• Journal of the Korea Society of Computer and Information
• /
• v.27 no.2
• /
• pp.33-42
• /
• 2022

In this paper, we proposed a deep learning model based on multi-scale dilated convolution feature fusion for the segmentation of black ice region in road image to send black ice warning to drivers in real time. In the proposed multi-scale dilated convolution feature fusion network, different dilated ratio convolutions are connected in parallel in the encoder blocks, and different dilated ratios are used in different resolution feature maps, and multi-layer feature information are fused together. The multi-scale dilated convolution feature fusion improves the performance by diversifying and expending the receptive field of the network and by preserving detailed space information and enhancing the effectiveness of diated convolutions. The performance of the proposed network model was gradually improved with the increase of the number of dilated convolution branch. The mIoU value of the proposed method is 96.46%, which was higher than the existing networks such as U-Net, FCN, PSPNet, ENet, LinkNet. The parameter was 1,858K, which was 6 times smaller than the existing LinkNet model. From the experimental results of Jetson Nano, the FPS of the proposed method was 3.63, which can realize segmentation of black ice field in real time.

• Proceedings of the KSRS Conference
• /
• 2005.10a
• /
• pp.79-82
• /
• 2005

Recently the necessity of road data is still being increased in industrial society, so there are many repairing and new constructions of roads at many areas. According to the development of government, city and region, the update and acquisition of road data for GIS (Geographical Information System) is very necessary. In this study, the fusion method with range data(3D Ground Coordinate System Data) and Intensity data in stand alone LiDAR data is used for road extraction and then digital image processing method is applicable. Up to date Intensity data of LiDAR is being studied. This study shows the possibility method for road extraction using Intensity data. Intensity and Range data are acquired at the same time. Therefore LiDAR does not have problems of multi-sensor data fusion method. Also the advantage of intensity data is already geocoded, same scale of real world and can make ortho-photo. Lastly, analysis of quantitative and quality is showed with extracted road image which compare with I: 1,000 digital map.

• The Journal of Korea Robotics Society
• /
• v.4 no.4
• /
• pp.312-319
• /
• 2009

This paper proposes a low-complexity indoor localization method of mobile robot under the dynamic environment by fusing the landmark image information from an ordinary camera and the distance information from sensor nodes in an indoor environment, which is based on sensor network. Basically, the sensor network provides an effective method for the mobile robot to adapt to environmental changes and guides it across a geographical network area. To enhance the performance of localization, we used an ordinary CCD camera and the artificial landmarks, which are devised for self-localization. Experimental results show that the real-time localization of mobile robot can be achieved with robustness and accurateness using the proposed localization method.

• Molecules and Cells
• /
• v.26 no.6
• /
• pp.517-529
• /
• 2008

Porosomes are supramolecular, lipoprotein structures at the cell plasma membrane, where membrane-bound secretory vesicles transiently dock and fuse to release inravesicular contents to the outside during cell secretion. The mouth of the porosome opening to the outside, range in size from 150 nm in diameter in acinar cells of the exocrine pancreas, to 12 nm in neurons, which dilates during cell secretion, returning to its resting size following completion of the process. In the past decade, the composition of the porosome, its structure and dynamics at nm resolution and in real time, and its functional reconstitution into artificial lipid membrane, have all been elucidated. In this mini review, the discovery of the porosome, its structure, function, isolation, chemistry, and reconstitution into lipid membrane, the molecular mechanism of secretory vesicle swelling and fusion at the base of porosomes, and how this new information provides a paradigm shift in our understanding of cell secretion, is discussed.

• International Journal of Naval Architecture and Ocean Engineering
• /
• v.6 no.3
• /
• pp.670-684
• /
• 2014

Ship Information Systems (SISs) have been one of the main research focuses in ship design and become a multidisciplinary area. With these growing research trends, it is important to consolidate the latest knowledge and information to keep up with the research needs. In this paper, the SIS and its different forms are introduced and discussed. The beginning of this paper discusses the history and evolution of SIS. The next part of this paper focuses on different fields and research areas such as networking technology, information fusion, information decision, message display, ship control in real-time SISs. A Semi-Physical Simulation Platform (SPSIM) designed for SIS research and its running effect through a new Fuzzy-PID fusion algorithm are introduced in this paper then. A brief literature survey and possible future direction concerning each topic is included.

• Journal of the Korean Society for Precision Engineering
• /
• v.12 no.3
• /
• pp.130-143
• /
• 1995

This paper presents a new methodology for on-line tool breadage detection by sensor fusion of an acoustic emission (AE) sensor and a built-in force sensor. A built-in piezoelectric force sensor, instead of a tool dynamometer, was used to measure the cutting force without altering the machine tool dynamics. The sensor was inserted in the tool turret housing of an NC lathe. FEM analysis was carried out to locate the most sensitive position for the sensor. A burst of AE signal was used as a triggering signal to inspect the cutting force. A sighificant drop of cutting force was utilized to detect tool breakage. The algorithm was implemented on a DSP board for in-process tool breakage detection. Experiental works showed an excellent monitoring capability of the proposed tool breakage detection system.

• Tribology and Lubricants
• /
• v.28 no.3
• /
• pp.130-135
• /
• 2012

This study presents a new methodology for realtime tool breakage detection by sensor fusion concept of two hall sensor and an acoustic emission (AE) sensor. Spindle induction motor torque of CNC Lathe during machining is estimated by two hall sensor. Estimated motor torque instead of a tool dynamometer was used to measure the cutting torque and tool breakage detection. A burst of AE signal was used as a triggering signal to inspect the cutting torque. A significant drop of cutting torque was utilized to detect tool breakage. The algorithm was implemented on a NI DAQ (Data Acquisition) board for in-process tool breakage detection. The result of experiment showed an excellent monitoring capability of the proposed tool breakage detection system. This system is available tool breakage monitoring through internet also provides this system's user with current cutting torque of induction motor.

• Journal of Institute of Control, Robotics and Systems
• /
• v.15 no.4
• /
• pp.365-371
• /
• 2009

A intelligent mobile hexapod robot with various types of sensors and wireless camera is introduced. We show this mobile robot can detect objects well by combining the results of active sensors and image processing algorithm. First, to detect objects, active sensors such as infrared rays sensors and supersonic waves sensors are employed together and calculates the distance in real time between the object and the robot using sensor's output. The difference between the measured value and calculated value is less than 5%. This paper suggests effective visual detecting system for moving objects with specified color and motion information. The proposed method includes the object extraction and definition process which uses color transformation and AWUPC computation to decide the existence of moving object. We add weighing values to each results from sensors and the camera. Final results are combined to only one value which represents the probability of an object in the limited distance. Sensor fusion technique improves the detection rate at least 7% higher than the technique using individual sensor.

• Proceedings of the Korean Vacuum Society Conference
• /
• 2012.02a
• /
• pp.526-526
• /
• 2012

DC magnetron discharges were studied using three-dimensional self-consistent particle-in-cell and Monte Carlo collisional (PIC-MCC) simulation codes. Two rectangular sputter sources (120 mm * 250 mm and 380 mm * 200 mm target sizes) were used in the simulation modeling. The number of incident ions to the Cu target as a function of position and simulation time was obtained. The target erosion profile was calculated by using the incident ions and the sputtering yields of the Cu target calculated with SRIM codes. The maximum ion density of the ion density distribution in the discharge was about $10^{10}cm^{-3}$ due to the calculation speed limit. The result may be less than one or two order of magnitude smaller than the real maximum ion density. However, the target erosion profiles of the two sputter sources were in good agreement with the measured target erosion profiles except for the erosion profile near the target surface, in which which the measured erosion width was broader than the simulation erosion width.

• Smart Structures and Systems
• /
• v.23 no.1
• /
• pp.61-69
• /
• 2019

Offshore structures are generally exposed to harsh environments such as strong tidal currents and wind loadings. Monitoring the structural soundness and integrity of offshore structures is crucial to prevent catastrophic collapses and to prolong their lifetime; however, it is intrinsically challenging because of the difficulties in accessing the critical structural members that are located under water for installing and repairing sensors and data acquisition systems. Virtual sensing technologies have the potential to alleviate such difficulties by estimating the unmeasured structural responses at the desired locations using other measured responses. Despite the usefulness of virtual sensing, its performance and applicability to the structural health monitoring of offshore structures have not been fully studied to date. This study investigates the use of virtual sensing of offshore structures. A Kalman filter based virtual sensing algorithm is developed to estimate responses at the location of interest. Further, this algorithm performs a multi-sensor data fusion to improve the estimation accuracy under non-stationary tidal loading. Numerical analysis and laboratory experiments are conducted to verify the performance of the virtual sensing strategy using a bottom-fixed offshore structural model. Numerical and experimental results show that the unmeasured responses can be reasonably recovered from the measured responses.