were performed for each grid cell using a scenario of crop management practices. After output files were created for grid cells that represent a paddy rice field in South Korea, each output file was aggregated into an output file in the netCDF format. It was found that the spatial pattern of crop yield was relatively similar to actual distribution of yields in Korea, although there were biases of crop yield depending on regions. It seemed that those differences resulted from uncertainties incurred in input data, e.g., transplanting date, cultivar in an area, as well as weather data. Our results indicated that a set of tools developed in this study would be useful for gridded simulation of different crop models. In the further study, it would be worthwhile to take into account compatibility to a modeling interface library for integrated simulation of an agricultural ecosystem.

• Economic and Environmental Geology
• /
• v.52 no.4
• /
• pp.313-322
• /
• 2019

SAR (Synthetic Aperture Radar) remote sensing data is a very useful tool for near-real-time identification of landslide affected areas that can occur over a large area due to heavy rains or typhoons. This study aims to develop an effective algorithm for automatically delineating landslide areas from the polarimetric SAR data acquired after the landslide event. To detect landslides from SAR observations, reduction of the speckle effects in the estimation of polarimetric SAR parameters and the orthorectification of geometric distortions on sloping terrain are essential processing steps. Based on the experimental analysis, it was found that the IDAN filter can provide a better estimation of the polarimetric parameters. In addition, it was appropriate to apply orthorectification process after estimating polarimetric parameters in the slant range domain. Furthermore, it was found that the polarimetric entropy is the most appropriate parameters among various polarimetric parameters. Based on those analyses, we proposed an automatic landslide detection algorithm using the histogram thresholding of the polarimetric parameters with the aid of terrain slope information. The landslide detection algorithm was applied to the ALOS-2 PALSAR-2 data which observed landslide areas in Japan triggered by Typhoon in September 2011. Experimental results showed that the landslide areas were successfully identified by using the proposed algorithm with a detection rate of about 82% and a false alarm rate of about 3%.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.16 no.6
• /
• pp.63-71
• /
• 2006

In this paper, we propose a secure solution for user authentication by using fingerprint verification on the sensor-client-server model, even with the client that is not necessarily trusted by the sensor holder or the server. To protect possible attacks launched at the untrusted client, our solution makes the fingerprint sensor validate the result computed by the client for the feature extraction. However, the validation should be simple so that the resource-constrained fingerprint sensor can validate it in real-time. To solve this problem, we separate the feature extraction into binarization and minutiae extraction, and assign the time-consuming binarization to the client. After receiving the result of binarization from the client, the sensor conducts a simple validation to check the result, performs the minutiae extraction with the received binary image from the client, and then sends the extracted minutiae to the server. Based on the experimental results, the proposed solution for fingerprint verification can be performed on the sensor-client-server model securely and in real-time with the aid of an untrusted client.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.8
• /
• pp.21-30
• /
• 2023

The study proposes a model that utilizes Python-based deep learning text classification techniques to detect the legality of illegal financial advertising posts on the internet. These posts aim to promote unlawful financial activities, including the trading of bank accounts, credit card fraud, cashing out through mobile payments, and the sale of personal credit information. Despite the efforts of financial regulatory authorities, the prevalence of illegal financial activities persists. By applying this proposed model, the intention is to aid in identifying and detecting illicit content in internet-based illegal financial advertisining, thus contributing to the ongoing efforts to combat such activities. The study utilizes convolutional neural networks(CNN) and recurrent neural networks(RNN, LSTM, GRU), which are commonly used text classification techniques. The raw data for the model is based on manually confirmed regulatory judgments. By adjusting the hyperparameters of the Korean natural language processing and deep learning models, the study has achieved an optimized model with the best performance. This research holds significant meaning as it presents a deep learning model for discerning internet illegal financial advertising, which has not been previously explored. Additionally, with an accuracy range of 91.3% to 93.4% in a deep learning model, there is a hopeful anticipation for the practical application of this model in the task of detecting illicit financial advertisements, ultimately contributing to the eradication of such unlawful financial advertisements.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.33 no.4
• /
• pp.671-685
• /
• 2023

By applying artificial intelligence to image editing technology, it has become possible to generate high-quality images with minimal traces of manipulation. However, since these technologies can be misused for criminal activities such as dissemination of false information, destruction of evidence, and denial of facts, it is crucial to implement strong countermeasures. In this study, image file and mobile forensic artifacts analysis were conducted for detecting image manipulation. Image file analysis involves parsing the metadata of manipulated images and comparing them with a Reference DB to detect manipulation. The Reference DB is a database that collects manipulation-related traces left in image metadata, which serves as a criterion for detecting image manipulation. In the mobile forensic artifacts analysis, packages related to image editing tools were extracted and analyzed to aid the detection of image manipulation. The proposed methodology overcomes the limitations of existing graphic feature-based analysis and combines with image processing techniques, providing the advantage of reducing false positives. The research results demonstrate the significant role of such methodology in digital forensic investigation and analysis. Additionally, We provide the code for parsing image metadata and the Reference DB along with the dataset of manipulated images, aiming to contribute to related research.

• Journal of Marine Bioscience and Biotechnology
• /
• v.2 no.3
• /
• pp.133-141
• /
• 2007

There are produced more than 600,000 tons of seaweeds every year along the coast of the Korea. Jeonnam province, south-west coast area, of Korea is producing 93% of total amounts of seaweeds. The laver, sea mustard, and tangleweed maintain stability in the output and has been exported as a simple product processing through drying or salting. It was evaluated the low value-added products and limited the expansion for the consumption of seaweeds. The seaweeds contains 40-60% carbohydrate and structurally different compared with land plant. The dietary fiber from seaweeds has been known the function of facilitating the bowl movement, excretion the heavy metal in the body, lowering the blood cholesterol level, anti-coagulant of blood, and anticancer. Especially, brown algae including sea mustard, seaweed fusiforme, and tangleweed contains alginic acid, laminarin, mannitol, fucoidan which are lowering the blood cholesterol level, lowering blood pressure, and fusion of blood clot. Agar-agar, carrageenan, and porphyran compound in red algae are known to antimutagenicity and anticoagulant function. In spite of potential of seaweed as a main bio-resource, there are lack of research to facilitate the consumption with its functional properties and consumers are unsatisfied with simple processing products. Also, the seaweed by-product dump into the sea and cause pollution of the seawater. Therefore, there are needed the scheme to promote the consumption of seaweeds. The development of value-added products, finding functional properties from seaweeds, development the functional feed for animal using seaweed by-products, and utilization of unused algae for food or other industrial uses will increase fisherman's income as well as serve as an aid for the people health due to its functional properties. Using by-product of seaweed and unexploited seaweed are needed to development of bio-degradable food packaging material and functional feed for animal.

• Investigative Magnetic Resonance Imaging
• /
• v.12 no.1
• /
• pp.8-19
• /
• 2008

Purpose : To investigate the 3-bond and spatial connectivity of human brain metabolites by scalar coupling and dipolar nuclear Overhauser effect/enhancement (NOE) interaction through 2D- correlation spectroscopy (COSY) and 2D- NOE spectroscopy (NOESY) techniques. Materials and Methods : All 2D experiments were performed on Bruker Avance 500 (11.8 T) with the zshield gradient triple resonance cryoprobe at 298 K. Human brain metabolites were prepared with 10% $D_2O$. Two-dimensional spectra with 2048 data points contains 320 free induction decay (FID) averaging. Repetition delay was 2 sec. The Top Spin 2.0 software was used for post-processing. Total 7 metabolites such as N-acetyl aspartate (NAA), creatine (Cr), choline (Cho), lutamine (Gln), glutamate (Glu), myo-inositol (Ins), and lactate (Lac) were included for major target metabolites. Results : Symmetrical 2D-COSY and 2D-NOESY pectra were successfully acquired: COSY cross peaks were observed in the only 1.0-4.5 ppm, however, NOESY cross peaks were observed in the 1.0-4.5 ppm and 7.9 ppm. From the result of the 2-D COSY data, cross peaks between the methyl protons ($CH_3$(3)) at 1.33 ppm and methine proton (CH(2)) at 4.11 ppm were observed in Lac. Cross peaks between the methylene protons (CH2(3,$H{\alpha}$)) at 2.50ppm and methylene protons ($CH_2$,(3,$H_B$)) at 2.70 ppm were observed in NAA. Cross peaks between the methine proton (CH(5)) at 3.27 ppm and the methine proton (CH(4,6)) at 3.59 ppm, between the methine proton (CH(1,3)) at 3.53 ppm and methine proton (CH(4,6)) at 3.59 ppm, and between the methine proton (CH(1,3)) at 3.53 ppm and methine proton (CH(2)) at 4.05 ppm were observed in Ins. From the result of 2-D NOESY data, cross peaks between the NH proton at 8.00 ppm and methyl protons ($CH_3$) were observed in NAA. Cross peaks between the methyl protons ($CH_3$(3)) at 1.33 ppm and methine proton (CH(2)) at 4.11 ppm were observed in Lac. Cross peaks between the methyl protons (CH3) at 3.03 ppm and methylene protons (CH2) at 3.93 ppm were observed in Cr. Cross peaks between the methylene protons ($CH_2$(3)) at 2.11 ppm and methylene protons ($CH_2$(4)) at 2.35 ppm, and between the methylene protons($CH_2$ (3)) at 2.11 ppm and methine proton (CH(2)) at 3.76 ppm were observed in Glu. Cross peaks between the methylene protons (CH2 (3)) at 2.14 ppm and methine proton (CH(2)) at 3.79 ppm were observed in Gln. Cross peaks between the methine proton (CH(5)) at 3.27 ppm and the methine proton (CH(4,6)) at 3.59 ppm, and between the methine proton (CH(1,3)) at 3.53 ppm and methine proton (CH(2)) at 4.05 ppm were observed in Ins. Conclusion : The present study demonstrated that in vitro 2D-COSY and NOESY represented the 3-bond and spatial connectivity of human brain metabolites by scalar coupling and dipolar NOE interaction. This study could aid in better understanding the interactions between human brain metabolites in vivo 2DCOSY study.