• Proceedings of the IEEK Conference
• /
• 2001.06d
• /
• pp.211-214
• /
• 2001

This paper presents an efficient fast motion estimation algorithm and image segmentation method for low bit-rate coding. First, with region split information, the algorithm splits the image having homogeneous and semantic regions like face and semantic regions in image. Then, in these regions, We find the motion vector using adaptive search window adjustment. Additionally, with this new segment based fast motion estimation, we reduce blocking artifacts by intensively coding our interesting region(face or arm) in input image. The simulation results show the improvement in coding performance and image quality.

• Nuclear Engineering and Technology
• /
• v.33 no.5
• /
• pp.514-525
• /
• 2001

Optimum characteristics of digital X-ray sensor components were analyzed to develop intra- oral dental digital X-ray image sensor using indirect method. Parametric analysis was carried out to optimize the phosphor thickness and the fiber optic plate (FOP) coupling to charge coupled device (CCD). X-ray absorption and light diffusion in the phosphor layer were analyzed by the Monte Carlo method. Real time X-ray image was obtained with prototype X- ray image sensor using general CCD camera with 1∼10 Ip/mm resolution. It has been previously shown that large resolution degradation in X-ray images was caused by miss alignment of FOP to CCD and optical adhesive selection. In this study, we reported that X-ray image quality was greatly improved by using optimized characteristics of alignment device and phosphor thickness.

• Nuclear Engineering and Technology
• /
• v.52 no.11
• /
• pp.2594-2600
• /
• 2020

A gamma camera system using radionuclide has a functional imaging technique and is frequently used in the field of nuclear medicine. In the gamma camera, it is extremely important to improve the image quality to ensure accurate detection of diseases. In this study, we designed a blind-deconvolution framework after a noise-reduction algorithm based on a non-local mean, which has been shown to outperform conventional methodologies with regard to the gamma camera system. For this purpose, we performed a simulation using the Monte Carlo method and conducted an experiment. The image performance was evaluated by visual assessment and according to the intensity profile, and a quantitative evaluation using a normalized noise-power spectrum was performed on the acquired image and the blind-deconvolution image after noise reduction. The result indicates an improvement in image performance for gamma camera images when our proposed algorithm is used.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 1996.06a
• /
• pp.233-236
• /
• 1996

A new method of blocking effects reduction is proposed in this paper for use in low bitrate image coding. We use 28 DCT kernel functions of which boundary values are linearly independent, and Gram-Schmidt process is applied to the boundary values in order to obtain 28 boundary-orthonormal basis images. Then we use these basis images to obtain the correction terms for blocking artifacts reduction. A threshold of block discontinuity is introduced for improvement of visual quality by reducing image blurring. We also investigate the number of basis images needed for efficient blocking artifacts reduction when the compression ratio changes.

• Proceedings of the IEEK Conference
• /
• 2007.07a
• /
• pp.209-210
• /
• 2007

Obtained disparity map from the stereo camera by using the several stereo matching algorithms carries lots of noise because of various causes. In our approach, mode filtering and noise elimination technique using the histogram and projection-based region merging methods are adopted for improving the quality of disparity map and image segmentation. The proposed algorithms are implemented in VHDL and the real-time experimentation shows the accurately divided objects.

• Journal of the Institute of Electronics Engineers of Korea SP
• /
• v.41 no.4
• /
• pp.87-92
• /
• 2004

Digital halftoning is to quantize a grayscale image to binary image. The error diffusion halftoning generates high quality bilevel image. But that also has some defects such as warms effect, sharpening and etc. To reduce these defects, Kite proposed the modified threshold modulation that has a parameter to control sharpening. Nevertheless some degradation left near edges with large luminance change. In this paver, we propose a method to control the parameter in proportional to local edge magnitude. The results of computer simulation show more reductions of the sharpening in the halftone image. Especially there are great improvement of quality near edges with large luminance change.

• Journal of Digital Convergence
• /
• v.16 no.4
• /
• pp.149-155
• /
• 2018

It has passed the era of Full High Definition Television (FHD) and ended the era of Ultra High Definition Television (UHD). We will talk about the problems caused by the difference in the number of pixels in the TVs and introduce a method to improve them. This method is a method of performing an interpolation method suitable for a given image with Weighted Essential Non-Oscillation (WENO). Thus, it is possible to reduce the distortion of the image and to ensure a better image quality. Therefore, if we use the WENO methodology, when we watch old video material on UHD TV, we can enjoy high definition of UHD TV without changing the number of pixels.

• The Journal of Korean Institute of Electromagnetic Engineering and Science
• /
• v.26 no.11
• /
• pp.1027-1030
• /
• 2015

In the airborne SAR, the motion errors induced by atmospheric turbulence decrease the resolution and increase the sidelobes. If the sidelobes are not properly compensated, the image quality is degraded. Thus, in this paper, we have introduced the sidelobe reduction method to increase the image quality. After we calculate the scene center based on the estimated squint angle for the flight path partitioned by the subaperture technique, we perform the motion compensation for the scene center. Then, after we perform the recursive sidelobe reduction for the region of interest in the reconstructed SAR image, we extend it for the full image.

• Korean Journal of Digital Imaging in Medicine
• /
• v.12 no.2
• /
• pp.139-143
• /
• 2010

Two-dimensional(2D) images like Multi Planar Reconstruction(MPR) Image or Maximum Intensity Projection(MIP) were used for the purpose of diagnosis, but MPR image's quality were limited due to its superior limit of Z-axis ability to produce permitted radiation exposure virtuous in the permitted time limit from the existing Spiral CT. However, in company with the development of the Multi Detector Computed Tomography(MDCT), we were able to get the Data with the equal amount of Voxel, also get varied reconstructions as in the aspect of our needs. This present study propose a reconstruction technique which is to extract a field using Region of interest(ROI) segmentation method for improvement of the quality of the medical image and after that reconstruct the concerned part using the four-directed symmetry method of the oval, than using the reconstructed data, reorganize the image by using the Curved MPR method. If current proposed method is used, it is highly effective because of its ability to accurately display the disease concerned part, which will reduce the decoding time and also effectively provide information based on the accuracy of the decode.

• Journal of Korean Society for Quality Management
• /
• v.51 no.1
• /
• pp.119-129
• /
• 2023

Purpose: This study proposes a classification model for implementing condition-based maintenance (CBM) by monitoring the real-time status of a machine using acceleration sensor data collected from a vehicle. Methods: The classification model's performance was improved by applying Fourier transform to convert the acceleration sensor data from the time domain to the frequency domain. Additionally, the Generative Adversarial Network (GAN) algorithm was used to augment images and further enhance the classification model's performance. Results: Experimental results demonstrate that the GAN algorithm can effectively serve as an image augmentation technique to enhance the performance of the classification model. Consequently, the proposed approach yielded a significant improvement in the classification model's accuracy. Conclusion: While this study focused on the effectiveness of the GAN algorithm as an image augmentation method, further research is necessary to compare its performance with other image augmentation techniques. Additionally, it is essential to consider the potential for performance degradation due to class imbalance and conduct follow-up studies to address this issue.