• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.36 no.6
• /
• pp.469-481
• /
• 2018

DL (Deep Learning) is getting popular in various fields to implement artificial intelligence that resembles human learning and cognition. DL based on complicate structure of the ANN (Artificial Neural Network) requires computing power and computation cost. Variety of DL models with improved performance have been developed with powerful computer specification. The main purpose of this paper is to detect buildings from aerial images and evaluate performance of Mask R-CNN (Region-based Convolutional Neural Network) developed by FAIR (Facebook AI Research) team recently. Mask R-CNN is a R-CNN that is evaluated to be one of the best ANN models in terms of performance for semantic segmentation with pixel-level accuracy. The performance of the DL models is determined by training ability as well as architecture of the ANN. In this paper, we characteristics of the Mask R-CNN with various types of the images and evaluate possibility of the generalization which is the ultimate goal of the DL. As for future study, it is expected that reliability and generalization of DL will be improved by using a variety of spatial information data for training of the DL models.

• Journal of Digital Convergence
• /
• v.19 no.2
• /
• pp.245-251
• /
• 2021

As the trend of steadily increasing the number of single or double household, there is a growing demand to see who is the outsider visiting the home during the free time. Various models of face recognition technology have been proposed through many studies, and Harr Cascade of OpenCV and Hog of Dlib are representative open source models. Among the two modes, Dlib's Hog has strengths in front of the indoor and at a limited distance, which is the focus of this study. In this paper, a face recognition visitor access system based on Dlib was designed and implemented. The whole system consists of a front module, a server module, and a mobile module, and in detail, it includes face registration, face recognition, real-time visitor verification and remote control, and video storage functions. The Precision, Specificity, and Accuracy according to the change of the distance threshold value were calculated using the error matrix with the photos published on the Internet, and compared with the results of previous studies. As a result of the experiment, it was confirmed that the implemented system was operating normally, and the result was confirmed to be similar to that reported by Dlib.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.24 no.10
• /
• pp.1331-1340
• /
• 2020

Object detection in maritime means that the captain detects floating objects that has a risk of colliding with the ship using the computer automatically and as accurately as human eyes. In conventional ships, the presence and distance of objects are determined through radar waves. However, it cannot identify the shape and type. In contrast, with the development of AI, cameras help accurately identify obstacles on the sea route with excellent performance in detecting or recognizing objects. The computer must calculate high-volume pixels to analyze digital images. However, the CPU is specialized for sequential processing; the processing speed is very slow, and smooth service support or security is not guaranteed. Accordingly, this study developed maritime object detection software and implemented it with FPGA to accelerate the processing of large-scale computations. Additionally, the system implementation was improved through embedded boards and FPGA interface, achieving 30 times faster performance than the existing algorithm and a three-times faster entire system.

• Journal of the Korea Society of Computer and Information
• /
• v.27 no.2
• /
• pp.153-161
• /
• 2022

Physical activities are decreasing and sitting time is increasing due to the automation, smartization, and intelligence of necessary household items throughout daily life. Recent healthcare studies have reported that the likelihood of obesity, diabetes, cardiovascular disease, and early death increases in proportion to sitting time. In this paper, we develop a sitting posture correction cushion in real time using capacitive pressure sensor based on conductive textile. It develops a pressure sensor using conductive textile, a key component of the posture correction cushion, and develops a low power-based pressure measurement circuit. It provides a function to transmit sensor values measured in real time to smartphones using BLE short-range wireless communication on the posture correction cushion, and develops a mobile application to check the condition of the sitting posture through these sensor values. In the mobile app, you can visualize your sitting posture and check it in real time, and if you keep it in the wrong posture for a certain period of time, you can notify it through an alarm. In addition, it is possible to visualize the sitting time and posture accuracy in a graph. Through the correction cushion in this paper, we experiment with how effective it is to correct the user's posture by recognizing the user's sitting posture, and present differentiation and excellence compared to other product.

• KIPS Transactions on Software and Data Engineering
• /
• v.12 no.2
• /
• pp.99-110
• /
• 2023

This study proposed a classification of malicious network traffic using the cyber threat framework(Mitre ATT&CK) and machine learning to solve the real-time traffic detection problems faced by current security monitoring systems. We applied a network traffic dataset called UNSW-NB15 to the Mitre ATT&CK framework to transform the label and generate the final dataset through rare class processing. After learning several boosting-based ensemble models using the generated final dataset, we demonstrated how these ensemble models classify network traffic using various performance metrics. Based on the F-1 score, we showed that XGBoost with no rare class processing is the best in the multi-class traffic environment. We recognized that machine learning ensemble models through Mitre ATT&CK label conversion and oversampling processing have differences over existing studies, but have limitations due to (1) the inability to match perfectly when converting between existing datasets and Mitre ATT&CK labels and (2) the presence of excessive sparse classes. Nevertheless, Catboost with B-SMOTE achieved the classification accuracy of 0.9526, which is expected to be able to automatically detect normal/abnormal network traffic.

• Korean Journal of Remote Sensing
• /
• v.39 no.2
• /
• pp.143-156
• /
• 2023

To precisely and reliably analyze the contents of the satellite imagery, recognizing the clouds which are the obstacle to gathering the useful information is essential. In recent times, deep learning yielded satisfactory results in various tasks, so many studies using deep neural networks have been conducted to improve the performance of cloud detection. However, existing methods for cloud detection have the limitation on increasing the performance due to the adopting the network models for semantic image segmentation without modification. To tackle this problem, we introduced the multi-branch neural architecture search to find optimal network structure for cloud detection. Additionally, the proposed method adopts the soft intersection over union (IoU) as loss function to mitigate the disagreement between the loss function and the evaluation metric and uses the various data augmentation methods. The experiments are conducted using the cloud detection dataset acquired by Arirang-3/3A satellite imagery. The experimental results showed that the proposed network which are searched network architecture using cloud dataset is 4% higher than the existing network model which are searched network structure using urban street scenes with regard to the IoU. Also, the experimental results showed that the soft IoU exhibits the best performance on cloud detection among the various loss functions. When comparing the proposed method with the state-of-the-art (SOTA) models in the field of semantic segmentation, the proposed method showed better performance than the SOTA models with regard to the mean IoU and overall accuracy.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.52 no.2
• /
• pp.96-109
• /
• 2024

The purpose of this study is to propose a method for evaluating the similarity of Show gardens using Deep Learning models, specifically VGG-16 and ResNet50. A model for judging the similarity of show gardens based on VGG-16 and ResNet50 models was developed, and was referred to as DRG (Deep Recognition of similarity in show Garden design). An algorithm utilizing GAP and Pearson correlation coefficient was employed to construct the model, and the accuracy of similarity was analyzed by comparing the total number of similar images derived at 1st (Top1), 3rd (Top3), and 5th (Top5) ranks with the original images. The image data used for the DRG model consisted of a total of 278 works from the Le Festival International des Jardins de Chaumont-sur-Loire, 27 works from the Seoul International Garden Show, and 17 works from the Korea Garden Show. Image analysis was conducted using the DRG model for both the same group and different groups, resulting in the establishment of guidelines for assessing show garden similarity. First, overall image similarity analysis was best suited for applying data augmentation techniques based on the ResNet50 model. Second, for image analysis focusing on internal structure and outer form, it was effective to apply a certain size filter (16cm × 16cm) to generate images emphasizing form and then compare similarity using the VGG-16 model. It was suggested that an image size of 448 × 448 pixels and the original image in full color are the optimal settings. Based on these research findings, a quantitative method for assessing show gardens is proposed and it is expected to contribute to the continuous development of garden culture through interdisciplinary research moving forward.

• Journal of Internet Computing and Services
• /
• v.25 no.2
• /
• pp.57-67
• /
• 2024

This paper explores the potential of electromyography (EMG) as a means of gesture recognition for user input in gesture-based interaction. EMG utilizes small electrodes within muscles to detect and interpret user movements, presenting a viable input method. To classify user gestures based on EMG data, machine learning techniques are employed, necessitating the preprocessing of raw EMG data to extract relevant features. EMG characteristics can be expressed through formulas such as Integrated EMG (IEMG), Mean Absolute Value (MAV), Simple Square Integral (SSI), Variance (VAR), and Root Mean Square (RMS). Additionally, determining the suitable time for gesture classification is crucial, considering the perceptual, cognitive, and response times required for user input. To address this, segment sizes ranging from a minimum of 100ms to a maximum of 1,000ms are varied, and feature extraction is performed to identify the optimal segment size for gesture classification. Notably, data learning employs overlapped segmentation to reduce the interval between data points, thereby increasing the quantity of training data. Using this approach, the paper employs four machine learning models (KNN, SVC, RF, XGBoost) to train and evaluate the system, achieving accuracy rates exceeding 96% for all models in real-time gesture input scenarios with a maximum segment size of 200ms.

• Smart Media Journal
• /
• v.13 no.6
• /
• pp.9-15
• /
• 2024

In recent trends, there has been an increase in 'Qshing' attacks, a hybrid form of phishing that exploits fake QR (Quick Response) codes impersonating government agencies to steal personal and financial information. Particularly, this attack method is characterized by its stealthiness, as victims can be redirected to phishing pages or led to download malicious software simply by scanning a QR code, making it difficult for them to realize they have been targeted. In this paper, we have developed a classification technique utilizing machine learning algorithms to identify the maliciousness of URLs embedded in QR codes, and we have explored ways to integrate this with existing QR code readers. To this end, we constructed a dataset from 128,587 malicious URLs and 428,102 benign URLs, extracting 35 different features such as protocol and parameters, and used AutoML to identify the optimal algorithm and hyperparameters, achieving an accuracy of approximately 87.37%. Following this, we designed the integration of the trained classification model with existing QR code readers to implement a service capable of countering Qshing attacks. In conclusion, our findings confirm that deriving an optimized algorithm for classifying malicious URLs in QR codes and integrating it with existing QR code readers presents a viable solution to combat Qshing attacks.

• Korean Journal of Cognitive Science
• /
• v.28 no.4
• /
• pp.193-221
• /
• 2017

Grouping episodes into semantically related categories is necessary for better mnemonic structure. However, the effect of grouping on memory of subordinate details was not clearly understood. In an fMRI study, we tested whether attending superordinate during semantic association disrupts or enhances subordinate episodic details. In each cycle of the experiment, five cue words were presented sequentially with two related detail words placed underneath for each cue. Participants were asked whether they could imagine a category that includes the previously shown cue words in each cycle, and their confidence on retrieval was rated. Participants were asked to perform cued recall tests on presented detail words after the session. Behavioral data showed that reaction times for categorization tasks decreased and confidence levels increased in the third trial of each cycle, thus this trial was considered to be an important insight where a semantic category was believed to be successfully established. Critically, the accuracy of recalling detail words presented immediately prior to third trials was lower than those of followed trials, indicating that subordinate details were disrupted during categorization. General linear model analysis of the trial immediately prior to the completion of categorization, specifically the second trial, revealed significant activation in the temporal gyrus and inferior frontal gyrus, areas of semantic memory networks. Representative Similarity Analysis revealed that the activation patterns of the third trials were more consistent than those of the second trials in the temporal gyrus, inferior frontal gyrus, and hippocampus. Our research demonstrates that semantic grouping can cause memories of subordinate details to fade, suggesting that semantic retrieval during categorization affects the quality of related episodic memory.