• 반도체디스플레이기술학회지
• /
• 제20권1호
• /
• pp.64-69
• /
• 2021

In this paper, a method for improving the defect classification performance in low contrast, ununiformity and featureless steel plate surfaces has been studied based on deep convolution neural network and transfer-learning neural network. The steel plate surface images have low contrast, ununiformity, and featureless, so that the contrast between defect and defect-free regions are not discriminated. These characteristics make it difficult to extract the feature of the surface defect image. A classifier based on a deep convolution neural network is constructed to extract features automatically for effective classification of images with these characteristics. As results of the experiment, AlexNet-based transfer-learning classifier showed excellent classification performance of 99.43% with less than 160 seconds of training time. The proposed classification system showed excellent classification performance for low contrast, ununiformity, and featureless surface images.

• 센서학회지
• /
• 제30권2호
• /
• pp.76-81
• /
• 2021

In this study, we propose a machine vision system with a high object recognition rate. By utilizing a multiple-exposure image sensing technique, the proposed deep learning-based machine vision system can cover a wide light intensity range without further learning processes on the various light intensity range. If the proposed machine vision system fails to recognize object features, the system operates in a multiple-exposure sensing mode and detects the target object that is blocked in the near dark or bright region. Furthermore, short- and long-exposure images from the multiple-exposure sensing mode are synthesized to obtain accurate object feature information. That results in the generation of a wide dynamic range of image information. Even with the object recognition resources for the deep learning process with a light intensity range of only 23 dB, the prototype machine vision system with the multiple-exposure imaging method demonstrated an object recognition performance with a light intensity range of up to 96 dB.

• Journal of Semiconductor Engineering
• /
• 제2권2호
• /
• pp.130-135
• /
• 2021

Recently, deep neural networks (DNNs) are actively used for action control so that an autonomous system, such as the robot, can perform human-like behaviors and operations. Unlike recognition tasks, the real-time operation is essential in action control, and it is too slow to use remote learning on a server communicating through a network. New learning techniques, such as reinforcement learning (RL), are needed to determine and select the correct robot behavior locally. In this paper, we propose an energy-efficient DNN processor with a LUT-based processing engine and near-zero skipper. A CNN-based facial emotion recognition and an RNN-based emotional dialogue generation model is integrated for natural HRI system and tested with the proposed processor. It supports 1b to 16b variable weight bit precision with and 57.6% and 28.5% lower energy consumption than conventional MAC arithmetic units for 1b and 16b weight precision. Also, the near-zero skipper reduces 36% of MAC operation and consumes 28% lower energy consumption for facial emotion recognition tasks. Implemented in 65nm CMOS process, the proposed processor occupies 1784×1784 um2 areas and dissipates 0.28 mW and 34.4 mW at 1fps and 30fps facial emotion recognition tasks.

• 정보처리학회논문지:소프트웨어 및 데이터공학
• /
• 제9권6호
• /
• pp.177-186
• /
• 2020

최근 딥러닝(Deep Learning) 기술의 발전에 따라 컴퓨터 비전(Computer Vision) 분야의 이미지 인식 성능이 향상되고 있으며, 또한 Serverless Computing이 이벤트 기반의 클라우드 애플리케이션 개발 및 서비스를 위한 차세대 클라우드 컴퓨팅 기술로 각광받고 있어 딥러닝과 Serverless Computing 기술을 접목하여 실생활에 이미지 인식 서비스를 사용하고자 하는 시도가 증가하고 있다. 따라서 본 논문에서는 Serverless Computing 기술을 활용하여 효율적인 딥러닝 기반 이미지 인식 서비스 시스템 개발 방법을 기술한다. 제안하는 시스템은 Serverless Computing 기반 AWS Lambda Server를 이용하여 적은 비용으로 대형 신경망 모델을 사용자에게 서비스할 수 있는 방법을 제안한다. 또한 AWS Lambda Server의 단점인 Cold Start Time 문제와 용량제한 문제를 해결하여 효과적으로 대형 신경망 모델을 사용하는 Serverless Computing 시스템을 구축할 수 있음을 보인다. 실험을 통해 AWS Lambda Serverless Computing 기술을 활용하여 본 논문에서 제안한 시스템이 비용 절감뿐만 아니라 처리 시간 및 용량제한 문제를 해결하여 대형 신경망 모델을 서비스하기에 효율적인 성능을 보임을 확인하였다.

• 정보과학회 논문지
• /
• 제44권11호
• /
• pp.1156-1164
• /
• 2017

기존의 버그 담당자 자동 배정 연구들은 대부분 기계학습 알고리즘을 기반으로 예측 시스템을 구축하는 방식이었다. 따라서, 고성능의 기계학습 모델을 적용하는 것이 담당자 자동 배정 시스템 성능의 핵심이 된다고 할 수 있으며 관련 연구에서는 높은 성능을 보이는 SVM, Naive Bayes 등의 기계학습 모델들이 주로 사용되고 있다. 본 논문에서는 기계학습 분야에서 최근 좋은 성능을 보이고 있는 딥 러닝을 버그 담당자 자동 배정에 적용하고 그 성능을 평가한다. 실험 결과, 딥 러닝 기반 Bug Triage 시스템이 활성 개발자 대상 실험에서 48%의 정확도를 달성했으며 이는 기존의 기계학습 대비 최대 69%향상된 결과이다.

• Journal of Positioning, Navigation, and Timing
• /
• 제10권1호
• /
• pp.49-54
• /
• 2021

Indoor positioning system becomes of increasing interests due to the demands for accurate indoor location information where Global Navigation Satellite System signal does not approach. Wi-Fi access points (APs) built in many construction in advance helps developing a Wi-Fi Received Signal Strength Indicator (RSSI) based indoor localization. This localization method first collects pairs of position and RSSI measurement set, which is called fingerprint database, and then estimates a user's position when given a query measurement set by comparing the fingerprint database. The challenge arises from nonlinearity and noise on Wi-Fi RSSI measurements and complexity of handling a large amount of the fingerprint data. In this paper, machine learning techniques have been applied to implement Wi-Fi based localization. However, most of existing indoor localizations focus on single position estimation. The main contribution of this paper is to develop multi-target localization by using deep neural, which is beneficial when a massive crowd requests positioning service. This paper evaluates the proposed multilocalization based on deep learning from a multi-story building, and analyses its learning effect as increasing number of target positions.

• Journal of Information Technology Applications and Management
• /
• 제29권3호
• /
• pp.43-55
• /
• 2022

In this study, with the goal of developing a deep learning-based product recommendation model for effective matching of influencers and products, a deep learning model with a collaborative filtering model combined with generalized matrix decomposition(GMF), a collaborative filtering model based on multi-layer perceptron (MLP), and neural collaborative filtering and generalized matrix Factorization (NeuMF), a hybrid model combining GMP and MLP was developed and tested. In particular, we utilize one-class problem free boosting (OCF-B) method to solve the one-class problem that occurs when training is performed only on positive cases using implicit feedback in the deep learning-based collaborative filtering recommendation model. In relation to model selection based on overall experimental results, the MLP model showed highest performance with weighted average precision, weighted average recall, and f1 score were 0.85 in the model (n=3,000, term=15). This study is meaningful in practice as it attempted to commercialize a deep learning-based recommendation system where influencer's promotion data is being accumulated, pactical personalized recommendation service is not yet commercially applied yet.

• 문화기술의 융합
• /
• 제9권3호
• /
• pp.647-654
• /
• 2023

딥러닝은 이미지, 텍스트와 같이 복잡한 데이터를 분류 및 인식하는데 유용한 방법으로 딥러닝 기법의 정확도는 딥러닝이 인터넷상의 AI 기반의 서비스를 유용하게 하는데 기초가 되었다. 그러나 딥러닝에서 훈련에 사용되는 방대한 양의 사용자 데이터는 사생활 침해 문제를 야기하였고 사진이나 보이스와 같이 사용자이 개인적이고 민감한 데이터를 수집한 기업들이 데이터들을 무기한으로 소유한다. 사용자들은 자신의 데이터를 삭제할 수 없고 사용되는 목적도 제한할 수 없다. 예를 들면, 환자 진료기록에 대한 딥러닝 기술을 적용하기 원하는 의료기관들과 같은 데이터소유자들은 사생활과 기밀유지 문제로 환자의 데이터를 공유할 수 없고 딥러닝 기술의 혜택을 받기 어렵다. 우리는 멀티 파티 시스템에서 다수의 작업자들이 입력 데이터집합을 공유하지 않고 신경망 모델을 공동으로 사용할 수 있는 프라이버시 보존 기술을 적용한 딥러닝 방법을 설계한다. 변형된 확률적 경사 하강에 기초한 최적화 알고리즘을 이용하여 하위 집합을 선택적으로 공유할 수 있는 방법을 이용하였고 결과적으로 개인정보를 보호하면서 학습 정확도를 증가시킨 학습을 할 수 있도록 하였다.

• 스마트미디어저널
• /
• 제11권2호
• /
• pp.39-52
• /
• 2022

In the context of the evolution of automation and intelligence, deep learning and machine learning algorithms have been widely applied in aquaculture in recent years, providing new opportunities for the digital realization of aquaculture. Especially, water quality management deserves attention thanks to its importance to food organisms. In this study, we proposed an end-to-end deep learning-based TabNet model for water quality prediction. From major indexes of water quality assessment, we applied novel deep learning techniques and machine learning algorithms in innovative fish aquaculture to predict the number of water cells counting. Furthermore, the application of deep learning in aquaculture is outlined, and the obtained results are analyzed. The experiment on in-house data showed an optimistic impact on the application of artificial intelligence in aquaculture, helping to reduce costs and time and increase efficiency in the farming process.

• 반도체디스플레이기술학회지
• /
• 제20권1호
• /
• pp.88-93
• /
• 2021

The automotive industry is developing day by day. Among them, it is very important to prevent accidents while driving. However, despite the importance of developing automobile industry technology, accidents due to road defects increase every year, especially in the rainy season. To this end, we proposed a road defect detection system for road management by converging deep learning and raspberry pi, which show various possibilities. In this paper, we developed a system that visually displays through a map after analyzing the images captured by the Raspberry Pi and the route GPS. The deep learning model trained for this system achieved 96% accuracy. Through this system, it is expected to manage road defects efficiently at a low cost.