• Korean Journal of Remote Sensing
• /
• v.38 no.6_1
• /
• pp.1047-1056
• /
• 2022

Sea ice, frozen sea water, in the Artic is a primary indicator of global warming. Due to its importance to the climate system, shipping-route navigation, and fisheries, Arctic sea ice prediction has gained increased attention in various disciplines. Recent advances in artificial intelligence (AI), motivated by a desire to develop more autonomous and efficient future predictions, have led to the development of new sea ice prediction models as alternatives to conventional numerical and statistical prediction models. This study aims to evaluate the performance of the two-stream convolutional long-and short-term memory (TS-ConvLSTM) AI model, which is designed for learning both global and local characteristics of the Arctic sea ice changes, for the minimum September Arctic sea ice from 2001 to 2021, and to show the possibility for an operational prediction system. Although the TS-ConvLSTM model generally increased the prediction performance as training data increased, predictability for the marginal ice zone, 5-50% concentration, showed a negative trend due to increasing first-year sea ice and warming. Additionally, a comparison of sea ice extent predicted by the TS-ConvLSTM with the median Sea Ice Outlooks (SIOs) submitted to the Sea Ice Prediction Network has been carried out. Unlike the TS-ConvLSTM, the median SIOs did not show notable improvements as time passed (i.e., the amount of training data increased). Although the TS-ConvLSTM model has shown the potential for the operational sea ice prediction system, learning more spatio-temporal patterns in the difficult-to-predict natural environment for the robust prediction system should be considered in future work.

• The Journal of the Convergence on Culture Technology
• /
• v.8 no.2
• /
• pp.385-391
• /
• 2022

Quality evaluation of beef carcasses is an important issue in the livestock industry. Recently, through the AI monitor system based on artificial intelligence, the quality manager can receive help in making accurate decisions based on the analysis of beef carcass images or result information. This artificial intelligence dataset is an important factor in judging performance. Existing datasets may have different surface orientation or resolution. In this paper, we proposed a two-stage classification model that can efficiently manage the grades of beef carcass image using deep learning. And to overcome the problem of the various conditions of the image, a new dataset of 1,300 images was constructed. The recognition rate of deep network for 5-grade classification using the new dataset was 72.5%. Two-stage evaluation is a method to increase reliability by taking advantage of the large difference between grades 1++, 1+, and grades 1 and 2 and 3. With two experiments using the proposed two stage model, the recognition rates of 73.7% and 77.2% were obtained. As this, The proposed method will be an efficient method if we have a dataset with 100% recognition rate in the first stage.

• Korean Journal of Remote Sensing
• /
• v.38 no.5_2
• /
• pp.765-779
• /
• 2022

This study evaluates the accuracy in identifying the burned area in South Korea using multi-temporal data from Sentinel-2 MSI and Landsat 8/9 OLI. Spectral indices such as the Difference Normalized Burn Ratio (dNBR), Relative Difference Normalized Burn Ratio (RdNBR), and Burned Area Index (BAI) were used to identify the burned area in the March 2022 forest fire in Uljin. Based on the results of six indices, the accuracy to detect the burned area was assessed for four satellites using Sentinel-2 and Landsat 8/9, respectively. Sentinel-2 and Landsat 8/9 produce images every 16 and 10 days, respectively, although it is difficult to acquire clear images due to clouds. Furthermore, using images taken before and after a forest fire to examine the burned area results in a rapid shift because vegetation growth in South Korea began in April, making it difficult to detect. Because Sentinel-2 and Landsat 8/9 images from February to May are based on the same date, this study is able to compare the indices with a relatively high detection accuracy and gets over the temporal resolution limitation. The results of this study are expected to be applied in the development of new indices to detect burned areas and indices that are optimized to detect South Korean forest fires.

• International journal of advanced smart convergence
• /
• v.11 no.1
• /
• pp.19-27
• /
• 2022

Across the world, 'housing' comprises a significant portion of wealth and assets. For this reason, fluctuations in real estate prices are highly sensitive issues to individual households. In Korea, housing prices have steadily increased over the years, and thus many Koreans view the real estate market as an effective channel for their investments. However, if one purchases a real estate property for the purpose of investing, then there are several risks involved when prices begin to fluctuate. The purpose of this study is to design a real estate price 'return rate' prediction model to help mitigate the risks involved with real estate investments and promote reasonable real estate purchases. Various approaches are explored to develop a model capable of predicting real estate prices based on an understanding of the immovability of the real estate market. This study employs the LSTM method, which is based on artificial intelligence and deep learning, to predict real estate prices and validate the model. LSTM networks are based on recurrent neural networks (RNN) but add cell states (which act as a type of conveyer belt) to the hidden states. LSTM networks are able to obtain cell states and hidden states in a recursive manner. Data on the actual trading prices of apartments in autonomous districts between January 2006 and December 2019 are collected from the Actual Trading Price Disclosure System of the Ministry of Land, Infrastructure and Transport (MOLIT). Additionally, basic data on apartments and commercial buildings are collected from the Public Data Portal and Seoul Metropolitan Government's data portal. The collected actual trading price data are scaled to monthly average trading amounts, and each data entry is pre-processed according to address to produce 168 data entries. An LSTM model for return rate prediction is prepared based on a time series dataset where the training period is set as April 2015~August 2017 (29 months), the validation period is set as September 2017~September 2018 (13 months), and the test period is set as December 2018~December 2019 (13 months). The results of the return rate prediction study are as follows. First, the model achieved a prediction similarity level of almost 76%. After collecting time series data and preparing the final prediction model, it was confirmed that 76% of models could be achieved. All in all, the results demonstrate the reliability of the LSTM-based model for return rate prediction.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.9
• /
• pp.1266-1271
• /
• 2022

Cats have strong wildness so they have a characteristic of hiding diseases well. The disease may have already worsened when the guardian finds out that the cat has a disease. It will be of great help in treating the cat's disease if the owner can recognize the cat's polydipsia, polyuria, and frequent urination more quickly. In this paper, 1) Efficient version of DeepLabCut for pose estimation, 2) YOLO v4 for object detection, 3) LSTM is used for behavior prediction, and 4) BoT-SORT is used for object tracking running on an artificial intelligence device. Using artificial intelligence technology, it predicts the cat's next, polyuria and frequency of urination through the analysis of the cat's behavior pattern from the home CCTV video and the weight sensor of the water bowl. And, through analysis of cat behavior patterns, we propose an application that reports disease prediction and abnormal behavior to the guardian and delivers it to the guardian's mobile and the server system.

• Proceedings of the Korea Water Resources Association Conference
• /
• 2023.05a
• /
• pp.30-30
• /
• 2023

This study aims to establish the multi-reservoir operation system model in the Upper Mun River Basin which includes 5 main dams namely, Mun Bon (MB), Lamchae (LC), Lam Takhong (LTK), Lam Phraphoeng (LPP), and Lower Lam Chiengkrai (LLCK) Dams. The knowledge and AI technology were applied aiming to develop innovative prototype for SMART dam-reservoir operation in future. Two different sorts of reservoir operation system model namely, Fuzzy Logic (FL) and Constraint Programming (CP) as well as the development of rainfall and reservoir inflow prediction models using Machine Learning (ML) technique were made to help specify the right amount of daily reservoir releases for the Royal Irrigation Department (RID). The model could also provide the essential information particularly for the Office of National Water Resource of Thailand (ONWR) to determine the short-term and long-term water resource management plan and strengthen water security against flood and drought in this region. The simulated results of base case scenario for reservoir operation in the Upper Mun from 2008 to 2021 indicated that in the same circumstances, FL and CP models could specify the new release schemes to increase the reservoir water storages at the beginning of dry season of approximately 125.25 and 142.20 MCM per year. This means that supplying the agricultural water to farmers in dry season could be well managed. In other words, water scarcity problem could substantially be moderated at some extent in case of incapability to control the expansion of cultivated area size properly. Moreover, using AI technology to determine the new reservoir release schemes plays important role in reducing the actual volume of water shortfall in the basin although the drought situation at LTK and LLCK Dams were still existed in some periods of time. Meanwhile, considering the predicted inflow and hydrologic factors downstream of 5 main dams by FL model and minimizing the flood volume by CP model could ensure that flood risk was considerably minimized as a result of new release schemes.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2022.05a
• /
• pp.165-167
• /
• 2022

The proportion of cat cats among companion animals has been increasing at an average annual rate of 25.4% since 2012. Cats have strong wildness compared to dogs, so they have a characteristic of hiding diseases well. Therefore, when the guardian finds out that the cat has a disease, the disease may have already worsened. Symptoms such as anorexia (eating avoidance), vomiting, diarrhea, polydipsia, and polyuria in cats are some of the symptoms that appear in cat diseases such as diabetes, hyperthyroidism, renal failure, and panleukopenia. It will be of great help in treating the cat's disease if the owner can recognize the cat's polydipsia (drinking a lot of water), polyuria (a large amount of urine), and frequent urination (urinating frequently) more quickly. In this paper, 1) Efficient version of DeepLabCut for posture prediction running on an artificial intelligence server, 2) yolov4 for object detection, and 3) LSTM are used for behavior prediction. Using artificial intelligence technology, it predicts the cat's next, polyuria and frequency of urination through the analysis of the cat's behavior pattern from the home CCTV video and the weight sensor of the water bowl. And, through analysis of cat behavior patterns, we propose an application that reports disease prediction and abnormal behavior to the guardian and delivers it to the guardian's mobile and the main server system.

• Phonetics and Speech Sciences
• /
• v.14 no.4
• /
• pp.45-56
• /
• 2022

This paper proposes an optimal methodology for automatically diagnosing and predicting the severity of the chronic kidney disease (CKD) using patients' utterances. In patients with CKD, the voice changes due to the weakening of respiratory and laryngeal muscles and vocal fold edema. Previous studies have phonetically analyzed the voices of patients with CKD, but no studies have been conducted to classify the voices of patients. In this paper, the utterances of patients with CKD were classified using the variety of utterance types (sustained vowel, sentence, general sentence), the feature sets [handcrafted features, extended Geneva Minimalistic Acoustic Parameter Set (eGeMAPS), CNN extracted features], and the classifiers (SVM, XGBoost). Total of 1,523 utterances which are 3 hours, 26 minutes, and 25 seconds long, are used. F1-score of 0.93 for automatically diagnosing a disease, 0.89 for a 3-classes problem, and 0.84 for a 5-classes problem were achieved. The highest performance was obtained when the combination of general sentence utterances, handcrafted feature set, and XGBoost was used. The result suggests that a general sentence utterance that can reflect all speakers' speech characteristics and an appropriate feature set extracted from there are adequate for the automatic classification of CKD patients' utterances.

• Journal of Internet of Things and Convergence
• /
• v.10 no.1
• /
• pp.69-75
• /
• 2024

This paper proposes an innovative cognitive IoT framework specifically designed for fog computing (FC) in the context of industrial Internet of Things (IIoT). The discourse in this paper is centered on the intricate design and functional architecture of the Cognitive IoT platform. A crucial feature of this platform is the integration of machine learning (ML) and artificial intelligence (AI), which enhances its operational flexibility and compatibility with a wide range of industrial applications. An exemplary application of this platform is highlighted through the Predictive Maintenance-as-a-Service (PdM-as-a-Service) model, which focuses on real-time monitoring of machine conditions. This model transcends traditional maintenance approaches by leveraging real-time data analytics for maintenance and management operations. Empirical results substantiate the platform's effectiveness within a fog computing milieu, thereby illustrating its transformative potential in the domain of industrial IoT applications. Furthermore, the paper delineates the inherent challenges and prospective research trajectories in the spheres of Cognitive IoT and Fog Computing within the ambit of Industrial Internet of Things (IIoT).

• Phonetics and Speech Sciences
• /
• v.16 no.1
• /
• pp.67-76
• /
• 2024

Recent advances in text-to-speech (TTS) technology have significantly improved the quality of synthesized speech, reaching a level where it can closely imitate natural human speech. Especially, TTS models offering various voice characteristics and personalized speech, are widely utilized in fields such as artificial intelligence (AI) tutors, advertising, and video dubbing. Accordingly, in this paper, we propose a one-shot multi-speaker TTS system that can ensure acoustic diversity and synthesize personalized voice by generating speech using unseen target speakers' utterances. The proposed model integrates a speaker encoder into a TTS model consisting of the FastSpeech2 acoustic model and the HiFi-GAN vocoder. The speaker encoder, based on the pre-trained RawNet3, extracts speaker-specific voice features. Furthermore, the proposed approach not only includes an English one-shot multi-speaker TTS but also introduces a Korean one-shot multi-speaker TTS. We evaluate naturalness and speaker similarity of the generated speech using objective and subjective metrics. In the subjective evaluation, the proposed Korean one-shot multi-speaker TTS obtained naturalness mean opinion score (NMOS) of 3.36 and similarity MOS (SMOS) of 3.16. The objective evaluation of the proposed English and Korean one-shot multi-speaker TTS showed a prediction MOS (P-MOS) of 2.54 and 3.74, respectively. These results indicate that the performance of our proposed model is improved over the baseline models in terms of both naturalness and speaker similarity.