• Journal of the Korea Convergence Society
• /
• v.9 no.4
• /
• pp.47-55
• /
• 2018

In this paper, we propose a novel face recognition(FR) method that takes advantage of combining weighted deep local features extracted from multiple Deep Convolutional Neural Networks(DCNNs) learned with a set of facial local regions. In the proposed method, the so-called weighed deep local features are generated from multiple DCNNs each trained with a particular face local region and the corresponding weight represents the importance of local region in terms of improving FR performance. Our weighted deep local features are applied to Joint Bayesian metric learning in conjunction with Nearest Neighbor(NN) Classifier for the purpose of FR. Systematic and comparative experiments show that our proposed method is robust to variations in pose, illumination, and expression. Also, experimental results demonstrate that our method is feasible for improving face recognition performance.

• Journal of Korea Multimedia Society
• /
• v.6 no.1
• /
• pp.48-57
• /
• 2003

This paper describes an approach for classifying myoelectric patterns using a multilayer perceptrons (MLP's) with genetic algorithm and hidden Markov models (HMM's) hybrid classifier. Genetic Algorithms play a role of selecting Multilayer Perceptron's optimized initial connection weights by its typical global search. The dynamic aspects of EMG are important for tasks such as continuous prosthetic control or various time length EMG signal recognition, which have not been successfully mastered by the most neural approaches. It is known that the hidden Markov model (HMM) is suitable for modeling temporal patterns. In contrast, the multilayer feedforward networks are suitable for static patterns. And, a lot of investigators have shown that the HMM's to be an excellent tool for handling the dynamical problems. Considering these facts, we suggest the combination of ANN and HMM algorithms that might lead to further improved EMG recognition systems.

• Journal of the Korea Convergence Society
• /
• v.9 no.6
• /
• pp.33-38
• /
• 2018

In this study, we try to detect new attacks for vehicle by learning only one class. We use Car-Hacking dataset, an intrusion detection dataset, which is used to evaluate classification performance. The dataset are created by logging CAN (Controller Area Network) traffic through OBD-II port from a real vehicle. The dataset have four attack types. One class classification is one of unsupervised learning methods that classifies attack class by learning only normal class. When using unsupervised learning, it difficult to achieve high efficiency because it does not use negative instances for learning. However, unsupervised learning has the advantage for classifying unlabeled data, which are new attacks. In this study, we use one class classifier to detect new attacks that are difficult to detect using signature-based rules on network intrusion detection system. The proposed method suggests a combination of parameters that detect all new attacks and show efficient classification performance for normal dataset.

• The Journal of Korea Institute of Information, Electronics, and Communication Technology
• /
• v.8 no.5
• /
• pp.373-383
• /
• 2015

This paper proposes a decision support system to recognizing retinal diseases. This paper uses a smartphone platform and cloud computing as the base of the system. A microscopic lens is attached int' the smartphone camera to capture the user retinal image for recognizing the user's retinal condition. An application is assembled in computer and then installed in to the smartphone. The application role is to connect between the system in smartphone and system in cloud, the application will send the retinal image to the cloud system to be classified. The paper uses OCFE (optimized classifier based on feature elimination) algorithm as the classifier. The retinal image is trained using combination of two ophthalmology databases DIARETDB1 v2.1 and STARE. Therefore, this system average accuracy is 88%, while the average error rate is 12%.

• The Journal of The Korea Institute of Intelligent Transport Systems
• /
• v.16 no.4
• /
• pp.1-12
• /
• 2017

Traffic accidents are caused by a combination of human factors, vehicle factors, and environmental factors. In the case of traffic accidents where rental cars are involved, the possibility and the severity of traffic accidents are expected to be different from those of other traffic accidents due to the unfamiliar environment of the driver. In this study, we developed a model to forecast the severity of rental car accidents by using Naive Bayes classifier for Busan, Gangneung, and Jeju city. In addition, we compared the prediction accuracy performance of two models where one model uses the variables of which statistical significance were verified in a prior study and another model uses the entire available variables. As a result of the comparison, it is shown that the prediction accuracy is higher when using the variables with statistical significance.

• Journal of Intelligence and Information Systems
• /
• v.28 no.3
• /
• pp.119-137
• /
• 2022

Recent recommendation system studies apply various deep learning models to represent user and item interactions better. One of the noteworthy studies is ONCF(Outer product-based Neural Collaborative Filtering) which builds a two-dimensional interaction map via outer product and employs CNN (Convolutional Neural Networks) to learn high-order correlations from the map. However, ONCF has limitations in recommendation performance due to the problems with CNN and the absence of side information. ONCF using CNN has an inductive bias problem that causes poor performances for data with a distribution that does not appear in the training data. This paper proposes to employ a Vision Transformer (ViT) instead of the vanilla CNN used in ONCF. The reason is that ViT showed better results than state-of-the-art CNN in many image classification cases. In addition, we propose a new architecture to reflect side information that ONCF did not consider. Unlike previous studies that reflect side information in a neural network using simple input combination methods, this study uses an independent auxiliary classifier to reflect side information more effectively in the recommender system. ONCF used a single latent vector for user and item, but in this study, a channel is constructed using multiple vectors to enable the model to learn more diverse expressions and to obtain an ensemble effect. The experiments showed our deep learning model improved performance in recommendation compared to ONCF.

• Korean Journal of Radiology
• /
• v.25 no.1
• /
• pp.43-54
• /
• 2024

Objective: To evaluate the added value of diffusion-weighted imaging (DWI)-based quantitative parameters to distinguish uterine sarcomas from atypical leiomyomas on preoperative magnetic resonance imaging (MRI). Materials and Methods: A total of 138 patients (age, 43.7 ± 10.3 years) with uterine sarcoma (n = 44) and atypical leiomyoma (n = 94) were retrospectively collected from four institutions. The cohort was randomly divided into training (84/138, 60.0%) and validation (54/138, 40.0%) sets. Two independent readers evaluated six qualitative MRI features and two DWI-based quantitative parameters for each index tumor. Multivariable logistic regression was used to identify the relevant qualitative MRI features. Diagnostic classifiers based on qualitative MRI features alone and in combination with DWI-based quantitative parameters were developed using a logistic regression algorithm. The diagnostic performance of the classifiers was evaluated using a cross-table analysis and calculation of the area under the receiver operating characteristic curve (AUC). Results: Mean apparent diffusion coefficient value of uterine sarcoma was lower than that of atypical leiomyoma (mean ± standard deviation, 0.94 ± 0.30 10^-3 mm²/s vs. 1.23 ± 0.25 10^-3 mm²/s; P < 0.001), and the relative contrast ratio was higher in the uterine sarcoma (8.16 ± 2.94 vs. 4.19 ± 2.66; P < 0.001). Selected qualitative MRI features included ill-defined margin (adjusted odds ratio [aOR], 17.9; 95% confidence interval [CI], 1.41-503, P = 0.040), intratumoral hemorrhage (aOR, 27.3; 95% CI, 3.74-596, P = 0.006), and absence of T2 dark area (aOR, 83.5; 95% CI, 12.4-1916, P < 0.001). The classifier that combined qualitative MRI features and DWI-based quantitative parameters showed significantly better performance than without DWI-based parameters in the validation set (AUC, 0.92 vs. 0.78; P < 0.001). Conclusion: The addition of DWI-based quantitative parameters to qualitative MRI features improved the diagnostic performance of the logistic regression classifier in differentiating uterine sarcomas from atypical leiomyomas on preoperative MRI.

• Journal of Korea Society of Industrial Information Systems
• /
• v.12 no.4
• /
• pp.15-24
• /
• 2007

The recognition of postal address images is indispensable for the automatic sorting of postal envelopes. The process of the address image recognition is composed of three steps-address image preprocessing, character recognition, address interpretation. The extracted character images from the preprocessing step are forwarded to the character recognition step, in which multiple candidate characters with reliability scores are obtained for each character image extracted. aracters with reliability scores are obtained for each character image extracted. Utilizing those character candidates with scores, we obtain the final valid address for the input envelope image through the address interpretation step. The envelope sorting rate depends on the performance of all three steps, among which character recognition step could be said to be very important. The good character recognizer would be the one which could produce valid candidates with very reliable scores to help the address interpretation step go easy. In this paper, we propose the method of generating character candidates with reliable recognition scores. We utilize the existing MLP(multilayered perceptrons) neural network of the address recognition system in the current automatic postal envelope sorters, as the classifier for the each image from the preprocessing step. The MLP is well known to be one of the best classifiers in terms of processing speed and recognition rate. The false alarm problem, however, might be occurred in recognition results, which made the address interpretation hard. To make address interpretation easy and improve the envelope sorting rate, we propose promising methods to reestimate the recognition score (confidence) of the existing MLP classifier: the generation method of the statistical recognition properties of the classifier and the method of the combination of the MLP and the subspace classifier which roles as a reestimator of the confidence. To confirm the superiority of the proposed method, we have used the character images of the real postal envelopes from the sorters in the post office. The experimental results show that the proposed method produces high reliability in terms of error and rejection for individual characters and non-characters.

• Journal of Microbiology and Biotechnology
• /
• v.8 no.5
• /
• pp.523-530
• /
• 1998

Rapid and quantitative analysis of physiological change and clavulanic acid production was studied by the combination of pyrolysis mass spectrometry (PyMS) and artificial neural network (ANN) in Streptomyces clavuligerus. Firstly, the continuous culture studies were carried out to get the physiological background and PyMS samples. Clavulanic acid production was inversely related to growth rate: Mycelium growth and $q_{cal}$ were optimum at 0.1 $h^{-1}\; and \;0.025 h^{-1}$ respectively. Changes in specific nutrient uptake rates ($q_{gly}$ and $q_{amn}$) also affected clavulanic acid production since clavulanic acid production appeared to be stimulated by the limitation of carbon and nitrogen. Fermentation broth containing mycelium taken from continuous cultures was analyzed by PyMS, and the PyMS spectra were analyzed with multivariate statistics. PCCV plots revealed that samples harvested under the same culture condition were clustered together but samples from different culture conditions formed separate clusters. To deconvolute the pyrolysis mass spectra so as to obtain quantitative information on the concentration of clavulanic acid, ANN was trained on Py MS data using a radial basis function classifier. The results showed that the physiological stages with different growth rate were successfully differentiated and it was possible to monitor the clavulanic acid production precisely and rapidly.

• Imaging Science in Dentistry
• /
• v.43 no.3
• /
• pp.153-161
• /
• 2013

Purpose: To prevent low bone mineral density (BMD), that is, osteoporosis, in postmenopausal women, it is essential to diagnose osteoporosis more precisely. This study presented an automatic approach utilizing a histogram-based automatic clustering (HAC) algorithm with a support vector machine (SVM) to analyse dental panoramic radiographs (DPRs) and thus improve diagnostic accuracy by identifying postmenopausal women with low BMD or osteoporosis. Materials and Methods: We integrated our newly-proposed histogram-based automatic clustering (HAC) algorithm with our previously-designed computer-aided diagnosis system. The extracted moment-based features (mean, variance, skewness, and kurtosis) of the mandibular cortical width for the radial basis function (RBF) SVM classifier were employed. We also compared the diagnostic efficacy of the SVM model with the back propagation (BP) neural network model. In this study, DPRs and BMD measurements of 100 postmenopausal women patients (aged >50 years), with no previous record of osteoporosis, were randomly selected for inclusion. Results: The accuracy, sensitivity, and specificity of the BMD measurements using our HAC-SVM model to identify women with low BMD were 93.0% (88.0%-98.0%), 95.8% (91.9%-99.7%) and 86.6% (79.9%-93.3%), respectively, at the lumbar spine; and 89.0% (82.9%-95.1%), 96.0% (92.2%-99.8%) and 84.0% (76.8%-91.2%), respectively, at the femoral neck. Conclusion: Our experimental results predict that the proposed HAC-SVM model combination applied on DPRs could be useful to assist dentists in early diagnosis and help to reduce the morbidity and mortality associated with low BMD and osteoporosis.