The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. ex. Some numerals are expressed as "XNUMX".
Copyrights notice
The original paper is in English. Non-English content has been machine-translated and may contain typographical errors or mistranslations. Copyrights notice
시끄럽고 흐릿한 이미지에 대한 모션 디블러링은 이미지 처리 커뮤니티에서 힘들고 근본적인 문제입니다. 여러 쌍의 잠상과 블러 커널이 동일한 흐릿한 이미지를 렌더링할 수 있으므로 이 문제의 최적화는 아직 해결되지 않았습니다. 이를 해결하기 위해 우리는 눈에 띄는 구조와 강화된 그라디언트의 데이터 기반 두꺼운 꼬리를 기반으로 잡음이 많고 흐릿한 이미지에 대한 효과적인 모션 디블러링 방법을 제시합니다. 구체적으로, 먼저 입력 이미지 노이즈를 제거하기 위한 전처리로 노이즈 제거를 사용하고 정확한 커널 추정을 위해 강한 에지를 복원합니다. 희박한 보완적 지식인 이미지 극단 채널 기반 사전(어두운 채널 사전 및 밝은 채널 사전)을 활용하여 눈에 띄는 구조를 추출합니다. 추출 기능의 매개변수를 조정하면 추출된 구조와 선명한 이미지 구조의 높은 근접성을 얻을 수 있습니다. 다음으로, 강화된 중간 이미지 그라디언트와 선명한 이미지의 무거운 꼬리 사전의 통합 항이 제안된 다음 이미지 복원 모델에 내장되어 흐릿한 이미지보다 선명한 이미지를 선호합니다. 합성 이미지와 실제 이미지에 대한 수많은 실험을 통해 제안된 방법이 최신 알고리즘에 비해 질적, 양적 측면에서 우수함을 확인했습니다.
Hongtian ZHAO
Shanghai Jiao Tong University
Shibao ZHENG
Shanghai Jiao Tong University
The copyright of the original papers published on this site belongs to IEICE. Unauthorized use of the original or translated papers is prohibited. See IEICE Provisions on Copyright for details.
부
Hongtian ZHAO, Shibao ZHENG, "Joint Extreme Channels-Inspired Structure Extraction and Enhanced Heavy-Tailed Priors Heuristic Kernel Estimation for Motion Deblurring of Noisy and Blurry Images" in IEICE TRANSACTIONS on Fundamentals,
vol. E103-A, no. 12, pp. 1520-1528, December 2020, doi: 10.1587/transfun.2020SMP0008.
Abstract: Motion deblurring for noisy and blurry images is an arduous and fundamental problem in image processing community. The problem is ill-posed as many different pairs of latent image and blur kernel can render the same blurred image, and thus, the optimization of this problem is still unsolved. To tackle it, we present an effective motion deblurring method for noisy and blurry images based on prominent structure and a data-driven heavy-tailed prior of enhanced gradient. Specifically, first, we employ denoising as a preprocess to remove the input image noise, and then restore strong edges for accurate kernel estimation. The image extreme channels-based priors (dark channel prior and bright channel prior) as sparse complementary knowledge are exploited to extract prominent structure. High closeness of the extracted structure to the clear image structure can be obtained via tuning the parameters of extraction function. Next, the integration term of enhanced interim image gradient and clear image heavy-tailed prior is proposed and then embedded into the image restoration model, which favors sharp images over blurry ones. A large number of experiments on both synthetic and real-life images verify the superiority of the proposed method over state-of-the-art algorithms, both qualitatively and quantitatively.
URL: https://global.ieice.org/en_transactions/fundamentals/10.1587/transfun.2020SMP0008/_p
부
@ARTICLE{e103-a_12_1520,
author={Hongtian ZHAO, Shibao ZHENG, },
journal={IEICE TRANSACTIONS on Fundamentals},
title={Joint Extreme Channels-Inspired Structure Extraction and Enhanced Heavy-Tailed Priors Heuristic Kernel Estimation for Motion Deblurring of Noisy and Blurry Images},
year={2020},
volume={E103-A},
number={12},
pages={1520-1528},
abstract={Motion deblurring for noisy and blurry images is an arduous and fundamental problem in image processing community. The problem is ill-posed as many different pairs of latent image and blur kernel can render the same blurred image, and thus, the optimization of this problem is still unsolved. To tackle it, we present an effective motion deblurring method for noisy and blurry images based on prominent structure and a data-driven heavy-tailed prior of enhanced gradient. Specifically, first, we employ denoising as a preprocess to remove the input image noise, and then restore strong edges for accurate kernel estimation. The image extreme channels-based priors (dark channel prior and bright channel prior) as sparse complementary knowledge are exploited to extract prominent structure. High closeness of the extracted structure to the clear image structure can be obtained via tuning the parameters of extraction function. Next, the integration term of enhanced interim image gradient and clear image heavy-tailed prior is proposed and then embedded into the image restoration model, which favors sharp images over blurry ones. A large number of experiments on both synthetic and real-life images verify the superiority of the proposed method over state-of-the-art algorithms, both qualitatively and quantitatively.},
keywords={},
doi={10.1587/transfun.2020SMP0008},
ISSN={1745-1337},
month={December},}
부
TY - JOUR
TI - Joint Extreme Channels-Inspired Structure Extraction and Enhanced Heavy-Tailed Priors Heuristic Kernel Estimation for Motion Deblurring of Noisy and Blurry Images
T2 - IEICE TRANSACTIONS on Fundamentals
SP - 1520
EP - 1528
AU - Hongtian ZHAO
AU - Shibao ZHENG
PY - 2020
DO - 10.1587/transfun.2020SMP0008
JO - IEICE TRANSACTIONS on Fundamentals
SN - 1745-1337
VL - E103-A
IS - 12
JA - IEICE TRANSACTIONS on Fundamentals
Y1 - December 2020
AB - Motion deblurring for noisy and blurry images is an arduous and fundamental problem in image processing community. The problem is ill-posed as many different pairs of latent image and blur kernel can render the same blurred image, and thus, the optimization of this problem is still unsolved. To tackle it, we present an effective motion deblurring method for noisy and blurry images based on prominent structure and a data-driven heavy-tailed prior of enhanced gradient. Specifically, first, we employ denoising as a preprocess to remove the input image noise, and then restore strong edges for accurate kernel estimation. The image extreme channels-based priors (dark channel prior and bright channel prior) as sparse complementary knowledge are exploited to extract prominent structure. High closeness of the extracted structure to the clear image structure can be obtained via tuning the parameters of extraction function. Next, the integration term of enhanced interim image gradient and clear image heavy-tailed prior is proposed and then embedded into the image restoration model, which favors sharp images over blurry ones. A large number of experiments on both synthetic and real-life images verify the superiority of the proposed method over state-of-the-art algorithms, both qualitatively and quantitatively.
ER -