• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.3
• /
• pp.1121-1141
• /
• 2020

We propose a robust visual object tracking algorithm fusing a convolutional neural network tracker trained offline from a large number of video repositories and a color histogram based tracker to track objects for mixing immersive audio. Our algorithm addresses the problem of occlusion and large movements of the CNN based GOTURN generic object tracker. The key idea is the offline training of a binary classifier with the color histogram similarity values estimated via both trackers used in this method to opt appropriate tracker for target tracking and update both trackers with the predicted bounding box position of the target to continue tracking. Furthermore, a histogram similarity constraint is applied before updating the trackers to maximize the tracking accuracy. Finally, we compute the depth(z) of the target object by one of the prominent unsupervised monocular depth estimation algorithms to ensure the necessary 3D position of the tracked object to mix the immersive audio into that object. Our proposed algorithm demonstrates about 2% improved accuracy over the outperforming GOTURN algorithm in the existing VOT2014 tracking benchmark. Additionally, our tracker also works well to track multiple objects utilizing the concept of single object tracker but no demonstrations on any MOT benchmark.

• Korean Journal of Environmental Biology
• /
• v.26 no.3
• /
• pp.203-213
• /
• 2008

Benthic macro invertebrate communities were collected at six different sampling sites in the Musucheon stream in Korea from July 2006 to July 2007, and ecological exergy values were calculated based on five different functional feeding groups (collector-gatherer, collector-filterer, predator, scrapper, and shredder) of benthic macro invertebrates. Each sampling site was categorized to three stream types (perennial, intermittent and drought) based on the water flow condition. Exergy values were low at all study sites right after a heavy rain and relatively higher in the perennial stream type than in the intermittent or the drought stream type. Self-Organizing Map (SOM), unsupervised artificial neural network, was implemented to pattern spatial and temporal dynamics of ecological exergy of the study sites. SOM classified samples into four clusters. The classification reflected the effects of floods and droughts on benthic macroinvertebrate communities, and was mainly related with the stream types of the sampling sites. Exergy values of each functional feeding group also responded differently according to the different stream types. Finally, the results showed that exergy is an effective ecological indicator, and patterning changes of exergy using SOM is an effective way to evaluate target ecosystems.

• Journal of Sensor Science and Technology
• /
• v.14 no.6
• /
• pp.409-414
• /
• 2005

For an intelligent vehicle driving which uses magnetic markers and magnetic sensors, we can get every kind of road information while moving the vehicle if we use the code that is encoded with N, S pole direction of markers. If we make it an only aim to move the vehicle, it becomes easy to control the vehicle the more we put markers close. By the way, to recognize the direction of a marker pole it is much better that the markers have no effect each other. To get road informations and move the vehicle autonomously we propose the methods of arranging magnetic sensors and algorithm of recognizing the position of the vehicle with those sensors. We verified the effectiveness of the methods with computer simulation.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.15 no.7 s.100
• /
• pp.821-828
• /
• 2005

In our life, we have used many digital appliances. They help us to improve the quality of life but sometimes give us unsatisfactory result. Because they produce specific noise. Especially vacuum cleaner produce much noise that is very annoying. So we need to study what sound metrics affect human sensibility. In this paper, we develop sound quality index for vacuum cleaner using the sound quality metrics defined in psychoacoustics. First, we carry out the subjective evaluation of vacuum cleaner sound to verify what vacuum sound feels good to human. And then artificial neural network estimated the complexity and the nonlinear characteristics of the relations between subjective evaluation and sound metrics. Finally the ANN is trained repeatedly to have a good performance for sound qualify index of the vacuum cleaner. As a result, the sound quality index of vacuum cleaner has a correlation of $93.5\%$ between the subjective evaluation and ANN. So, there exist three factors that Is loudness, sharpness, roughness which affect the sound quality of vacuum cleaner.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.14 no.2
• /
• pp.119-127
• /
• 2004

Butterfly valves are popularly used in service in the industrial and water works pipeline systems with large diameter because of its lightweight, simple structure and the rapidity of its manipulation. Sometimes cavitation can occur. resulting in noise, vibration and rapid deterioration of the valve trim, and do not allow further operation. Thus, the monitoring of cavitation is of economic interest and is very importance in industry. This paper proposes a condition monitoring scheme using statistical feature evaluation and support vector machine (SVM) to detect the cavitation conditions of butterfly valve which used as a flow control valve at the pumping stations. The stationary features of vibration signals are extracted from statistical moments. The SVMs are trained, and then classify normal and cavitation conditions of control valves. The SVMs with the reorganized feature vectors can distinguish the class of the untrained and untested data. The classification validity of this method is examined by various signals that are acquired from butterfly valves in the pumping stations and compared the classification success rate with those of self-organizing feature map neural network.

• Transactions of Materials Processing
• /
• v.27 no.3
• /
• pp.154-159
• /
• 2018

In the rectangular bar multi-stage drawing process, the cross-section dimensional accuracy of the rectangular bar varies depending on the aspect ratio and process conditions. It is very important to predict the dimensional error of the cross-section occurring in the multi-stage drawing process according to the aspect ratio of the rectangular bar and the half die angle of each pass. In this study, a process map for improving the dimensional accuracy according to the aspect ratio was derived in the drawing process of a rectangular bar. FE-simulation of the multi-stage shape drawing process was carried out with four types of rectangular bar. The results of the FE-simulation were trained to the nonlinear relationship between the shape parameters using an Artificial Neural Network (ANN), and the process maps were derived from them. The optimum half die angles were determined from the process maps on the dimensional accuracy. The validity of the suggested process map for aspect ratios 1.25~2:1 were verified through FE-simulation and experimentation.

• Journal of Soil and Groundwater Environment
• /
• v.21 no.3
• /
• pp.21-34
• /
• 2016

In the present study, we applied various machine learning techniques comparatively for prediction of subsurface structures based on multiple secondary information (i.e., well-logging data). The machine learning techniques employed in this study are Naive Bayes classification (NB), artificial neural network (ANN), support vector machine (SVM) and logistic regression classification (LR). As an alternative model, conventional hidden Markov model (HMM) and modified hidden Markov model (mHMM) are used where additional information of transition probability between primary properties is incorporated in the predictions. In the comparisons, 16 boreholes consisted with four different materials are synthesized, which show directional non-stationarity in upward and downward directions. Futhermore, two types of the secondary information that is statistically related to each material are generated. From the comparative analysis with various case studies, the accuracies of the techniques become degenerated with inclusion of additive errors and small amount of the training data. For HMM predictions, the conventional HMM shows the similar accuracies with the models that does not relies on transition probability. However, the mHMM consistently shows the highest prediction accuracy among the test cases, which can be attributed to the consideration of geological nature in the training of the model.

• Proceedings of the Korea Committee for Ocean Resources and Engineering Conference
• /
• 2006.11a
• /
• pp.298-302
• /
• 2006

As the age of information technology is coming, companies stress the need of knowledge management. Companies construct ERP system including knowledge management. But, it is not easy to formalize knowledge in organization. we focused on data mining system by using genetic programming. But, we don't have enough data to perform the learning process of genetic programming. We have to reduce input parameter(s) or increase number of learning or training data. In order to do this, the enhanced data mining system by using GP combined with SOM(Self organizing map) is adopted in this paper. We can reduce the number of learning data by adopting SOM.

• Journal of the Korean Data and Information Science Society
• /
• v.28 no.6
• /
• pp.1217-1227
• /
• 2017

Deep learning has been studied explosively and has achieved excellent performance in areas like image and speech recognition, the application areas in which computations have been challenges with ordinary machine learning techniques. The theoretical study of deep learning has also been researched toward improving the performance. In this paper, we try to find a key of the success of the deep learning in rich and efficient expressiveness of the deep learning function, and analyze the theoretical studies related to it.

• Journal of Institute of Control, Robotics and Systems
• /
• v.4 no.4
• /
• pp.487-493
• /
• 1998

Camera calibration is the process of determining the coordinate relationship between a camera image and its real world space. Accurate calibration of a camera is necessary for the applications that involve quantitative measurement of camera images. However, if the camera plane is parallel or near parallel to the calibration board on which 2 dimensional objects are defined(this is called "ill-conditioned"), existing solution procedures are not well applied. In this paper, we propose a neural network-based approach to camera calibration for 2D images formed by a mono-camera or a pair of cameras. Multi-layer perceptrons are developed to transform the coordinates of each image point to the world coordinates. The validity of the approach is tested with data points which cover the whole 2D space concerned. Experimental results for both mono-camera and stereo-camera cases indicate that the proposed approach is comparable to Tsai's method［8］. Especially for the stereo camera case, the approach works better than the Tsai's method as the angle between the camera optical axis and the Z-axis increases. Therefore, we believe the approach could be an alternative solution procedure for the ill -conditioned camera calibration.libration.