• Journal of the Korea Society of Computer and Information
• /
• v.26 no.4
• /
• pp.11-19
• /
• 2021

An abnormal object refers to a person, an object, or a mechanical device that performs abnormal and unusual behavior and needs observation or supervision. In order to detect this through artificial intelligence algorithm without continuous human intervention, a method of observing the specificity of temporal features using optical flow technique is widely used. In this study, an abnormal situation is identified by learning an algorithm that translates an input image frame to an optical flow image using a Generative Adversarial Network (GAN). In particular, we propose a technique that improves the pre-processing process to exclude unnecessary outliers and the post-processing process to increase the accuracy of identification in the test dataset after learning to improve the performance of the model's abnormal behavior identification. UCSD Pedestrian and UMN Unusual Crowd Activity were used as training datasets to detect abnormal behavior. For the proposed method, the frame-level AUC 0.9450 and EER 0.1317 were shown in the UCSD Ped2 dataset, which shows performance improvement compared to the models in the previous studies.

• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.31 no.6
• /
• pp.1215-1225
• /
• 2021

Although Artificial Intelligence (AI) techniques have shown impressive performance in various fields, they are vulnerable to adversarial examples which induce misclassification by adding human-imperceptible perturbations to the input. Previous studies to defend the adversarial examples can be classified into three categories: (1) model retraining methods; (2) input transformation methods; and (3) adversarial examples detection methods. However, even though the defense methods against adversarial examples have constantly been proposed, there is no research to classify the type of adversarial attack. In this paper, we proposed an adversarial attack family classification method based on dimensionality reduction and clustering. Specifically, after extracting adversarial perturbation from adversarial example, we performed Linear Discriminant Analysis (LDA) to reduce the dimensionality of adversarial perturbation and performed K-means algorithm to classify the type of adversarial attack family. From the experimental results using MNIST dataset and CIFAR-10 dataset, we show that the proposed method can efficiently classify five tyeps of adversarial attack(FGSM, BIM, PGD, DeepFool, C&W). We also show that the proposed method provides good classification performance even in a situation where the legitimate input to the adversarial example is unknown.

• Journal of the korean academy of Pediatric Dentistry
• /
• v.49 no.2
• /
• pp.131-139
• /
• 2022

This study aimed to evaluate the effectiveness of deep convolutional neural networks (CNNs) for diagnosis of interproximal caries in pediatric intraoral radiographs. A total of 500 intraoral radiographic images of first and second primary molars were used for the study. A CNN model (Resnet 50) was applied for the detection of proximal caries. The diagnostic accuracy, sensitivity, specificity, receiver operating characteristic (ROC) curve, and area under ROC curve (AUC) were calculated on the test dataset. The diagnostic accuracy was 0.84, sensitivity was 0.74, and specificity was 0.94. The trained CNN algorithm achieved AUC of 0.86. The diagnostic CNN model for pediatric intraoral radiographs showed good performance with high accuracy. Deep learning can assist dentists in diagnosis of proximal caries lesions in pediatric intraoral radiographs.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.40 no.4
• /
• pp.251-260
• /
• 2022

Although the number of trees affected by pine wilt disease is decreasing, the affected area is expanding across the country. Recently, with the development of deep learning technology, it is being rapidly applied to the detection study of pine wilt nematodes and dead trees. The purpose of this study is to efficiently acquire deep learning training data and acquire accurate true values to further improve the detection ability of U-Net models through learning. To achieve this purpose, by using a filtering method applying a step-by-step deep learning algorithm the ambiguous analysis basis of the deep learning model is minimized, enabling efficient analysis and judgment. As a result of the analysis the U-Net model using the true values analyzed by period in the detection and performance improvement of dead pine trees of wilt nematode using the U-Net algorithm had a recall rate of -0.5%p than the U-Net model using the previously provided true values, precision was 7.6%p and F-1 score was 4.1%p. In the future, it is judged that there is a possibility to increase the precision of wilt detection by applying various filtering techniques, and it is judged that the drone surveillance method using drone orthographic images and artificial intelligence can be used in the pine wilt nematode disaster prevention project.

• Journal of the Korean Society of Marine Environment & Safety
• /
• v.28 no.7
• /
• pp.1120-1128
• /
• 2022

In the event of a marine accident, the longer the exposure time to the sea increases, the faster the chance of survival decreases. However, because the search area of the sea is extremely wide compared to that of land, marine object detection technology based on the sensor mounted on a satellite or an aircraft must be applied rather than ship for an efficient search. The purpose of this study was to rapidly detect an object in the ocean using a hyperspectral image sensor mounted on an aircraft. The image captured by this sensor has a spatial resolution of 8,241 × 1,024, and is a large-capacity data comprising 127 spectra and a resolution of 0.7 m per pixel. In this study, a marine object detection model was developed that combines a seawater identification algorithm using DBSCAN and a density-based land removal algorithm to rapidly analyze large data. When the developed detection model was applied to the hyperspectral image, the performance of analyzing a sea area of about 5 km² within 100 s was confirmed. In addition, to evaluate the detection accuracy of the developed model, hyperspectral images of the Mokpo, Gunsan, and Yeosu regions were taken using an aircraft. As a result, ships in the experimental image could be detected with an accuracy of 90 %. The technology developed in this study is expected to be utilized as important information to support the search and rescue activities of small ships and human life.

• Proceedings of the Korean Institute of Navigation and Port Research Conference
• /
• 2023.05a
• /
• pp.100-103
• /
• 2023

Recently, development of maritime autonomoust surface ships and eco-friendly ships, production and evaluation research considering various marine environments is needed in the field of optimal routes as the demand for accurate and detailed real-time marine environment prediction information expands. An algorithm that can calculate the optimal route while reducing the risk of the marine environment and uncertainty in energy consumption in smart ships was developed in 2 stages. In the first stage, a profile was created by combining marine environmental information with ship location and status information within the Automatic Ship Identification System(AIS). In the second stage, a model was developed that could define the marine environment energy map using the configured profile results, A regression equation was generated by applying Random Forest among machine learning techniques to reflect about 600,000 data. The Random Forest coefficient of determination (R²) was 0.89, showing very high reliability. The Dijikstra shortest path algorithm was applied to the marine environment prediction at June 1 to 3, 2021, and to calculate the optimal safety route and express it on the map. The route calculated by the random forest regression model was streamlined, and the route was derived considering the state of the marine environment prediction information. The concept of route calculation based on real-time marine environment prediction information in this study is expected to be able to calculate a realistic and safe route that reflects the movement tendency of ships, and to be expanded to a range of economic, safety, and eco-friendliness evaluation models in the future.

• The korean journal of orthodontics
• /
• v.54 no.1
• /
• pp.48-58
• /
• 2024

Objective: To quantify the effects of midline-related landmark identification on midline deviation measurements in posteroanterior (PA) cephalograms using a cascaded convolutional neural network (CNN). Methods: A total of 2,903 PA cephalogram images obtained from 9 university hospitals were divided into training, internal validation, and test sets (n = 2,150, 376, and 377). As the gold standard, 2 orthodontic professors marked the bilateral landmarks, including the frontozygomatic suture point and latero-orbitale (LO), and the midline landmarks, including the crista galli, anterior nasal spine (ANS), upper dental midpoint (UDM), lower dental midpoint (LDM), and menton (Me). For the test, Examiner-1 and Examiner-2 (3-year and 1-year orthodontic residents) and the Cascaded-CNN models marked the landmarks. After point-to-point errors of landmark identification, the successful detection rate (SDR) and distance and direction of the midline landmark deviation from the midsagittal line (ANS-mid, UDM-mid, LDM-mid, and Me-mid) were measured, and statistical analysis was performed. Results: The cascaded-CNN algorithm showed a clinically acceptable level of point-to-point error (1.26 mm vs. 1.57 mm in Examiner-1 and 1.75 mm in Examiner-2). The average SDR within the 2 mm range was 83.2%, with high accuracy at the LO (right, 96.9%; left, 97.1%), and UDM (96.9%). The absolute measurement errors were less than 1 mm for ANS-mid, UDM-mid, and LDM-mid compared with the gold standard. Conclusions: The cascaded-CNN model may be considered an effective tool for the auto-identification of midline landmarks and quantification of midline deviation in PA cephalograms of adult patients, regardless of variations in the image acquisition method.

• Information Systems Review
• /
• v.25 no.1
• /
• pp.129-143
• /
• 2023

In the era of big data, the field of artificial intelligence is showing remarkable growth, and in particular, the image classification learning methods by deep learning are becoming an important area. Various studies have been actively conducted to further improve the performance of CNNs, which have been widely used in image classification, among which a representative method is the Convolutional Recurrent Neural Network (CRNN) algorithm. The CRNN algorithm consists of a combination of CNN for image classification and RNNs for recognizing time series elements. However, since the inputs used in the RNN area of CRNN are the flatten values extracted by applying the convolution and pooling technique to the image, pixel values in the same phase in the image appear in different order. And this makes it difficult to properly learn the sequence of arrangements in the image intended by the RNN. Therefore, this study aims to improve image classification performance by proposing a novel hybrid method of CNN and RNN applying the concepts of encoder and decoder. In this study, the effectiveness of the new hybrid method was verified through various experiments. This study has academic implications in that it broadens the applicability of encoder and decoder concepts, and the proposed method has advantages in terms of model learning time and infrastructure construction costs as it does not significantly increase complexity compared to conventional hybrid methods. In addition, this study has practical implications in that it presents the possibility of improving the quality of services provided in various fields that require accurate image classification.

• Korean Journal of Radiology
• /
• v.21 no.8
• /
• pp.987-997
• /
• 2020

Objective: Measurement of the liver and spleen volumes has clinical implications. Although computed tomography (CT) volumetry is considered to be the most reliable noninvasive method for liver and spleen volume measurement, it has limited application in clinical practice due to its time-consuming segmentation process. We aimed to develop and validate a deep learning algorithm (DLA) for fully automated liver and spleen segmentation using portal venous phase CT images in various liver conditions. Materials and Methods: A DLA for liver and spleen segmentation was trained using a development dataset of portal venous CT images from 813 patients. Performance of the DLA was evaluated in two separate test datasets: dataset-1 which included 150 CT examinations in patients with various liver conditions (i.e., healthy liver, fatty liver, chronic liver disease, cirrhosis, and post-hepatectomy) and dataset-2 which included 50 pairs of CT examinations performed at ours and other institutions. The performance of the DLA was evaluated using the dice similarity score (DSS) for segmentation and Bland-Altman 95% limits of agreement (LOA) for measurement of the volumetric indices, which was compared with that of ground truth manual segmentation. Results: In test dataset-1, the DLA achieved a mean DSS of 0.973 and 0.974 for liver and spleen segmentation, respectively, with no significant difference in DSS across different liver conditions (p = 0.60 and 0.26 for the liver and spleen, respectively). For the measurement of volumetric indices, the Bland-Altman 95% LOA was -0.17 ± 3.07% for liver volume and -0.56 ± 3.78% for spleen volume. In test dataset-2, DLA performance using CT images obtained at outside institutions and our institution was comparable for liver (DSS, 0.982 vs. 0.983; p = 0.28) and spleen (DSS, 0.969 vs. 0.968; p = 0.41) segmentation. Conclusion: The DLA enabled highly accurate segmentation and volume measurement of the liver and spleen using portal venous phase CT images of patients with various liver conditions.

• Journal of Digital Convergence
• /
• v.21 no.3
• /
• pp.33-39
• /
• 2023

This study is a performance evaluation of a blood sugar monitoring system that combines a PPG sensor, which is an evaluation device for blood glucose monitoring, and a DNN algorithm when monitoring capillary blood glucose. The study is a researcher-led clinical trial conducted on participants from September 2023 to November 2023. PPG-BGMS compared predicted blood sugar levels for evaluation using 1-minute heart rate and heart rate variability information and the DNN prediction algorithm with capillary blood glucose levels measured with a blood glucose meter of the standard personal blood sugar management system. Of the 100 participants, 50 had type 2 diabetes (T2DM), and the average age was 67 years (range, 28 to 89 years). It was found that 100% of the predicted blood sugar level of PPG-BGMS was distributed in the A+B area of the Clarke error grid and Parker(Consensus) error grid. The MARD value of PPG-BGMS predicted blood glucose is 5.3 ± 4.0%. Consequentially, the non-blood-based PPG-BGMS was found to be non-inferior to the instantaneous blood sugar level of the clinical standard blood-based personal blood glucose measurement system.