• The Journal of the Korea Contents Association
• /
• v.21 no.6
• /
• pp.701-715
• /
• 2021

The drastic advance of recent deep learning technologies is heavily dependent on training datasets which are essential to train models by themselves with less human efforts. In comparison with the work to design deep learning models, preparing datasets is a long haul; at the moment, in the domain of vision intelligent, datasets are still being made by handwork requiring a lot of time and efforts, where workers need to directly make labels on each image usually with GUI-based labeling tools. In this paper, we overview the current status of vision datasets focusing on what datasets are being shared and how they are prepared with various labeling tools. Particularly, in order to relieve the repetitive and tiring labeling work, we present an interactive smart image annotating system with which the annotation work can be transformed from the direct human-only manual labeling to a correction-after-checking by means of a support of automatic labeling. In an experiment, we show that automatic labeling can greatly improve the productivity of datasets especially reducing time and efforts to specify regions of objects found in images. Finally, we discuss critical issues that we faced in the experiment to our annotation system and describe future work to raise the productivity of image datasets creation for accelerating AI technology.

• Journal of Digital Convergence
• /
• v.19 no.3
• /
• pp.187-194
• /
• 2021

OSINT(Open Source Intelligence), rapidly increasing on the surface web in various forms, can also be used for criminal investigations by using profiling. This technique has become quite common in foreign investigative agencies such as the United States. On the other hand, in Korea, it is not used a lot, and there is a large deviation in the quantity and quality of information acquired according to the experience and knowledge level of investigator. Unlike Bazzell's most well-known model, we designed a Korean-style OSINT-based profiling technique that considers the Korean web environment and provides timeline information, focusing on the improved workflow. The database schema to improve the efficiency of profiling is also presented. Using this, we can obtain search results that guarantee a certain level of quantity and quality. And it can also be used as a standard training course. To increase the effectiveness and efficiency of criminal investigations using this technique, it is necessary to strengthen the legal basis and to introduce automation technologies.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.1
• /
• pp.469-481
• /
• 2022

This study is for increasing evacuation safety by analyzing RSET(the required safe escape time) through the arrangement of personnel by floor and by room while evacuating in a Girls' High School Dormitory. For this study, PAPS(Physical Activity Promotion System) results that have not been studied so far were analyzed and reflected in evacuation simulations on the premise that individual student's physical strength can affect evacuation. Based on the PAPS results, four scenarios were applied. In addition, evacuation simulation using the pathfinder program was conducted in two situations: the evacuation route was assigned or not. Scenario 4 was the fastest at 168.5 seconds of RSET in assigning evacuation routes among scenarios. As a result of this study, the arrangement of students focusing on improving their academic ability and student life guidance excluding student physical strength has problem. In order to solve this problem, it is effective to place C group students(low grade on PAPS) on low floors and A group students(high grade on PAPS) on high floors and to assign evacuation routes in each room. In the future, the following ways need to be more studied. A study on how to increase evacuation safety through practical evacuation training, the way of assessing evacuation safety reflecting the lifestyle and physical strength of girls, the evacuation route assignment according to the fire occurrence point, and the method to secure evacuation routes in the event of a fire near stairs or entrances should be conducted.

• Journal of the Korea institute for structural maintenance and inspection
• /
• v.26 no.2
• /
• pp.28-36
• /
• 2022

As port structures are exposed to various extreme external loads such as wind (typhoons), sea waves, or collision with ships; it is important to evaluate the structural safety periodically. To monitor the port structure, especially the rubber fender, a fender segmentation system using a vision sensor and deep learning method has been proposed in this study. For fender segmentation, a new deep learning network that improves the encoder-decoder framework with the receptive field block convolution module inspired by the eccentric function of the human visual system into the DenseNet format has been proposed. In order to train the network, various fender images such as BP, V, cell, cylindrical, and tire-types have been collected, and the images are augmented by applying four augmentation methods such as elastic distortion, horizontal flip, color jitter, and affine transforms. The proposed algorithm has been trained and verified with the collected various types of fender images, and the performance results showed that the system precisely segmented in real time with high IoU rate (84%) and F1 score (90%) in comparison with the conventional segmentation model, VGG16 with U-net. The trained network has been applied to the real images taken at one port in Republic of Korea, and found that the fenders are segmented with high accuracy even with a small dataset.

• Journal of the Korean Society of Surveying, Geodesy, Photogrammetry and Cartography
• /
• v.40 no.3
• /
• pp.239-247
• /
• 2022

Cities are becoming more complex due to rapid industrialization and population growth in modern times. In particular, urban areas are rapidly changing due to housing site development, reconstruction, and demolition. Thus accurate road information is necessary for various purposes, such as High Definition Map for autonomous car driving. In the case of the Republic of Korea, accurate spatial information can be generated by making a map through the existing map production process. However, targeting a large area is limited due to time and money. Road, one of the map elements, is a hub and essential means of transportation that provides many different resources for human civilization. Therefore, it is essential to update road information accurately and quickly. This study uses Semantic Segmentation algorithms Such as LinkNet, D-LinkNet, and NL-LinkNet to extract roads from drone images and then apply hyperparameter optimization to models with the highest performance. As a result, the LinkNet model using pre-trained ResNet-34 as the encoder achieved 85.125 mIoU. Subsequent studies should focus on comparing the results of this study with those of studies using state-of-the-art object detection algorithms or semi-supervised learning-based Semantic Segmentation techniques. The results of this study can be applied to improve the speed of the existing map update process.

• Journal of Internet Computing and Services
• /
• v.23 no.3
• /
• pp.35-45
• /
• 2022

Even in recent years, treatment of first-aid patients is still often delayed due to a shortage of medical resources in marginalized areas. Research on automating the analysis of medical data to solve the problems of inaccessibility for medical services and shortage of medical personnel is ongoing. Computer vision-based medical inspection automation requires a lot of cost in data collection and labeling for training purposes. These problems stand out in the works of classifying lesion that are rare, or pathological features and pathogenesis that are difficult to clearly define visually. Anomaly detection is attracting as a method that can significantly reduce the cost of data collection by adopting an unsupervised learning strategy. In this paper, we propose methods for detecting abnormal images on chest X-RAY images as follows based on existing anomaly detection techniques. (1) Normalize the brightness range of medical images resampled as optimal resolution. (2) Some feature vectors with high representative power are selected in set of patch features extracted as intermediate-level from lesion-free images. (3) Measure the difference from the feature vectors of lesion-free data selected based on the nearest neighbor search algorithm. The proposed system can simultaneously perform anomaly classification and localization for each image. In this paper, the anomaly detection performance of the proposed system for chest X-RAY images of PA projection is measured and presented by detailed conditions. We demonstrate effect of anomaly detection for medical images by showing 0.705 classification AUROC for random subset extracted from the PadChest dataset. The proposed system can be usefully used to improve the clinical diagnosis workflow of medical institutions, and can effectively support early diagnosis in medically poor area.

• Journal of Korea Entertainment Industry Association
• /
• v.15 no.3
• /
• pp.41-50
• /
• 2021

While artificial intelligence is rapidly replacing human jobs, the art field where human creativity is important is considered an exception. There are currently several AI mastering services in the field of mastering music, a profession at the border between art and technology. In general, the quality of AI mastering is considered to be inferior to the work of a human professional mastering engineer. In this paper, acoustic analysis, listening experiments, and expert interviews were conducted to compare AI and human experts. In the acoustic analysis, In the analysis of audio, there was no significant difference between the results of professional mastering engineers and the results of artificial intelligence. In the listening experiment, the non-musicians could not distinguish between the sound quality of the professional mastering engineer's work and the artificial intelligence work. The group of musicians showed a preference for a specific sound source, but the preference for a specific mastering did not appear significantly. In an expert interview, In expert interviews, respondents answered that there was no significant difference in quality between the two mastering services, and the biggest difference was the communication method between the mastering service provider and the user. In addition, as data increases, it is expected that artificial intelligence mastering will achieve rapid quality improvement and further improvement in communication.

• Korean Journal of Remote Sensing
• /
• v.38 no.6_4
• /
• pp.1925-1934
• /
• 2022

Monitoring nuclear activity for inaccessible areas using remote sensing technology is essential for nuclear non-proliferation. In recent years, deep learning has been actively used to detect nuclear-activity-related small objects. However, high-resolution satellite imagery containing small objects can result in class imbalance. As a result, there is a performance degradation problem in detecting small objects. Therefore, this study aims to improve detection accuracy by analyzing the effect of the ratio of small objects related to nuclear activity in the input data for the performance of the deep learning model. To this end, six case datasets with different ratios of small object pixels were generated and a U-Net model was trained for each case. Following that, each trained model was evaluated quantitatively and qualitatively using a test dataset containing various types of small object classes. The results of this study confirm that when the ratio of object pixels in the input image is adjusted, small objects related to nuclear activity can be detected efficiently. This study suggests that the performance of deep learning can be improved by adjusting the object pixel ratio of input data in the training dataset.

• The Journal of the Korea Contents Association
• /
• v.22 no.5
• /
• pp.746-764
• /
• 2022

Communication breakdown has been mentioned causing a heavy burden for dementia caregivers. This study aims to inspect and structure the process and results of communication between people with dementia and their caregivers. The impeding/facilitating elements of communication are also extracted. Interviews with 21 of dementia care experts about the direct and indirect experiences of communication with people with dementia were analyzed based on the grounded theory. Results show that combination of the cognitive and communication decline of the people with dementia, confusing environment and caregivers' inappropriate attitude and lack of communication skills leads to communication breakdown and relations severance. Minimal contacts and task-oriented conversation results in conflicts and people with dementia's increasing agitation, anxiety and violent behaviors while understanding of individuality and listening with heart lead to recovered lucidity in the state of serious dementia, recovered pleasure and voluntary participation in the daily activities for people with dementia. Core paradigm was defined as 'Person Centered Care through relation formation'. There are 4 types of communication with people with dementia : partnering, patronizing, conflicting, avoiding types. Researchers suggest that Person Centered based communication skills be educated and trained for dementia caregivers.

• Journal of Korea Entertainment Industry Association
• /
• v.14 no.6
• /
• pp.301-309
• /
• 2020

The purpose of this study was to determine whether swiss ball exercise combined with taping would improve low back pain, disability and quality of life in women with pregnancy-related Low Back Pain (PR-LBP). Sixteen patients with PR-LBP were recruited and were randomly divided into two groups: taping and swiss ball exercise group (n=8) and taping and stretching group (n=8). The taping and swiss ball exercise group was treated with swiss ball exercise with kinesio taping, while the taping and stretching group received only taping. The taping and swiss ball exercise group performed swiss ball exercise for 30 minutes a day, 5 times a week for 4weeks, while the taping and stretching group conducted stretching exercise for the same amount of time. VAS was used to assess pain level of low back. Disability was measured using a Roland Morris Disability Questionnaire(RMDQ). Quality of life were measured by SF-36. The pain intensity of low back in the taping and swiss ball exercise group improved significantly greater than the taping and stretching group (p<0.05). Significant improvement in the disability was observed in the taping and swiss ball exercise group compared to the taping and stretching group (p<0.05). The SF 36 in the taping and swiss ball exercise group improved significantly greater than the taping and stretching group (p<0.05). Our findings indicate that swiss ball exercise combined with taping is beneficial and effective to improve low back health and quality of life in women with PR-LBP.