• Journal of KIISE
• /
• v.42 no.10
• /
• pp.1222-1230
• /
• 2015

Segmentation of the anterior cruciate ligament (ACL) in knee MRI remains a challenging task due to its inhomogeneous signal intensity and low contrast with surrounding soft tissues. In this paper, we propose a multi-atlas-based segmentation of the ACL in knee MRI with locally-aligned probabilistic atlas (PA) in an iterative graph cuts framework. First, a novel PA generation method is proposed with global and local multi-atlas alignment by means of rigid registration. Second, with the generated PA, segmentation of the ACL is performed by maximum-aposteriori (MAP) estimation and then by graph cuts. Third, refinement of ACL segmentation is performed by improving shape prior through mask-based PA generation and iterative graph cuts. Experiments were performed with a Dice similarity coefficients of 75.0%, an average surface distance of 1.7 pixels, and a root mean squared distance of 2.7 pixels, which increased accuracy by 12.8%, 22.7%, and 22.9%, respectively, from the graph cuts with patient-specific shape constraints.

• Journal of Korea Spatial Information System Society
• /
• v.9 no.2
• /
• pp.25-34
• /
• 2007

Existing approaches that select an order for the join of three or more data streams have always used the simple heuristics. For their disadvantage - only one factor is considered and that is join selectivity or arrival rate, these methods lead to poor performance and inefficiency In some applications. The graph-based sliding window multi -join algorithm with optimal join sequence is proposed in this paper. In this method, sliding window join graph is set up primarily, in which a vertex represents a join operator and an edge indicates the join relationship among sliding windows, also the vertex weight and the edge weight represent the cost of join and the reciprocity of join operators respectively. Then the optimal join order can be found in the graph by using improved MVP algorithm. The final result can be produced by executing the join plan with the nested loop join procedure, The advantages of our algorithm are proved by the performance comparison with existing join algorithms.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.21 no.5
• /
• pp.564-568
• /
• 2011

Microblogs became popular information delivery ways due to the spread of smart phones. They have the characteristic of reflecting the interests of users more quickly than other medium. Particularly, in case of the subject which attracts many users, microblogs can supply rich information originated from various information sources. Nevertheless, it has been considered as a hard problem to obtain useful information from microblogs because too much noises are in them. So far, various methods are proposed to extract and track some subjects from particular documents, yet these methods do not work effectively in case of microblogs which consist of short phrases. In this paper, we propose a graph-based topic extraction and partitioning method to understand interests of users about a certain keyword. The proposed method contains the process of generating a keyword graph using the co-occurrences of terms in the microblogs, and the process of splitting the graph by using a network partitioning method. When we applied the proposed method on some keywords. our method shows good performance for finding a topic about the keyword and partitioning the topic into sub-topics.

• KIISE Transactions on Computing Practices
• /
• v.21 no.4
• /
• pp.327-332
• /
• 2015

Constructing a k-nearest neighbor (k-NN) graph is a primitive operation in the field of recommender systems, information retrieval, data mining and machine learning. Although there have been many algorithms proposed for constructing a k-NN graph, either the existing approaches cannot be used for various types of similarity measures, or the performance of the approaches is decreased as the number of nodes or dimensions increases. In this paper, we present a novel algorithm for k-NN graph construction based on "balanced" canopy clustering. The experimental results show that irrespective of the number of nodes or dimensions, our algorithm is at least five times faster than the brute-force approach while retaining an accuracy of approximately 92%.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.245-248
• /
• 2009

Recently, Many studies have been progressing to path-finding with dynamic information on the Terminal Based Navigation System(TBNS). However, the most of existing algorithms are based on $A{\ast}$ algorithm. Path-finding algorithms which use heuristic function may occur a problem of the increase of exploring cost in case of that there is no way determined by heuristic function or there are 2 way more which have almost same cost. In this paper, two abstract graph(AG) that are different method of construction, Homogeneous Node merging($AG^H$) and Connected Node Merging($AG^C$), are implemented. The abstract graph is a simple graph of real road network. The method of using the abstract graph is proposed for reducing dependency of heuristic and exploring cost. In result of evaluation of performance, $AG^C$ has better performance than $AG^H$ at construction cost but $AG^C$ has worse performance than $AG^H$ exploring cost.

• Journal of the Korea Society of Computer and Information
• /
• v.11 no.4 s.42
• /
• pp.87-96
• /
• 2006

Although bytecode has many good features, it has slow execution speed and it is not an ideal representation for program analysis or optimization. For analysises and optimizations. bytecode must be translated to a Static Single Assignment Form(SSA Form) But when bytecode is translated a SSA Form it has lost type informations of son variables. For resolving these problem in this paper, we create extended control flow graph on bytecode. Also we convert the control flow graph to SSA Form for static analysis. Calculation about many informations such as dominator, immediate dominator. dominance frontier. ${\phi}$-Function. renaming are required to convert to SSA Form. To obtain appropriate type for generated SSA Form, we proceed the followings. First. we construct call graph and derivation graph of classes. And the we collect information associated with each node. After finding equivalence nodes and constructing Strongly Connected Component based on the collected informations. we assign type to each node.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.26 no.8
• /
• pp.1103-1110
• /
• 2022

Due to the rapid development of AI-related industries, countless edge devices are working in the real world. Since data generated within the smart space consisted of these devices is beyond imagination, it is becoming increasingly difficult for edge devices to process. To solve this issue, EdgeCPS has appeared. EdgeCPS is a technology to support harmonious execution of various application services including AI applications through interworking between edge devices and edge servers, and augmenting resources/functions. Therefore, we propose a knowledge-sharing graph-based componentized AI application support system applicable to the EdgeCPS platform. The graph is designed to effectively store information which are essential elements for creating AI applications. In order to easily change resource/function augmentation under the support of the EdgeCPS platform, AI applications are operated as components. The application support system is linked with the knowledge graph so that users can easily create and test applications, and visualizes the execution aspect of the application to users as a pipeline.

• Physical Therapy Rehabilitation Science
• /
• v.9 no.4
• /
• pp.287-294
• /
• 2020

Objective: This study aimed to investigate the effects of an ultrasound-based bar graph proportional to the quadriceps muscle thickness as a real-time visual feedback training tool in the elderly. Design: Cross-sectional study. Methods: Twenty-four elderly persons participated in this study and were randomly divided into three groups: oral training group (n=8, group 1), ultrasound imaging group (n=8, group 2), and graph group (n=8, group 3). In the pre condition, all participants performed maximal voluntary isometric contraction (MVIC) of the quadriceps with knee extension three times with oral training. In the post condition, group 1 performed MVIC of the quadriceps with oral training, group 2 performed MVIC of the quadriceps with real-time visual feedback using ultrasound imaging, and group 3 performed MVIC of the quadriceps with real-time visual feedback using a bar graph proportional to the quadriceps muscle thickness, three times for all groups. Muscle thickness, activity (mean, peak), tone, stiffness, logarithmic decrement, relaxation, and creep were measured in both conditions in all participants. Results: Visual feedback with a bar graph showed significant effects on muscle thickness, mean muscle activity, and peak muscle activity compared with oral training and visual feedback with ultrasound imaging (p<0.05). Conclusions: Isometric training of the quadriceps with real-time visual feedback using a bar graph proportional to the quadriceps muscle thickness may be more effective than other methods in improving muscle thickness and muscle activity. This study presented a tool that can help increase muscle thickness in the elderly.

• The Journal of the Korea Contents Association
• /
• v.22 no.8
• /
• pp.33-43
• /
• 2022

Labeled graphs are used to represent entities, their relationships, and their structures in real data such as knowledge graphs and protein interactions. With the rapid development of IT and the explosive increase in data, there has been a need for a subgraph matching technology to provide information that the user is interested in. In this paper, we propose an approximate Top-k labeled subgraph matching scheme that considers the semantic similarity of labels and the difference in graph structure. The proposed scheme utilizes a learning model using FastText in order to consider the semantic similarity of a label. In addition, the label similarity graph(LSG) is used for approximate subgraph matching by calculating similarity values between labels in advance. Through the LSG, we can resolve the limitations of the existing schemes that subgraph expansion is possible only if the labels match exactly. It supports structural similarity for a query graph by performing searches up to 2-hop. Based on the similarity value, we provide k subgraph matching results. We conduct various performance evaluations in order to show the superiority of the proposed scheme.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.12
• /
• pp.3383-3397
• /
• 2023

Scene graphs serve as semantic abstractions of images and play a crucial role in enhancing visual comprehension and reasoning. However, the performance of Scene Graph Generation is often compromised when working with biased data in real-world situations. While many existing systems focus on a single stage of learning for both feature extraction and classification, some employ Class-Balancing strategies, such as Re-weighting, Data Resampling, and Transfer Learning from head to tail. In this paper, we propose a novel approach that decouples the feature extraction and classification phases of the scene graph generation process. For feature extraction, we leverage a transformer-based architecture and design an adaptive calibration function specifically for predicate classification. This function enables us to dynamically adjust the classification scores for each predicate category. Additionally, we introduce a Distribution Alignment technique that effectively balances the class distribution after the feature extraction phase reaches a stable state, thereby facilitating the retraining of the classification head. Importantly, our Distribution Alignment strategy is model-independent and does not require additional supervision, making it applicable to a wide range of SGG models. Using the scene graph diagnostic toolkit on Visual Genome and several popular models, we achieved significant improvements over the previous state-of-the-art methods with our model. Compared to the TDE model, our model improved mR@100 by 70.5% for PredCls, by 84.0% for SGCls, and by 97.6% for SGDet tasks.