中国科学院机构知识库网格系统: 检索

浏览/检索结果: 共6条，第1-6条

帮助

条数/页：排序方式：
	Degradation regression with uncertainty for blind super-resolution 期刊论文 OAI收割 NEUROCOMPUTING, 2024, 卷号: 582, 页码: 14 作者: Li, Shang; Zhang, Guixuan; Luo, Zhengxiong; Liu, Jie; Zeng, Zhi \| 收藏 \| 浏览/下载：38/0 \| 提交时间：2024/07/03 Blind super-resolution Degradation estimation Uncertainty learning
	End-to-End Alternating Optimization for Real-World Blind Super Resolution 期刊论文 OAI收割 INTERNATIONAL JOURNAL OF COMPUTER VISION, 2023, 页码: 18 作者: Luo, Zhengxiong; Huang, Yan; Li, Shang; Wang, Liang; Tan, Tieniu \| 收藏 \| 浏览/下载：16/0 \| 提交时间：2023/11/17 Blind super resolution Degradation estimation Alternating optimization Restorer Estimator
	Narrowing the Gap: Improved Detector Training With Noisy Location Annotations 期刊论文 OAI收割 IEEE TRANSACTIONS ON IMAGE PROCESSING, 2022, 卷号: 31, 页码: 6369-6380 作者: Wang, Shaoru; Gao, Jin; Li, Bing; Hu, Weiming \| 收藏 \| 浏览/下载：33/0 \| 提交时间：2022/11/14 Annotations Noise measurement Detectors Task analysis Training Object detection Degradation Object detection noisy label Bayesian estimation teacher-student learning
	Baselines Extraction from Curved Document Images via Slope Fields Recovery 期刊论文 OAI收割 IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, 2020, 卷号: 42, 期号: 4, 页码: 793-808 作者: Meng, Gaofeng; Pan, Chunhong; Xiang, Shiming; Wu, Ying \| 收藏 \| 浏览/下载：39/0 \| 提交时间：2020/06/02 Estimation Image segmentation Layout Distortion Strips Image quality Degradation Document image processing curved baselines extraction slope fields recovery geometric distortion rectification
	Remote sensing image restoration using estimated point spread function (EI CONFERENCE) 会议论文 OAI收割 2010 International Conference on Information, Networking and Automation, ICINA 2010, October 17, 2010 - October 19, 2010, Kunming, China 作者: Yang L.; Yang L. 收藏 \| 浏览/下载：33/0 \| 提交时间：2013/03/25 In order to reduce image blur caused by the degradation phenomenon in the imaging process the acquired images of the space remote sensing camera are restored. First the frequency-domain notch filter is adopted to remove strip noises in the images. Then degradation function which is referred to as the point spread function (PSF) of the imaging system is estimated using the knife-edge method. To improve the accuracy of the estimation the estimated PSF is adjusted with Gaussian fitting. Finally the images are restored by Wiener filtering with the fitted PSF. The restoration results of the remote sensing images show that almost all strip noises are eliminated by the notch filter. After denoising and restoration the variance of the remote sensing image worked with in this paper increases 30.979 and the gray mean gradient increases 3.312. Due to Gaussian fitting the accuracy of the PSF estimation is heightened. Image restoration with the final PSF is benefit to interpreting and analyzing the remote sensing images. After restoration the contrasts of the restored images are increased and the visual effects become clearer. 2010 IEEE.
	Estimation algorithm of mobile OFDM CFOs based on step-size (EI CONFERENCE) 会议论文 OAI收割 2009 International Conference on Information Engineering and Computer Science, ICIECS 2009, December 19, 2009 - December 20, 2009, Wuhan, China Lin L.; Qiao Y.-F.; Su W.-X. 收藏 \| 浏览/下载：26/0 \| 提交时间：2013/03/25 In order to maintain orthogonality among subcarriers for mobile OFDM a CFOs step-size estimation model is established to avoid severe performance degradation caused by ICI and algorithmic stability step-size parameter the error of CFOs estimate and etc are investigated. First after an averaging window algorithm is introduced the mathematic model of step-size estimation algorithm is obtained through step-size parameter and time-varying CFOs. Then based on the results of MATLAB simulation the determination of step-size parameter the advantages of step-size algorithm and precision are discussed. Finally time sequence plots of added CP interleaver and subcarrier modulation with VHDL are presented. Experimental results indicate that the maximum of CFOs errors is 0.00987 in 3 variance interval when mobile velocity is 400 km/h according with verdicted principle that in order for the ICI effects to be negligible the error of CFOs estimation must be accurate within 1-2% of the subcarrier spacing and programs can correctly implement on circuit board. 2009 IEEE.