• Journal of Korea Multimedia Society
• /
• v.25 no.1
• /
• pp.39-51
• /
• 2022

Image-based gender classification and age estimation of human are classic problems in computer vision. Most of researches in this field focus just only one task of either gender classification or age estimation and most of the reported methods for each task focus on accuracy performance and are not computationally light. Thus, running both tasks together simultaneously on low cost mobile or embedded systems with limited cpu processing speed and memory capacity are practically prohibited. In this paper, we propose a novel light-weight gender classification and age estimation method based on ensemble multitasking deep learning with light-weight processing neural network architecture, which processes both gender classification and age estimation simultaneously and in real-time even for embedded systems. Through experiments over various well-known datasets, it is shown that the proposed method performs comparably to the state-of-the-art gender classification and/or age estimation methods with respect to accuracy and runs fast enough (average 14fps) on a Jestson Nano embedded board.

• Health Policy and Management
• /
• v.23 no.1
• /
• pp.35-43
• /
• 2013

Background: Diagnosis procedure combination (DPC) has recently been introduced in Korea as a demonstration project and it has aimed the improvement of accuracy in bundled payment instead of Diagnosis related group (DRG). The purpose of this study is to investigate that the model of end-stage liver disease (MELD) score as the severity classification of liver diseases is adequate for improving reimbursement of DPC. Methods: The subjects of this study were 329 patients of liver disease (Korean DRG ver. 3.2 H603) who had discharged from National Health Insurance Corporation Ilsan Hospital which is target hospital of DPC demonstration project, between January 1, 2007 and July 31, 2010. We tested the cost differences by severity classifications which were DRG severity classification and clinical severity classification-MELD score. We used a multiple regression model to find the impacts of severity on total medical cost controlling for demographic factor and characteristics of medical services. The within group homogeneity of cost were measured by calculating the coefficient of variation and extremal quotient. Results: This study investigates the relationship between medical costs and other variables especially severity classifications of liver disease. Length of stay has strong effect on medical costs and other characteristics of patients or episode also effect on medical costs. MELD score for severity classification explained the variation of costs more than DRG severity classification. Conclusion: The accuracy of DRG based payment might be improved by using various clinical data collected by clinical situations but it should have objectivity with considering availability. Adequate compensation for severity should be considered mainly in DRG based payment. Disease specific severity classification would be an alternative like MELD score for liver diseases.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.12 no.2
• /
• pp.103-108
• /
• 2002

Manufacturing the complex surface plates in Stern and Stem is major factor in computing the processing cost of a ship. If these parts are effectively classified, it helps to compute the processing cost and find the way of cut-down on the processing costs. This study is intended to effectively classify surface plates. To solve this problem, we apply Pattern Classification of Neural-Networks.

• Korean Journal of Computational Design and Engineering
• /
• v.17 no.4
• /
• pp.225-233
• /
• 2012

A comprehensive research effort in order to develop and disseminate modernized Korean housing (Hanok) has recently been initiated by Korean government. This large scale research project encompasses a wide spectrum of housing development including public policy, architectural plans, modules, construction materials and methods, prefabricated assemblies, automated production, construction management, and advanced information systems. For the purpose of integrating and automating the whole processes from an industry perspective, it is of great importance to develop a standard classification system and project numbering system (PNS) for the modernized Korean housing. This paper focuses on the standard classification systems and PNS for cost and schedule control. The distinct characteristics and managerial requirements were explored and embedded into the proposed classifications for modernized Hanok.

• The Journal of Information Systems
• /
• v.18 no.3
• /
• pp.375-399
• /
• 2009

In this study, we propose a integrated model of logistic regression, artificial neural networks, support vector machines(SVM), with case-based reasoning(CBR). To predict respondents in the direct marketing is the binary classification problem as like bankruptcy prediction, IDS, churn management and so on. To solve the binary problems, we employed logistic regression, artificial neural networks, SVM. and CBR. CBR is a problem-solving technique and shows significant promise for improving the effectiveness of complex and unstructured decision making, and we can obtain excellent results through CBR in this study. Experimental results show that the classification accuracy of integration model using CBR is superior to logistic regression, artificial neural networks and SVM. When we apply the customer response model to predict respondents in the direct marketing, we have to consider from the view point of profit/cost about the misclassification.

• Proceedings of the Korean Society for Noise and Vibration Engineering Conference
• /
• 2008.04a
• /
• pp.809-816
• /
• 2008

With increasing demands for reducing cost of maintenance which can detect machine fault automatically; low cost and intelligent functionality sensors are required. Rapid developments, in semiconductor, computing, and communication have led to a new generation of sensor called "smart" sensors with functionality and intelligence. The purpose of this research is comparison of machine fault classification between general analyzer signals and smart sensor signals. Three types of sensors are used in induction motors faults diagnosis, which are vibration, current and flux. Classification results are satisfied.

• Journal of Korean Society of Water and Wastewater
• /
• v.29 no.6
• /
• pp.651-657
• /
• 2015

Recently, the need for asset management(AM) plan introduction to reduce increasing O&M cost with aging water facilities is on the rise. Therefore, asset inventory classification is necessary as the first step for AM plan construction. In this study, all assets of YW water treatment plant(WTP) were classified as 5 steps. In addition, specific code name was given to each asset which can increase compatibility in constructing the AM programs among WTPs. In the future, codes for attribute and status of asset will be allocated, which can facilitate proper AM operation.

• International Journal of Fuzzy Logic and Intelligent Systems
• /
• v.9 no.2
• /
• pp.141-146
• /
• 2009

Support vector machine is a classification algorithm based on statistical learning theory. It has shown many results with good performances in the data mining fields. But there are some problems in the algorithm. One of the problems is its heavy computing cost. So we have been difficult to use the support vector machine in the dynamic and online systems. To overcome this problem we propose to use stratified sampling of statistical sampling theory. The usage of stratified sampling supports to reduce the size of training data. In our paper, though the size of data is small, the performance accuracy is maintained. We verify our improved performance by experimental results using data sets from UCI machine learning repository.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.7
• /
• pp.2154-2168
• /
• 2022

Our work addresses the issues associated with usage of the semantic features by Bag of Words model, which requires construction of the dictionary. Extracting the relevant features and clustering them into code book or dictionary is computationally intensive and requires large storage area. Hence we propose to use a simple distribution of multiple shape based features, which is a mixture of gradients, radius and slope angles requiring very less computational cost and storage requirements but can serve as an equivalent image representative. The experimental work conducted on PASCAL VOC 2007 dataset exhibits marginally closer performance in terms of accuracy with the Bag of Word model using Self Organizing Map for clustering and very significant computational gain.

• International journal of advanced smart convergence
• /
• v.8 no.4
• /
• pp.75-81
• /
• 2019

Though recent advancement of deep learning methods have provided satisfactory results from large data domain, somehow yield poor performance on few-shot classification tasks. In order to train a model with strong performance, i.e. deep convolutional neural network, it depends heavily on huge dataset and the labeled classes of the dataset can be extremely humongous. The cost of human annotation and scarcity of the data among the classes have drastically limited the capability of current image classification model. On the contrary, humans are excellent in terms of learning or recognizing new unseen classes with merely small set of labeled examples. Few-shot learning aims to train a classification model with limited labeled samples to recognize new classes that have neverseen during training process. In this paper, we increase the backbone depth of the embedding network in orderto learn the variation between the intra-class. By increasing the network depth of the embedding module, we are able to achieve competitive performance due to the minimized intra-class variation.