• 전기전자학회논문지
• /
• 제23권4호
• /
• pp.1408-1414
• /
• 2019

전이학습을 수행한 심층 인공신경망을 압축센싱 심혈관 자기공명영상에 적용하였다. 전이학습은 선행학습 신경망의 구조나 필터 커널, 가중치를 현재의 학습이나 응용에 활용하는 방법이다. 전이학습은 학습 속도를 향상시키고, 학습 데이터가 제한적일 때 신경망의 일반화에 도움이 된다. 8명의 건강한 지원자가 참여한 심장 자기공명영상 실험에서 전이학습을 수행한 신경망은 단독학습 신경망에 비해 학습시간이 5배 이상 단축되었다. 시험 데이터에 대해서도 전이학습을 수행한 신경망은 전이학습을 수행하지 않은 신경망에 비하여 낮은 정규화 평균제곱오차와 향상된 재구성 영상화질을 보였다.

• Investigative Magnetic Resonance Imaging
• /
• 제19권1호
• /
• pp.19-30
• /
• 2015

Purpose: A new compressed sensing technique by iterative truncation of small transformed coefficients (ITSC) is proposed for fast cardiac CINE MRI. Materials and Methods: The proposed reconstruction is composed of two processes: truncation of the small transformed coefficients in the r-f domain, and restoration of the measured data in the k-t domain. The two processes are sequentially applied iteratively until the reconstructed images converge, with the assumption that the cardiac CINE images are inherently sparse in the r-f domain. A novel sampling strategy to reduce the normalized mean square error of the reconstructed images is proposed. Results: The technique shows the least normalized mean square error among the four methods under comparison (zero filling, view sharing, k-t FOCUSS, and ITSC). Application of ITSC for multi-slice cardiac CINE imaging was tested with the number of slices of 2 to 8 in a single breath-hold, to demonstrate the clinical usefulness of the technique. Conclusion: Reconstructed images with the compression factors of 3-4 appear very close to the images without compression. Furthermore the proposed algorithm is computationally efficient and is stable without using matrix inversion during the reconstruction.

• 한국위성정보통신학회논문지
• /
• 제9권4호
• /
• pp.104-110
• /
• 2014

본 논문에서는 무인항공기인 방송용 멀티콥터를 이용한 Full-HD급 이상 화질의 이미지를 효율적으로 전송하기 위해 이미지 압축 센싱 기법을 적용하고, Sparse 신호의 효율적 복원을 위해 Turbo 알고리즘과 Markov chain Monte Carlo (MCMC) 알고리즘의 복원 성능을 모의실험을 통해 비교 분석하였다. 제안된 복원 기법은 압축 센싱에 기반하여 데이터 용량을 줄이고 빠르고 오류 없는 원신호 복원에 중점을 두었다. 다수의 이미지 파일로 모의실험을 진행한 결과 Loopy belief propagation(BP) 기반의 Turbo 복원 알고리즘이 Gibbs sampling기반 알고리즘을 수행하는 MCMC 알고리즘 보다 평균 복원 연산 시간, NMSE 값에서 우수하여 보다 효율적인 복원 방법으로 생각된다.

• Current Optics and Photonics
• /
• 제5권6호
• /
• pp.606-616
• /
• 2021

A photonic integrated interferometric imaging system possesses the characteristics of small-scale, low weight, low power consumption, and better image quality. It has potential application for replacing conventional large space telescopes. In this paper, the principle of photonic integrated interferometric imaging is investigated. A novel lenslet array arrangement and lenslet pairing approach are proposed, which are helpful in improving spatial frequency coverage. For the novel lenslet array arrangement, two short interference arms were evenly distributed between two adjacent long interference arms. Each lenslet in the array would be paired twice through the novel lenslet pairing approach. Moreover, the image reconstruction model for optical interferometric imaging based on compressed sensing was established. Image simulation results show that the peak signal to noise ratio (PSNR) of the reconstructed image based on compressive sensing is about 10 dB higher than that of the direct restored image. Meanwhile, the normalized mean square error (NMSE) of the direct restored image is approximately 0.38 higher than that of the reconstructed image. Structural similarity index measure (SSIM) of the reconstructed image based on compressed sensing is about 0.33 higher than that of the direct restored image. The increased spatial frequency coverage and image reconstruction approach jointly contribute to better image quality of the photonic integrated interferometric imaging system.

• Investigative Magnetic Resonance Imaging
• /
• 제25권1호
• /
• pp.10-22
• /
• 2021

Purpose: To overcome the difficulty in building a large data set with a high-quality in medical imaging, a concept of 'blended-transfer learning' (BTL) using a combination of both source data and target data is proposed for the target task. Materials and Methods: Source and target tasks were defined as training of the source and target networks to reconstruct cardiac CINE images from undersampled data, respectively. In transfer learning (TL), the entire neural network (NN) or some parts of the NN after conducting a source task using an open data set was adopted in the target network as the initial network to improve the learning speed and the performance of the target task. Using BTL, an NN effectively learned the target data while preserving knowledge from the source data to the maximum extent possible. The ratio of the source data to the target data was reduced stepwise from 1 in the initial stage to 0 in the final stage. Results: NN that performed BTL showed an improved performance compared to those that performed TL or standalone learning (SL). Generalization of NN was also better achieved. The learning curve was evaluated using normalized mean square error (NMSE) of reconstructed images for both target data and source data. BTL reduced the learning time by 1.25 to 100 times and provided better image quality. Its NMSE was 3% to 8% lower than with SL. Conclusion: The NN that performed the proposed BTL showed the best performance in terms of learning speed and learning curve. It also showed the highest reconstructed-image quality with the lowest NMSE for the test data set. Thus, BTL is an effective way of learning for NNs in the medical-imaging domain where both quality and quantity of data are always limited.

• 한국정보통신학회논문지
• /
• 제18권7호
• /
• pp.1557-1564
• /
• 2014

본 논문에서는 일반화된 공간천이변조시스템에서 신호 복원 성능의 개선을 위하여 병렬 직교매칭퍼슛 기술을 이용한 신호 검출기법을 제안하고 그 성능을 분석한다. 일반화된 공간천이변조 시스템에서 수신신호의 복원은 압축 센싱에서 성긴신호 복원과 매우 유사하다. 성긴 신호 복원에서 자주 사용되는 직교매칭퍼슛 기법은 매 반복과정에서 수신 신호와 채널 행렬과의 상관도가 높은 인덱스를 송신신호의 Nonzero 인덱스로 1개씩 선택한다. 반면 제안된 POMP기법에서는 수신신호를 이용하여 첫 번째 반복과정에서 채널행렬과의 상관도가 높은 인덱스를 복수(M)개 선택한 후, 선택된 M개의 인덱스를 초기 인덱스로 하는 M개의 OMP과정을 병렬적으로 수행한다. 최종적으로 각 OMP과정에서 복원된 신호 중 수신된 신호와 복원신호사이의 잔차 (Residual)가 가장 작은 후보 신호를 최종 복원 신호로 선택한다. 본 논문에서는 POMP기법에 양자화기법을 결합한 알고리즘도 함께 제안한다. 제안된 POMP알고리즘은 OMP대비 M배의 복잡도를 갖지만 신호 복원 성능은 매우 탁월하다.

• 대한원격탐사학회지
• /
• 제26권3호
• /
• pp.297-304
• /
• 2010

Since the change in Doppler centroid according to moving targets brings alteration to the phase in azimuth differential signals of synthetic aperture radar (SAR) data, one can measure the velocity of the moving targets using this effect. In this study, we will investigate theoretically measuring the velocity of an object from azimuth differential signals by using range compressed data which is the interim outcome of treatment from the simulated SAR raw data of moving targets on the background of sea clutter. Also, it will provide evaluation for the elements that affect the estimation error of velocity from a single SAR sensor. By making RADARSAT-1 simulated image as a specific case, the research includes comparisons for the means of velocity measurement classified by the directions of movement in the four following cases. 1. A case of a single target without currents, 2. A case of a single target with tidal currents of 0.5 m/s, 1 m/s, and 3 m/s, 3. A case of two targets on a same azimuth line moving in a same direction and velocity, 4. A case of a single target contiguous to land where radar backscatter is strong. As a result, when two moving targets exist in SAR image outside the range of approximately 256 pixels, the velocity of the object can be measured with high accuracy. However, when other moving targets exist in the range of approximately 128 pixels or when the target was contiguous to the land of strong backscatter coefficient (NRCS: normalized radar cross section), the estimated velocity was in error by 10% at the maximum. This is because in the process of assuming the target's location, an error occurs due to the differential signals affected by other scatterers.