• 제어로봇시스템학회:학술대회논문집
• /
• 1990.10b
• /
• pp.1440-1445
• /
• 1990

The system with softfware packages for control system design unifying and encompassing rule based control and conventional control based on numerical models were developed. Users who are not familiar with control theory, numerical computing, and artificial intelligence (AI) can perform system analysis, control design and development of AI control system without difficulty.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.15 no.6
• /
• pp.2115-2127
• /
• 2021

Although significant progress has been made in synthesizing visually realistic face images by Generative Adversarial Networks (GANs), there still lacks effective approaches to provide fine-grained control over the generation process for semantic facial attribute editing. In this work, we propose a novel cross channel self-attention based generative adversarial network (CCA-GAN), which weights the importance of multiple channels of features and archives pixel-level feature alignment and conversion, to reduce the impact on irrelevant attributes while editing the target attributes. Evaluation results show that CCA-GAN outperforms state-of-the-art models on the CelebA dataset, reducing Fréchet Inception Distance (FID) and Kernel Inception Distance (KID) by 15~28% and 25~100%, respectively. Furthermore, visualization of generated samples confirms the effect of disentanglement of the proposed model.

• Journal of Information Processing Systems
• /
• v.17 no.3
• /
• pp.571-585
• /
• 2021

Applications for object detection are expanding as it is automated through artificial intelligence-based processing, such as deep learning, on a large volume of images and videos. High dependence on training data and a non-transparent way to find answers are the common characteristics of deep learning. Attacks on training data and training models have emerged, which are closely related to the nature of deep learning. Privacy, integrity, and robustness for the extracted information are important security issues because deep learning enables object recognition in images and videos. This paper summarizes the security issues that need to be addressed for future applications and analyzes the state-of-the-art security studies related to robustness, privacy, and integrity of object detection for images and videos.

• Electronics and Telecommunications Trends
• /
• v.37 no.5
• /
• pp.22-32
• /
• 2022

Optical character recognition is a primary technology required in different fields, including digitizing archival documents, industrial automation, automatic driving, video analytics, medicine, and financial institution, among others. It was created in 1928 using pattern matching, but with the advent of artificial intelligence, it has since evolved into a high-performance character recognition technology. Recently, methods for detecting curved text and characters existing in a complicated background are being studied. Additionally, deep learning models are being developed in a way to recognize texts in various orientations and resolutions, perspective distortion, illumination reflection and partially occluded text, complex font characters, and special characters and artistic text among others. This report reviews the recent deep learning-based text detection and recognition methods and their various applications.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.18 no.3
• /
• pp.101-108
• /
• 2023

Optimizing the performance of deep neural networks on embedded systems is a challenging task that requires efficient compilers and runtime systems. We propose a TVM-based approach that consists of three steps: quantization, auto-scheduling, and ahead-of-time compilation. Our approach reduces the computational complexity of models without significant loss of accuracy, and generates optimized code for various hardware platforms. We evaluate our approach on three representative CNNs using ImageNet Dataset on the NVIDIA Jetson AGX Xavier board and show that it outperforms baseline methods in terms of processing speed.

• Annual Conference of KIPS
• /
• 2023.05a
• /
• pp.678-680
• /
• 2023

본 논문은 정확하면서도 효율적인 한국어 문장 분류 기법에 대해서 논의한다. 최근 자연어처리 분야에서 사전 학습된 언어 모델(Pre-trained Language Models, PLM)은 미세조정(fine-tuning)을 통해 문장 분류 하위 작업(downstream task)에서 성공적인 결과를 보여주고 있다. 하지만, 이러한 미세조정은 하위 작업이 바뀔 때마다 사전 학습된 언어 모델의 전체 매개변수(model parameters)를 학습해야 한다는 단점을 갖고 있다. 본 논문에서는 이러한 문제를 해결할 수 있도록 도메인 적응기(domain adapter)를 활용한 한국어 문장 분류 프레임워크인 DAKS(Domain Adaptation-based Korean Sentence classification framework)를 제안한다. 해당 프레임워크는 학습되는 매개변수의 규모를 크게 줄임으로써 효율적인 성능을 보였다. 또한 문장 분류를 위한 특징(feature)으로써 한국어 사전학습 모델(KLUE-RoBERTa)의 다양한 은닉 계층 별 은닉 상태(hidden states)를 활용하였을 때 결과를 비교 분석하고 가장 적합한 은닉 계층을 제시한다.

• Journal of information and communication convergence engineering
• /
• v.22 no.3
• /
• pp.256-266
• /
• 2024

Sasang constitutional medicine (SCM) is one of the best traditional therapeutic approaches used in Korea. SCM prioritizes personalized treatment that considers the unique constitution of an individual and encompasses their physical characteristics, personality traits, and susceptibility to specific diseases. Facial features are essential for diagnosing Sasang constitutional types (SCTs). This study aimed to develop a real-time artificial intelligence-based model for diagnosing SCTs using facial images, building an SCTs prediction model based on a machine learning method. Facial features from all images were extracted to develop this model using feature engineering and machine learning techniques. The fusion of these features was used to train the AI model. We used four machine learning algorithms, namely, random forest (RF), multilayer perceptron (MLP), gradient boosting machine (GBM), and extreme gradient boosting (XGB), to investigate SCTs. The GBM outperformed all the other models. The highest accuracy achieved in the experiment was 81%, indicating the robustness of the proposed model and suitability for real-time applications.

• International Journal of Internet, Broadcasting and Communication
• /
• v.16 no.3
• /
• pp.27-32
• /
• 2024

Reliable and fine-grained musical metadata are required for efficient search of rapidly increasing music files. In particular, since the primary motive for listening to music is its emotional effect, diversion, and the memories it awakens, emotion classification along with genre classification of music is crucial. In this paper, as an initial approach towards a "ground-truth" dataset for music emotion and genre classification, we elaborately generated a music corpus through labeling of a large number of ordinary people. In order to verify the suitability of the dataset through the classification results, we extracted features according to MPEG-7 audio standard and applied different machine learning models based on statistics and deep neural network to automatically classify the dataset. By using standard hyperparameter setting, we reached an accuracy of 93% for genre classification and 80% for emotion classification, and believe that our dataset can be used as a meaningful comparative dataset in this research field.

• Journal of Sensor Science and Technology
• /
• v.33 no.5
• /
• pp.248-254
• /
• 2024

Artificial intelligence techniques have improved fire-detection methods; however, false alarms still occur. Conventional methods detect fires using current sensors, which can lead to detection errors due to temporary environmental changes or noise. Thus, fire-detection methods must include a trend analysis of past information. We propose a deep-learning-based fire detection method using multi-sensor data and Kendall's tau. The proposed system used a BiLSTM model to predict fires using pre-processed multi-sensor data and extracted trend information. Kendall's tau indicates the trend of a time-series data as a score; therefore, it is easy to obtain a target pattern. The experimental results showed that the proposed system with trend values recorded an accuracy of 99.93% for BiLSTM and GRU models in a 20-tap moving average filter and 40% fire threshold. Thus, the proposed trend approach is more accurate than that of conventional approaches.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.1
• /
• pp.283-291
• /
• 2021

In this paper, we propose the development and application of a software education program for underachievers. The software education program for underachieving students was developed in consideration of the characteristics of learner's suffering from underachievement and the educational effects of software education, and is meaningful in that it proposes a plan to improve the learning gap in distance learning. Learners can acquire digital literacy and learning skills by solving structured tasks in the form of courseware, intelligent tutoring, debugging, and artificial intelligence learning models in educational programs. Based on the effects of software education, such as enhancing logical thinking ability and problem solving ability, this program provides opportunities to solve fusion tasks to underachievers. Based on this, it is expected that it can have a positive effect on the overall academic work.