• Journal of the Korean Society of Radiology
• /
• v.9 no.3
• /
• pp.139-145
• /
• 2015

When examining a patient using SPECT, gamma rays emitted from the body decrease or scatter. And when they reach the detector they spread in accordance with physical characteristics and geometric shapes of the scanner, quantitative analysis was difficult. For exact quantitative analysis of gamma rays emitted from the body, so that they must be considered to correction about PVE(partial volume effect). In this paper, sinogram filter was implemented to solve comprehensive PVE of SPECT. According to the results in which implemented filter was applied, partial volume effect caused by SPECT was removed. To compare proposed method and conventional method, PSNR(Peak Signal to Noise Ratio) was executed. As a result, proposed method was indicated as 7dB, conventional method was indicated as 14db respectively. dB(decibel) level of the proposed methods is lower, since the MSE(mean square error) becomes greater because scattered ray was removed, PSNR value is low. Therefore, by applying the proposed method for removing the PVE of SPECT imaging method, the image quality is improved.

• Journal of radiological science and technology
• /
• v.44 no.6
• /
• pp.655-661
• /
• 2021

Positron emission tomography (PET) images is affected by acquisition time, short acquisition times results in low gamma counts leading to degradation of image quality by statistical noise. Noise2Void(N2V) is self supervised denoising model that is convolutional neural network (CNN) based deep learning. The purpose of this study is to evaluate denoising performance of N2V for PET image with a short acquisition time. The phantom was scanned as a list mode for 10 min using Biograph mCT40 of PET/CT (Siemens Healthcare, Erlangen, Germany). We compared PET images using NEMA image-quality phantom for standard acquisition time (10 min), short acquisition time (2min) and simulated PET image (S2 min). To evaluate performance of N2V, the peak signal to noise ratio (PSNR), normalized root mean square error (NRMSE), structural similarity index (SSIM) and radio-activity recovery coefficient (RC) were used. The PSNR, NRMSE and SSIM for 2 min and S2 min PET images compared to 10min PET image were 30.983, 33.936, 9.954, 7.609 and 0.916, 0.934 respectively. The RC for spheres with S2 min PET image also met European Association of Nuclear Medicine Research Ltd. (EARL) FDG PET accreditation program. We confirmed generated S2 min PET image from N2V deep learning showed improvement results compared to 2 min PET image and The PET images on visual analysis were also comparable between 10 min and S2 min PET images. In conclusion, noisy PET image by means of short acquisition time using N2V denoising network model can be improved image quality without underestimation of radioactivity.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.43 no.1 s.343
• /
• pp.65-70
• /
• 2006

An image sensor has limited dynamic range while the human eye has logarithmic response over wide range of light intensity. Although the sensor gain can be set high to identify details in darker area on the image, this results in saturation in brighter area. The gamma correction is essential to fit the human eye response. However, the digital gamma correction degrades image quality especially for darker area on the image due to the limited ADC resolution and the dynamic range. This Paper proposes a CMOS image sensor (CIS) with a nonlinear analog-to-digital converter (AU) which performs analog gamma correction. The CIS with the proposed nonlinear analog-to-digital conversion scheme was fabricated with a $0.35{\mu}m$ CMOS process. The analog gamma correction using the proposed nonlinear ADC CIS provides the 2.2dB peak-signal-to-noise-ratio(PSM) improved image qualify than conventional digital gamma correction. The PSNR of the image obtain from the digital gamma correction is 25.6dB while it is 27.8dB for analog gamma correction. The PSNR improvement over digital gamma correction is about $28.8\%$.

• The Journal of the Acoustical Society of Korea
• /
• v.37 no.1
• /
• pp.12-20
• /
• 2018

Efforts have been made to reconstruct low-resolution underwater images to high-resolution ones by using the image SR (Super-Resolution) method, all to improve efficiency when acquiring side-scan sonar images. As side-scan sonar images are similar with the optical images with respect to exploiting 2-dimensional signals, conventional image restoration methods for optical images can be considered as a solution. One of the most typical super-resolution methods for optical image is a sparse coding and there are studies for verifying applicability of sparse coding method for underwater images by analyzing sparsity of underwater images. Sparse coding is a method that obtains recovered signal from input signal by linear combination of dictionary and sparse coefficients. However, it requires huge computational load to accurately estimate sparse coefficients. In this study, a sparse coding based underwater image super-resolution method is applied while a selective reconstruction method for object region is suggested to reduce the processing time. For this method, this paper proposes an edge detection and object and non object region classification method for underwater images and combine it with sparse coding based image super-resolution method. Effectiveness of the proposed method is verified by reducing the processing time for image reconstruction over 32 % while preserving same level of PSNR (Peak Signal-to-Noise Ratio) compared with conventional method.

• Journal of the Korean Society of Radiology
• /
• v.18 no.2
• /
• pp.93-99
• /
• 2024

The digital medical imaging, especially, computed tomography (CT), should necessarily be considered in terms of noise distribution caused by converting to X-ray photon to digital imaging signal. Recently, the denoising technique based on deep learning architecture is increasingly used in the medical imaging field. Here, we evaluated noise reduction effect according to various noise types based on the U-net deep learning model in the lung CT images. The input data for deep learning was generated by applying Gaussian noise, Poisson noise, salt and pepper noise and speckle noise from the ground truth (GT) image. In particular, two types of Gaussian noise input data were applied with standard deviation values of 30 and 50. There are applied hyper-parameters, which were Adam as optimizer function, 100 as epochs, and 0.0001 as learning rate, respectively. To analyze the quantitative values, the mean square error (MSE), the peak signal to noise ratio (PSNR) and coefficient of variation (COV) were calculated. According to the results, it was confirmed that the U-net model was effective for noise reduction all of the set conditions in this study. Especially, it showed the best performance in Gaussian noise.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.37 no.2A
• /
• pp.128-134
• /
• 2012

In this paper, we propose a novel approach to enhance the recognition performance of a far and partially occluded three-dimensional (3-D) target in computational curving-effective integral imaging (CEII) by using the direct pixel-mapping (DPM) method. With this scheme, the elemental image array (EIA) originally picked up from a far and partially occluded 3-D target can be converted into a new EIA just like the one virtually picked up from a target located close to the lenslet array. Due to this characteristic of DPM, resolution and quality of the reconstructed target image can be highly enhanced, which results in a significant improvement of recognition performance of a far 3-D object. Experimental results reveal that image quality of the reconstructed target image and object recognition performance of the proposed system have been improved by 1.75 dB and 4.56% on the average in PSNR (peak-to-peak signal-to-noise ratio) and NCC (normalized correlation coefficient), respectively, compared to the conventional system.

• The Journal of the Acoustical Society of Korea
• /
• v.15 no.5
• /
• pp.107-112
• /
• 1996

In this paper we propose a new fast codebook training algorithm for reducing the searching time of LBG algorithm. For each training data, the proposed algorithm stores the indexes of codewords that are close to that training data in the first iteration. It reduces computation time by searching only those codewords, the indexes of which are stored for each training data. Compared to one of the previous fast training algorithm, FSLBG, it obtains a better codebook with less exccution time. In our experiment, the performance of the codebook generated by the proposed algorithm in terms of peak signal-to-noise ratio(TSNR) is very close to that of LBG algorithm. However, the codewords to be searched for each training data of the proposed algorithm is only about 6%, for a codebook size of 256 and 1.6%, for a codebook size of 1.24, of LBG algorithm.

• Journal of the Korea Society of Computer and Information
• /
• v.18 no.11
• /
• pp.59-68
• /
• 2013

In this paper, an effective scheme was proposed to estimate simple convex surface and simple concave surface which exist in image. This scheme is applied to input image to estimate simple convex surface or simple concave surface. When simple convex surface or simple concave surface exists, another proposed efficient interpolation scheme is used for the interpolated pixel to have the characteristics of simple convex surface or simple concave surface. The magnified image using the proposed schemes is more similar to the real image than the magnified image using the previous schemes. The PSNR values of the magnified images using the proposed schemes are greater than those of the magnified images using the previous interpolation schemes.

• Journal of the Korean Society of Radiology
• /
• v.13 no.4
• /
• pp.509-516
• /
• 2019

In the linear transformation method, the original image is divided into a plurality of range blocks, and a partial transform system for finding an optimal domain block existing in the image for each range block is used to adjust the performance of the compression ratio and the picture quality, The nonlinear transformation method uses only the rotation transformation among eight shuffle transforms. Since the search is performed only in the limited domain block, the coding time is faster than the linear transformation method of searching the domain block for any block in the image, Since the optimal domain block for the range block can not be selected in the image, the performance may be lower than other methods. Therefore, the nonlinear transformation method improves the performance by increasing the approximation degree of the brightness coefficient conversion instead of selecting the optimal domain block, The smaller the size of the block, the higher the PSNR value, The higher the compression ratio is increased groups were quadtree block divided to encode the image at best.

• International Journal of Computer Science & Network Security
• /
• v.21 no.9
• /
• pp.275-280
• /
• 2021

With the development of communication networks, the processes of exchanging and transmitting information rapidly developed. As millions of images are sent via social media every day, also wireless sensor networks are now used in all applications to capture images such as those used in traffic lights, roads and malls. Therefore, there is a need to reduce the size of these images while maintaining an acceptable degree of quality. In this paper, we use Python software to apply K-mean Clustering algorithm to compress RGB images. The PSNR, MSE, and SSIM are utilized to measure the image quality after image compression. The results of compression reduced the image size to nearly half the size of the original images using k = 64. In the SSIM measure, the higher the K, the greater the similarity between the two images which is a good indicator to a significant reduction in image size. Our proposed compression technique powered by the K-Mean clustering algorithm is useful for compressing images and reducing the size of images.