• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2009.10a
• /
• pp.574-578
• /
• 2009

Real-time object detection for distinguishing a moving object of interests from the background image in still image or video image sequence is an essential step to a correct object tracking and recognition. Moving cast shadow can be misclassified as part of objects or moving objects because the shadow region is included in the moving object region after object segmentation. For this reason, an algorithm for shadow removal plays an important role in the results of accurate moving object detection and tracking systems. To handle with the problems, an accurate algorithm based on the features of moving object and shadow in color space is presented in this paper. Experimental results show that the proposed algorithm is effective to detect a moving object and to remove shadow in test video sequences.

• The Journal of Information Systems
• /
• v.19 no.1
• /
• pp.1-17
• /
• 2010

This paper proposes an updating policy of indoor moving object databases (IMODB) for location-based services. our method applies the Ka1man filter on the recently collected measured positions to estimate the moving object's position and velocity at the moment of the most recent measurement, and extrapolate the current position with the estimated position and velocity. If the distance between the extrapolated current position and the measured current position is within the threshold, in other words if they are close then we skip updating the IMODB. When the IMODB needs to know the moving object's position at a certain moment T, it applies the Kalman filter on the series of the measurements received before T and extrapolates the position at T with the estimations obtained by the Kalman filter. In order to verify the efficiency of our updating method, we performed the experiments of applying our method on the series of measured positions obtained by applying the fingerprinting indoor positioning method while we are actually walking through the test bed. In the analysis of the test results, we estimated the communication saving rate of our method and the error increment rate caused by the communication saving.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.17 no.7
• /
• pp.1841-1857
• /
• 2023

CenterNet, a novel object detection algorithm without anchor based on key points, regards the object as a single center point for prediction and directly regresses the object's height and width. However, because the objects have different sizes, directly regressing their height and width will make the model difficult to converge and lose the intrinsic relationship between object's width and height, thereby reducing the stability of the model and the consistency of prediction accuracy. For this problem, we proposed an algorithm based on the regression of the diagonal half-length and the center angle, which significantly compresses the solution space of the regression components and enhances the intrinsic relationship between the decoded components. First, encode the object's width and height into the diagonal half-length and the center angle, where the center angle is the angle between the diagonal and the vertical centreline. Secondly, the predicted diagonal half-length and center angle are decoded into two length components. Finally, the position of the object bounding box can be accurately obtained by combining the corresponding center point coordinates. Experiments show that, when using CenterNet as the improved baseline and resnet50 as the Backbone, the improved model achieved 81.6% and 79.7% mAP on the VOC 2007 and 2012 test sets, respectively. When using Hourglass-104 as the Backbone, the improved model achieved 43.3% mAP on the COCO 2017 test sets. Compared with CenterNet, the improved model has a faster convergence rate and significantly improved the stability and prediction accuracy.

• Journal of the Korea Institute of Building Construction
• /
• v.8 no.5
• /
• pp.43-51
• /
• 2008

This study examines shear capacity performance and structural characteristics of reinforced concrete beam using carbon fiber sheet(CFS), g)ass fiber sheet(GFS), glass fiber steel plate(GSP) and carbon fiber bar CB) which are reinforcing materials for reinforced concrete beam in order to produce similar condition to repair and reinforce actual structure and aims to provide data available In designing and constructing reinforced concrete structures under the structural damage. This study obtains the following conclusions. After considering the shear experiment results. it was indicated that the CB reinforced test object was the best in the shear capacity improvement and ductility capacity as it was contained in the concrete and was all operated, Also, GFS reinforced test object indicated the reduced flexural capacity but good shear capacity. GSP reinforced test object had bigger reinforcing strength than other reinforcing test objects. On the other hand, it showed the lowest reinforcement effect as compared section thickness of reinforced material because it showed the bigger relativity a section thickness of reinforced material. If the adherence to the concrete is improved, it will seem to show bigger reinforcement effect.

• Proceedings of the Korea Concrete Institute Conference
• /
• 2000.10b
• /
• pp.1261-1266
• /
• 2000

In the impact echo method, a stress pulse is introduced into an object at on accessible surface by a transmitter. The pulse propagates into the test object and is reflected by flaws or interfaces. In this paper, void and crack locations of concrete specimens were detected using impact echo method. In their modal identification procedures, the double least squares solution for Ibrahim Time Domain technique was used.

• Journal of Institute of Control, Robotics and Systems
• /
• v.18 no.7
• /
• pp.673-679
• /
• 2012

This paper introduces robust tracking algorithm for fast and erratic moving object. CAMSHIFT algorithm has less computation and efficient performance for object tracking. However, the method fails to track a object if it moves out of search window by fast velocity and/or large movement. The size of the search window in CAMSHIFT algorithm should be selected manually also. To solve these problems, we propose an efficient prediction technique for fast movement of object using Kalman Filter with automatic initial setting and variable configuration technique for search window. The proposed method is compared to the traditional CAMSHIFT algorithm for searching and tracking performance of objects on test image frames.

• Proceedings of the IEEK Conference
• /
• 2008.06a
• /
• pp.837-838
• /
• 2008

This paper presents a new signature and Fourier descriptor based algorithm for recognizing a rotated object and its rotation angle. Fourier descriptor is used to represent an object using its frequence parameters which are not influenced by rotation. once the object is recognized, the point with the largest auto-correlation coefficient which can be calculated from signature of the object is used to find angle of the object. The outstanding performance of the proposed algorithm has been tested with the test images where more than 10 2D objects arbitrarily located on a table.

• Proceedings of the IEEK Conference
• /
• 1999.06a
• /
• pp.615-618
• /
• 1999

In this paper, a object detection and tracking algorithm is presented which exhibits robust properties for image sequences with complex background. The proposed algorithm is composed of three parts: moving object detection, object tracking, and motion analysis. The moving object detection algorithm is implemented using a temporal median background method which is suitable for real-time applications. In the motion analysis, we propose a new technique for removing a temporal clutter, such as a swaying plant or a light reflection of a background object. In addition, we design a multiple vehicle tracking system based on Kalman filtering. Computer simulation of the proposed scheme shows its robustness for MPEG-7 test image sequences.

• Journal of the Korean Institute of Telematics and Electronics B
• /
• v.31B no.8
• /
• pp.87-98
• /
• 1994

Object-oriented analysis-synthesis coding subdivides each image of a sequence into moving objects and compensates the motion of each object. Thus it can reconstruct real motion better than conventional motion-compensated coding techniques at very-low-bit-rates. It uses a mapping parameter technique for estimating motion information of each object. Since a mapping parameter technique uses gradient operators it is sensitive to redundant details and noise. To accurately determine mapping parameters, we propose a new analysis method using integral projections for estimation of gradient values. Also to reconstruct correctly the local motion the proposed algorithm divides an image into segmented objects each of which having uniform motion information while the conventional one assumes a large object having the same motion information. Computer simulation results with several test sequences show that the proposed image analysis method in object-oriented analysis-synthesis coding shows better performance than the conventional one.

• ETRI Journal
• /
• v.17 no.3
• /
• pp.45-63
• /
• 1995

This paper proposes an algorithm to improve the correlation between giga-hertz transverse electromagnetic (GTEM) cell and open area test site (OATS). It is based on the dipole modeling process of an unknown source object in a GTEM cell and on the evaluation of the approximate far field equations correlated with measured GTEM powers at output port of the GTEM cell. In this algorithm, the relative phase differences between dipole moments play an important part in modeling the test object as a set of dipoles and offer stable calculation of emission values. The radiated emission test using this algorithm requires fifteen orientations of equipment under test, but the increased orientations as compared with the previous method have little effect on the time needed for testing. Radiation from a notebook computer has been tested for statistical analysis of the correlation between GTEM data and OATS data. The emission test results of the notebook computer show that the mean, the standard deviation, and the correlation coefficient are -0.62, 1.99, and +0.85, respectively. These figures indicate that this algorithm provides improved accuracy in the measurement of electromagnetic emissions over the previous method.