Article
Mathematics
Shuren Qi, Yushu Zhang, Chao Wang, Rushi Lan
Summary: In this paper, a simple and effective method is proposed to define invariants from blurred images without data augmentation and deblurring. This method utilizes the properties of fractional moments under projection operators to design invariants with blur invariance and rotation invariance. Experiment results demonstrate the good properties and potential of this method for small-scale robust vision problems.
Article
Computer Science, Artificial Intelligence
Chunzhi Gu, Xuequan Lu, Ying He, Chao Zhang
Summary: The article introduces a novel image deblurring method that does not require estimating blur kernels. By analyzing a pair of images captured easily in low-light conditions, the method utilizes GMM and EM algorithm to estimate parameters, incorporating bilateral terms and a detail layer to preserve sharp features. Experimental results demonstrate that the method outperforms existing techniques in terms of robustness, visual quality, and quantitative metrics.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Geochemistry & Geophysics
Cheng Su, Zeyu Xu, Fan Shen, Xiaocan Zhang
Summary: This letter presents an automatic detection method for identifying blurred areas within remote sensing images, and explores the relationship between the level of visual blurring and the difficulty of automatic detection.
IEEE GEOSCIENCE AND REMOTE SENSING LETTERS
(2023)
Article
Computer Science, Information Systems
Joshua Santoso, Williem, In Kyu Park
Summary: In this paper, a novel joint framework called D2R (Deblurring-to-Reconstruction) is proposed to address the problems of both human motion deblurring and 3D holistic body reconstruction simultaneously. By generating a new dataset containing sharp/blur image pairs and corresponding 3D body pose/shape, and by utilizing an additional structure-aware module for alternate training, the proposed method achieves superior results in terms of 3D holistic human reconstruction when the input image is deblurred.
Article
Computer Science, Information Systems
Lichun Mei, Huaiye Wang, Caiyun Wang, Yuanfu Zhao, Jun Zhang, Xiaoxia Zhao
Summary: A fast and robust template matching scheme, called MSTMM, is proposed to address the difficulty in matching under occluded scenarios caused by nonlinear intensity differences and structure differences between visible and thermal infrared images. The scheme utilizes Expanded Slice Transform with Adaptive Gray Level (EST-AGL) for extracting information about pixel distribution and maps EST-AGL matrices to integers or floats. The similarity between visible and thermal infrared images is evaluated using the Normalized Cross Correlation (NCC) algorithm. Experimental results show that the proposed MSTMM scheme achieves efficient template matching with a high success rate and reduced computational complexity.
Article
Computer Science, Artificial Intelligence
Hojatollah Zamani, Arash Amini
Summary: This paper addresses the problem of ellipse recovery from blurred shape images by representing ellipses as zero-level-sets of bivariate polynomials of degree 2 and detecting them using image moments. The approach is more robust against noise and able to detect ellipses even when they intersect or parts of the boundaries are lost. Simulation results confirm the superiority of the proposed method against existing techniques.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Optics
Hao Xu, Haiquan Hu, Shiqi Chen, Zhihai Xu, Qi Li, Tingting Jiang, Yueting Chen
Summary: This paper presents a joint imaging system of diffractive imaging and clear imaging, as well as a convolutional neural network-based method for hyperspectral image (HSI) reconstruction. Experimental results show that the proposed method can reconstruct HSIs with high spatial resolution and spectral accuracy.
OPTICS AND LASERS IN ENGINEERING
(2023)
Article
Computer Science, Artificial Intelligence
Hanyu Hong, JiaoWei Shi, Ziyuan Liu, Yaozong Zhang, Jinmeng Wu
Summary: This paper proposes a real-time algorithm based on curvature feature in the polar coordinate system to detect the critical part of reconnaissance balloons, achieving high precision results with a real-time detection on DSP. The method surpasses existing comparison algorithms in terms of processing speed and detection precision.
JOURNAL OF REAL-TIME IMAGE PROCESSING
(2021)
Article
Computer Science, Artificial Intelligence
Wende Dong, Shuyin Tao, Guili Xu, Yueting Chen
Summary: This paper proposes a regularized blind deconvolution method for restoring Poissonian blurred images, which achieves high quality restored images by combining L-0 norm, total variation, and negative logarithmic Poisson log-likelihood.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Optics
Yanling Chen, Lianglun Cheng, Heng Wu, Fei Mo, Ziyang Chen
Summary: The proposed method utilizes an iterative differential thermal information filter to fuse infrared and visible images, resulting in a fusion image with prominent thermal targets from the infrared image and detailed information from the visible image. Experimental results demonstrate the advantages and effectiveness of the method compared to deep learning and non deep learning-based methods.
OPTICS AND LASERS IN ENGINEERING
(2022)
Article
Instruments & Instrumentation
Zijie Zhou, Baofeng Zhang, Xiao Yu
Summary: In this paper, an immune coordination deep network (ICDNet) based on the biological immune coordination mechanism is proposed for hand heat trace extraction. By dividing the fuzzy area of the heat trace and extracting features, as well as obtaining prior information of the trace, the infrared heat trace target can be accurately extracted.
INFRARED PHYSICS & TECHNOLOGY
(2022)
Article
Spectroscopy
Zhou Zi-jie, Zhang Bao-feng, Yu Xiao
Summary: By utilizing the algorithm based on neural immune network, the extraction of infrared pipeline targets in complex backgrounds can be more accurate, with higher accuracy and lower error rate compared to traditional methods.
SPECTROSCOPY AND SPECTRAL ANALYSIS
(2021)
Article
Engineering, Multidisciplinary
Xiao Yu, YuHua Lu, Qiang Gao
Summary: This paper presents a neural immune ensemble learning algorithm for diagnosing abnormal high temperature regions in infrared steam pipeline images, which can effectively diagnose high temperature areas with higher accuracy compared to other image classification algorithms, meeting the needs of actual engineering applications.
INTERNATIONAL JOURNAL OF PRESSURE VESSELS AND PIPING
(2021)
Article
Instruments & Instrumentation
Lichun Mei, Caiyun Wang, Huaiye Wang, Yuanfu Zhao, Jun Zhang, Xiaoxia Zhao
Summary: This paper proposes a scheme called Hybrid Matching by Pixel Distribution Mapping (HMPDM) that combines traditional methods and neural network methods to achieve fast template matching. The scheme overcomes the challenge of nonlinear intensity differences between multi-modal images by extracting and mapping pixel distribution information. The experimental results demonstrate the real-time and accurate performance of the scheme.
INFRARED PHYSICS & TECHNOLOGY
(2022)
Article
Chemistry, Analytical
Luo Tian, Kepeng Qiu, Yufeng Zhao, Peng Wang
Summary: In this paper, an effective deblurring method that utilizes inertial sensors to aid in the deblurring process is proposed. The method applies a total-least-squares based iterative optimization scheme to handle the image deblurring problem involving blur kernel errors, and learns relating priors by neural networks. Experimental results show that the proposed approach achieves higher accuracy and robustness in edge detection of motion-blurred images compared to existing representative deblurring methods.
Article
Instruments & Instrumentation
Zengrun Wen, Xiulin Fan, Kaile Wang, Weiming Wang, Song Gao, Wenjing Hao, Yuanmei Gao, Yangjian Cai, Liren Zheng
Summary: This study presents a transition from Q-switched state to continuous wave state in an erbium-doped fiber laser, accompanied by irregular mode-hopping. The results showed that the transition between these two states can be achieved by adjusting the pump power. Modulation peaks were observed on both the Q-switched pulse train and the continuous wave background, and the central wavelength fluctuated.
INFRARED PHYSICS & TECHNOLOGY
(2024)