• ETRI Journal
• /
• v.46 no.3
• /
• pp.526-537
• /
• 2024

Monitoring fetal growth in utero is crucial to anomaly diagnosis. However, current computer-vision models struggle to accurately assess the key metrics (i.e., head circumference and occipitofrontal and biparietal diameters) from ultrasound images, largely owing to a lack of training data. Mitigation usually entails image augmentation (e.g., flipping, rotating, scaling, and translating). Nevertheless, the accuracy of our task remains insufficient. Hence, we offer a U-Net fetal head measurement tool that leverages a hybrid Dice and binary cross-entropy loss to compute the similarity between actual and predicted segmented regions. Ellipse-fitted two-dimensional ultrasound images acquired from the HC18 dataset are input, and their lower feature layers are reused for efficiency. During regression, a novel region of interest pooling layer extracts elliptical feature maps, and during segmentation, feature pyramids fuse field-layer data with a new scale attention method to reduce noise. Performance is measured by Dice similarity, mean pixel accuracy, and mean intersection-over-union, giving 97.90%, 99.18%, and 97.81% scores, respectively, which match or outperform the best U-Net models.

• Earthquakes and Structures
• /
• v.18 no.1
• /
• pp.27-44
• /
• 2020

This work presents a comparison of three performance-based seismic design methods (PBSD) as applied to plane steel frames having eccentric braces (EBFs) and buckling restrained braces (BRBFs). The first method uses equivalent modal damping ratios (ξ_k), referring to an equivalent multi-degree-of-freedom (MDOF) linear system, which retains the mass, the elastic stiffness and responds in the same way as the original non-linear MDOF system. The second method employs modal strength reduction factors (${\bar{q}}_k$) resulting from the corresponding modal damping ratios. Contrary to the behavior factors of code based design methods, both ξ_k and ${\bar{q}}_k$ account for the first few modes of significance and incorporate target deformation metrics like inter-storey drift ratio (IDR) and local ductility as well as structural characteristics like structural natural period, and soil types. Explicit empirical expressions of ξ_k and ${\bar{q}}_k$, recently presented by the present authors elsewhere, are also provided here for reasons of completeness and easy reference. The third method, developed here by the authors, is based on a hybrid force/displacement (HFD) seismic design scheme, since it combines the force-base design (FBD) method with the displacement-based design (DBD) method. According to this method, seismic design is accomplished by using a behavior factor (q_h), empirically expressed in terms of the global ductility of the frame, which takes into account both non-structural and structural deformation metrics. These expressions for q_h are obtained through extensive parametric studies involving non-linear dynamic analysis (NLDA) of 98 frames, subjected to 100 far-fault ground motions that correspond to four soil types of Eurocode 8. Furthermore, these factors can be used in conjunction with an elastic acceleration design spectrum for seismic design purposes. Finally, a comparison among the above three seismic design methods and the Eurocode 8 method is conducted with the aid of non-linear dynamic analyses via representative numerical examples, involving plane steel EBFs and BRBFs.

• Imaging Science in Dentistry
• /
• v.52 no.4
• /
• pp.383-391
• /
• 2022

Purpose: Periodontitis, the most prevalent chronic inflammatory condition affecting teeth-supporting tissues, is diagnosed and classified through clinical and radiographic examinations. The staging of periodontitis using panoramic radiographs provides information for designing computer-assisted diagnostic systems. Performing image segmentation in periodontitis is required for image processing in diagnostic applications. This study evaluated image segmentation for periodontitis staging based on deep learning approaches. Materials and Methods: Multi-Label U-Net and Mask R-CNN models were compared for image segmentation to detect periodontitis using 100 digital panoramic radiographs. Normal conditions and 4 stages of periodontitis were annotated on these panoramic radiographs. A total of 1100 original and augmented images were then randomly divided into a training (75%) dataset to produce segmentation models and a testing (25%) dataset to determine the evaluation metrics of the segmentation models. Results: The performance of the segmentation models against the radiographic diagnosis of periodontitis conducted by a dentist was described by evaluation metrics(i.e., dice coefficient and intersection-over-union [IoU] score). MultiLabel U-Net achieved a dice coefficient of 0.96 and an IoU score of 0.97. Meanwhile, Mask R-CNN attained a dice coefficient of 0.87 and an IoU score of 0.74. U-Net showed the characteristic of semantic segmentation, and Mask R-CNN performed instance segmentation with accuracy, precision, recall, and F1-score values of 95%, 85.6%, 88.2%, and 86.6%, respectively. Conclusion: Multi-Label U-Net produced superior image segmentation to that of Mask R-CNN. The authors recommend integrating it with other techniques to develop hybrid models for automatic periodontitis detection.

• Journal of the Korean Geographical Society
• /
• v.52 no.1
• /
• pp.123-141
• /
• 2017

This study predicted urban green spaces for 2020 based on two scenarios keeping or freeing the green-belt in the Daegu metropolitan city using a hybrid Cellular Automata(CA)-Markov model and analyzed the spatial dynamics of urban green spaces between 2009 and 2020 using a land cover change detection technique and spatial metrics. Markov chain analysis was employed to derive the transition probability for projecting land cover change into the future for 2020 based on two land cover maps in 1998 and 2009 provided by the Ministry of Environment. Multi-criteria evaluation(MCE) was adopted to develop seven suitability maps which were empirically derived in relation to the six restriction factors underlying the land cover change between the years 1998 and 2009. A hybrid CA-Markov model was then implemented to predict the land cover change over an 11 year period to 2020 based on two scenarios keeping or freeing the green-belt. The projected land cover for 2009 was cross-validated with the actual land cover in 2009 using Kappa statistics. Results show that urban green spaces will be remarkably fragmented in the suburban areas such as Dalseong-gun, Seongseo, Ansim and Chilgok in the year 2020 if the Daegu metropolitan city keeps its urbanization at current pace and in case of keeping the green-belt. In case of freeing the green-belt, urban green spaces will be fragmented on the fringes of the green-belt. It is thus required to monitor urban green spaces systematically considering the spatial change patterns identified by this study for sustainably managing them in the Daegu metropolitan city in the near future.

• Journal of Intelligence and Information Systems
• /
• v.27 no.1
• /
• pp.23-46
• /
• 2021

Collaborative filtering, which is often used in personalization recommendations, is recognized as a very useful technique to find similar customers and recommend products to them based on their purchase history. However, the traditional collaborative filtering technique has raised the question of having difficulty calculating the similarity for new customers or products due to the method of calculating similaritiesbased on direct connections and common features among customers. For this reason, a hybrid technique was designed to use content-based filtering techniques together. On the one hand, efforts have been made to solve these problems by applying the structural characteristics of social networks. This applies a method of indirectly calculating similarities through their similar customers placed between them. This means creating a customer's network based on purchasing data and calculating the similarity between the two based on the features of the network that indirectly connects the two customers within this network. Such similarity can be used as a measure to predict whether the target customer accepts recommendations. The centrality metrics of networks can be utilized for the calculation of these similarities. Different centrality metrics have important implications in that they may have different effects on recommended performance. In this study, furthermore, the effect of these centrality metrics on the performance of recommendation may vary depending on recommender algorithms. In addition, recommendation techniques using network analysis can be expected to contribute to increasing recommendation performance even if they apply not only to new customers or products but also to entire customers or products. By considering a customer's purchase of an item as a link generated between the customer and the item on the network, the prediction of user acceptance of recommendation is solved as a prediction of whether a new link will be created between them. As the classification models fit the purpose of solving the binary problem of whether the link is engaged or not, decision tree, k-nearest neighbors (KNN), logistic regression, artificial neural network, and support vector machine (SVM) are selected in the research. The data for performance evaluation used order data collected from an online shopping mall over four years and two months. Among them, the previous three years and eight months constitute social networks composed of and the experiment was conducted by organizing the data collected into the social network. The next four months' records were used to train and evaluate recommender models. Experiments with the centrality metrics applied to each model show that the recommendation acceptance rates of the centrality metrics are different for each algorithm at a meaningful level. In this work, we analyzed only four commonly used centrality metrics: degree centrality, betweenness centrality, closeness centrality, and eigenvector centrality. Eigenvector centrality records the lowest performance in all models except support vector machines. Closeness centrality and betweenness centrality show similar performance across all models. Degree centrality ranking moderate across overall models while betweenness centrality always ranking higher than degree centrality. Finally, closeness centrality is characterized by distinct differences in performance according to the model. It ranks first in logistic regression, artificial neural network, and decision tree withnumerically high performance. However, it only records very low rankings in support vector machine and K-neighborhood with low-performance levels. As the experiment results reveal, in a classification model, network centrality metrics over a subnetwork that connects the two nodes can effectively predict the connectivity between two nodes in a social network. Furthermore, each metric has a different performance depending on the classification model type. This result implies that choosing appropriate metrics for each algorithm can lead to achieving higher recommendation performance. In general, betweenness centrality can guarantee a high level of performance in any model. It would be possible to consider the introduction of proximity centrality to obtain higher performance for certain models.

• Journal of the Korea Society of Computer and Information
• /
• v.15 no.11
• /
• pp.67-73
• /
• 2010

In this paper, a new face detection method that is more accurate than the conventional methods is proposed. This method utilizes a hybrid of Hausdorff distance based on the geometric similarity between the two sets of points and the LBP distance based on the distribution of local micro texture of an image. The parameters for normalization and the optimal blending factor of the two different metrics were calculated from training sample images. Popularly used face database was used to show that the proposed method is more effective and robust to the variation of the pose, illumination, and back ground than the methods based on the Hausdorff distance or LBP distance. In the particular case, the average error distance between the detected and the true face location was reduced to 47.9% of the result of LBP method, and 22.8% of the result of Hausdorff method.

• Journal of Electrical Engineering and Technology
• /
• v.8 no.1
• /
• pp.190-196
• /
• 2013

This paper proposes a new guidance law, which considers the Field of View (FOV) of the seeker when a missile has a strapdown seeker mounted instead of a gimbal seeker. When a strapdown seeker, which has a narrow FOV, is used for tracking a target, the FOV of the seeker is an important consideration for guidance performance metrics such as miss distance. We propose a new guidance law called hybrid guidance (HG) to address the shortcomings of conventional guidance laws such as proportional navigation guidance (PNG), which cannot maintain lock-on conditions against high speed targets due to the narrow FOV of the strapdown seeker. The aim of the HG law is to null miss distance and to maintain the look angle within the FOV of the strapdown seeker. In order to achieve this goal, we combine two guidance laws in the HG law. One is a PNG law to null the LOS rate, and the other is a sliding mode guidance (SMG) law derived to keep the look angle within the FOV by employing a Lyapunov-like function with a sliding mode control methodology. We also propose a method to switch these two guidance laws at certain look angles for better guidance performance.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.13 no.7
• /
• pp.1411-1421
• /
• 2009

As the wireless internet grows exponentially, the recent trend has an increasing demand for wireless network and multimedia services. RTP is used to support the multimedia communication over the Internet and it supports the flexibility and adaptability over a wide range. However, RTP has a limitation that it cannot support end-to-end QoS guarantee in a wireless home network which has low throughput and high delay. In this paper, we propose the architecture of a real-time multimedia communication and design and implement the hybrid flow control in the architecture. The hybrid flow control mechanism is based on modified AIMD using metrics such as the network state information and the user properties. We implement the porposed flow control using JMF to evaluate the performance of the proposed flow control. The experimental results show that the proposed flow control has better performance than the AIMD.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.8
• /
• pp.2140-2156
• /
• 2023

The Wireless Sensor Network (WSN), is constructed out of teeny-tiny sensor nodes that are very low-cost, have a low impact on the environment in terms of the amount of power they consume, and are able to successfully transmit data to the base station. The primary challenges that are presented by WSN are those that are posed by the distance between nodes, the amount of energy that is consumed, and the delay in time. The sensor node's source of power supply is a battery, and this particular battery is not capable of being recharged. In this scenario, the amount of energy that is consumed rises in direct proportion to the distance that separates the nodes. Here, we present a Hybrid Firefly Glow-Worm Swarm Optimization (HF-GSO) guided routing strategy for preserving WSNs' low power footprint. An efficient fitness function based on firefly optimization is used to select the Cluster Head (CH) in this procedure. It aids in minimising power consumption and the occurrence of dead sensor nodes. After a cluster head (CH) has been chosen, the Glow-Worm Swarm Optimization (GSO) algorithm is used to figure out the best path for sending data to the sink node. Power consumption, throughput, packet delivery ratio, and network lifetime are just some of the metrics measured and compared between the proposed method and methods that are conceptually similar to those already in use. Simulation results showed that the proposed method significantly reduced energy consumption compared to the state-of-the-art methods, while simultaneously increasing the number of functioning sensor nodes by 2.4%. Proposed method produces superior outcomes compared to alternative optimization-based methods.

• International conference on construction engineering and project management
• /
• 2022.06a
• /
• pp.792-799
• /
• 2022

Building Information Modeling (BIM) technology is a key component of modern construction engineering and project management workflows. As-is BIM models that represent the spatial reality of a project site can offer crucial information to stakeholders for construction progress monitoring, error checking, and building maintenance purposes. Geometric methods for automatically converting raw scan data into BIM models (Scan-to-BIM) often fail to make use of higher-level semantic information in the data. Whereas, semantic segmentation methods only output labels at the point level without creating object level models that is necessary for BIM. To address these issues, this research proposes a hybrid semantic-geometric approach for clutter-resistant floorplan generation from laser-scanned building point clouds. The input point clouds are first pre-processed by normalizing the coordinate system and removing outliers. Then, a semantic segmentation network based on PointNet++ is used to label each point as ceiling, floor, wall, door, stair, and clutter. The clutter points are removed whereas the wall, door, and stair points are used for 2D floorplan generation. A region-growing segmentation algorithm paired with geometric reasoning rules is applied to group the points together into individual building elements. Finally, a 2-fold Random Sample Consensus (RANSAC) algorithm is applied to parameterize the building elements into 2D lines which are used to create the output floorplan. The proposed method is evaluated using the metrics of precision, recall, Intersection-over-Union (IOU), Betti error, and warping error.