• IEMEK Journal of Embedded Systems and Applications
• /
• v.11 no.3
• /
• pp.153-162
• /
• 2016

This paper proposes an optimal dynamic resource allocation method in IoT (Internet of Things) parking guidance system using Q-learning resource allocation model. In the proposed method, a resource allocation using a forecasting model based on Q-learning is employed for optimal utilization of parking guidance system. To demonstrate efficiency and availability of the proposed method, it is verified by computer simulation and practical testbed. Through simulation results, this paper proves that the proposed method can enhance total throughput, decrease penalty fee issued by SLA (Service Level Agreement) and reduce response time with the dynamic number of users.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.14 no.6
• /
• pp.295-311
• /
• 2019

The smart product market is growing year by year and is being used in many areas. There are various ways of interacting with smart products and users by inputting voice recognition, touch and finger movements. It is most important to detect an accurate hand region as a whole step to recognize hand movement. In this paper, we propose a method to detect accurate hand region in real time in various environments. A conventional method of detecting a hand region includes a method using depth information of a multi-sensor camera, a method of detecting a hand through machine learning, and a method of detecting a hand region using a color model. Among these methods, a method using a multi-sensor camera or a method using a machine learning requires a large amount of calculation and a high-performance PC is essential. Many computations are not suitable for embedded systems, and high-end PCs increase or decrease the price of smart products. The algorithm proposed in this paper detects the hand region using the color model, corrects the problems of the existing hand detection algorithm, and detects the accurate hand region based on various experimental environments.

• Information Systems Review
• /
• v.21 no.1
• /
• pp.23-38
• /
• 2019

The social network service (SNS) is one of the important marketing channels, so many companies actively exploit SNSs by posting SNS messages with appropriate content and style for their customers. In this paper, we focused on the psychological distances embedded in the SNS messages and developed a method to measure the psychological distance in SNS message by mixing a traditional content analysis, natural language processing (NLP), and machine learning. Through a traditional content analysis by human coding, the psychological distance was extracted from the SNS message, and these coding results were used for input data for NLP and machine learning. With NLP, word embedding was executed and Bag of Word was created. The Support Vector Machine, one of machine learning techniques was performed to train and test the psychological distance in SNS message. As a result, sensitivity and precision of SVM prediction were significantly low because of the extreme skewness of dataset. We improved the performance of SVM by balancing the ratio of data by upsampling technique and using data coded with the same value in first content analysis. All performance index was more than 70%, which showed that psychological distance can be measured well.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2019.11a
• /
• pp.110-113
• /
• 2019

본 논문에서는 임베디드 시스템에서의 양자화 기계학습을 수행할 경우 발생하는 양자화 오차를 효과적으로 보상하기 위한 방법론을 제안한다. 경사 도함수(Gradient)를 사용하는 기계학습이나 비선형 신호처리 알고리즘에서 양자화 오차는 경사 도함수의 조기 소산(Early Vanishing Gradient)을 야기하여 전체적인 알고리즘의 성능 하락을 가져온다. 이를 보상하기 위하여 경사 도함수의 최대 성분에 대하여 직교하는 방향의 보상 탐색 벡터를 유도하여 양자화 오차로 인한 성능 하락을 보상하도록 한다. 또한, 기존의 고정 학습률 대신, 내부 순환(Inner Loop) 없는 비선형 최적화 알고리즘에 기반한 적응형 학습률 결정 알고리즘을 제안한다. 실험결과 제안한 방식의 알고리즘을 비선형 최적화 문제에 적용할 시 양자화 오차로 인한 성능 하락을 최소화시킬 수 있음을 확인하였다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2017.07a
• /
• pp.9-10
• /
• 2017

본 논문에서 소음, 진동을 이용한 딥러닝 기반 기계 고장진단 임베디드 시스템을 제안하였다. 제안된 시스템은 기계로부터 취득된 소리와 진동을 바탕으로 학습한 DNN모델을 통해 실시간으로 기계 고장을 진단한다. 딥러닝 기술을 사용하여 학습에 따라 적용대상이 변경될 수 있도록 함으로써 특정 기계에 종속적이지 않고 가변적으로 다양한 기계에 대해 고장 예지 및 건전성 관리를 제공하도록 설계하였으며, 이를 증명하기 위해 액추에이터를 환풍기로 설정하여 정상상태와 4가지 비정상상태의 5가지상태를 학습하여 실험한 결과 93%의 정확도를 얻었다.

• IEMEK Journal of Embedded Systems and Applications
• /
• v.17 no.2
• /
• pp.77-83
• /
• 2022

In this paper, the two-stage object detection approach is proposed to implement a deep learning-based label inspection system on edge computing environments. Since the label printed on the products during the production process contains important information related to the product, it is significantly to check the label information is correct. The proposed system uses the lightweight deep learning model that able to employ in the low-performance edge computing devices, and the two-stage object detection approach is applied to compensate for the low accuracy relatively. The proposed Two-Stage object detection approach consists of two object detection networks, Label Area Detection Network and Character Detection Network. Label Area Detection Network finds the label area in the product image, and Character Detection Network detects the words in the label area. Using this approach, we can detect characters precise even with a lightweight deep learning models. The SF-YOLO model applied in the proposed system is the YOLO-based lightweight object detection network designed for edge computing devices. This model showed up to 2 times faster processing time and a considerable improvement in accuracy, compared to other YOLO-based lightweight models such as YOLOv3-tiny and YOLOv4-tiny. Also since the amount of computation is low, it can be easily applied in edge computing environments.

• ETRI Journal
• /
• v.45 no.2
• /
• pp.318-328
• /
• 2023

Recently, embedded systems, such as mobile platforms, have multiple processing units that can operate in parallel, such as centralized processing units (CPUs) and neural processing units (NPUs). We can use deep-learning compilers to generate machine code optimized for these embedded systems from a deep neural network (DNN). However, the deep-learning compilers proposed so far generate codes that sequentially execute DNN operators on a single processing unit or parallel codes for graphic processing units (GPUs). In this study, we propose PartitionTuner, an operator scheduler for deep-learning compilers that supports multiple heterogeneous PUs including CPUs and NPUs. PartitionTuner can generate an operator-scheduling plan that uses all available PUs simultaneously to minimize overall DNN inference time. Operator scheduling is based on the analysis of DNN architecture and the performance profiles of individual and group operators measured on heterogeneous processing units. By the experiments for seven DNNs, PartitionTuner generates scheduling plans that perform 5.03% better than a static type-based operator-scheduling technique for SqueezeNet. In addition, PartitionTuner outperforms recent profiling-based operator-scheduling techniques for ResNet50, ResNet18, and SqueezeNet by 7.18%, 5.36%, and 2.73%, respectively.

• Journal of Broadcast Engineering
• /
• v.24 no.1
• /
• pp.77-86
• /
• 2019

This paper proposes using machine-learning technology to analyze and classify historical collected documents based on them. Data is collected based on keywords associated with a specific domain and the non-conceptuals such as special characters are removed. Then, tag each word of the document collected using a Korean-language morpheme analyzer with its nouns, verbs, and sentences. Embedded documents using Doc2Vec model that converts documents into vectors. Measure the similarity between documents through the embedded model and learn the document classifier using the machine running algorithm. The highest performance support vector machine measured 0.83 of F1-score as a result of comparing the classification model learned.

• Steel and Composite Structures
• /
• v.49 no.1
• /
• pp.109-123
• /
• 2023

Headed studs welded to steel beams and embedded within the concrete of deck slabs are vital components of modern composite floor systems, where safety and economy depend on the accurate predictions of the stud shear resistance. The multitude of existing deck profiles and the complex behavior of studs in deck slab ribs makes developing accurate and reliable mechanical or empirical design models challenging. The paper addresses this issue by presenting a machine learning (ML) model developed from the natural gradient boosting (NGBoost) algorithm capable of producing probabilistic predictions and a database of 464 push-out tests, which is considerably larger than the databases used for developing existing design models. The proposed model outperforms models based on other ML algorithms and existing descriptive equations, including those in EC4 and AISC 360, while offering probabilistic predictions unavailable from other models and producing higher shear resistances for many cases. The present study also showed that the stud shear resistance is insensitive to the concrete elastic modulus, stud welding type, location of slab reinforcement, and other parameters considered important by existing models. The NGBoost model was interpreted by evaluating the feature importance and dependence determined with the SHapley Additive exPlanations (SHAP) method. The model was calibrated via reliability analyses in accordance with the Eurocodes to ensure that its predictions meet the required reliability level and facilitate its use in design. An interactive open-source web application was created and deployed to the cloud to allow for convenient and rapid stud shear resistance predictions with the developed model.

• Journal of Internet of Things and Convergence
• /
• v.9 no.6
• /
• pp.1-9
• /
• 2023

Malware files containing concealed malicious scripts have recently been identified within MS Office documents frequently. In response, this paper describes the design and implementation of a system that automatically detects malicious digital files using machine learning techniques. The system is proficient in identifying malicious scripts within MS Office files that exploit the OLE VBA macro functionality, detecting malicious scripts embedded within the CDH/LFH/ECDR internal field values through OOXML structure analysis, and recognizing abnormal CDH/LFH information introduced within the OOXML structure, which is not conventionally referenced. Furthermore, this paper presents a mechanism for utilizing the VirusTotal malicious script detection feature to autonomously determine instances of malicious tampering within MS Office files. This leads to the design and implementation of a machine learning-based integrated software. Experimental results confirm the software's capacity to autonomously assess MS Office file's integrity and provide enhanced detection performance for arbitrary MS Office files when employing the optimal machine learning model.