• Proceedings of the Korea Information Processing Society Conference
• /
• 2007.11a
• /
• pp.90-93
• /
• 2007

우리나라만의 독창적인 의료체계인 사상체질은 의학적 본질의 우수성에도 불구하고 크게 대중화 되지 않았으며 인지도 또한 높지 못하다. 이는 사상체질에서 가장 중요한 부분이 사상체질의 정확한 분류인데 현재 임상현장에서 행해지고 있는 사상체질 분류 방식은 임상의의 경험과 주관적 소견에 의해 분류되고 있기 때문에 진단 결과에 객관성이 없고 정확도가 낮게 평가되고 있는 실정이다. 이를 위해 사상체질 진단 방법 중 하나인 용모사기론을 IT공학의 영상처리에 적용하여 안면 영상 분석을 통해 사상체질 분류를 수행하고자 한다. 이를 위해 본 논문에서는 사상의학적 원전과 기존의 방법들을 조사, 연구하여 사상체질 분류의 중요한 요소를 결정하고 시스템 구현을 목표로 실험을 통해 사상체질 분류의 유의성을 갖는 측정 요소에 대해 검증해 보고자 한다.

• International Journal of Advanced Culture Technology
• /
• v.8 no.4
• /
• pp.119-128
• /
• 2020

Object detection is an important field of computer vision and is applied to applications such as security, autonomous driving, and face recognition. Recently, as the application of artificial intelligence technology including deep learning has been applied in various fields, it has become a more powerful tool that can learn meaningful high-level, deeper features, solving difficult problems that have not been solved. Therefore, deep learning techniques are also being studied in the field of object detection, and algorithms with excellent performance are being introduced. In this paper, a deep learning-based object detection algorithm used to detect multiple objects in an image is investigated, and future development directions are presented.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2022.06a
• /
• pp.224-227
• /
• 2022

객체 분류는 입력으로 주어진 이미지에 포함된 객체의 종류를 판단하는 기술이다. 대표적인 딥러닝 기반의 객체 분류 방법으로서 Faster R-CNN[2], YOLO[3] 등의 모델이 개발되었으나, 여전히 성능 향상의 여지가 있다. 본 연구에서는 각도 마진 손실 함수를 기존의 몇 가지 객채 분류 모델에 적용하여 성능 향상을 유도한다. 각도 마진 손실 함수는 얼굴 인식 모델인 SphereFace [4]에서 제안한 방법으로, 얼굴 인식과 같이 단일 도메인의 데이터셋을 분류하는 문제를 풀기 위해 제안되었다. 이는 기존 소프트맥스 함수에서 클래스 결정 경계선에 마진을 주는 방식으로 클래스 간의 구분 능력을 향상시킨다. 본 논문은 각도 마진 손실 함수를 CIFAR10, CIFAR100 데이터셋의 분류 문제에 적용하였으며 ResNet, EfficientNet, MobileNet 등의 백본 네트워크로 실험하여 평균적으로 mAP 성능이 향상되는 것을 확인하였다.

• International Journal of Computer Science & Network Security
• /
• v.23 no.1
• /
• pp.125-139
• /
• 2023

Facial expression recognition, a topical problem in the field of computer vision and pattern recognition, is a direct means of recognizing human emotions and behaviors. This paper first summarizes the datasets commonly used for expression recognition and their associated characteristics and presents traditional machine learning algorithms and their benefits and drawbacks from three key techniques of face expression; image pre-processing, feature extraction, and expression classification. Deep learning-oriented expression recognition methods and various algorithmic framework performances are also analyzed and compared. Finally, the current barriers to facial expression recognition and potential developments are highlighted.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.732-734
• /
• 2023

Facial Action Units Detection (FAUs) problem focuses on identifying various detail units expressing on the human face, as defined by the Facial Action Coding System, which constitutes a fine-grained classification problem. This is a challenging task in computer vision. In this study, we propose a Prompt Tuning approach to address this problem, involving a 2-step training process. Our method demonstrates its effectiveness on the Affective in the Wild dataset, surpassing other existing methods in terms of both accuracy and efficiency.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.11a
• /
• pp.458-460
• /
• 2023

가공식품이라는 특정 도메인에서는 기존 검색엔진에서 많이 활용되는 BM25 만을 가지고 최저가 검색하는 데는 어려움이 있다. 본 논문에서는 BM25 외에도 검색의 정확성을 높이기 위해 HuggingFace 에 공개되어 있는 KoELECTRA 를 활용하여 개체명 인식(Named Entity Recognition 과 이진 분류모델(Binary Classification)을 Fine-tuning 하고 BM25 와 연계하여 구축한 검색시스템을 제안한다. 기존의 BM25 대비 성능 평가를 통해 효과를 검증하였다.

• ETRI Journal
• /
• v.46 no.3
• /
• pp.513-525
• /
• 2024

Deep neural networks trained on labeled medical data face major challenges owing to the economic costs of data acquisition through expensive medical imaging devices, expert labor for data annotation, and large datasets to achieve optimal model performance. The heterogeneity of diseases, such as Alzheimer's disease, further complicates deep learning because the test cases may substantially differ from the training data, possibly increasing the rate of false positives. We propose a reconstruction-based self-supervised anomaly detection model to overcome these challenges. It has a dual-subnetwork encoder that enhances feature encoding augmented by skip connections to the decoder for improving the gradient flow. The novel encoder captures local and global features to improve image reconstruction. In addition, we introduce an entropy-based image conversion method. Extensive evaluations show that the proposed model outperforms benchmark models in anomaly detection and classification using an encoder. The supervised and unsupervised models show improved performances when trained with data preprocessed using the proposed image conversion method.

• Journal of IKEEE
• /
• v.21 no.4
• /
• pp.388-396
• /
• 2017

Since there are many variables such as various poses, illuminations and occlusions in a face detection problem, a high performance detection system is required. Although CNN is excellent in image classification, CNN operatioin requires high-performance hardware resources. But low cost low power environments are essential for small and mobile systems. So in this paper, the CPU-FPGA integrated system is designed based on 3-stage cascade CNN architecture using small size FPGA. Adaptive Region of Interest (ROI) is applied to reduce the number of CNN operations using face information of the previous frame. We use a Field Programmable Gate Array(FPGA) to accelerate the CNN computations. The accelerator reads multiple featuremap at once on the FPGA and performs a Multiply-Accumulate (MAC) operation in parallel for convolution operation. The system is implemented on Altera Cyclone V FPGA in which ARM Cortex A-9 and on-chip SRAM are embedded. The system runs at 30FPS with HD resolution input images. The CPU-FPGA integrated system showed 8.5 times of the power efficiency compared to systems using CPU only.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.40 no.1
• /
• pp.19-29
• /
• 2003

This paper presents a novel hybrid classifier for face recognition with artificially generated virtual training samples. We utilize both the nearest neighbor approach in feature angle space and a connectionist model to obtain a synergy effect by combining the results of two heterogeneous classifiers. First, a classifier called the nearest feature angle (NFA), based on angular information, finds the most similar feature to the query from a given training set. Second, a classifier has been developed based on the recall of stored frontal projection of the query feature. It uses a frontal recall network (FRN) that finds the most similar frontal one among the stored frontal feature set. For FRN, we used an ensemble neural network consisting of multiple multiplayer perceptrons (MLPs), each of which is trained independently to enhance generalization capability. Further, both classifiers used the virtual training set generated adaptively, according to the spatial distribution of each person's training samples. Finally, the results of the two classifiers are combined to comprise the best matching class, and a corresponding similarit measure is used to make the final decision. The proposed classifier achieved an average classification rate of 96.33% against a large group of different test sets of images, and its average error rate is 61.5% that of the nearest feature line (NFL) method, and achieves a more robust classification performance.

• Journal of IKEEE
• /
• v.21 no.4
• /
• pp.397-403
• /
• 2017

In the modern society, psychological diseases and impulsive crimes due to stress are occurring. In order to reduce the stress, the existing treatment methods consisted of continuous visit counseling to determine the psychological state and prescribe medication or psychotherapy. Although this face-to-face counseling method is effective, it takes much time to determine the state of the patient, and there is a problem of treatment efficiency that is difficult to be continuously managed depending on the individual situation. In this paper, we propose an artificial intelligence emotion management system that emotions of user monitor in real time and induced to a table state. The system measures multiple bio-signals based on the PPG and the GSR sensors, preprocesses the data into appropriate data types, and classifies four typical emotional states such as pleasure, relax, sadness, and horror through the SVM algorithm. We verify that the emotion of the user is guided to a stable state by providing a real-time emotion management service when the classification result is judged to be a negative state such as sadness or fear through experiments.