检索结果-内蒙古大学图书馆

Institution of Engineering and Technology International Communication Conference on Wireless Mobile and Computing

作者： Liu, Yi Wang, Yongfang Zhang, Zhaoyang School of Communication and Information Engineering Shanghai University Shanghai 200072 China Key Laboratory of Advanced Display and System Application Ministry of Education Shanghai 200072 China

ISBN: (纸本)9781849191388

Color differences between different views can cause performance degradation in multiview video coding. In order to overcome this problem, in this paper a local color correction method based on histogram segmentation is proposed. In the proposed method, we correct the color changes in the different segmentation regions between the current picture and the reference picture. Extensive experimental results demonstrate that the proposed method outperforms the traditional global linear color correction methods with objective mean of disparity values and subjective visual quality, and achieves better correction results.

关键词： Color

来源：评论

学校读者我要写书评

暂无评论

A Generic Image Feature Extraction Method Towards Multiple Vision Tasks 9th

A Generic Image Feature Extraction Method Towards Multiple V...

引用

9th International Forum on Digital Multimedia Communication, IFTC 2022

作者： Wei, Jiao An, Ping Huang, Kunqiang Yang, Chao Ma, Ran School of Communication and Information Engineering Shanghai University Shanghai200444 China Key Laboratory for Advanced Display and System Application Ministry of Education Shanghai200072 China

ISBN: (纸本)9789819908554

With the developments of artificial intelligence and image processing technology, the study of multiple vision tasks has become a hot topic in the field of computer vision, and the key to implementing vision tasks lies in the feature extraction of images. To meet the demands of modern life for multiple vision tasks and promote the deployment of artificial intelligence on edge devices, we propose an image feature extraction method applied to multiple vision tasks, which promotes the deployment of artificial intelligence on edge devices. Firstly, the function of the secondary task in multi-task learning is combined with a feature migration strategy to build a feature migration model for multiple vision tasks. Then, a generic feature extraction model based on feature fusion and feature migration is established under the guidance of image classification. The experimental results show that the proposed feature migration model with image classification as the secondary task can improve the mAP (mean average precision) of the target detection by 5.5%, and the feature migration model with target detection as the secondary task can improve the average accuracy of the image classification by 1%. The proposed generic feature extraction model can effectively implement the image classification and target detection tasks. © 2023, The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.

关键词： Feature extraction

来源：评论

学校读者我要写书评

暂无评论

Depth map upsampling using segmentation and edge information 1

引用

8th International Conference on Image and Graphics, ICIG 2015

作者： Zheng, Shuai An, Ping Zuo, Yifan Zou, Xuemei Wang, Jianxin School of Communication and Information Engineering Shanghai University Shanghai200072 China Key Laboratory of Advanced Display and System Application of the Ministry of Education Shanghai200072 China

ISBN: (数字)9783319219639

ISBN: (纸本)9783319219622

A depth upsampling method based on Markov Random Field is proposed, considering the depth and color information. First, the initial interpolated depth map is inaccurate and oversmooth, we use a rectangle window centered on every pixel to search the maximum and minimum depth value of the depth map to find out the edge pixels. Then, we use the depth information to guide the segmentation of the color image and build different data terms and smoothness terms for the edge and non-edge pixels. The result depth map is piecewise smooth and the edge is sharp. In the meanwhile, the result is good where the color information is consistent while the depth is not or where the depth information is consistent while the color is not. Experiments show that the proposed method performs better than other upsampling methods in terms of mean absolute difference (MAD). © Springer International Publishing Switzerland 2015.

关键词： Color

来源：评论

学校读者我要写书评

暂无评论

Multi-depth Hologram Generation with Unsupervised-learning Based Computer-generated Holography

Multi-depth Hologram Generation with Unsupervised-learning B...

引用

Digital Holography and Three-Dimensional Imaging, DH 2022

作者： Shui, Xinghua Zheng, Huadong Department of Precision Mechanical Engineering Shanghai University Shanghai200444 China Key Laboratory of Advanced Display and System Application Ministry of Education Shanghai200072 China

Unsupervised-learning based computer-generated holography provides an approach for 2D hologram generation. We propose an unsupervised learning network for multi-depth hologram generation with fully utilizing the diffe... 详细信息

ISBN: (纸本)9781957171128

关键词： Holograms

来源：评论

学校读者我要写书评

暂无评论

Optimizing structure for constructing a highly efficient inverted top-emitting organic light-emitting diode with stable electroluminescent spectra

引用

Journal of Semiconductors 2014年第2期35卷 24-27页

作者：张小文许积文王华魏斌曾华荣蒋雪茵张志林 Guangxi Key Laboratory of Information Materials Guilin University of Electronic Technology Key Laboratory of Advanced Display and System Applications Ministry of EducationShanghai University Key Laboratory of Inorganic Functional Materials and Devices Shanghai Institute of CeramicsChinese Academy of Sciences

We demonstrate a highly efficient inverted top-emitting organic light-emitting diode （TOLED） having stable electroluminescent spectra and color coordination with variation of viewing angles by simply tuning the reso- nance wavelength corresponding to the free emission of the emitter. Using a doped fluorescent emitting system, the inverted TOLED exhibits an enhanced maximum current efficiency of 19 cd/A and a power efficiency of 17 lm/W, which are much higher than those （11 cd/A and 5 lm/W） of the counterpart with normal structure, although both TOLEDs behave with similar stable electroluminescent spectra characteristics. The results indicate that we provide a simple and effective method of constructing an excellent inverted TOLED for potentially practical applications.

关键词： OLED inverted structure stable electroluminescent spectrum microcavity effect

来源：评论

学校读者我要写书评

暂无评论

Modulation waveform cycle normalization based unipolar SPWM controller

引用

2013 3rd International Symposium on Chemical Engineering and Material Properties, ISCEMP 2013

作者： Jiang, Chao Hu, Yue Li Zhang, Ke Ke Cao, Yang Key Laboratory of Advanced Display and System Application Shanghai University Shanghai 200072 China School of Mechatronic Engineering and Automation Shanghai University Shanghai 200072 China

ISBN: (纸本)9783037858363

The thesis mainly proposed a novel design method for Sinusoidal Pulse Width Modulation (SPWM) controller based on modulation waveform cycle normalization. T is the period of sinusoidal modulation waveform, and T/n is the cycle of triangular carrier. Shift registers and adders of SPWM controller were designed to translate the 2n numerical value for SPWM waveform on-off switching points, which were calculated offline, to control the level of the outpout SPWM waveform,which is the key point of modulation waveform cycle normalization, hi the thesis RTL function simulation verification by ModelSim and testing for FPGA have been achieved successfully. The results show that the novel SPWM controller is reliable, which has practical significance. © (2013) Trans Tech Publications, Switzerland.

关键词： Shift registers

来源：评论

学校读者我要写书评

暂无评论

Performance analysis of region of interest video transmission system

Performance analysis of region of interest video transmissio...

引用

IET Conference on Wireless, Mobile and Sensor Networks 2007, CCWMSN'07

作者： Wang, Yongfang Zhang, Zhaoyang Liu, Zhi School of Communication and Information Engineering Shanghai University Shanghai 200072 China Key Laboratory of Advanced Display and System Application Ministry of Education Shanghai 200072 China

ISBN: (纸本)9780863418365

We investigate the video transmission efficiency of channel code, modulation mode and layered space time code for region of interest video transmission system. Excellent channel code achieves more coding gains, suitable modulation mode adaptively overcomes different channel fading and layered space time code achieves not only coding gain but only diversity gain. Based on the above analysis, we propose an adaptive video transmission system according to channel environment and video constructed quality require. We derive an analytic bound for the error probability, and conduct experiments for H.264 interest of region video streaming. It is shown that the proposed system enhances the PSNR performance of region of interest significantly by adaptive choice of channel code, modulation mode, and layered space time code in the wireless channel.

关键词： Modulation

来源：评论

学校读者我要写书评

暂无评论

View-specific based error concealment scheme for multi-view plus depth video 13th

View-specific based error concealment scheme for multi-view ...

引用

13th International Forum of Digital TV and Wireless Multimedia Communication, IFTC 2016

作者： Ma, Ran Hu, Xiangyu Liu, Deyang Hou, Yu An, Ping School of Communication and Information Engineering Shanghai University Shanghai200072 China Key Laboratory of Advanced Display and System Application Ministry of Education Shanghai200072 China

ISBN: (纸本)9789811042102

Multi-view video plus depth (MVD) is an efficient format of 3D video. MVD video can be encoded by either H.264/AVC or HEVC standard to gain higher compression ratio, which benefits their broadcasting over the Internet. However, the encoded and transferred MVD video tends to develop worse visual quality degradation caused by lossy network channel. Therefore, error concealment is here to help refrain this problem. In this paper, we propose a classification-related error concealment method for MVD video. Within our method, motion-based classification is used to judge whether the corrupted blocks are static or not. If so, the blocks from reference frames are adopted to conceal the corrupted blocks. Otherwise, view-specific based concealment procedures, which are designed in accordance with view features, are used to conceal the corrupted blocks in different views. Experimental results on AVC-based Test Model (ATM) show the superiority of this concealment scheme to several other error concealment methods in PSNR along with acceptable execution time increase. © Springer Nature Singapore Pte Ltd. 2017.

关键词： Errors

来源：评论

学校读者我要写书评

暂无评论

Thick films of polymer Direct-conversion X-ray detectors 23

Thick films of polymer Direct-conversion X-ray detectors

引用

23rd IEEE Electronics Packaging Technology Conference, EPTC 2021

作者： Chen, Hu Li, Yi Hao, Zhao Wu, Xingyang Zhang, Jianhua Shanghai University Key Laboratory of Advanced Display and System Application Ministry of Education Shanghai200040 China Shanghai University School of Microelectronics Shanghai201800 China

ISBN: (纸本)9781665416191

Organic conjugated polymers have been widely used in the field of X-ray detectors because of their simple processing, mechanical flexibility and low cost. In this paper, we report a vertical structure X-ray detector using PDMS material which is fabricated by simple process steps. The response of pure PDMS material to X-ray is verified, and the effects of different process conditions on the device performance were tested. The detector with film forming speed 2000 rpm and curing temperature 60C still has high sensitivity left (434.5 mu {CGy}_{ {air }}{-1} {cm}{-2}right) to low dose rate left(23.9 mu {Gy}_{ {air }}{S}{-1}right) when the applied bias voltage is 150 V. Finally, the device provides an idea for the development of low cost and high performance flexible direct X-ray detectors. © 2021 IEEE.

关键词： X ray detectors

来源：评论

学校读者我要写书评

暂无评论

A new hybrid error concealment for multi-view video

A new hybrid error concealment for multi-view video

引用

IEEE China Summit and International Conference on Signal and Information Processing, ChinaSIP 2015

作者： Yu, Hou Ran, Ma Panpan, Li Ping, An Chao, Yang School of Communication and Information Engineering Shanghai University Shanghai200072 China Key Laboratory of Advanced Display and System Application of the Ministry of Education Shanghai200072 China

ISBN: (纸本)9781479919482

The paper proposes a new hybrid error concealment for the multi-view video (MVV). Firstly, the lost macroblocks (MBs) are classified into the static MBs and the motive MBs by the motion information. Then the static MBs are repaired by image inpainting, which is improved to apply to inter and inter-view frame. The motive MBs are fixed by outer boundary matching algorithm (OBMA), in which candidate motion vector (MV) set is also expanded to inter and interview frame. The experimental results show that the hybrid algorithm can conceal the lost MBs with lower complexity and improve the subjective and objective quality, compared with traditional concealment algorithm. © 2015 IEEE.

关键词： Image enhancement

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：