• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.31 no.5
• /
• pp.987-999
• /
• 2021

Recently, Federated learning has become an issue due to privacy invasion caused by data. Federated learning is safe from privacy violations because it does not need to be collected into a server and does not require learning data. As a result, studies on application methods for utilizing distributed devices and data are underway. However, Federated learning is no longer safe as research on the reconstruction attack to restore learning data from gradients transmitted in the Federated learning process progresses. This paper is to verify numerically and visually how well data reconstruction attacks work in various data situations. Considering that the attacker does not know how the data is constructed, divide the data with the class from when only one data exists to when multiple data are distributed within the class, and use MNIST data as an evaluation index that is MSE, LOSS, PSNR, and SSIM. The fact is that the more classes and data, the higher MSE, LOSS, and PSNR and SSIM are, the lower the reconstruction performance, but sufficient privacy invasion is possible with several reconstructed images.

• Journal of the Korea Convergence Society
• /
• v.11 no.12
• /
• pp.15-22
• /
• 2020

In this paper, we proposed a deep learning based super-resolution method that combines Channel Attention and Spatial Attention feature enhancement methods. It is important to restore high-frequency components, such as texture and features, that have large changes in surrounding pixels during super-resolution processing. We proposed a super-resolution method using feature enhancement that combines Channel Attention and Spatial Attention. The existing CNN (Convolutional Neural Network) based super-resolution method has difficulty in deep network learning and lacks emphasis on high frequency components, resulting in blurry contours and distortion. In order to solve the problem, we used an emphasis block that combines Channel Attention and Spatial Attention to which Skip Connection was applied, and a Residual Block. The emphasized feature map extracted by the method was extended through Sub-pixel Convolution to obtain the super resolution. As a result, about PSNR improved by 5%, SSIM improved by 3% compared with the conventional SRCNN, and by comparison with VDSR, about PSNR improved by 2% and SSIM improved by 1%.

• Journal of the Korean Society of Radiology
• /
• v.15 no.3
• /
• pp.281-291
• /
• 2021

This study aimed to measure the quantitative changes in images according to the use of compressed sensing in expressing the slow flow rate in TOF MRA test using magnetic resonance imaging. This study set different blood flow rate sections by using auto-injector and flow phantom and compared changes in the SNR, CNR, SSIM, and RMSE measurements by different CS factors between TOF with CS and TOF without CS. One-way ANOVA was performed to test the effect on the image induced by the increase of the CS factor. The results revealed that TOF MRA with CS significantly decreased scan time without significantly affecting SNR and CNR compared to TOF MRA with CS. On the other hand, the differences in SSIM and RMSE between TOF with CS and TOF without CS increased as the CS factor increased. Therefore, it is necessary to efficiently reduce scan time by adapting the CS technique while considering the appropriate range of the CS factor. Additionally, more studies are needed to evaluate CS factors and the similarity precision of images further.

• Journal of the Korean Society of Radiology
• /
• v.17 no.2
• /
• pp.215-221
• /
• 2023

Among brain CT scan conditions including the lens, the tube voltage was changed to 80, 100, and 120 kVp and applied. The change in dose was analyzed using lead, lead goggles and barium sulfate silicon shielding materials, and the degree of influence of the shielding materials on image quality was compared and analyzed by applying the SNR, CNR, and SSIM index analysis methods. As a result, it was analyzed that although the dose was reduced by applying all shielding materials, the difference in dose reduction was not large (P > 0.05). In addition, as for the change in image quality due to the application of the shielding material, SNR and CNR were the highest when lead goggles were applied, and the structural similarity was measured to be the best as it was closest to the reference value of 1 in SSIM analysis. Therefore, based on the results of this study, it is thought that if more diverse shielding materials and clinical test results are derived and applied, it will be helpful for the clinical application criteria in the case of shielding utilization inspection.

• Journal of the Korean Society of Radiology
• /
• v.17 no.3
• /
• pp.417-422
• /
• 2023

In this study, normal saline was diluted with the contrast medium at a certain ratio for the purpose of reducing the image quality poor and side effects caused by the contrast medium during CT examination. At this time, by finding the energy level of DECT that can compensate for the decrease in contrast of the image according to the degree of dilution, the usefulness of applying DECT for compensating the difference in image contrast was investigated through comparative analysis by applying SNR, CNR, and SSIM. As a result, when a dilution ratio of 4 (contrast medium): 6 (normal saline) and the energy level of DECT of 65 keV were applied, the contrast difference was the most similar to that when using the undiluted contrast medium. At this time, SNR was 813.71 ± 37.6, CNR was the highest at 921.87 ± 17.1, and SSIM index was measured at 0.851, which is the most similar to 1. The results of this study are meaningful in providing basic information for finding the appropriate dilution rate and energy level for each examination site through future clinical studies. It is believed that it can be reduced.

• Journal of Multimedia Information System
• /
• v.4 no.4
• /
• pp.311-316
• /
• 2017

High Efficiency Video Coding (HEVC) is the newest video coding standard for improvement in video data compression. This new standard provides a significant improvement in picture quality, especially for high-resolution videos. A quadtree-based structure is created for the encoding and decoding processes and the rate-distortion (RD) cost is calculated for all possible dimensions of coding units in the quadtree. To get the best combination of the block an optimization process is performed in the encoder, called rate distortion optimization (RDO). In this work we are proposing a novel approach to enhance the overall RDO process of HEVC encoder. The proposed algorithm is performed in two steps. In the first step, like HEVC, it performs general rate distortion optimization. The second step is an extra checking where a SSIM based cost is evaluated. Moreover, a fast SSIM (FSSIM) calculation technique is also proposed in this paper.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.6 no.1
• /
• pp.426-445
• /
• 2012

Conventional rate control (RC) schemes for H.264 video coding usually regulate output bit rate to match channel bandwidth by adjusting quantization parameter (QP) at fixed full frame rate, and the passive frame skipping to avoid buffer overflow usually occurs when scene changes or high motions exist in video sequences especially at low bit rate, which degrades spatial-temporal quality and causes jerky effect. In this paper, an active content adaptive frame skipping scheme is proposed instead of passive methods, which skips subjectively trivial frames by structural similarity (SSIM) measurement between the original frame and the interpolated frame via motion vector (MV) copy scheme. The saved bits from skipped frames are allocated to coded key ones to enhance their spatial quality, and the skipped frames are well recovered based on MV copy scheme from adjacent key ones at the decoder side to maintain constant frame rate. Experimental results show that the proposed active SSIM-based frameskip scheme acquires better and more consistent spatial-temporal quality both in objective (PSNR) and subjective (SSIM) sense with low complexity compared to classic fixed frame rate control method JVT-G012 and prior objective metric based frameskip method.

• 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.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.16 no.9
• /
• pp.2882-2903
• /
• 2022

Image super-resolution (SR) processing is of great value in the fields of digital image processing, intelligent security, film and television production and so on. This paper proposed a densely connected deep learning network based on cascade architecture, which can be used to solve the problem of super-resolution in the field of image quality enhancement. We proposed a more efficient residual scaling dense block (RSDB) and the multi-channel cascade architecture to realize more efficient feature reuse. Also we proposed a hybrid loss function based on L₁ error and L_∞ error to achieve better L_∞ error performance. The experimental results show that the overall performance of the network is effectively improved on cascade architecture and residual scaling. Compared with the residual dense net (RDN), the PSNR / SSIM of the new method is improved by 2.24% / 1.44% respectively, and the L_∞ performance is improved by 3.64%. It shows that the cascade connection and residual scaling method can effectively realize feature reuse, improving the residual convergence speed and learning efficiency of our network. The L_∞ performance is improved by 11.09% with only a minimal loses of 1.14% / 0.60% on PSNR / SSIM performance after adopting the new loss function. That is to say, the L_∞ performance can be improved greatly on the new loss function with a minor loss of PSNR / SSIM performance, which is of great value in L_∞ error sensitive tasks.

• Journal of the Korea Institute of Military Science and Technology
• /
• v.25 no.5
• /
• pp.476-486
• /
• 2022

Synthetic dynamic infrared image generation from the given virtual environment is being the primary goal to simulate the output of the infra-red(IR) camera installed on a vehicle to evaluate the control algorithm for various search & reconnaissance missions. Due to the difficulty to obtain actual IR data in complex environments, Artificial intelligence(AI) has been used recently in the field of image data generation. In this paper, CycleGAN technique is applied to obtain a more realistic synthetic IR image. We added the Structural Similarity Index Measure(SSIM) loss function to the L1 loss function to generate a more realistic synthetic IR image when the CycleGAN image is generated. From the simulation, it is applicable to the guided-missile flight simulation tests by using the synthetic infrared image generated by the proposed technique.