Deep residual deconvolutional networks for defocus blur detection
Abstract Accurate defocus blur detection has instigated wide research interest for the last few years. However, it is still a meaningful yet challenging machine vision task, and most methods rely on prior knowledge. Convolutional neural networks have proved the huge success for different tasks withi...
| Main Authors: | , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2021-02-01
|
| Series: | IET Image Processing |
| Subjects: | |
| Online Access: | https://doi.org/10.1049/ipr2.12057 |
| _version_ | 1811344406900375552 |
|---|---|
| author | Kai Zeng Yaonan Wang Jianxu Mao Xianen Zhou |
| author_facet | Kai Zeng Yaonan Wang Jianxu Mao Xianen Zhou |
| author_sort | Kai Zeng |
| collection | DOAJ |
| description | Abstract Accurate defocus blur detection has instigated wide research interest for the last few years. However, it is still a meaningful yet challenging machine vision task, and most methods rely on prior knowledge. Convolutional neural networks have proved the huge success for different tasks within the computer vision, and machine learning flew. A simple yet effective method of defocus blur detection was proposed in this paper, which by applying the deep residual convolutional encoder‐decoder network. The aims of DRDN is to automatically generate pixel‐level predictions for defocus blur images, and reconstruct output detection results of the same size as the input, which by performing several deconvolution operations at multiple scales through the transposed convolution, and skip connection. Afterwards, we used the slide window detection strategy and traversed the input image with a certain stride. Experiments on challenging benchmarks of defocus blur detection show that our algorithm achieved state‐of‐the‐art performance, and powerfully balanced the detection accuracy, and detection time. |
| first_indexed | 2024-04-13T19:46:51Z |
| format | Article |
| id | doaj.art-0592a3a2281e406b83061bfa4d4239ed |
| institution | Directory Open Access Journal |
| issn | 1751-9659 1751-9667 |
| language | English |
| last_indexed | 2024-04-13T19:46:51Z |
| publishDate | 2021-02-01 |
| publisher | Wiley |
| record_format | Article |
| series | IET Image Processing |
| spelling | doaj.art-0592a3a2281e406b83061bfa4d4239ed2022-12-22T02:32:41ZengWileyIET Image Processing1751-96591751-96672021-02-0115372473410.1049/ipr2.12057Deep residual deconvolutional networks for defocus blur detectionKai Zeng0Yaonan Wang1Jianxu Mao2Xianen Zhou3The School of Electrical and Information Engineering Hunan University ChinaThe School of Electrical and Information Engineering Hunan University ChinaThe School of Electrical and Information Engineering Hunan University ChinaThe School of Electrical and Information Engineering Hunan University ChinaAbstract Accurate defocus blur detection has instigated wide research interest for the last few years. However, it is still a meaningful yet challenging machine vision task, and most methods rely on prior knowledge. Convolutional neural networks have proved the huge success for different tasks within the computer vision, and machine learning flew. A simple yet effective method of defocus blur detection was proposed in this paper, which by applying the deep residual convolutional encoder‐decoder network. The aims of DRDN is to automatically generate pixel‐level predictions for defocus blur images, and reconstruct output detection results of the same size as the input, which by performing several deconvolution operations at multiple scales through the transposed convolution, and skip connection. Afterwards, we used the slide window detection strategy and traversed the input image with a certain stride. Experiments on challenging benchmarks of defocus blur detection show that our algorithm achieved state‐of‐the‐art performance, and powerfully balanced the detection accuracy, and detection time.https://doi.org/10.1049/ipr2.12057Optical, image and video signal processingComputer vision and image processing techniquesNeural nets |
| spellingShingle | Kai Zeng Yaonan Wang Jianxu Mao Xianen Zhou Deep residual deconvolutional networks for defocus blur detection IET Image Processing Optical, image and video signal processing Computer vision and image processing techniques Neural nets |
| title | Deep residual deconvolutional networks for defocus blur detection |
| title_full | Deep residual deconvolutional networks for defocus blur detection |
| title_fullStr | Deep residual deconvolutional networks for defocus blur detection |
| title_full_unstemmed | Deep residual deconvolutional networks for defocus blur detection |
| title_short | Deep residual deconvolutional networks for defocus blur detection |
| title_sort | deep residual deconvolutional networks for defocus blur detection |
| topic | Optical, image and video signal processing Computer vision and image processing techniques Neural nets |
| url | https://doi.org/10.1049/ipr2.12057 |
| work_keys_str_mv | AT kaizeng deepresidualdeconvolutionalnetworksfordefocusblurdetection AT yaonanwang deepresidualdeconvolutionalnetworksfordefocusblurdetection AT jianxumao deepresidualdeconvolutionalnetworksfordefocusblurdetection AT xianenzhou deepresidualdeconvolutionalnetworksfordefocusblurdetection |