• Korean Journal of Artificial Intelligence
• /
• v.11 no.4
• /
• pp.1-8
• /
• 2023

In this paper, we propose a black ice detection platform framework using Convolutional Neural Networks (CNNs). To overcome black ice problem, we introduce a real-time based early warning platform using CNN-based architecture, and furthermore, in order to enhance the accuracy of black ice detection, we apply a multi-scale dilation convolution feature fusion (MsDC-FF) technique. Then, we establish a specialized experimental platform by using a comprehensive dataset of thermal road black ice images for a training and evaluation purpose. Experimental results of a real-time black ice detection platform show the better performance of our proposed network model compared to conventional image segmentation models. Our proposed platform have achieved real-time segmentation of road black ice areas by deploying a road black ice area segmentation network on the edge device Jetson Nano devices. This approach in parallel using multi-scale dilated convolutions with different dilation rates had faster segmentation speeds due to its smaller model parameters. The proposed MsCD-FF Net(2) model had the fastest segmentation speed at 5.53 frame per second (FPS). Thereby encouraging safe driving for motorists and providing decision support for road surface management in the road traffic monitoring department.

• Imaging Science in Dentistry
• /
• v.53 no.4
• /
• pp.271-281
• /
• 2023

Purpose: The objective of this scoping review was to investigate the applicability and performance of various convolutional neural network (CNN) models in tooth numbering on panoramic radiographs, achieved through classification, detection, and segmentation tasks. Materials and Methods: An online search was performed of the PubMed, Science Direct, and Scopus databases. Based on the selection process, 12 studies were included in this review. Results: Eleven studies utilized a CNN model for detection tasks, 5 for classification tasks, and 3 for segmentation tasks in the context of tooth numbering on panoramic radiographs. Most of these studies revealed high performance of various CNN models in automating tooth numbering. However, several studies also highlighted limitations of CNNs, such as the presence of false positives and false negatives in identifying decayed teeth, teeth with crown prosthetics, teeth adjacent to edentulous areas, dental implants, root remnants, wisdom teeth, and root canal-treated teeth. These limitations can be overcome by ensuring both the quality and quantity of datasets, as well as optimizing the CNN architecture. Conclusion: CNNs have demonstrated high performance in automated tooth numbering on panoramic radiographs. Future development of CNN-based models for this purpose should also consider different stages of dentition, such as the primary and mixed dentition stages, as well as the presence of various tooth conditions. Ultimately, an optimized CNN architecture can serve as the foundation for an automated tooth numbering system and for further artificial intelligence research on panoramic radiographs for a variety of purposes.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.24 no.1
• /
• pp.141-147
• /
• 2024

Recently, there has been growing interest in personalized services tailored to an individual's preferences. This has led to ongoing research aimed at recognizing and leveraging an individual's personality traits. Among various methods for personality assessment, the OCEAN model stands out as a prominent approach. In utilizing OCEAN for personality recognition, a multi modal artificial intelligence model that incorporates linguistic, paralinguistic, and non-linguistic information is often employed. This paper examines the impact of the margin value set for extracting facial areas from video data on the accuracy of a personality recognition model that uses facial expressions to determine OCEAN traits. The study employed personality recognition models based on 2D Patch Partition, R2plus1D, 3D Patch Partition, and Video Swin Transformer technologies. It was observed that setting the facial area extraction margin to 60 resulted in the highest 1-MAE performance, scoring at 0.9118. These findings indicate the importance of selecting an optimal margin value to maximize the efficiency of personality recognition models.

• Journal of the Korea Society of Computer and Information
• /
• v.29 no.9
• /
• pp.79-87
• /
• 2024

This study aims to improve communication for people with hearing impairments by developing artificial intelligence models that recognize and classify emotions from voice data. To achieve this, we utilized three major AI models: CNN-Transformer, HuBERT-Transformer, and Wav2Vec 2.0, to analyze users' voices in real-time and classify their emotions. To effectively extract features from voice data, we applied transformation techniques such as Mel-Frequency Cepstral Coefficient (MFCC), aiming to accurately capture the complex characteristics and subtle changes in emotions within the voice. Experimental results showed that the HuBERT-Transformer model demonstrated the highest accuracy, proving the effectiveness of combining pre-trained models and complex learning structures in the field of voice-based emotion recognition. This research presents the potential for advancements in emotion recognition technology using voice data and seeks new ways to improve communication and interaction for individuals with hearing impairments, marking its significance.

• Journal of the Korea Society of Computer and Information
• /
• v.28 no.12
• /
• pp.147-153
• /
• 2023

As it spreads to all industries of artificial intelligence technology, AIaaS equipped with artificial intelligence services is emerging. In particular, non-IT companies are suffering from the absence of software experts, difficulties in training big data models, and difficulties in collecting and analyzing various types of data. AIaaS makes it easier and more economical for users to build a system by providing various IT resources necessary for artificial intelligence software development as well as functions necessary for artificial intelligence software in the form of a service. Therefore, the supply and demand for such cloud-based AIaaS services will increase rapidly. However, the quality of services provided by AIaaS becomes an important factor in what is required as the supply and demand for AIaaS increases. However, research on a comprehensive and practical quality evaluation metric to measure this is currently insufficient. Therefore, in this paper, we develop and propose a usability, replacement, scalability, and publicity metric, which are the four metrics necessary for measuring reusability, based on implementation, convenience, efficiency, and accessibility, which are characteristics of AIaaS, for reusability evaluation among the service quality measurement factors of AIaaS. The proposed metrics can be used as a tool to predict how much services provided by AIaaS can be reused for potential users in the future.

• Smart Structures and Systems
• /
• v.26 no.4
• /
• pp.403-418
• /
• 2020

In recent years, the use of Fiber Reinforced Polymers (FRPs) as one of the most common ways to increase the strength of concrete samples, has been introduced. Evaluation of the final strength of these specimens is performed with different experimental methods. In this research, due to the variety of models, the low accuracy and impact of different parameters, the use of new intelligence methods is considered. Therefore, using artificial intelligent-based models, a new solution for evaluating the bond strength of FRP is presented in this paper. 150 experimental samples were collected from previous studies, and then two new hybrid models of Imperialist Competitive Algorithm (ICA)-Artificial Neural Network (ANN) and Artificial Bee Colony (ABC)-ANN were developed. These models were evaluated using different performance indices and then, a comparison was made between the developed models. The results showed that the ICA-ANN model's ability to predict the bond strength of FRP is higher than the ABC-ANN model. Finally, to demonstrate the capabilities of this new model, a comparison was made between the five experimental models and the results were presented for all data. This comparison showed that the new model could offer better performance. It is concluded that the proposed hybrid models can be utilized in the field of this study as a suitable substitute for empirical models.

• Proceedings of the KSRS Conference
• /
• 2000.04a
• /
• pp.9-9
• /
• 2000

For the planning of future land use for economic activities, an essential component is the identification of the vulnerable areas for natural hazard and environmental impacts from the activities. Also, exploration for mineral and energy resources is carried out by a step by step approach. At each step, a selection of the target area for the next exploration strategy is made based on all the data harnessed from the previous steps. The uncertainty of the selected target area containing undiscovered resources is a critical factor for estimating the exploration risk. We have developed not only spatial prediction models based on adapted artificial intelligence techniques to predict target and vulnerable areas but also validation techniques to estimate the uncertainties associated with the predictions. The prediction models will assist the scientists and decision-makers to make two critical decisions: (i) of the selections of the target or vulnerable areas, and (ii) of estimating the risks associated with the selections.

• Journal of Distribution Science
• /
• v.15 no.4
• /
• pp.69-78
• /
• 2017

Purpose - The study was AI as exploratory study on artificial intelligence (AI) shopping information services, to explore the possibility of a new business of the distribution industry. For research, we compare to IBM of consumer awareness surveys an AI shopping information service for retailing channel and target goods group. Finally, we present to service scenario for distribution service using AI. Research design, data, and methodology - First, to identify possible the success of the information service shopping using AI, AI technology for the consumer is very important for the acceptance of judgement. Therefore, we explored the possibility of AI information service for business as a shopping. The experimental data were used to interpret the meaning of the relevant literature and the IBM Institute of Business Value (IBV) Report 2015. This research is based on the use of a technical acceptance model (TAM) to determine whether the consumer would adopt the 'AI shopping information service' technology. Step 1 of the process assumes that the consumer adopts AI technology. In step 2, consumers find their preference channels and goods targeted at them as per their preferences. Finally Step 3, we present scenario for 'AI shopping information service' based on the results of Step 1 and 2. Results - Consumers have expressed their high interests in the new shopping information services, especially the on/off line distribution channels can use shopping information to increase the efficiency in provision of goods. Digital channel (such as SNS, online shopping etc.) is especially high value goods such as cars, furniture, and home appliances by displaying it to an appropriate product group. Conclusions - The study reveals the potential for the use of new business models such as 'AI shopping information service' by the distribution industry. We present seven scenario related AI application refer from IBM suggestion, and the findings would enable the distribution industry to approach target consumers with their products, especially high value goods. 'Shopping advisor' is considered to the most effective. In order to apply to the other field of the distribution industry business, which utilizes AI technology, it should be accompanied by additional empirical data analysis should be undertaken.

• International Journal of Advanced Culture Technology
• /
• v.9 no.2
• /
• pp.100-105
• /
• 2021

Contactless service is rapidly emerging as a new growth strategy due to consumers who are reluctant to the face-to-face situation in the global pandemic of coronavirus disease 2019 (COVID-19), and various technologies are being developed to support the fast-growing contactless service market. In particular, the restaurant industry is one of the most desperate industrial fields requiring technologies for contactless service, and the representative technical case should be a kiosk, which has the advantage of reducing labor costs for the restaurant owners and provides psychological relaxation and satisfaction to the customer. In this paper, we propose a solution to the restaurant's store operation through the unmanned kiosk using a state-of-the-art artificial intelligence (AI) technology of image recognition. Especially, for the products that do not have barcodes in bakeries, fresh foods (fruits, vegetables, etc.), and autonomous restaurants on highways, which cause increased labor costs and many hassles, our proposed system should be very useful. The proposed system recognizes products without barcodes on the ground of image-based AI algorithm technology and makes automatic payments. To test the proposed system feasibility, we established an AI vision system using a commercial camera and conducted an image recognition test by training object detection AI models using donut images. The proposed system has a self-learning system with mismatched information in operation. The self-learning AI technology allows us to upgrade the recognition performance continuously. We proposed a fully automated payment system with AI vision technology and showed system feasibility by the performance test. The system realizes contactless service for self-checkout in the restaurant business area and improves the cost-saving in managing human resources.

• Journal of Convergence for Information Technology
• /
• v.12 no.2
• /
• pp.157-162
• /
• 2022

In a society full of knowledge and information, digital literacy and artificial intelligence (AI) education that can utilize AI technology is needed to solve numerous everyday problems based on computational thinking. In this study, data-centered AI education was conducted while teaching computer programming to non-computer programming students at universities, and the correlation between major factors related to academic performance was analyzed in addition to student satisfaction surveys. The results indicated that there was a strong correlation between grades and problem-solving ability-based tasks, and learning satisfaction. Multiple regression analysis also showed a significant effect on grades (F=225.859, p<0.001), and student satisfaction was high. The non-computer programming students were also able to understand the importance of data and the concept of AI models, focusing on specific examples of project types, and confirmed that they could use AI smoothly in their fields of interest. If further cases of AI education are explored and students' AI education is activated, it will be possible to suggest its direction that can collaborate with experts through interest in AI technology.