• International journal of advanced smart convergence
• /
• v.11 no.2
• /
• pp.22-29
• /
• 2022

In this paper, for the purpose of designing an real-time unmanned monitoring system, the YOLOv5s (small) object detection model was applied on the NVIDIA TX2^TM AI (Artificial Intelligence) edge computing platform in order to design the fundamental function of an unmanned monitoring system that can detect objects in real time. YOLOv5s was applied to the our real-time unmanned monitoring system based on the performance evaluation of object detection algorithms (for example, R-CNN, SSD, RetinaNet, and YOLOv5). In addition, the performance of the four YOLOv5 models (small, medium, large, and xlarge) was compared and evaluated. Furthermore, based on these results, the YOLOv5s model suitable for the design purpose of this paper was ported to the NVIDIA TX2^TM AI edge computing system and it was confirmed that it operates normally. The real-time unmanned monitoring system designed as a result of the research can be applied to various application fields such as an security or monitoring system. Future research is to apply NMS (Non-Maximum Suppression) modification, model reconstruction, and parallel processing programming techniques using CUDA (Compute Unified Device Architecture) for the improvement of object detection speed and performance.

• The Transactions of the Korea Information Processing Society
• /
• v.2 no.1
• /
• pp.1-11
• /
• 1995

Object-oriented database systems represent the complex data model of real- world and manage the complex data. An object manager handles the manipulation and access of all objects in object-oriented database management systems. In this paper, we not only propose the design rules of an object manager interface(OMI) which is specified the calling function and the semantics of an object manager for object-oriented query processing, but also implement the OMI based on the rules. The OMI implemented supports the independently developing environment of query processing and object management modules, and can be applicable to various applications, and also provides the efficient access method to users.

• The Journal of Korea Robotics Society
• /
• v.7 no.1
• /
• pp.9-19
• /
• 2012

This paper proposes a performance index for the multiple shape object handling of dual arm manipulator to determine whether a robot is good or not. When the dual-arm manipulator grasps a fixed object and is posed, the dual-arm manipulator should procure a space to freely control the manipulator. As a performance evaluation parameter, each joint torque from current sensor signal is utilized. From the current information, torque and energy for each joint are estimated. In this paper an performance index for an unstructured object is defined by an energy-cost function, and stability analysis for each motion is derived by the maximum force to the object. The maximum force to the object is computed by the inertia of object and acceleration information of end-effector. The acceleration data are derived by the double derivation of each encoder signal. Manipulability measure which implies how efficiently the dual-arm manipulator can move with the grasped object, can be represented by the intersection of the two manipulability ellipsoids for the left and right arms. Effectiveness of the proposed algorithm has been verified through the practical simulations and real experiments.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.7 no.12
• /
• pp.3180-3199
• /
• 2013

While non-predefined object segmentation (NDOS) distinguishes an arbitrary self-assumed object from its background, predefined object segmentation (DOS) pre-specifies the target object. In this paper, a new and novel method to segment predefined objects is presented, by globally optimizing an orientation-based objective function that measures the fitness of the object boundary, in a discretized parameter space. A specific object is explicitly described by normalized discrete sets of boundary points and corresponding normal vectors with respect to its plane shape. The orientation factor provides robust distinctness for target objects. By considering the order of transformation elements, and their dependency on the derived over-segmentation outcome, the domain of translations and scales is efficiently discretized. A branch and bound algorithm is used to determine the transformation parameters of a shape model corresponding to a target object in an image. The results tested on the PASCAL dataset show a considerable achievement in solving complex backgrounds and unclear boundary images.

• The Transactions of the Korea Information Processing Society
• /
• v.2 no.3
• /
• pp.299-313
• /
• 1995

Object-Oriented design method decomposes the program into object that establish the relationship between objects. It is the technique of object centered decomposition unlike function oriented legacy system. Since existing Object- Oriented design method doesn't fully explain of the design method of module and decomposition, and need to implement the Object-Oriented software as a total system init. It has many problems that whole object design method and application programs should be analyzed for the understanding an Object-Oriented software component. Therefore, this study provides a procedure for Object-Oriented module design and criteria of decomposition to practically support the complex and large system development. Futhermore, it shows a consistency between design model and implementation through a case study applying these module decomposition criteria to the real system.

• The Mathematical Education
• /
• v.56 no.4
• /
• pp.387-405
• /
• 2017

The purpose of this study is to investigate the change of students 'perception and expression about the motion of object following distance function $={x \atop 3}$ and distance function $y=\frac{x^3}{3}+3$ according to the necessity of research on students' perception and expression about integral constant. In this paper, we present the recognition and the expression of the difference of the constant in the relationship between the distance function and the speed function of the students, while examining the process of constructing the speed function and the inverse process of the distance function. This provides implications for the relationship between the derivative and the indefinite integral corresponding to the inverse process. In particular, in a teaching experiment, a constructive activity was performed to analyze the motion of two distance functions, where the student had a difference of the constant term. At this time, the students used the expression 'starting point' for the constants in the distance function, and the motion was interpreted by using the meaning. This can be seen as a unique 'students' mathematics' in the process of analyzing the motion of objects. These scenes, in introducing the notion of the relation between differential and indefinite integral, it is beyond the comprehension of the integral constant as a computational procedure, so that the learner can understand the meaning of the integral constant in relation to the motion of the object. It is expected that it will be a meaningful basic research on the relationship between differential and integral.

• The KSFM Journal of Fluid Machinery
• /
• v.10 no.2 s.41
• /
• pp.22-31
• /
• 2007

This paper describes the shape optimization of rotor blade in a transonic axial compressor rotor. Three surrogate models, Kriging, radial basis neural network and response surface methods, are introduced to find optimum blade shape and to compare the characteristics of object function at each optimal design condition. Blade sweep, lean and skew are considered as design variables and adiabatic efficiency is selected as an objective function. Throughout the shape optimization of the compressor rotor, the predicted adiabatic efficiency has almost same value for three surrogate models. Among the three design variables, a blade sweep is the most sensitive on the object function. It is noted that the blade swept to backward and skewed to the blade pressure side is more effective to increase the adiabatic efficiency in the axial compressor Flow characteristics of an optimum blade are also compared with the results of reference blade.

• Proceedings of the Korean Society of Precision Engineering Conference
• /
• 2003.06a
• /
• pp.1471-1474
• /
• 2003

The correlation between the object function and the design parameter is shown on this paper by using the characteristic function for tile mixed result of the structure analysis. tile buckling analysis and the table of orthogonal array according to the original crane's dimensional change. About the above two object functions, the effective of design change according to the change of design parameters could be estimated. Also, the crane's weight is reduced up to 20.58 percent maintaining the structural stability according to the thickness of plate.

• Journal of the Korean Society of Radiology
• /
• v.18 no.1
• /
• pp.21-26
• /
• 2024

The one of the famous deep learning models for object detection task is you only look once version 5 (YOLOv5) framework based on the one stage architecture. In addition, YOLOv5 model indicated high performance for accurate lesion detection using the bottleneck CSP layer and skip connection function. The purpose of this study was to evaluate the performance of YOLOv5 framework according to various hyperparameters in position emission tomogrpahy (PET) phantom images. The dataset was obtained from QIN PET segmentation challenge in 500 slices. We set the bounding box to generate ground truth dataset using labelImg software. The hyperparameters for network train were applied by changing optimization function (SDG, Adam, and AdamW), activation function (SiLU, LeakyRelu, Mish, and Hardwish), and YOLOv5 model size (nano, small, large, and xlarge). The intersection over union (IOU) method was used for performance evaluation. As a results, the condition of outstanding performance is to apply AdamW, Hardwish, and nano size for optimization function, activation function and model version, respectively. In conclusion, we confirmed the usefulness of YOLOv5 network for object detection performance in nuclear medicine images.

• Bulletin of the Korean Mathematical Society
• /
• v.47 no.6
• /
• pp.1213-1224
• /
• 2010

Using Ahlfors' covering surface method, some properties on the strong Borel direction of algebroid function of finite order are obtained. The main object of this paper is to prove existence theorem of a strong Borel direction and the existence of filling discs in such direction which briefly extends some results of meromorphic function.