Filters








1,276 Hits in 4.6 sec

Pyramid Real Image Denoising Network [article]

Yiyun Zhao, Zhuqing Jiang, Aidong Men, Guodong Ju
2019 arXiv   pre-print
To tackle the issue of blind denoising, in this paper, we propose a novel pyramid real image denoising network (PRIDNet), which contains three stages.  ...  Second, at the multi-scale denoising stage, pyramid pooling is utilized to extract multi-scale features.  ...  Therefore multi-scale features can not be adaptively expressed. To address these issues, we propose a pyramid real image denoising network (PRIDNet) as shown in Fig. 2 .  ... 
arXiv:1908.00273v2 fatcat:ww4xrbmwbncxhp76te2tajbf44

NTIRE 2020 Challenge on Real Image Denoising: Dataset, Methods and Results [article]

Abdelrahman Abdelhamed, Mahmoud Afifi, Radu Timofte, Michael S. Brown, Yue Cao, Zhilu Zhang, Wangmeng Zuo, Xiaoling Zhang, Jiye Liu, Wendong Chen, Changyuan Wen, Meng Liu (+78 others)
2020 arXiv   pre-print
This challenge has two tracks for quantitatively evaluating image denoising performance in (1) the Bayer-pattern rawRGB and (2) the standard RGB (sRGB) color spaces.  ...  The proposed methods by the participating teams represent the current state-of-the-art performance in image denoising targeting real noisy images.  ...  Tyan Parallel U-net for Real Image Denoising The team proposed parallel U-net for considering global and pixel-wise denoising at the same time.  ... 
arXiv:2005.04117v1 fatcat:iwtpyxikerbqhhvkpmwghqxeke

PNEN: Pyramid Non-Local Enhanced Networks [article]

Feida Zhu, Chaowei Fang, Kai-Kuang Ma
2020 arXiv   pre-print
Additionally, the pyramid non-local block can be directly incorporated into convolution neural networks for other image restoration tasks.  ...  We integrate it into two existing methods for image denoising and single image super-resolution, achieving consistently improved performance.  ...  In this paper, we propose a deep pyramid non-local enhanced network (PNEN) for edge-preserving image smoothing.  ... 
arXiv:2008.09742v1 fatcat:nf3vtbjxxbcgtjgato2qrjaffi

Image Denoising and Ring Artifacts Removal for Spectral CT via Deep Neural Network

Xiaojie Lv, Xuezhi Ren, Peng He, Mi Zhou, Zourong Long, Xiaodong Guo, Chengyu Fan, Biao Wei, Peng Feng
2020 IEEE Access  
Then we use the data set to train our network for image denoising and ring artifacts removal.  ...  And the performance of the FCPRN is better than that of some networks for CT image denoising.  ...  For more information, see https://creativecommons.org/licenses/by/4.0/ This article has been accepted for publication in a future issue of this journal, but has not been fully edited.  ... 
doi:10.1109/access.2020.3044708 fatcat:piakw5i2knanvnuh4fhldcmioa

Transform Domain Pyramidal Dilated Convolution Networks For Restoration of Under Display Camera Images [article]

Hrishikesh P.S., Densen Puthussery, Melvin Kuriakose, Jiji C.V
2020 arXiv   pre-print
Two different networks are proposed for the restoration of images taken with two types of UDC technologies.  ...  The first method uses a pyramidal dilated convolution within a wavelet decomposed convolutional neural network for pentile-organic LED (P-OLED) based display system.  ...  The DCT domain features are then fused with the pixel domain features obtained from the parallel pixel domain branch as shown in Fig. 6 . 6 Training loss plot of (a) PDCRN for P-OLED (b) PDCRN with dual  ... 
arXiv:2009.09393v1 fatcat:agwxfj7rgjdjvexb6fbxamonji

Human Detection via Image Denoising for 5G-Enabled Intelligent Applications

Hui Li, Hang Zhou, Xiaoguo Liang, Fen Cai, Lingwei Xu, Wei Kong, Ying Guo, Junjuan Xia
2021 Wireless Communications and Mobile Computing  
and HRFPN are used to extract and fuse high-resolution features of denoised images, respectively, to obtain high-quality feature representation, and finally, a cascaded object detector is used for classification  ...  A number of video images will be compressed for efficient transmission; the resulting incomplete feature representation of images will drop the human detection performance.  ...  Acknowledgments The work is supported by the Opening Foundation of Fujian Provincial Key Laboratory of Data Intensive Computing under Grant BD202001 and Opening Foundation of Key Laboratory of Computer Network  ... 
doi:10.1155/2021/5344890 fatcat:5ritk5ft4jazxk3gpchjnuot5m

ERNet Family: Hardware-Oriented CNN Models for Computational Imaging Using Block-Based Inference [article]

Chao-Tsung Huang
2020 arXiv   pre-print
The results also show that, for block-based inference, ERNet can outperform the state-of-the-art FFDNet and EDSR-baseline models for image denoising and super-resolution respectively.  ...  Convolutional neural networks (CNNs) demand huge DRAM bandwidth for computational imaging tasks, and block-based processing has recently been applied to greatly reduce the bandwidth.  ...  In contrast, the truncated-pyramid inference flow in [17] proposes to recompute these features for saving SRAM area.  ... 
arXiv:1910.05787v2 fatcat:vdbovuj6kzgznmmz6d75wibtfq

Fault Diagnosis of Rolling Bearings Based on a Residual Dilated Pyramid Network and Full Convolutional Denoising Autoencoder

Hongmei Shi, Jingcheng Chen, Jin Si, Changchang Zheng
2020 Sensors  
In order to obtain efficient expression of data denoising feature of encoding network, time-frequency images are first input into the encoding-decoding network for unsupervised pre-training.  ...  In this regard, this paper proposes a new intelligent diagnosis algorithm for rolling bearing faults based on a residual dilated pyramid network and full convolutional denoising autoencoder (RDPN-FCDAE  ...  Figure 5 . 5 Dilated Pyramid Network (DPN). DPN obtains the multi-scale feature pyramid by exploiting multiple parallel filters with different dilated rates.  ... 
doi:10.3390/s20205734 pmid:33050210 fatcat:nqxkoxf2abfvhjcmdp56dz2zcq

eCNN

Chao-Tsung Huang, Yu-Chun Ding, Huan-Ching Wang, Chi-Wen Weng, Kai-Ping Lin, Li-Wei Wang, Li-De Chen
2019 Proceedings of the 52nd Annual IEEE/ACM International Symposium on Microarchitecture - MICRO '52  
We apply a block-based inference flow which can eliminate all the DRAM bandwidth for feature maps and accordingly propose a hardware-oriented network model, ERNet, to optimize image quality based on hardware  ...  Convolutional neural networks (CNNs) have recently demonstrated superior quality for computational imaging applications.  ...  Therefore, two specific features for computational imaging networks are not considered for optimization: 1) the spatial resolution of feature maps is not aggressively downsampled and 2) the models are  ... 
doi:10.1145/3352460.3358263 dblp:conf/micro/HuangDWWLWC19 fatcat:u3n4eq42orazrpehal6swwxu4y

New Techniques for Preserving Global Structure and Denoising with Low Information Loss in Single-Image Super-Resolution

Yijie Bei, Alex Damian, Shijia Hu, Sachit Menon, Nikhil Ravi, Cynthia Rudin
2018 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
Furthermore, we present new neural network architectures that specifically address the two challenges listed above: denoising and preservation of large-scale structure.  ...  This work identifies and addresses two important technical challenges in single-image super-resolution: (1) how to upsample an image without magnifying noise and (2) how to preserve large scale structure  ...  The original image is downsampled several times, with each downsampled image being fed through a parallel superresolution network.  ... 
doi:10.1109/cvprw.2018.00132 dblp:conf/cvpr/BeiDHMRR18 fatcat:xx2mnqdkmffzvaiep6rsfwdnu4

New Techniques for Preserving Global Structure and Denoising with Low Information Loss in Single-Image Super-Resolution [article]

Yijie Bei, Alex Damian, Shijia Hu, Sachit Menon, Nikhil Ravi, Cynthia Rudin
2018 arXiv   pre-print
Furthermore, we present new neural network architectures that specifically address the two challenges listed above: denoising and preservation of large-scale structure.  ...  This work identifies and addresses two important technical challenges in single-image super-resolution: (1) how to upsample an image without magnifying noise and (2) how to preserve large scale structure  ...  Allow the residual blocks in a super-resolution network (such as EDSR) to simultaneously denoise the input images and extract features.  ... 
arXiv:1805.03383v2 fatcat:ytogw2jrnze73ifnndpwsjkuay

Fast, Trainable, Multiscale Denoising [article]

Sungjoon Choi, John Isidoro, Pascal Getreuer, Peyman Milanfar
2018 arXiv   pre-print
Denoising is a fundamental imaging problem. Versatile but fast filtering has been demanded for mobile camera systems.  ...  We consider the fixed-scale denoising filter as a building block for multiscale training and inference. We begin by taking noisy input images and forming pyramids by downsampling by factors of two.  ...  INTRODUCTION Image denoising is known as a challenging problem that has been explored for many decades.  ... 
arXiv:1802.06130v1 fatcat:rrfwxyxutzez3ojpzb3sumnxlm

A deep convolutional neural network using directional wavelets for low-dose X-ray CT reconstruction

Eunhee Kang, Junhong Min, Jong Chul Ye
2017 Medical Physics (Lancaster)  
Moreover, our CNN is designed to have various types of residual learning architecture for faster network training and better denoising.  ...  Conventional model-based denoising approaches are, however, computationally very expensive, and image domain denoising approaches hardly deal with CT specific noise patterns.  ...  The denoised images by the newly trained network with 3mm images better preserve the fine image details than those of the previous network.  ... 
doi:10.1002/mp.12344 pmid:29027238 fatcat:effgydeo45dxzagasdc7qyi77a

Exploring Inter-frequency Guidance of Image for Lightweight Gaussian Denoising [article]

Zhuang Jia
2021 arXiv   pre-print
With the convolutional neural networks showing strong capability in computer vision tasks, the performance of image denoising has also been brought up by CNN based methods.  ...  Image denoising is of vital importance in many imaging or computer vision related areas.  ...  Gaussian Denoising As for the effectiveness and simplicity for analysis, Gaussian denoising of images has been studied for a long time.  ... 
arXiv:2112.11779v1 fatcat:fnlsp23ajbhrre4owrujfoxyvq

A Novel Framework for Detection of Cervical Cancer

V. Pushpalatha
2018 Asian Journal of Engineering and Applied Technology  
Colposcopy images of cervix are analyzed in this study for the recognition of cervical cancer.  ...  This framework comprises of five phases, (i) Dual tree discrete wavelet transform to pre-process the image (ii) Curvelet transform and contour transform to enhance the image (iii) K-means for segmentation  ...  RELATED WORKS Deep features based convolutional neural network is compared with seven classic classifiers fed with hand crafted pyramid features for classification of cervigram images in [3] and it is  ... 
doi:10.51983/ajeat-2018.7.2.1016 fatcat:5hp7pwl37bdlxanzxp5bga4twq
« Previous Showing results 1 — 15 out of 1,276 results