• Journal of the Institute of Convergence Signal Processing
• /
• v.24 no.2
• /
• pp.119-125
• /
• 2023

In this study, we photograph driving obstacle objects such as crosswalks, side spheres, manholes, braille blocks, partial ramps, temporary safety barriers, stairs, and inclined curb that hinder or cause inconvenience to the movement of the vulnerable using electric mobility aids. We develop an optimal AI model that classifies photographed objects and automatically recognizes them, and implement an algorithm that can efficiently determine obstacles in front of electric mobility aids. In order to enable object detection to be AI learning with high probability, the labeling form is labeled as a polygon form when building a dataset. It was developed using a Mask R-CNN model in Detectron2 framework that can detect objects labeled in the form of polygons. Image acquisition was conducted by dividing it into two groups: the general public and the transportation weak, and image information obtained in two areas of the test bed was secured. As for the parameter setting of the Mask R-CNN learning result, it was confirmed that the model learned with IMAGES_PER_BATCH: 2, BASE_LEARNING_RATE 0.001, MAX_ITERATION: 10,000 showed the highest performance at 68.532, so that the user can quickly and accurately recognize driving risks and obstacles.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.37 no.3
• /
• pp.259-266
• /
• 2013

An automated cell tracking system is used to automatically analyze and track the changes in cell behavior in time-lapse cell images acquired using a microscope with a cell culture. Clustering is the partial overlapping of neighboring cells in the process of cell change. Separating clusters into individual cells is very important for cell tracking. In this study, we proposed an algorithm for separating clusters by using ellipse fitting based on a direct least square method. We extracted the contours of clusters, divided them into line segments, and then produced their fitted ellipses using a direct least square method for each line segment. All of the fitted ellipses could be used to separate their corresponding clusters. In experiments, our algorithm separated clusters with average precisions of 91% for two overlapping cells, 84% for three overlapping cells, and about 73% for four overlapping cells.

• Journal of Trauma and Injury
• /
• v.29 no.4
• /
• pp.99-104
• /
• 2016

Purpose: There are many patients visited to ED in an alcohol intoxicated state. For these patients, it is difficult to predict by only clinical examination whether he/she would have brain lesion. The purpose of this study is to research whether it is possible to predict brain lesion by only clinical examination findings, with comparing patients with/without actual brain lesions. Methods: A retrospective study was performed at a university hospital for the period 11 months with the medical records. As for the inclusion group, head trauma patients with objectively proved drunk, judging by their blood ethanol concentration, and performed the brain CT were selected. In terms of medical record, Glasgow coma scale (GCS), the presence of neurologic abnormalities, the presence of lesion on brain CT of the patients, were examined. From laboratory results, blood ethanol concentration, platelet count, prothrombin time (PT), activated partial thromboplastin time (aPTT) and glucose concentration were identified. Results: For this study, there were total 80 patients of inclusion group. There was no statistically significant difference in terms of GCS score and neurological examination abnormalities, between the group with brain lesion and the group without brain lesion on brain CT. Conclusion: Alcohol intoxicated patient with head trauma visits the ED, it is not possible to distinguish or determine whether brain lesion exists or not by only clinical findings. In order to check the lesion existence, the image examination, therefore, should be considered and performed.

• Journal of Digital Convergence
• /
• v.14 no.10
• /
• pp.415-421
• /
• 2016

There has been an enormous increase not only in the international economic exchanges but also in the cultural exchanges. However, studies to settle the cultural differences have been studied a lot with in economic and business partial but research of eliminating cultural differences for travelers has not been studied. Therefore, in this study, we tried to develop the service design to eliminate the cultural differences. To set the spatial range, we choose airport which can be the hub and the door of the travelers at the same time. Moreover, we tried to understand the meaning of communication and sharing to resolve differences in culture through literature studies. Further, we analyzed the airport application and examples of Wi-Fi linked services, so can be applied to the proposal. Travelers share the cultural differences by themselves; through the Wi-Fi service in the airport, and the existing airport application. We expect to be able to resolve the cultural differences between nations and develop not only their prosperity but also the national image by this service.

• Smart Structures and Systems
• /
• v.20 no.4
• /
• pp.397-413
• /
• 2017

Precast concrete (PC) members are currently being employed for general construction or partial replacement to reduce construction period. As assembly work in PC construction requires connecting PC members accurately, measuring the 6-DOF (degree of freedom) relative displacement is essential. Multiple planar markers and camera-based displacement measurement systems can monitor the 6-DOF relative displacement of PC members. Conventional methods, such as direct linear transformation (DLT) for homography estimation, which are applied to calculate the 6-DOF relative displacement between the camera and marker, have several major problems. One of the problems is that when the marker is partially hidden, the DLT method cannot be applied to calculate the 6-DOF relative displacement. In addition, when the images of markers are blurred, error increases with the DLT method which is employed for its estimation. To solve these problems, a hybrid method, which combines the advantages of the DLT and MCL (Monte Carlo localization) methods, is proposed. The method evaluates the 6-DOF relative displacement more accurately compared to when either the DLT or MCL is used alone. Each subsystem captures an image of a marker and extracts its subpixel coordinates, and then the data are transferred to a main system via a wireless communication network. In the main system, the data from each subsystem are used for 3D visualization. Thereafter, the real-time movements of the PC members are displayed on a tablet PC. To prove the feasibility, the hybrid method is compared with the DLT method and MCL in real experiments.

• Journal of Cadastre & Land InformatiX
• /
• v.46 no.1
• /
• pp.87-100
• /
• 2016

The acquisition of the forest resource information has depended on a partial sampling method or aerial photographs which demand a lot of effort and time because of the vast areas and the difficult approach. For the acquisition of the forest resource information, there have been the optical remote-sensing and the multi-spectrum image to offer only horizontal distributions of trees, but a new technological approach, such as Airborne LiDAR, is more necessary to acquire directly three dimensional information related to the forest terrains and trees' features. This paper proposes an algorithm for the forest information extraction such as trees' individual numbers and the heights of trees by using LiDAR data. Especially, this proposed algorithm adopts a region growing method for the extraction of the vegetation-point and extracts the forest information using morphological features of trees.

• Journal of Food Hygiene and Safety
• /
• v.5 no.4
• /
• pp.179-186
• /
• 1990

This study was performed for assessing carcinogenicity of several synthetic hormones; Diethylstilbesterol (DES), 17-ethinylestradiol ($EE_2$) and Bovine somatotrophin (BST). Six weeks old F344 rats were divided into five groups and given an intraperitoneally injection of 200 mg of diethylnitrosamine (DENA). At two week after beginnig of experiment, DES, $EE_2$, BST. Phenobarbital were administered to group 1, 2, 3 and 4 respectively, group 4 is positive control and group 5 is negative control. At the same time, all groups received a single i.p. injection of D-galactosamine at a dose of 300 mg/kg and underwent 2/3 partial hepatectomy at week 5. All rats were sacrificed at the end of week 8 for assessment of liver lesion development. The liver was processed for immunohistochmical staining for GST-P and quantitatively analyzed by image analyzer. It was concluded that two synthetic estrogen hormones (DES, $EE_2$) was different significantly (p < 0.01) but BST was not different as compared with control group. Therefore, we thought that DES, $EE_2$ was promoting effects and BST was not in rat hepatocarcinogenesis.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.43 no.3 s.309
• /
• pp.8-15
• /
• 2006

Obstacle detection is an important task for many mobile robot applications. The methods using stereo vision and optical flow are computationally expensive. Therefore, this paper presents a vision-based obstacle detection method using only two view images. The method uses a single passive camera and odometry, performs in real-time. The proposed method is an obstacle detection method using 3D reconstruction from taro views. Processing begins with feature extraction for each input image using Dr. Lowe's SIFT(Scale Invariant Feature Transform) and establish the correspondence of features across input images. Using extrinsic camera rotation and translation matrix which is provided by odometry, we could calculate the 3D position of these corresponding points by triangulation. The results of triangulation are partial 3D reconstruction for obstacles. The proposed method has been tested successfully on an indoor mobile robot and is able to detect obstacles at 75msec.

• Journal of the Institute of Electronics Engineers of Korea SC
• /
• v.43 no.5 s.311
• /
• pp.1-9
• /
• 2006

Because most applications of vision-based object tracking demonstrate satisfactory operations only under very constrained environments that have simplifying assumptions or specific visual attributes, these approaches can't track target objects for the highly variable, unstructured, and dynamic environments like a traffic scene. An adaptive fusion framework is essential that takes advantage of the richness of visual information such as color, appearance shape and so on, especially at cluttered and dynamically changing scenes with partial occlusion[1]. This paper develops a particle filter based adaptive fusion framework and improves the robustness and adaptation of this framework by adding a new distinctive visual attribute, an image feature descriptor using SIFT (Scale Invariant Feature Transform)[2] and adding an automatic teaming scheme of the SIFT feature library according to viewpoint, illumination, and background change. The proposed algorithm is applied to track various traffic objects like vehicles, pedestrians, and bikes in a driver assistance system as an important component of the Intelligent Transportation System.

• Journal of The Korean Astronomical Society
• /
• v.36 no.spc1
• /
• pp.93-99
• /
• 2003

Various attempts have been made to explain the: pronounced seasonal and universal time (UT) variations of geomagnetic indices. As one of such attempts, we analyze the hourly-averaged auroral electroject indices obtained during the past 20 years. The AU and AL indices maximize during summer and equinoctial months, respectively. By normalizing the contribution of the solar conductivity enhancement to the AU index, or to the eastward electrojet, it is found that the AU also follows the same semiannual variation pattern of the AL index, suggesting that the electric field is the main modulator of the semiannual magnetic variation. The fact that the variation pattern of the yearly-mean AU index follows the mirror image of the AL index provides another indication that the electric field is the main modulator of magnetic disturbance. The pronounced UT variations of the auroral electrojet indices are also noted. To determine the magnetic activity dependence, the probability of recording a given activity level of AU and AL during each UT is examined. The UT variation of the AL index, thus obtained, shows a maximum at around 1200-1800 UT and a minimum around 0000-0800 UT particularly during winter. It is closely associated with the rotation of the geomagnetic pole around the rotational axis, which results in the change of the solar-originated ionospheric conductivity distribution over the polar region. On the other hand the UT variation is prominent during disturbed periods, indicating that the latitudinal mismatch between the AE stations and the auroral electrojet belt is responsible for it. Although not as prominent as the AL index, the probability distribution of the AU also shows two UT peaks. We confirm that the Dst index shows more prominent seasonal variation than the AE indices. However, the UT variation of the Dst index is only noticeable during the main phase of a magnetic storm. It is a combined result of the uneven distribution of the Dst stations and frequent developments of the partial ring current and substorm wedge current preferentially during the main phase.