• Journal of Internet Computing and Services
• /
• v.18 no.1
• /
• pp.65-75
• /
• 2017

The Detection Model is the model to find the result of a certain purpose using artificial intelligent, data mining, intelligent algorithms In Cyber Security, it usually uses to detect intrusion, malwares, cyber incident, and attacks etc. There are an amount of unlabeled data that are collected in a real environment such as security data. Since the most of data are not defined the class labels, it is difficult to know type of data. Therefore, the label determination process is required to detect and analysis with accuracy. In this paper, we proposed a KDFL(K-means and D-S Fusion based Labeling) method using D-S inference and k-means(unsupervised) algorithms to decide label of data records by fusion, and a detection model architecture using a proposed labeling method. A proposed method has shown better performance on detection rate, accuracy, F1-measure index than other methods. In addition, since it has shown the improved results in error rate, we have verified good performance of our proposed method.

• Journal of the Korea Academia-Industrial cooperation Society
• /
• v.18 no.12
• /
• pp.157-163
• /
• 2017

The object segmentation approach has the merit of reducing the processing cost required to detect moths of interest, because it applies a moth detection algorithm to the segmented objects after segmenting the objects individually in the moth image. In this paper, an object segmentation method for moth detection in pheromone trap images is proposed. Our method consists of preprocessing, thresholding, morphological filtering, and object labeling processes. Thresholding in the process is a critical step significantly influencing the performance of object segmentation. The proposed method can threshold very elaborately by reflecting the local properties of the moth images. We performed thresholding using global and local versions of Ostu's method and, used the proposed method for the moth images of Carposina sasakii acquired on a pheromone trap placed in an orchard. It was demonstrated that the proposed method could reflect the properties of light and background on the moth images. Also, we performed object segmentation and moth classification for Carposina sasakii images, where the latter process used an SVM classifier with training and classification steps. In the experiments, the proposed method performed the detection of Carposina sasakii for 10 moth images and achieved an average detection rate of 95% of them. Therefore, it was shown that the proposed technique is an effective monitoring method of Carposina sasakii in an orchard.

• Journal of Korea Society of Industrial Information Systems
• /
• v.12 no.3
• /
• pp.116-123
• /
• 2007

Despite the number of backlight manufacturer is increased as the market of flat panel display equipments and related development devices is enlarged, the inspection based on the human eye is still used in many backlight production lines. The defects such as particle, spot and scratch on the light emitting surface of the backlight prevent the LCD device from displaying the colors correctly. From that manual inspection it is difficult to maintain the quality of backlight consistently because the accuracy and the speed of the inspection may change with the physical condition of the operater. In this paper we studied on the development of automatic backlight surface defect inspection system. For this, we made up of the computer vision system and we developed the main program with various user interfaces to operate the inspection system effectively. And we developed the image processing module to extract the defect information. Furthermore, we presented the labeling process to reconstruct defect regions using the labeling table and the defect index. From the experimental results, we found that our system can detect all defect regions identified from human eye and it is sufficient to substitute for the conventional surface inspection.

• Journal of KIISE:Software and Applications
• /
• v.33 no.11
• /
• pp.942-952
• /
• 2006

We propose an automatic segmentation method for identifying pulmonary structures using gray-level information of chest CT images. Our method consists of following five steps. First, to segment pulmonary structures based on the difference of gray-level value, we select the threshold using optimal thresholding. Second, we separate the thorax from the background air and then the lungs and airways from the thorax by applying the inverse operation of 2D region growing in chest CT images. To eliminate non-pulmonary structures which has similar intensities with the lungs, we use 3D connected component labeling. Third, we segment the trachea and left and right mainstem bronchi using 3D branch-based region growing in chest CT images. Fourth, we can obtain accurate lung boundaries by subtracting the result of third step from the result of second step. Finally, we select the threshold in accordance with histogram analysis and then segment radio-dense pulmonary vessels by applying gray-level thresholding to the result of the second step. To evaluate the accuracy of proposed method, we make a visual inspection of segmentation result of lungs, airways and pulmonary vessels. We compare the result of the conventional region growing with the result of proposed 3D branch-based region growing. Experimental results show that our proposed method extracts lung boundaries, airways, and pulmonary vessels automatically and accurately.

• Journal of KIISE:Software and Applications
• /
• v.34 no.7
• /
• pp.646-654
• /
• 2007

In recent years, the use of text inserted into TV contents has grown to provide viewers with better visual understanding. In this paper, video text is defined as superimposed text region located of the bottom of video. Video text extraction is the first step for video information retrieval and video indexing. Most of video text detection and extraction methods in the previous work are based on text color, contrast between text and background, edge, character filter, and so on. However, the video text extraction has big problems due to low resolution of video and complex background. To solve these problems, we propose a method to extract text from videos using the Harris corner detector. The proposed algorithm consists of four steps: corer map generation using the Harris corner detector, extraction of text candidates considering density of comers, text region determination using labeling, and post-processing. The proposed algorithm is language independent and can be applied to texts with various colors. Text region update between frames is also exploited to reduce the processing time. Experiments are performed on diverse videos to confirm the efficiency of the proposed method.

• Journal of IKEEE
• /
• v.23 no.1
• /
• pp.112-119
• /
• 2019

In this paper, we propose a deep learning structure suitable for embedded system. The flame detection process of the proposed deep learning structure consists of four steps : flame area detection using flame color model, flame image classification using deep learning structure for flame color specialization, $N{\times}N$ cell separation in detected flame area, flame image classification using deep learning structure for flame shape specialization. First, only the color of the flame is extracted from the input image and then labeled to detect the flame area. Second, area of flame detected is the input of a deep learning structure specialized in flame color and is classified as flame image only if the probability of flame class at the output is greater than 75%. Third, divide the detected flame region of the images classified as flame images less than 75% in the preceding section into $N{\times}N$ units. Fourthly, small cells divided into $N{\times}N$ units are inserted into the input of a deep learning structure specialized to the shape of the flame and each cell is judged to be flame proof and classified as flame images if more than 50% of cells are classified as flame images. To verify the effectiveness of the proposed deep learning structure, we experimented with a flame database of ImageNet. Experimental results show that the proposed deep learning structure has an average resource occupancy rate of 29.86% and an 8 second fast flame detection time. The flame detection rate averaged 0.95% lower compared to the existing deep learning structure, but this was the result of light construction of the deep learning structure for application to embedded systems. Therefore, the deep learning structure for flame detection proposed in this paper has been proved suitable for the application of embedded system.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.19 no.6
• /
• pp.96-106
• /
• 2020

To solve the limitation of traffic monitoring that occur from embedded sensor such as loop and piezo sensors, the thermal imaging camera was installed on the roadside. As the length of Class 1(passenger car) is getting longer, it is becoming difficult to classify from Class 3(2-axle truck) by using an embedded sensor. The collected images were labeled to generate training data. A total of 17,536 vehicle images (640x480 pixels) training data were produced. CNN (Convolutional Neural Network) was used to achieve vehicle classification based on thermal image. Based on the limited data volume and quality, a classification accuracy of 97.7% was achieved. It shows the possibility of traffic monitoring system based on AI. If more learning data is collected in the future, 12-class classification will be possible. Also, AI-based traffic monitoring will be able to classify not only 12-class, but also new various class such as eco-friendly vehicles, vehicle in violation, motorcycles, etc. Which can be used as statistical data for national policy, research, and industry.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.11
• /
• pp.471-480
• /
• 2023

The Korean construction order volume in South Korea grew significantly from 91.3 trillion won in public orders in 2013 to a total of 212 trillion won in 2021, particularly in the private sector. As the size of the domestic and overseas markets grew, the scale and complexity of EPC (Engineering, Procurement, Construction) projects increased, and risk management of project management and ITB (Invitation to Bid) documents became a critical issue. The time granted to actual construction companies in the bidding process following the EPC project award is not only limited, but also extremely challenging to review all the risk terms in the ITB document due to manpower and cost issues. Previous research attempted to categorize the risk terms in EPC contract documents and detect them based on AI, but there were limitations to practical use due to problems related to data, such as the limit of labeled data utilization and class imbalance. Therefore, this study aims to develop an AI model that can categorize the contract terms based on the FIDIC Yellow 2017(Federation Internationale Des Ingenieurs-Conseils Contract terms) standard in detail, rather than defining and classifying risk terms like previous research. A multi-text classification function is necessary because the contract terms that need to be reviewed in detail may vary depending on the scale and type of the project. To enhance the performance of the multi-text classification model, we developed the ELECTRA PLM (Pre-trained Language Model) capable of efficiently learning the context of text data from the pre-training stage, and conducted a four-step experiment to validate the performance of the model. As a result, the ensemble version of the self-developed ITB-ELECTRA model and Legal-BERT achieved the best performance with a weighted average F1-Score of 76% in the classification of 57 contract terms.

• Journal of Intelligence and Information Systems
• /
• v.19 no.3
• /
• pp.1-23
• /
• 2013

To discover significant social issues such as unemployment, economy crisis, social welfare etc. that are urgent issues to be solved in a modern society, in the existing approach, researchers usually collect opinions from professional experts and scholars through either online or offline surveys. However, such a method does not seem to be effective from time to time. As usual, due to the problem of expense, a large number of survey replies are seldom gathered. In some cases, it is also hard to find out professional persons dealing with specific social issues. Thus, the sample set is often small and may have some bias. Furthermore, regarding a social issue, several experts may make totally different conclusions because each expert has his subjective point of view and different background. In this case, it is considerably hard to figure out what current social issues are and which social issues are really important. To surmount the shortcomings of the current approach, in this paper, we develop a prototype system that semi-automatically detects social issue keywords representing social issues and problems from about 1.3 million news articles issued by about 10 major domestic presses in Korea from June 2009 until July 2012. Our proposed system consists of (1) collecting and extracting texts from the collected news articles, (2) identifying only news articles related to social issues, (3) analyzing the lexical items of Korean sentences, (4) finding a set of topics regarding social keywords over time based on probabilistic topic modeling, (5) matching relevant paragraphs to a given topic, and (6) visualizing social keywords for easy understanding. In particular, we propose a novel matching algorithm relying on generative models. The goal of our proposed matching algorithm is to best match paragraphs to each topic. Technically, using a topic model such as Latent Dirichlet Allocation (LDA), we can obtain a set of topics, each of which has relevant terms and their probability values. In our problem, given a set of text documents (e.g., news articles), LDA shows a set of topic clusters, and then each topic cluster is labeled by human annotators, where each topic label stands for a social keyword. For example, suppose there is a topic (e.g., Topic1 = {(unemployment, 0.4), (layoff, 0.3), (business, 0.3)}) and then a human annotator labels "Unemployment Problem" on Topic1. In this example, it is non-trivial to understand what happened to the unemployment problem in our society. In other words, taking a look at only social keywords, we have no idea of the detailed events occurring in our society. To tackle this matter, we develop the matching algorithm that computes the probability value of a paragraph given a topic, relying on (i) topic terms and (ii) their probability values. For instance, given a set of text documents, we segment each text document to paragraphs. In the meantime, using LDA, we can extract a set of topics from the text documents. Based on our matching process, each paragraph is assigned to a topic, indicating that the paragraph best matches the topic. Finally, each topic has several best matched paragraphs. Furthermore, assuming there are a topic (e.g., Unemployment Problem) and the best matched paragraph (e.g., Up to 300 workers lost their jobs in XXX company at Seoul). In this case, we can grasp the detailed information of the social keyword such as "300 workers", "unemployment", "XXX company", and "Seoul". In addition, our system visualizes social keywords over time. Therefore, through our matching process and keyword visualization, most researchers will be able to detect social issues easily and quickly. Through this prototype system, we have detected various social issues appearing in our society and also showed effectiveness of our proposed methods according to our experimental results. Note that you can also use our proof-of-concept system in http://dslab.snu.ac.kr/demo.html.