• Journal of Convergence for Information Technology
• /
• v.11 no.3
• /
• pp.20-33
• /
• 2021

This study is to suggest improvements of personal information protection in South Korea, according to requiring the safety of process and protection of personal information. Accordingly, based on data collection and analysis through literature research, this study derived the issues and suitable standards of personal information for major artificial intelligence services. In addition, this cases studies were reviewed, focusing on the legal compliance and porcessing compliance for personal information proection in major countries. And it suggested the improvement plan applied in South Korea. As the results, in legal compliance, it is required reorganization of related laws, responsibility and compliance to develop and provide AI, and operation of risk management for personal information protection laws in AI services. In terms of processing compliance, first, in pre-processing and refining, it is necessary to standardize data set reference models, control data set quality, and voluntarily label AI applications. Second, in development and utilization of algorithm, it is need to establish and apply a clear regulation of the algorithm. As such, South Korea should apply suitable improvement tasks for personal information protection of safe AI service.

• Journal of Intelligence and Information Systems
• /
• v.29 no.2
• /
• pp.171-187
• /
• 2023

When a person's face is recognized through a recording device such as a low-pixel surveillance camera, it is difficult to capture the face due to low image quality. In situations where it is difficult to recognize a person's face, problems such as not being able to identify a criminal suspect or a missing person may occur. Existing studies on face recognition used refined datasets, so the performance could not be measured in various environments. Therefore, to solve the problem of poor face recognition performance in low-quality images, this paper proposes a method to generate high-quality images by performing image quality improvement on low-quality facial images considering various environments, and then improve the performance of facial feature point detection. To confirm the practical applicability of the proposed architecture, an experiment was conducted by selecting a data set in which people appear relatively small in the entire image. In addition, by choosing a facial image dataset considering the mask-wearing situation, the possibility of expanding to real problems was explored. As a result of measuring the performance of the feature point detection model by improving the image quality of the face image, it was confirmed that the face detection after improvement was enhanced by an average of 3.47 times in the case of images without a mask and 9.92 times in the case of wearing a mask. It was confirmed that the RMSE for facial feature points decreased by an average of 8.49 times when wearing a mask and by an average of 2.02 times when not wearing a mask. Therefore, it was possible to verify the applicability of the proposed method by increasing the recognition rate for facial images captured in low quality through image quality improvement.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.174-176
• /
• 2022

In this paper, we present a mobile application for 'personalized customized learning' for children with articulation disorders using an artificial intelligence (AI) algorithm. A dataset (Data Set) to analyze, judge, and predict the learner's articulation situation and degree. In particular, we designed a prototype model by looking at how AI can be improved and advanced compared to existing applications from the UX/UI (GUI) aspect. So far, the focus has been on visual experience, but now it is an important time to process data and provide a UX/UI (GUI) experience to users. The UX/UI (GUI) of the proposed mobile application was to be provided according to the learner's articulation level and situation by using CRNN (Convolution Recurrent Neural Network) of DeepLearning and Auto Encoder GPT-3 (Generative Pretrained Transformer). The use of artificial intelligence algorithms will provide a learning environment with a high degree of perfection to children with articulation disorders, thereby enhancing the learning effect. I hope that you do not have any fear or discomfort in conversation by improving the perfection of articulation with 'personalized and customized learning'.

• Journal of Digital Convergence
• /
• v.19 no.5
• /
• pp.307-313
• /
• 2021

In this paper, a lightweight network with fewer parameters compared to the existing object detection method is proposed. In the case of the currently used detection model, the network complexity has been greatly increased to improve accuracy. Therefore, the proposed network uses EfficientNet as a feature extraction network, and the subsequent layers are formed in a pyramid structure to utilize low-level detailed features and high-level semantic features. An attention process was applied between pyramid structures to suppress unnecessary noise for prediction. All computational processes of the network are replaced by depth-wise and point-wise convolutions to minimize the amount of computation. The proposed network was trained and evaluated using the PASCAL VOC dataset. The features fused through the experiment showed robust properties for various objects through a refinement process. Compared with the CNN-based detection model, detection accuracy is improved with a small amount of computation. It is considered necessary to adjust the anchor ratio according to the size of the object as a future study.

• Journal of the Society of Disaster Information
• /
• v.17 no.2
• /
• pp.245-253
• /
• 2021

Purpose: This study develops a risk prediction model that predicts the risk of a fire site by using initial information such as building information and reporter acquisition information, and supports effective mobilization of fire fighting resources and the establishment of damage minimization strategies for appropriate responses in the early stages of a disaster. Method: In order to identify the variables related to the fire damage scale on the fire statistics data, a correlation analysis between variables was performed using a machine learning algorithm to examine predictability, and a learning data set was constructed through preprocessing such as data standardization and discretization. Using this, we tested a plurality of machine learning algorithms, which are evaluated as having high prediction accuracy, and developed a risk prediction model applying the algorithm with the highest accuracy. Result: As a result of the machine learning algorithm performance test, the accuracy of the random forest algorithm was the highest, and it was confirmed that the accuracy of the intermediate value was relatively high for the risk class. Conclusion: The accuracy of the prediction model was limited due to the bias of the damage scale data in the fire statistics, and data refinement by matching data and supplementing the missing values was necessary to improve the predictive model performance.

• Korean Journal of Clinical Pharmacy
• /
• v.29 no.4
• /
• pp.231-237
• /
• 2019

Objective: The aim of this study was to analyze the status of split tablet prescription in South Korea. Methods: We conducted this analysis using 2016 National Patient Sample data from the Health Insurance Review and Assessment Service. We computed split tablet prescription rates by sex and age and determined which medicine and medical specialties had the highest split tablet prescribing rates. Results: The proportion of prescriptions that included split tablets was 15.6% (n=6,687,35). The proportion of prescriptions that included split tablets was higher for females (56.7%) than for males (43.3%), while that of prescriptions including split tablets versus total prescriptions for each sex was higher for males (16.4%) than for females (14.9%) (p<0.001). In the age group under 19 years, the proportion of prescriptions including split tablets (53.7%) was more than half of the total. The highest tablet splitting rate was found to be 89.9% for formoterol fumarate (40 ㎍), and pseudoephedrine hydrochloride (60 mg) had the highest number of prescriptions. Pediatrics (65.6%) was the medical field with the highest rate of split tablet prescription. Conclusion: Split tablets were most prescribed to pediatric patients. To minimize the use of split tablets, it is necessary to develop lower dose tablets and establish a policy that promotes prescription of these lower-dose tablets.

• Journal of Broadcast Engineering
• /
• v.26 no.4
• /
• pp.419-428
• /
• 2021

Shadow removal plays a key role for the pre-processing of image processing techniques such as object tracking and detection. With the advances of image recognition based on deep convolution neural networks, researches for shadow removal have been actively conducted. In this paper, we propose a novel method for shadow removal, which utilizes self attention distillation to extract semantic features. The proposed method gradually refines results of shadow detection, which are extracted from each layer of the proposed network, via top-down distillation. Specifically, the training procedure can be efficiently performed by learning the contextual information for shadow removal without shadow masks. Experimental results on various datasets show the effectiveness of the proposed method for shadow removal under real world environments.

• Journal of the Society of Disaster Information
• /
• v.18 no.2
• /
• pp.364-373
• /
• 2022

Purpose: The purpose of this paper is to develope smoke detection using AI model for detecting the initial fire in underground utility tunnels using CCTV Method: To improve detection performance of smoke which is high irregular, a deep learning model for fire detection was trained to optimize smoke detection. Also, several approaches such as dataset cleansing and gradient exploding release were applied to enhance model, and compared with results of those. Result: Results show the proposed approaches can improve the model performance, and the final model has good prediction capability according to several indexes such as mAP. However, the final model has low false negative but high false positive capacities. Conclusion: The present model can apply to smoke detection in underground utility tunnel, fixing the defect by linking between the model and the utility tunnel control system.

• Journal of the Korean Institute of Landscape Architecture
• /
• v.51 no.3
• /
• pp.166-178
• /
• 2023

This study aims to introduce and assess CNN Deep Learning methods to analyze visual landscape images on social media with embedded user perceptions and experiences. This study analyzed visual landscape images by focusing on a healing place. For the study, seven adjectives related to healing were selected through text mining and consideration of previous studies. Subsequently, 50 evaluators were recruited to build a Deep Learning image. Evaluators were asked to collect three images most suitable for 'healing', 'healing landscape', and 'healing place' on portal sites. The collected images were refined and a data augmentation process was applied to build a CNN model. After that, 15,097 images of 'healing' and 'healing landscape' on portal sites were collected and classified to analyze the visual landscape of a healing place. As a result of the study, 'quiet' was the highest in the category except 'other' and 'indoor' with 2,093 (22%), followed by 'open', 'joyful', 'comfortable', 'clean', 'natural', and 'beautiful'. It was found through research that CNN Deep Learning is an analysis method that can derive results from visual landscape image analysis. It also suggested that it is one way to supplement the existing visual landscape analysis method, and suggests in-depth and diverse visual landscape analysis in the future by establishing a landscape image learning dataset.

• Journal of Broadcast Engineering
• /
• v.24 no.6
• /
• pp.919-927
• /
• 2019

In this paper, we propose an accurate camera calibration method for acquiring multiview stereoscopic images. Generally, camera calibration is performed by using checkerboard structured patterns. The checkerboard pattern simplifies feature point extraction process and utilizes previously recognized lattice structure, which results in the accurate estimation of relations between the point on 2-dimensional image and the point on 3-dimensional space. Since estimation accuracy of camera parameters is dependent on feature matching, accurate detection of checkerboard corner is crucial. Therefore, in this paper, we propose the method that performs accurate camera calibration method through accurate detection of checkerboard corners. Proposed method detects checkerboard corner candidates by utilizing 1-dimensional gaussian filters with succeeding corner refinement process to remove outliers from corner candidates and accurately detect checkerboard corners in sub-pixel unit. In order to verify the proposed method, we check reprojection errors and camera location estimation results to confirm camera intrinsic parameters and extrinsic parameters estimation accuracy.