선택한 단어 수는 입니다.
최소 단어 이상 선택하여야 합니다.
최소 단어 이상 선택하여야 합니다.
선택한 단어 수는 30입니다.
최대 10 단어까지만 선택 가능합니다.
최대 10 단어까지만 선택 가능합니다.
문제 정의
- We performed experiments to demonstrate the efficiency of noise reduction with the FNLM noise reduction algorithm in the gamma camera system. In particular, the comparison between the proposed reduction algorithm and the conventional methods was not implemented, since the purpose of this research is to cope with the various problems using image-processing. The results confirmed that the proposed FNLM noise reduction algorithm produced appropriate image quality with acceptable image-processing time.
제안 방법
- There have been many studies on the FNLM noise reduction algorithm in X-ray imaging systems, but few on quantitative evaluations of gamma-ray imaging systems for creating the algorithm-applied images. Therefore, the purpose of this work was to evaluate image performance related to the noise characteristics using the proposed FNLM noise reduction algorithm (compared with the median filter, Wiener filter, and TV noise reduction algorithm) in a gamma camera system using a low-energy, high-resolution parallel-hole collimator. The normalized noise power spectrum (NNPS), contrast-to-noise ratio (CNR), and coefficient of variation (COV) for the algorithm were evaluated in an experimental gamma-ray image.
대상 데이터
- The time required to denoise the phantom image in the gamma camera system was calculated for each noise reduction method, with the fastest being the FNLM algorithm in all phantom images. The computer system used for the experiments was an Intel dual core CPU (Q9550 2.8 GHz) and NVIDIA GeForce graphic card (9500 GS). The average processing time was approximately 0.
이론/모형
- 5. In the cold sphere and rod phantoms, the evaluated CNR increases from the median filter to the Wiener filter, via the TV noise reduction algorithm, to the proposed FNLM noise reduction algorithm in that order. In the cold sphere phantom, when the conventional methods and the FNLM noise reduction algorithm were compared, the CNR using the proposed algorithm was approximately 2.
- There is now extensive research that shows the effectiveness of the abovementioned noise reduction algorithms in medical images. In this study, a fast nonlocal means (FNLM)-based noise reduction algorithm was designed using an acceleration function. There have been many studies on the FNLM noise reduction algorithm in X-ray imaging systems, but few on quantitative evaluations of gamma-ray imaging systems for creating the algorithm-applied images.
- In particular, the temporal resolution can be improved because of the term of the acceleration function considering the Euclidean distance in the designed algorithm. The time required to denoise the phantom image in the gamma camera system was calculated for each noise reduction method, with the fastest being the FNLM algorithm in all phantom images. The computer system used for the experiments was an Intel dual core CPU (Q9550 2.
- However, one of the concerns with using the NLM noise reduction algorithm is the low time resolution in image processing. To address the low time resolution problem, the acceleration function was used with Euclidean distance based on comparing two neighborhood pixels with the fast Fourier transform and summed squared image [17].
- To compare the image performances of the noise reduction algorithms, conventional methods using the median filter, Wiener filter, and TV noise reduction algorithm were also used.
성능/효과
- 29 times higher than those of the noisy image, median filter, Wiener filter, and TV noise reduction algorithms, respectively. In the cold rod phantom, when the conventional methods and the FNLM noise reduction algorithm were compared, the CNR using the proposed algorithm was approximately 2.21, 1.41, 1.31, and 1.26 times higher than those of the noisy image, median filter, Wiener filter, and TV noise reduction algorithms, respectively.
- 36 times better than those of the noisy image, median filter, Wiener filter, and TV noise reduction algorithms, respectively. In the cold rod phantom, when the conventional methods and the FNLM noise reduction algorithm were compared, the COV using the proposed algorithm was approximately 7.73, 4.45, 4.09, and 2.82 times better than those of the noisy image, median filter, Wiener filter, and TV noise reduction algorithms, respectively.
- In the cold sphere and rod phantoms, the evaluated CNR increases from the median filter to the Wiener filter, via the TV noise reduction algorithm, to the proposed FNLM noise reduction algorithm in that order. In the cold sphere phantom, when the conventional methods and the FNLM noise reduction algorithm were compared, the CNR using the proposed algorithm was approximately 2.27, 1.46, 1.35, and 1.29 times higher than those of the noisy image, median filter, Wiener filter, and TV noise reduction algorithms, respectively. In the cold rod phantom, when the conventional methods and the FNLM noise reduction algorithm were compared, the CNR using the proposed algorithm was approximately 2.
- In addition, the evaluated COV goes from the proposed FNLM noise reduction algorithm to the TV noise reduction algorithm, via the Wiener filter, to the median filter, in increasing order, in the cold sphere and rod phantoms. In the cold sphere phantom, when the conventional methods and the FNLM noise reduction algorithm were compared, the COV using the proposed algorithm was approximately 8.18, 4.91, 4.64, and 2.36 times better than those of the noisy image, median filter, Wiener filter, and TV noise reduction algorithms, respectively. In the cold rod phantom, when the conventional methods and the FNLM noise reduction algorithm were compared, the COV using the proposed algorithm was approximately 7.
- 5 mm-1 spatial frequency) lower than that of the median or Wiener filter. It was proved that the proposed FNLM noise reduction algorithm can effectively eliminate the noise intensity by a much greater degree in the cold region phantom using the gamma camera system and achieve improved noise characteristics. In addition, the result confirmed that the calculated NNPS values decreased sharply with the spatial frequency at approximately 1.
- The feasibility of the FNLM noise reduction algorithm in Deluxe Jaszcazk phantom images with the gamma camera system was confirmed. Figs.
- In particular, the comparison between the proposed reduction algorithm and the conventional methods was not implemented, since the purpose of this research is to cope with the various problems using image-processing. The results confirmed that the proposed FNLM noise reduction algorithm produced appropriate image quality with acceptable image-processing time.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.