• Investigative Magnetic Resonance Imaging
• /
• v.23 no.1
• /
• pp.1-16
• /
• 2019

Dynamic contrast enhanced (DCE) magnetic resonance (MR) imaging plays an important role in non-invasive detection and characterization of primary and metastatic lesions in the liver. Recently, efforts have been made to improve spatial and temporal resolution of DCE liver MRI for arterial phase imaging. Review of recent publications related to arterial phase imaging of the liver indicates that there exist primarily two approaches: breath-hold and free-breathing. For breath-hold imaging, acquiring multiple arterial phase images in a breath-hold is the preferred approach over conventional single-phase imaging. For free-breathing imaging, a combination of three-dimensional (3D) stack-of-stars golden-angle sampling and compressed sensing parallel imaging reconstruction is one of emerging techniques. Self-gating can be used to decrease respiratory motion artifact. This article introduces recent MRI technologies relevant to hepatic arterial phase imaging, including differential subsampling with Cartesian ordering (DISCO), golden-angle radial sparse parallel (GRASP), and X-D GRASP. This article also describes techniques related to dynamic 3D image reconstruction of the liver from golden-angle stack-of-stars data.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2023.05a
• /
• pp.479-481
• /
• 2023

Generative Adversarial Networks (GANs) have facilitated substantial improvement in single-image super-resolution (SR) by enabling the generation of photo-realistic images. However, the high memory requirements of GAN-based SRs (mainly generators) lead to reduced performance and increased energy consumption, making it difficult to implement them onto resource-constricted devices. In this study, we propose an efficient and compressed architecture for the SR-GAN (generator) model using the model compression technique Knowledge Distillation. Our approach involves the transmission of knowledge from a heavy network to a lightweight one, which reduces the storage requirement of the model by 58% with also an increase in their performance. Experimental results on various benchmarks indicate that our proposed compressed model enhances performance with an increase in PSNR, SSIM, and image quality respectively for x4 super-resolution tasks.

• Imaging Science in Dentistry
• /
• v.50 no.4
• /
• pp.331-337
• /
• 2020

Purpose: As cone-beam computed tomography (CBCT) has become the most widely used 3-dimensional (3D) imaging modality in the dental field, storage space and costs for large-capacity data have become an important issue. Therefore, if 3D data can be stored at a clinically acceptable compression rate, the burden in terms of storage space and cost can be reduced and data can be managed more efficiently. In this study, a deep learning network for super-resolution was tested to restore compressed virtual CBCT images. Materials and Methods: Virtual CBCT image data were created with a publicly available online dataset (CQ500) of multidetector computed tomography images using CBCT reconstruction software (TIGRE). A very deep super-resolution (VDSR) network was trained to restore high-resolution virtual CBCT images from the low-resolution virtual CBCT images. Results: The images reconstructed by VDSR showed better image quality than bicubic interpolation in restored images at various scale ratios. The highest scale ratio with clinically acceptable reconstruction accuracy using VDSR was 2.1. Conclusion: VDSR showed promising restoration accuracy in this study. In the future, it will be necessary to experiment with new deep learning algorithms and large-scale data for clinical application of this technology.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.17 no.7
• /
• pp.1687-1693
• /
• 2013

This paper addresses an adaptive postprocessing technique to remove blocking effects of the highly compressed images. The proposed technique removes blocking effects selectively by applying filters with different strength according to block analysis based on the directional activity. One-dimensional filters which are used to remove grid noises accomplish the adaptive filtering to the signal itself as well as to the directionality of the block. Moreover, we propose a detection method of the staircase noises and corner outliers and a two-dimensional directional filter to remove them. Experimental results for various images and bitrates show that the proposed method outperforms the conventional methods in PSNR for the objective performance and GBIM for the subjective quality evaluation.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.8 no.4
• /
• pp.1499-1513
• /
• 2014

This paper proposes a novel blind image watermarking scheme exploiting Block Truncation Coding (BTC). Most of existing BTC-based watermarking or data hiding methods embed information in BTC compressed images by modifying the BTC encoding stage or BTC-compressed data, resulting in watermarked images with bad quality. Other than existing BTC-based watermarking schemes, our scheme does not really perform the BTC compression on images during the embedding process but uses the parity of BTC quantization data to guide the watermark embedding and extraction processes. In our scheme, we use a binary image as the original watermark. During the embedding process, the original cover image is first partitioned into non-overlapping $4{\times}4$ blocks. Then, BTC is performed on each block to obtain its BTC quantized high mean and low mean. According to the parity of high mean and the parity of low mean, two watermark bits are embedded in each block by modifying the pixel values in the block to make sure that the parity of high mean and the parity of low mean in the modified block are equal to the two watermark bits. During the extraction process, BTC is first performed on each block to obtain its high mean and low mean. By checking the parity of high mean and the parity of low mean, we can extract the two watermark bits in each block. The experimental results show that the proposed watermarking method is fragile to most image processing operations and various kinds of attacks while preserving the invisibility very well, thus the proposed scheme can be used for image authentication.

• Journal of Biomedical Engineering Research
• /
• v.35 no.5
• /
• pp.151-159
• /
• 2014

Computed tomography (CT) is one of the most widely used medical imaging modality. However, substantial x-ray dose exposed to the human subject during the CT scan is a great concern. Region-of-interest (ROI) CT is considered to be a possible solution for its potential to reduce the x-ray dose to the human subject. In most of ROI-CT scans, the ROI is set to a circular shape whose diameter is often considerably smaller than the full field-of-view (FOV). However, an arbitrarily shaped ROI is very desirable to reduce the x-ray dose more than the circularly shaped ROI can do. We propose a new method to make a non-circular convex-shaped ROI along with the image reconstruction method. To make a ROI with an arbitrary convex shape, dynamic collimations are necessary to minimize the x-ray dose at each angle of view. In addition to the dynamic collimation, we get the ROI projection data with slightly lower sampling rate in the view direction to further reduce the x-ray dose. We reconstruct images from the ROI projection data in the compressed sensing (CS) framework assisted by the exterior projection data acquired from the pilot scan to set the ROI. To validate the proposed method, we used the experimental micro-CT projection data after truncating them to simulate the dynamic collimation. The reconstructed ROI images showed little errors as compared to the images reconstructed from the full-FOV scan data as well as little artifacts inside the ROI. We expect the proposed method can significantly reduce the x-ray dose in CT scans if the dynamic collimation is realized in real CT machines.

• ETRI Journal
• /
• v.27 no.1
• /
• pp.53-63
• /
• 2005

A compressed video bitstream is sensitive to errors that may severely degrade the reconstructed images even when the bit error rate is small. One approach to combat the impact of such errors is the use of error concealment at the decoder without increasing the bit rate or changing the encoder. For spatial-error concealment, we propose a method featuring edge continuity and texture preservation as well as low computation to reconstruct more visually acceptable images. Aiming at temporal error concealment, we propose a two-step algorithm based on block matching principles in which the assumption of smooth and uniform motion for some adjacent blocks is adopted. As simulation results show, the proposed spatial and temporal methods provide better reconstruction quality for damaged images than other methods.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2003.05a
• /
• pp.268-272
• /
• 2003

Large medical images in PACS are compressed for saving storage space and improving network speed. The integrated compression environment was designed and developed for uniting of various compression methods. Various compression algorithm-RLE compression, lossless JEPG, JPEG, was built into it, complying with DICOM. A image compression using DWT was also implemented in it. And a unified algorithm of lossless compression and lossy compression was designed to improve images quality and to make compression ratios high. And integrated compression environment was operating together with a database program for efficient and user-friendly management.

• Journal of the Korea Computer Industry Society
• /
• v.2 no.11
• /
• pp.1465-1476
• /
• 2001

For the applications like video indexing, text understanding, and automatic captions localization system, real-time localization of captions is an essential task. This paper presents a algorithm for localization of captions in MPEG compression images based on I frame. In this algorithm, caption text regions are segmented from background images using their distinguishing texture characteristics and chrominance information. Unlike previously published algorithms which fully decompress the video sequence before extracting the text regions, this algorithm locates candidate caption text region directly in the DCT compressed domain.

• Proceedings of the IEEK Conference
• /
• 2000.07a
• /
• pp.19-22
• /
• 2000

In this paper, we propose a watermarking method for still images using Discrete Cosine Transform (DCT). Watermarking is a copyright protection technique for digital contents by hiding secret information into the contents. The proposed method embeds the watermark information into DCT coefficients. To obtain a watermarked image that is not only high quality but also has robustness for compression, we considered a method to change the degree of embedding by utilizing the activity of each DCT block. The simulation results show that the proposed scheme can obtain huh quality watermarked images and we can extract most of the embedded data even if they are compressed by JPEG scheme.