• Journal of Computing Science and Engineering
• /
• 제6권3호
• /
• pp.219-226
• /
• 2012

Recently, shape analysis of human organs has achieved much attention, owing to its potential to localize structural abnormalities. For a group-wise shape analysis, it is important to accurately restore the shape of a target structure in each subject and to build the inter-subject shape correspondences. To accomplish this, we propose a shape modeling method based on the Laplacian deformation framework. We deform a template model of a target structure in the segmented images while restoring subject-specific shape features by using Laplacian surface representation. In order to build the inter-subject shape correspondences, we implemented the progressive weighting scheme for adaptively controlling the rigidity parameter of the deformable model. This weighting scheme helps to preserve the relative distance between each point in the template model as much as possible during model deformation. This area-preserving deformation allows each point of the template model to be located at an anatomically consistent position in the target structure. Another advantage of our method is its application to human organs of non-spherical topology. We present the experiments for evaluating the robustness of shape modeling against large variations in shape and size with the synthetic sets of the second cervical vertebrae (C2), which has a complex shape with holes.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2009년도 IWAIT
• /
• pp.136-141
• /
• 2009

This paper proposed a framework for handover method in continuously tracking a person of interest across cooperative pan-tilt-zoom (PTZ) cameras. The algorithm here is based on a robust non-parametric technique for climbing density gradients to find the peak of probability distributions called the mean shift algorithm. Most tracking algorithms use only one cue (such as color). The color features are not always discriminative enough for target localization because illumination or viewpoints tend to change. Moreover the background may be of a color similar to that of the target. In our proposed system, the continuous person tracking across cooperative PTZ cameras by mean shift tracking that using color and shape histogram to be feature distributions. Color and shape distributions of interested person are used to register the target person across cameras. For the first camera, we select interested person for tracking using skin color, cloth color and boundary of body. To handover tracking process between two cameras, the second camera receives color and shape cues of a target person from the first camera and using linear color calibration to help with handover process. Our experimental results demonstrate color and shape feature in mean shift algorithm is capable for continuously and accurately track the target person across cameras.

• 대한원격탐사학회:학술대회논문집
• /
• 대한원격탐사학회 2006년도 춘계학술대회 논문집
• /
• pp.7-14
• /
• 2006

Detection and identification of targets from remotely sensed imagery are of great interest for civilian and military application. This paper presents an algorithm for target detection in high spatial resolution imagery based on the spectral and the dimensional characteristics of the reference target. In this algorithm, the spectral and the dimensional information of the reference target is extracted automatically from the sample image of the reference target. Then in the entire image, the candidate target pixels are extracted based on the spectral characteristics of the reference target. Finally, groups of candidate pixels which form isolated spatial objects of similar size to that of the reference target are extracted as detected targets. The experimental test results showed that even though the algorithm detected spatial objects which has different shape as targets if the spectral and the dimensional characteristics are similar to that of the reference target, it could detect 97.5% of the targets in the image. Using hyperspectral image and utilizing the shape information are expected to increase the performance of the proposed algorithm.

• 대한원격탐사학회지
• /
• 제22권1호
• /
• pp.41-47
• /
• 2006

Detection and identification of targets from remotely sensed imagery are of great interest for civilian and military application. This paper presents an algorithm for target detection in high spatial resolution imagery based on the spectral and the dimensional characteristics of the reference target. In this algorithm, the spectral and the dimensional information of the reference target is extracted automatically from the sample image of the reference target. Then in the entire image, the candidate target pixels are extracted based on the spectral characteristics of the reference target. Finally, groups of candidate pixels which form isolated spatial objects of similar size to that of the reference target are extracted as detected targets. The experimental test results showed that even though the algorithm detected spatial objects which has different shape as targets if the spectral and the dimensional characteristics are similar to that of the reference target, it could detect 97.5% of the targets in the image. Using hyperspectral image and utilizing the shape information are expected to increase the performance of the proposed algorithm.

• 한국군사과학기술학회지
• /
• 제20권1호
• /
• pp.40-47
• /
• 2017

To classify the vehicle targets automatically using thermal images, there are usually two main categories of feature extraction method, local and shape feature extraction methods. Since thermal images have less texture information than color images, the shape feature extraction method is useful when the segmentation results are correct. However, if there are some errors in target segmentation, the shape feature may contain some errors, then the classification accuracy can be decreased. To overcome these problems, in this paper, we propose the segmentation reliability estimation method for target classification. The segmentation reliability can be estimated by using the difference information of average intensities and edge energies between the target and the background area. The estimated segmentation reliability is applied in the decision level fusion method of classification results using both shape and local features. Experiment results using the thermal images of the vehicle targets (main battle tank, armored personnel carrier, military truck, and an estate car) show that the proposed classification method and the segmentation reliability estimation method have a good performance in classification accuracy.

• 한국정밀공학회지
• /
• 제25권1호
• /
• pp.160-170
• /
• 2008

Shape morphing is the process of transforming a source shape, through intermediate shapes, into a target shape. Two main problems to be considered in three dimensional shape morphing are vertex correspondence and path interpolation. In this paper, an approach which uses the linear interpolation of the Laplacian coordinates of the source and target meshes is introduced for the determination of more plausible path when two topologically identical shapes are morphed. When two shapes to be morphed are different in shape and topology, a new method which combines shape deformation theory based on Laplacian coordinate and mean value coordinate with distance field theory is proposed for the efficient treatment of vertex correspondence and path interpolation problems. The validity and effectiveness of the suggested method was demonstrated by using it to morph large and complex polygon models including male and female whole body models.

• 제어로봇시스템학회:학술대회논문집
• /
• 제어로봇시스템학회 1996년도 Proceedings of the Korea Automatic Control Conference, 11th (KACC); Pohang, Korea; 24-26 Oct. 1996
• /
• pp.12-15
• /
• 1996

In this paper, a new shape control law is derived as a result of introducing the parametric curve representation. This control alw is based on the estimation of the curve parameters corresponding to the target joint positions and the target tip position. Estimating target curve parameters makes it possible to find, easily, a simple shape control law by the Lyapunov design method.

• 한국군사과학기술학회지
• /
• 제26권3호
• /
• pp.246-251
• /
• 2023

RADAR(Radio Detection and Ranging) is an important system for surveillance and reconnaissance by detecting a reflected signal which obtains the range from the radar to the target, and the velocity of the target. The magnitude of the reflected signal varies due to the radar cross section of the target, characteristic of the transmission and reception antenna, distance between the radar and the target, and power and wavelength of the transmitted signal. Thus, the RCS is the important characteristic of the target to determine if the target can be observed by the RADAR system. It is based on the material and shape of the target. We have measured the reflection signal of a simple square-shaped (20 × 20 cm) target made of a new material, a gallium-based liquid metal alloy and compared that of well-known metals including copper, aluminum. The magnitude of reflected signal of the aluminum target was the largest and it was 2.4 times larger than that of the liquid metal target. We also investigated the effect of the shape by measuring reflectance of the F-22 3D model(~1/95 ratio) target covered with/without copper, aluminium, and liquid metal. The largest magnitude of the reflected signal measured from side-view with the copper-covered F-22 model was 2.6 times greater than that of liquid metal. The reflectance study of the liquid metal would be helpful for liquid metal-based frequency selective surface or metamaterials.

• 소성∙가공
• /
• 제24권1호
• /
• pp.28-36
• /
• 2015

After a short review of the iterative optimal blank method, a new method of measuring the shape error for stamped parts with 3D contour lines, which is an essential component of the optimal blank design, is proposed. When the contour line of the target shape does not exist in a plane, but exists in 3D space, especially when the shape of the target contour line is very complicated as in the real automotive parts, then the measurement of the shape error is critical. In the current study, a method of shape error measurement based on the minimum distance is suggested as an evolution of the radius vector method. With the proposed method, the optimal blank shapes of real automotive parts were found and compared to the results of the radius vector method. From the current investigation the new method is found to resolve the issues with the radius vector method.

• 대한전자공학회:학술대회논문집
• /
• 대한전자공학회 2002년도 하계종합학술대회 논문집(3)
• /
• pp.29-32
• /
• 2002

This paper suggest target detection algorithm for mobile robot control using color and shape recognition. In many cases, ultrasonic sensor(USS) is used in mobile robot system to measure the distance between obstacles. But with only USS, it may have many restrictions. So we attached CCD camera to mobile robot to overcome its restrictions. If visual information is given to robot system then robot system will be able to accomplish more complex mission successfully. With acquired vision data, robot looks for target by color and recognize its shape.