• International journal of advanced smart convergence
• /
• v.12 no.4
• /
• pp.134-141
• /
• 2023

We present a novel method aimed at refining ground truth data through regularization and modification, particularly applicable when working with the original ground truth set. Enhancing the performance of deep neural networks is achieved by applying regularization techniques to the existing ground truth data. In many machine learning tasks requiring pixel-level segmentation sets, accurately delineating objects is vital. However, it proves challenging for thin and elongated objects such as blood vessels in X-ray coronary angiography, often resulting in inconsistent generation of ground truth data. This method involves an analysis of the quality of training set pairs - comprising images and ground truth data - to automatically regulate and modify the boundaries of ground truth segmentation. Employing the active contour model and a recursive ground truth generation approach results in stable and precisely defined boundary contours. Following the regularization and adjustment of the ground truth set, there is a substantial improvement in the performance of deep neural networks.

• International journal of advanced smart convergence
• /
• v.13 no.2
• /
• pp.187-194
• /
• 2024

For precise cardiac diagnostics and treatment, we introduce a novel method for patient-specific mapping between myocardial and coronary anatomy, leveraging local variations in myocardial thickness. This complex system integrates and automates multiple sophisticated components, including left ventricle segmentation, myocardium segmentation, long-axis estimation, coronary artery tracking, and advanced geodesic Voronoi distance mapping. It meticulously accounts for variations in myocardial thickness and precisely delineates the boundaries between coronary territories according to the conventional 17-segment myocardial model. Each phase of the system provides a step-by-step approach to automate coronary artery mapping onto the myocardium. This innovative method promises to transform cardiac imaging by offering highly precise, automated, and patient-specific analyses, potentially enhancing the accuracy of diagnoses and the effectiveness of therapeutic interventions for various cardiac conditions.

• Electronics and Telecommunications Trends
• /
• v.38 no.6
• /
• pp.119-127
• /
• 2023

Market-oriented networks support various tasks in the market domain. We analyze trends in structural changes of such networks to adapt to preferences in general market movements. This analysis is different from conventional ones that focus on specific technologies. Instead, we focus on the paradigm shift of network technology from connectivity functionalities to platforms supporting business domains for direct modeling. Moreover, we analyze current development efforts of technologies based on popular and realistic solutions such as FIWARE, 5GinFIRE, IBN, IDN, and HNSP. Remarkably, we detail HNSP as an open research and development platform to experiment with business models and enable co-building with developers. We observe a clear paradigm shift of communications technology from a closed to an open job-shop style.

• Journal of the Institute of Convergence Signal Processing
• /
• v.20 no.3
• /
• pp.159-164
• /
• 2019

Detecting wildfire using only color in image information is a very difficult issue. This paper proposes an algorithm to detect forest fire area by analyzing color and motion of the area in the video including forest fire. The proposed algorithm removes the background region using the Gaussian Mixture based background segmentation algorithm, which does not depend on the lighting conditions. In addition, the RGB channel is changed to an HSV channel to extract flame candidates based on color. The extracted flame candidates judge that it is not a flame if the area moves while labeling and tracking. If the flame candidate areas extracted in this way are in the same position for more than 2 minutes, it is regarded as flame. Experimental results using the implemented algorithm confirmed the validity.

• Journal of Nuclear Fuel Cycle and Waste Technology(JNFCWT)
• /
• v.20 no.3
• /
• pp.365-369
• /
• 2022

Operating and decommissioning nuclear power plants generates radioactive waste. This radioactive waste can be categorized into several different levels, for example, low, intermediate, and high, according to the regulations. Currently, low and intermediate-level waste are stored in conventional 200-liter drums to be disposed. However, in Korea, the disposal of intermediate-level radioactive waste is virtually impossible as there are no available facilities. Furthermore, large-sized intermediate-level radioactive waste, such as reactor internals from decommissioning, need to be segmented into smaller sizes so they can be adequately stored in the conventional drums. This segmentation process requires additional costs and also produces secondary waste. Therefore, this paper suggests repurposing the no-longer-used spent nuclear fuel casks. The casks are larger in size than the conventional drums, thus requiring less segmentation of waste. Furthermore, the safety requirements of the spent nuclear fuel casks are severer than those of the drums. Hence, repurposed spent nuclear fuel casks could better address potential risks such as dropping, submerging, or a fire. In addition, the spent nuclear fuel casks need to be disposed in compliance with the regulations for low level radioactive waste. This cost may be avoided by repurposing the casks.

• The KIPS Transactions:PartB
• /
• v.18B no.4
• /
• pp.213-220
• /
• 2011

This paper proposes an effective, four-stage smoke detection method based on video that provides emergency response in the event of unexpected hazards in early fire-alarming systems. In the first phase, an approximate median method is used to segment moving regions in the present frame of video. In the second phase, a color segmentation of smoke is performed to select candidate smoke regions from these moving regions. In the third phase, a feature extraction algorithm is used to extract five feature parameters of smoke by analyzing characteristics of the candidate smoke regions such as area randomness and motion of smoke. In the fourth phase, extracted five parameters of smoke are used as an input for a K-nearest neighbor (KNN) algorithm to identify whether the candidate smoke regions are smoke or non-smoke. Experimental results indicate that the proposed four-stage smoke detection method outperforms other algorithms in terms of smoke detection, providing a low false alarm rate and high reliability in open and large spaces.

• Korean Chemical Engineering Research
• /
• v.57 no.2
• /
• pp.274-282
• /
• 2019

In recent years, big data analysis has been expanded to include automatic control through reinforcement learning as well as prediction through modeling. Research on the utilization of image data is actively carried out in various industrial fields such as chemical, manufacturing, agriculture, and bio-industry. In this paper, we applied NASNet, which is an AutoML reinforced learning algorithm, to DeepU-Net neural network that modified U-Net to improve image semantic segmentation performance. We used BRATS2015 MRI data for performance verification. Simulation results show that DeepU-Net has more performance than the U-Net neural network. In order to improve the image segmentation performance, remove dropouts that are typically applied to neural networks, when the number of kernels and filters obtained through reinforcement learning in DeepU-Net was selected as a hyperparameter of neural network. The results show that the training accuracy is 0.5% and the verification accuracy is 0.3% better than DeepU-Net. The results of this study can be applied to various fields such as MRI brain imaging diagnosis, thermal imaging camera abnormality diagnosis, Nondestructive inspection diagnosis, chemical leakage monitoring, and monitoring forest fire through CCTV.

• The Journal of Korean Institute of Information Technology
• /
• v.17 no.6
• /
• pp.17-25
• /
• 2019

Accurate tracking of the sinkholes that are appearing frequently now is an important method of protecting human and property damage. Although many sinkhole detection systems have been proposed, it is still far from completely solved especially in-depth area. Furthermore, detection of sinkhole algorithms experienced the problem of unstable result that makes the system difficult to fire a warning in real-time. In this paper, we proposed a method of sinkhole tracking by deep learning and data association, that takes advantage of the recent development of CNN transfer learning. Our system consists of three main parts which are binary segmentation, sinkhole classification, and sinkhole tracking. The experiment results show that the sinkhole can be tracked in real-time on the dataset. These achievements have proven that the proposed system is able to apply to the practical application.

• Journal of Satellite, Information and Communications
• /
• v.8 no.4
• /
• pp.100-104
• /
• 2013

In this paper, we present an optimal threshold setting method for flame detection of infrared thermal image. Conventional infrared flame detection methods used fixed intensity threshold to segment candidate flame regions and further processing is performed to decide correct flame detection. So flame region segmentation step using the threshold is important processing for fire detection algorithm. The threshold should be change in input image depends on camera types and operation conditions. We have analyzed the conventional thresholds composed of fixed-intensity, average, standard deviation, maximum value. Finally, we extracted that the optimal threshold value is more than summation of average and standard deviation, and less than maximum value. it will be enhance flame detection rate than conventional fixed-threshold method.

• Journal of Satellite, Information and Communications
• /
• v.9 no.2
• /
• pp.33-37
• /
• 2014

This paper propose a method to sort flame regions and non-flame regions in a color image based on temperature Characteristics of flame. The traditional algorithms simply detect flame regions those are colored between yellow and red and there are lot of false detection in this method. But the colors of real flame are fallen between white and red and flame color variation over the flame. In this paper, it reduce false detection by separating colors according to temperature Characteristics of flame. The proposed method firstly finds a color model to express the temperature Characteristics of fire and then the color model is non-linearly quantized based on color values and analyzed using histogram and finally detect the candidate flame regions. The proposed method has 71.8% of matching rate and if it is compared with non-matching rate of traditional algorithms, the non-matching rate is improved by 27 times than others.