• 한국축산식품학회지
• /
• 제38권2호
• /
• pp.362-375
• /
• 2018

This study proposed a rapid microscopic examination method for pork freshness evaluation by using the self-assembled hyperspectral microscopic imaging (HMI) system with the help of feature extraction algorithm and pattern recognition methods. Pork samples were stored for different days ranging from 0 to 5 days and the freshness of samples was divided into three levels which were determined by total volatile basic nitrogen (TVB-N) content. Meanwhile, hyperspectral microscopic images of samples were acquired by HMI system and processed by the following steps for the further analysis. Firstly, characteristic hyperspectral microscopic images were extracted by using principal component analysis (PCA) and then texture features were selected based on the gray level co-occurrence matrix (GLCM). Next, features data were reduced dimensionality by fisher discriminant analysis (FDA) for further building classification model. Finally, compared with linear discriminant analysis (LDA) model and support vector machine (SVM) model, good back propagation artificial neural network (BP-ANN) model obtained the best freshness classification with a 100 % accuracy rating based on the extracted data. The results confirm that the fabricated HMI system combined with multivariate algorithms has ability to evaluate the fresh degree of pork accurately in the microscopic level, which plays an important role in animal food quality control.

• 한국항해항만학회:학술대회논문집
• /
• 한국항해항만학회 2015년도 춘계학술대회
• /
• pp.310-311
• /
• 2015

선박 항적 데이터는 해상교통관제센터에 의해 실시간으로 모니터링 되고 수집되어 진다. 이러한 데이터를 기반으로 선박의 항적 패턴분석과 항적 모델을 추출하여 해상교통관제사의 의사결정에 기여하고자 한다. 항적 데이터의 처리와 가공, 항적 모델링을 위하여 SVM알고리즘이 사용되었으며, 적정 파라미터 선정을 위하여 k-fold cross validation이 사용되었다. 제안된 항적 데이터 모델링을 통하여 이상거동 선박의 사전 판별, 선박의 추측위치 계산 등에 응용하여 해상교통과제사의 의사결정을 지원하고자 한다.

• 정보관리학회지
• /
• 제35권2호
• /
• pp.37-62
• /
• 2018

문헌정보학 분야의 국내 학술지 논문으로 구성된 문헌집합을 대상으로 기계학습에 기초한 자동분류의 성능에 영향을 미치는 요소들을 검토하였다. 특히, "정보관리학회지"에 수록된 논문에 주제 범주를 자동 할당하는 분류 성능 측면에서 용어 가중치부여 기법, 학습집합 크기, 분류 알고리즘, 범주 할당 방법 등 주요 요소들의 특성을 다각적인 실험을 통해 살펴보았다. 결과적으로 분류 환경 및 문헌집합의 특성에 따라 각 요소를 적절하게 적용하는 것이 효과적이며, 보다 단순한 모델의 사용으로 상당히 좋은 수준의 성능을 도출할 수 있었다. 또한, 국내 학술지 논문의 분류는 특정 논문에 하나 이상의 범주를 할당하는 복수-범주 분류(multi-label classification)가 실제 환경에 부합한다고 할 수 있다. 따라서 이러한 환경을 고려하여 단순하고 빠른 분류 알고리즘과 소규모의 학습집합을 사용하는 최적의 분류 모델을 제안하였다.

• Genomics & Informatics
• /
• 제4권4호
• /
• pp.161-166
• /
• 2006

The establishment of DNA microarray technology has enabled high-throughput analysis and molecular profiling of various types of cancers. By using the gene expression data from microarray analysis we are able to investigate diagnostic applications at the molecular level. The most important step in the application of microarray technology to cancer diagnostics is the selection of specific markers from gene expression profiles. In order to select markers of Immortalization and transformation we used c-myc and $H-ras^{V12}$ oncogene-transfected NIH3T3 cells as our model system. We have identified 8751 differentially expressed genes in the immortalization/transformation model by multivariate permutation F-test (95% confidence, FDR<0.01). Using the support vector machine algorithm, we selected 13 discriminative genes which could be used to predict immortalization and transformation with perfect accuracy. We assayed $H-ras^{V12}$-transfected 'transformed' cells to validate our immortalization/transformation dassification system. The selected molecular markers generated valuable additional information for tumor diagnosis, prognosis and therapy development.

• 대한산업공학회지
• /
• 제39권4호
• /
• pp.271-277
• /
• 2013

In spite of the emphasis on quality control in auto-industry, most of subcontract enterprises still lack a systematic in-process quality monitoring system for predicting the product/part quality for their customers. While their manufacturing processes have been getting automated and computer-controlled ever, there still exist many uncertain parameters and the process controls still rely on empirical works by a few skilled operators and quality experts. In this paper, a real-time product quality monitoring system for auto-manufacturing industry is presented to provide the systematic method of predicting product qualities from real-time production data. The proposed framework consists of a product quality ontology model for complex manufacturing supply chain environments, and a real-time quality prediction tool using support vector machine algorithm that enables the quality monitoring system to classify the product quality patterns from the in-process production data. A door trim production example is illustrated to verify the proposed quality prediction model.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권10호
• /
• pp.3708-3728
• /
• 2021

Indian herbal plants are used in agriculture and in the food, cosmetics, and pharmaceutical industries. Laboratory-based tests are routinely used to identify and classify similar herb species by analyzing their internal cell structures. In this paper, we have applied computer vision techniques to do the same. The original leaf image was preprocessed using the Chan-Vese active contour segmentation algorithm to efface the background from the image by setting the contraction bias as (v) -1 and smoothing factor (µ) as 0.5, and bringing the initial contour close to the image boundary. Thereafter the segmented grayscale image was fed to a leaky capacitance fired neuron model (LCFN), which differentiates between similar herbs by combining different groups of pixels in the leaf image. The LFCN's decay constant (f), decay constant (g) and threshold (h) parameters were empirically assigned as 0.7, 0.6 and h=18 to generate the 1D feature vector. The LCFN time sequence identified the internal leaf structure at different iterations. Our proposed framework was tested against newly collected herbal species of natural images, geometrically variant images in terms of size, orientation and position. The 1D sequence and shape features of aloe, betel, Indian borage, bittergourd, grape, insulin herb, guava, mango, nilavembu, nithiyakalyani, sweet basil and pomegranate were fed into the 5-fold Bayesian regularization neural network (BRNN), K-nearest neighbors (KNN), support vector machine (SVM), and ensemble classifier to obtain the highest classification accuracy of 91.19%.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제15권6호
• /
• pp.2168-2187
• /
• 2021

The smart grid replaces the traditional power structure with information inventiveness that contributes to a new physical structure. In such a field, malicious information injection can potentially lead to extreme results. Incorrect, FDI attacks will never be identified by typical residual techniques for false data identification. Most of the work on the detection of FDI attacks is based on the linearized power system model DC and does not detect attacks from the AC model. Also, the overwhelming majority of current FDIA recognition approaches focus on FDIA, whilst significant injection location data cannot be achieved. Building on the continuous developments in deep learning, we propose a Deep Learning based Locational Detection technique to continuously recognize the specific areas of FDIA. In the development area solver gap happiness is a False Data Detector (FDD) that incorporates a Convolutional Neural Network (CNN). The FDD is established enough to catch the fake information. As a multi-label classifier, the following CNN is utilized to evaluate the irregularity and cooccurrence dependency of power flow calculations due to the possible attacks. There are no earlier statistical assumptions in the architecture proposed, as they are "model-free." It is also "cost-accommodating" since it does not alter the current FDD framework and it is only several microseconds on a household computer during the identification procedure. We have shown that ANN-MLP, SVM-RBF, and CNN can conduct locational detection under different noise and attack circumstances through broad experience in IEEE 14, 30, 57, and 118 bus systems. Moreover, the multi-name classification method used successfully improves the precision of the present identification.

• 센서학회지
• /
• 제31권6호
• /
• pp.433-440
• /
• 2022

Recently, owing to global warming, average summer temperatures are increasing and the number of hot days is increasing is increasing, which leads to an increase in heat stroke. In particular, outdoor workers directly exposed to the heat are at higher risk of heat stroke; therefore, preventing heat-related illnesses and managing safety have become important. Although various wearable devices have been developed to prevent heat stroke for outdoor workers, applying various sensors to the safety helmets that workers must wear is an excellent alternative. In this study, we developed a smart helmet that measures various vital signs of the wearer such as body temperature, heart rate, and sweat rate; external environmental signals such as temperature and humidity; and movement signals of the wearer such as roll and pitch angles. The smart helmet can acquire the various data by connecting with a smartphone application. Environmental data can check the status of heat wave advisory, and the individual vital signs can monitor the health of workers. In addition, we developed an algorithm that classifies the risk of heat-related illness as normal and abnormal by inputting a set of vital signs of the wearer using a support vector machine technique, which is a machine learning technique that allows for rapid binary classification with high reliability. Furthermore, the classified results suggest that the safety manager can supervise the prevention of heat stroke by receiving feedback from the control system.

• Smart Structures and Systems
• /
• 제29권1호
• /
• pp.17-28
• /
• 2022

The cable component of cable-stayed bridges is gradually impacted by weather conditions, vehicle loads, and material corrosion. The stayed cable is a critical load-carrying part that closely affects the operational stability of a cable-stayed bridge. Damaged cables might lead to the bridge collapse due to their tension capacity reduction. Thus, it is necessary to develop structural health monitoring (SHM) techniques that accurately identify damaged cables. In this work, a combinational identification method of three efficient techniques, including statistical analysis, clustering, and neural network models, is proposed to detect the damaged cable in a cable-stayed bridge. The measured dataset from the bridge was initially preprocessed to remove the outlier channels. Then, the theory and application of each technique for damage detection were introduced. In general, the statistical approach extracts the parameters representing the damage within time series, and the clustering approach identifies the outliers from the data signals as damaged members, while the deep learning approach uses the nonlinear data dependencies in SHM for the training model. The performance of these approaches in classifying the damaged cable was assessed, and the combinational identification method was obtained using the voting ensemble. Finally, the combination method was compared with an existing outlier detection algorithm, support vector machines (SVM). The results demonstrate that the proposed method is robust and provides higher accuracy for the damaged cable detection in the cable-stayed bridge.

• Design & Manufacturing
• /
• 제16권4호
• /
• pp.1-6
• /
• 2022

Fine blanking is a press processing technology that can process most of the product thickness into a smooth surface with a single stroke. In this fine blanking process, shear is an essential step. The punches and dies used in the shear are subjected to impacts of tens to hundreds of gravitational accelerations, depending on the type and thickness of the material. Therefore, among the components of the fine blanking mold (dies), punches and dies are the parts with the shortest lifespan. In the actual production site, various types of tool damage occur such as wear of the tool as well as sudden punch breakage. In this study, machine learning algorithms were used to predict these problems in advance. The dataset used in this paper consisted of the signal of the vibration sensor installed in the tool and the measured burr size (tool wear). Various features were extracted so that artificial intelligence can learn effectively from signals. It was trained with 5 features with excellent distinguishing performance, and the SVM algorithm performance was the best among 33 learning models. As a result of the research, the vibration signal at the time of imminent tool replacement was matched with an accuracy of more than 85%. It is expected that the results of this research will solve problems such as tool damage due to accidental punch breakage at the production site, and increase in maintenance costs due to prediction errors in punch exchange cycles due to wear.