• Journal of Advanced Navigation Technology
• /
• v.14 no.2
• /
• pp.284-288
• /
• 2010

Video surveillance is widely used in establishing the societal security network. In this paper, intrusion detection based on visual information acquired by static camera is proposed. Proposed approach uses background model constructed by approximated median filter(AMF) to find a foreground candidate, and detected object is calculated by analyzing motion information. Motion detection is determined by the relative size of 2D object in RGB space, finally, the threshold value for detecting object is determined by heuristic method. Experimental results showed that the performance of intrusion detection is better one when the spatio-temporal candidate informations change abruptly.

• The Transactions of the Korea Information Processing Society
• /
• v.4 no.5
• /
• pp.1257-1266
• /
• 1997

The concepts of class, inheritance and information hicing and so on provide the great strengthes of object-oriented languages, but they also introduce diffculties in porfram analysis and understanding. Particulary, it is move difficult to umderstand the dyamic aspects than the static ones of object-oriented programs. The dyamicaspects can be understood by recognizing the event's reciprocal action among the classes. In this paper, it will be supplied to the reprecentation of event abstraction which is useful for understanding the object-oriented programs.And the clustering concept with the events will be applied to abstract the events. By clustering the events, user can get the information about function of the classes and the reteival of the class library.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.14 no.12
• /
• pp.4776-4794
• /
• 2020

We propose a rapid adaptive learning framework for streaming object detection, called EER-ASSL. The method combines the expected error reduction (EER) dependent rollback learning and the active semi-supervised learning (ASSL) for a rapid adaptive CNN detector. Most CNN object detectors are built on the assumption of static data distribution. However, images are often noisy and biased, and the data distribution is imbalanced in a real world environment. The proposed method consists of collaborative sampling and EER-ASSL. The EER-ASSL utilizes the active learning (AL) and rollback based semi-supervised learning (SSL). The AL allows us to select more informative and representative samples measuring uncertainty and diversity. The SSL divides the selected streaming image samples into the bins and each bin repeatedly transfers the discriminative knowledge of the EER and CNN models to the next bin until convergence and incorporation with the EER rollback learning algorithm is achieved. The EER models provide a rapid short-term myopic adaptation and the CNN models an incremental long-term performance improvement. EER-ASSL can overcome noisy and biased labels in varying data distribution. Extensive experiments shows that EER-ASSL obtained 70.9 mAP compared to state-of-the-art technology such as Faster RCNN, SSD300, and YOLOv2.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.6
• /
• pp.793-799
• /
• 2004

Many researches and validations for the complexity metrics of the object-oriented systems have been studied. Most of them are aimed for the measurement of the partial aspects of the systems, for example, the coupling between objects, the complexity of inheritance structures, the cohesion of methods, and so on. But the software practitioners want to measure the complexity of overall system, not partial. We studied the complexity of the overall structures of object-oriented systems by analyzing the class diagram of UML. The class diagram is composed of classes and their relations. There are three kinds of relations, association, generalization, and aggregation, which are making the structure of object-oriented systems to be difficult to understand. We proposed a heuristic metric to measure the complexity of object-oriented systems by putting together the three kinds of the relations. This metric will be helpful to the software developers for their designing tasks by evaluating the complexity of the structures of object-oriented system and redesigning tasks of the system.

• The Journal of the Convergence on Culture Technology
• /
• v.4 no.1
• /
• pp.291-297
• /
• 2018

In this paper, we present an automatic image object tracking system for Studio cameras on the stage. For object tracking, we use the OpenCV-based Python program using PC, Raspberry Pi 3 and mobile devices. There are many methods of image object tracking such as mean-shift, CAMshift (Continuously Adaptive Mean shift), background modelling using GMM(Gaussian mixture model), template based detection using SURF(Speeded up robust features), CMT(Consensus-based Matching and Tracking) and TLD methods. CAMshift algorithm is very efficient for real-time tracking because of its fast and robust performance. However, in this paper, we implement an image object tracking system for studio cameras based CMT algorithm. This is an optimal image tracking method because of combination of static and adaptive correspondences. The proposed system can be applied to an effective and robust image tracking system for continuous object tracking on the stage in real time.

• Journal of KIISE:Software and Applications
• /
• v.29 no.1_2
• /
• pp.114-125
• /
• 2002

In this paper, we are proposing a hierarchical segmentation method that first segments the video data into units of shots by detecting cut and dissolve, and then decides types of camera operations or object movements in each shot. In our previous work[1], each picture group is divided into one of the three detailed categories, Shot(in case of scene change), Move(in case of camera operation or object movement) and Static(in case of almost no change between images), by analysing DC(Direct Current) component of I(Intra) frame. In this process, we have designed two-stage hierarchical neural network with inputs of various multiple features combined. Then, the system detects the accurate shot position, types of camera operations or object movements by searching P(Predicted), B(Bi-directional) frames of the current picture group selectively and hierarchically. Also, the statistical distributions of macro block types in P or B frames are used for the accurate detection of cut position, and another neural network with inputs of macro block types and motion vectors method can reduce the processing time by using only DC coefficients of I frames without decoding and by searching P, B frames selectively and hierarchically. The proposed method classified the picture groups in the accuracy of 93.9-100.0% and the cuts in the accuracy of 96.1-100.0% with three different together is used to detect dissolve, types of camera operations and object movements. The proposed types of video data. Also, it classified the types of camera movements or object movements in the accuracy of 90.13% and 89.28% with two different types of video data.

• Journal of information and communication convergence engineering
• /
• v.5 no.2
• /
• pp.83-87
• /
• 2007

This research studies real time interaction and dynamic data shared through 3D scenes in virtual network environments. In a distributed virtual environment of client-server structure, consistency is maintained by the static information exchange; as jerks occur by packet delay when updating messages of dynamic data exchanges are broadcasted disorderly, the network bottleneck is reduced by predicting the movement path by using the Dead-reckoning algorithm. In Dead-reckoning path prediction, the error between the estimated and the actual static values which is over the threshold based on the shared object location requires interpolation and multicasting of the previous location by the ESPDU of DIS. The shared dynamic data of the 3D virtual environment is implementation using the VRML.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.11
• /
• pp.2989-3000
• /
• 1999

Until now, most Intelligent Tutoring Systems are lacking in the modularity, the extensibility of the system, and the flexibility in the dynamic environment due to the static exchanges of knowledge among modules. To overcome these flexibility in the dynamic due to the static exchanges of knowledge among modules. To overcome these problems, we will suggest, in this paper, a Distance Intelligent Tutoring Framework, called DELFOM, based on the multiagent to cope with the various and complicated learner's requests. We could make different types of learning systems by simply changing the contents of DELFOM External that is variant part of DELFOM. This framework, therefore, provides software reuse and the extensibility based on object-oriented paradigm. And we will propose two different distance learning systems using DELFOM. Therefore this framework gives the developer/the learner the effective and easy development/learning environment. DELFOM is implemented using CORBA and Java for the network transparency and platform independence.

• Proceedings of the Korea Concrete Institute Conference
• /
• 2001.11a
• /
• pp.965-970
• /
• 2001

The object of this research is to evaluate the seismic performance of existing RC bridge piers that were constructed before the adoption of the seismic design provision of Korea Bridge Design Specification in 1992. In this research, adopted test parameters were limited ductile design or non-seismic design, aspect ratio, confinement steel type, loading pattern, lap-spliced ratio for longitudinal reinforcement. This study has been performed to verify the effect of test parameter by quasi-static test. Quasi-static test has been done to investigate the physical seismic performance of RC bridge piers, such as lateral force-displacement hysteretic curve, envelope curve etc. It has been observed that seismic performance of lap-spliced test specimen, non-seismically designed specimens, was significantly reduced.

• Tribology and Lubricants
• /
• v.19 no.5
• /
• pp.259-267
• /
• 2003

A brake device for the high-speed impacting object is designed using an axial crushing of thin-walled metal cylinder. Thickness of the cylinder is increased smoothly from the impacting end to the fixed end, resulting in the truncated cone shape. Truncated cone shape minimizes the imperfection-sensitivity of the structure and ensures that plastic hinges are formed sequentially from impacting end. This prevents the undesirable sudden rise in the first peak-crushing load. Several specimens with different conic angles, mean thickness of the wall, and materials were designed and quasi-static compression tests were performed on them. Results indicate that adoption of appropriate conic angle prevents simultaneous wrinkles generation and sudden rise of crushing load and that appropriate conic angle differs in each case, depending on the geometry and material property of the cylinder. Finite element analysis was performed for static compression of the cylinder and its accuracy was checked for the future application.