• The Journal of the Korea institute of electronic communication sciences
• /
• v.15 no.1
• /
• pp.85-92
• /
• 2020

Deep learning shows outstanding performance in image and video analysis, such as object classification, object detection and semantic segmentation. In this paper, it is analyzed that the performances of deep learning models can be affected by characteristics of train dataset. It is proposed as a method for selecting activation function and optimization algorithm of deep learning to classify facial expression. Classification performances are compared and analyzed by applying various algorithms of each component of deep learning model for CK+, MMI, and KDEF datasets. As results of simulation, it is shown that genetic algorithm can be an effective solution for optimizing components of deep learning model.

• Journal of Electrical Engineering and Technology
• /
• v.12 no.2
• /
• pp.718-725
• /
• 2017

The method of stator segmentation is generally taken to enhance the electromagnetic performance of surface-mounted permanent magnet (SPM) machine and reduce its production cost. Based on the model with single slot, the expressions of cogging torque in machine with uniform or non-uniform segmentations are deduced and the optimal combination is given. Moreover, this paper discusses a structured skewing method and put forward a novel stator structure model to reduce the cogging torque in segmented permanent magnet machine. The model can reduce the cogging torque amplitude by shifting a proper angle of slot-opening. The shifting angle formula for analysis can also be suitable for other permanent machine with segmented stator. Finally the results of finite element simulation are given to prove that the method is effective and feasible.

• Journal of the Korean Society of Manufacturing Technology Engineers
• /
• v.22 no.3
• /
• pp.452-458
• /
• 2013

Over the past several years, many studies have been carried out in the field of 3D data inspection systems. Several attempts have been made to improve the quality of manufactured parts. The introduction of laser sensors for inspection has made it possible to acquire data at a remarkably high speed. In this paper, a robust inspection technique for detecting defects in 3D pressed parts using laser-scanned data is proposed. Point cloud data are segmented for the extraction of features. These segmented features are used for shape matching during the localization process. An iterative closest point (ICP) algorithm is used for the localization of the scanned model and CAD model. To achieve a higher accuracy rate, the ICP algorithm is modified and then used for matching. To enhance the speed of the matching process, aKd-tree algorithm is used. Then, the deviation of the scanned points from the CAD model is computed.

• Journal of the Korea Computer Graphics Society
• /
• v.4 no.1
• /
• pp.31-40
• /
• 1998

Active Contour Model(ACM) is an efficient method for segmenting an object. The main shortcoming of ACM is that its result is very dependent on the shape and location of an initial contour. To overcome this shortcoming, a new segmentation algorithm is proposed in this paper. The proposed algorithm uses B-splines to describe the active contour and applies Simulated Annealing (SA) and Genetic Algorithm(GA) as energy minimization techniques. We tried to overcome the initialization problem of traditional ACM and compared the result of ACM using GA and that using SA with 2D synthetic binary images. CT and MR images.

• Journal of Korean Society for Geospatial Information Science
• /
• v.7 no.1 s.13
• /
• pp.75-86
• /
• 1999

Three dimensional city model is composed of man-made and natural features, among these, most of man-made features are buildings. Therefore, it is very important to extract the building informations accurately and promptly to update the existing database. To achieve this, DTM can be reconstructed using building Information which is extracted from DTM, then this can be used as three dimensional city model. Thus, this paper aims to extract building boundaries and heights from high resolution DTM and edge informations of aerial photograph using mathematical morphology and image segmentation. We found that it is possible to extract buildings using opening operation in mathematical morphology and to improve the accuracy of building extraction using edge informations from aerial photograph.

• Journal of the Korean Institute of Telematics and Electronics S
• /
• v.36S no.11
• /
• pp.95-101
• /
• 1999

In this paper, we propose an adaptive method of tracking multiple moving objects using contour and features in surrounding conditions. We use an adaptive background model for robust processing in surrounding conditions. Object segmentation model detects pixels thresholded from local difference image between background and current image and extracts connected regions. Data association problem is solved by using feature extraction and object recognition model in searching window. We use Kalman filters for real-time tracking. The results of simulation show that the proposed method is good for tracking multiple moving objects in highway image sequences.

• Korean Journal of Remote Sensing
• /
• v.36 no.5_2
• /
• pp.929-937
• /
• 2020

In this manuscript, the UNet++ model, which is one of the representative deep learning techniques for semantic segmentation, was used to detect changes in temporal satellite images. To analyze the learning results according to various loss functions, we evaluated the change detection results using trained UNet++ models by binary cross entropy and the Jaccard coefficient. In addition, the learning results of the deep learning model were analyzed compared to existing pixel-based change detection algorithms by using WorldView-3 images. In the experiment, it was confirmed that the performance of the deep learning model could be determined depending on the characteristics of the loss function, but it showed better results compared to the existing techniques.

• Journal of Korea Multimedia Society
• /
• v.9 no.9
• /
• pp.1150-1159
• /
• 2006

This paper proposed a curve progress control function of the area base instead of the existing edge indication function, in order to detect the brain ventricle area by utilizing a geodesic active contour model. The proposed curve progress control function is very effective in detecting the brain ventricle area and this function is based on the average brightness of the brain ventricle area which appears brighter in MRI images. Compared numerically by using various measures, the proposed method in this paper can detect brain ventricle areas better than the existing method. By examining images of normal and diseased brain's images by brain tumor, we compared the several brain ventricle detection algorithms with proposed method visually and verified the effectiveness of the proposed method.

• The Journal of Asian Finance, Economics and Business
• /
• v.7 no.7
• /
• pp.279-291
• /
• 2020

This study aims to develop the concept of innovation models with the marketing channel construct approach, marketing innovation, product segmentation, and customer insight; as well as improvements to the theory of resource-based combined with the method of service-dominant logic. This study approach is based on quantitative descriptive conducted with three stages of testing scenarios. The first test is the mapping of the innovation model construct through testing the validity and reliability with the moderation of customer orientation variables. The second scenario examines the relationship of influence between the independent variables on the dependent variable of 29 hypothetical analysis equation modeling. The unit of analysis was conducted on 497 SMEs involved in the food and beverage sectors, with the criteria being SMEs must have a rating of 4-5 points on the Go-Food applications software. The results shown that: 1) the construct used to develop an innovative model both directly and via moderation is positive and significant; 2) Through a complicated relationship that involves all components of the variable, it outlines a positive and significant effect except for the path of analysis (μ5). The theoretical and managerial implications state that the service-dominant logic approach and resource-based view theory have extreme reliability and interrelations.

• Electronics and Telecommunications Trends
• /
• v.33 no.5
• /
• pp.103-110
• /
• 2018

Studies on artificial intelligence have been developed for the past couple of decades. After a few periods of prosperity and recession, a new machine learning method, so-called Deep Learning, has been introduced. This is the result of high-quality big- data, an increase in computing power, and the development of new algorithms. The main targets for deep learning are 1D audio and 2D images. The application domain is being extended from a discriminative model, such as classification/segmentation, to a generative model. Currently, deep learning is used for processing 3D data. However, unlike 2D, it is not easy to acquire 3D learning data. Although low-cost 3D data acquisition sensors have become more popular owing to advances in 3D vision technology, the generation/acquisition of 3D data remains a very difficult problem. Moreover, it is not easy to directly apply an existing network model, such as a convolution network, owing to the variety of 3D data representations. In this paper, we summarize the 3D deep learning technology that have started to be developed within the last 2 years.