检索结果-内蒙古大学图书馆

Learning Weighting Map for Bit-Depth Expansion within a Rational Range

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Liu, Yuqing Jia, Qi Zhang, Jian Fan, Xin Wang, Shanshe Ma, Siwei Gao, Wen The School of Software Dalian University of Technology Dalian116620 China International School of Information Science and Engineering Dalian University of Technology Dalian116620 China The School of Electronic and Computer Engineering Peking University Shenzhen Graduate School Shenzhen China The School of Electronics Engineering and Computer Science Institute of Digital Media Peking University Beijing100871 China

Bit-depth expansion (BDE) is one of the emerging technologies to display high bit-depth (HBD) image from low bit-depth (LBD) source. Existing BDE methods have no unified solution for various BDE situations, and directly learn a mapping for each pixel from LBD image to the desired value in HBD image, which may change the given high-order bits and lead to a huge deviation from the ground truth. In this paper, we design a bit restoration network (BRNet) to learn a weight for each pixel, which indicates the ratio of the replenished value within a rational range, invoking an accurate solution without modifying the given high-order bit information. To make the network adaptive for any bit-depth degradation, we investigate the issue in an optimization perspective and train the network under progressive training strategy for better performance. Moreover, we employ Wasserstein distance as a visual quality indicator to evaluate the difference of color distribution between restored image and the ground truth. Experimental results show our method can restore colorful images with fewer artifacts and false contours, and outperforms state-of-the-art methods with higher PSNR/SSIM results and lower Wasserstein distance. The source code will be made available at https://***/yuqing-liu-dut/ bit-depth-expansion Copyright © 2022, The Authors. All rights reserved.

关键词： Image reconstruction

Design and Development An Low Cost Intelligent Irrigation System in Bamboo Farm 10

学校读者我要写书评

暂无评论

Design and Development An Low Cost Intelligent Irrigation Sy...

10th IEEE Global Conference on Consumer Electronics, GCCE 2021

作者： Yen, Ping-Yi Fanjiang, Yong-Yi Hung, Chi-Huang Jun-Bin, Lai Ku Pao Home Economics Commercial High School Graduate Institute of Applied Science And Engineering Fu Jen Catholic University Department of Data Processing New Taipei City Taiwan Fu Jen Catholic University Graduate Institute of Applied Science And Engineering Dept. of Computer Science And Information Engineering New Taipei City Taiwan Fu Jen Catholic University Lee-Ming Institute of Technology Graduate Institute of Applied Science And Engineering Dept. of Information Technology New Taipei City Taiwan Nation University of Business Department of Information Management Taipei City Taiwan

ISBN: (纸本)9781665436762

Smart agriculture has not been an unreachable technology in recent years. From the lighting control during cultivation, temperature and humidity control and other technologies have become quite easy. However, most of these technologies be used in greenhouses, and they are not as simple as expected in an open environment. In this research, a smart irrigation control performed in a bamboo farm. The water content of the soil understood through a sensing module and transmitted back to the central controller. Farmers do not necessarily have to open the water tank to discharge water as before, and then close the water tank based on experience. In order to achieve a simple and intelligent agricultural irrigation system. © 2021 IEEE.

关键词： Water tanks

Classification for Glass Bottles Based on Improved Selective Search Algorithm

学校读者我要写书评

暂无评论

computers, Materials & Continua 2020年第7期64卷 233-251页

作者： Shuqiang Guo Baohai Yue Manyang Gao Xinxin Zhou Bo Wang School of Computer Science Northeast Electric Power UniversityJilin132000China Information and Communication Company State Grid Tianjin Electric Power CompanyTianjin300000China Graduate School of Science and Engineering Iwate UniversityMorioka020-8550Japan

The recycling of glass bottles can reduce the consumption of resources and contribute to environmental *** present,the classification of recycled glass bottles is difficult due to the many differences in specifications and *** paper proposes a classification algorithm for glass bottles that is divided into two stages,namely the extraction of candidate regions and the classification of *** the candidate region extraction stage,aiming at the problem of the large time overhead caused by the use of the SIFT(scale-invariant feature transform)descriptor in SS(selective search),an improved feature of HLSN(Haar-like based on SPP-Net)is *** integral graph is introduced to accelerate the process of forming an HBSN vector,which overcomes the problem of repeated texture feature calculation in overlapping regions by *** the classification stage,the improved SS algorithm is used to extract target *** target regions are merged using a non-maximum suppression algorithm according to the classification scores of the respective regions,and the merged regions are classified using the trained *** demonstrate that,compared with the original SS,the improved SS algorithm increases the calculation speed by 13.8%,and its classification accuracy is 89.4%.Additionally,the classification algorithm for glass bottles has a certain resistance to noise.

关键词： Classification of glass bottle HBSN feature improved selective search algorithm LightGBM

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Liu, Yuqing Jia, Qi Zhang, Jian Fan, Xin Wang, Shanshe Ma, Siwei Gao, Wen School of Software Dalian University of Technology Dalian116620 China International School of Information Science and Engineering Dalian University of Technology Dalian116620 China School of Electronic and Computer Engineering Peking University Shenzhen Graduate School Shenzhen China School of Electronics Engineering and Computer Science Institute of Digital Media Peking University Beijing100871 China

As a highly ill-posed issue, single image super-resolution (SISR) has been widely investigated in recent years. The main task of SISR is to recover the information loss caused by the degradation procedure. According to the Nyquist sampling theory, the degradation leads to aliasing effect and makes it hard to restore the correct textures from low-resolution (LR) images. In practice, there are correlations and self-similarities among the adjacent patches in the natural images. This paper considers the self-similarity and proposes a hierarchical image super-resolution network (HSRNet) to suppress the influence of aliasing. We consider the SISR issue in the optimization perspective, and propose an iterative solution pattern based on the half-quadratic splitting (HQS) method. To explore the texture with local image prior, we design a hierarchical exploration block (HEB) and progressive increase the receptive field. Furthermore, multi-level spatial attention (MSA) is devised to obtain the relations of adjacent feature and enhance the high-frequency information, which acts as a crucial role for visual experience. Experimental result shows HSRNet achieves better quantitative and visual performance than other works, and remits the aliasing more effectively. Copyright © 2022, The Authors. All rights reserved.

关键词： Textures

Improving Generalization of Metric Learning via Listwise Self-distillation

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Zeng, Zelong Yang, Fan Wang, Zheng Satoh, Shin'ichi The Department of Information and Communication Engineering Graduate School of Information Science and Technology The University of Tokyo Japan The Digital Content and Media Sciences Research Division National Institute of Informatics Japan The School of Computer Science National Engineering Research Center for Multimedia Software Wuhan University China

Most deep metric learning (DML) methods employ a strategy that forces all positive samples to be close in the embedding space while keeping them away from negative ones. However, such a strategy ignores the internal relationships of positive (negative) samples and often leads to overfitting, especially in the presence of hard samples and mislabeled samples. In this work, we propose a simple yet effective regularization, namely Listwise Self-Distillation (LSD), which progressively distills a model's own knowledge to adaptively assign a more appropriate distance target to each sample pair in a batch. LSD encourages smoother embeddings and information mining within positive (negative) samples as a way to mitigate overfitting and thus improve generalization. Our LSD can be directly integrated into general DML frameworks. Extensive experiments show that LSD consistently boosts the performance of various metric learning methods on multiple datasets. Copyright © 2022, The Authors. All rights reserved.

关键词： Distillation

Covariation and Constant Modulus Decomposition Based Interference Resistant Access System in Smart Grid

学校读者我要写书评

暂无评论

Covariation and Constant Modulus Decomposition Based Interfe...

IEEE Conference on Vehicular Technology (VTC)

作者： Yuan Zhang Dongyang Xu Pinyi Ren James A. Ritcey Keping Yu Joel J. P. C. Rodrigues School of Information and Communications Engineering Xi'an Jiaotong University Xi'an China National Mobile Communications Research Laboratory Southeast University China Department of Electrical and Computer Engineering University of Washington Seattle WA USA Graduate School of Science and Engineering Hosei University Tokyo Japan College of Computer Science and Technology China University of Petroleum (East China) China Instituto de Telecomunicações Covilhã Portugal

The reduced-capability new radio (NR RedCap) was introduced in 3GPP Rel-17 to cater to the use cases that are not yet best served by current NR specifications, such as smart grid and industrial wireless sensors. For the grant-free access system in smart grid, the resistance to impulse noise is a key issue. By using fractional low-order covariance and constant modulus based tensor decomposition, this paper skillfully enables user identification in this scenario while suppressing the effect of impulse noise. The proposed scheme uses spread spectrum signal as the pilot signal. And the user identity is represented jointly by the spread spectrum sequence and information codes. In this condition, we start by transforming the pilot signals into a tensor. The fractional low-order covariance is then used to suppress the impulse noise, and the constant modulus is used to improve the performance of the algorithm during the iterative process of tensor decomposition. Finally the sensor identity is confirmed by the decomposition result. Simulation results show that the proposed scheme can greatly improve the performance of user identification under impulse noise channel. Specifically, the identification rate of the proposed algorithm valued 99.815% outperformed that of AMP valued 89.1471% when generalized signal-to-noise ratio GSNR = 0 dB. In addition, the proposed scheme can also correctly estimate the channel gain from the sensors to the base station in impulsive noise environment.

关键词：

Relationship between number of annotations and accuracy in segmentation of the erector spinae muscle using Bayesian U-Net in torso CT images

学校读者我要写书评

暂无评论

Relationship between number of annotations and accuracy in s...

International Forum on Medical Imaging in Asia 2021, IFMIA 2021

作者： Wakamatsu, Yuichi Kamiya, Naoki Zhou, Xiangrong Hara, Takeshi Fujita, Hiroshi Department of Media Informatics Graduate School of Information Science and Technology Aichi Prefectural University 1522-3 Ibaragabasama Nagakute-shi Aichi480-1198 Japan Department of Information Systems Graduate School of Information Science and Technology Aichi Prefectural University 1522-3 Ibaragabasama Nagakute-shi Aichi480-1198 Japan Department of Electrical Electronic and Computer Engineering Faculty of Engineering Gifu University 1-1 Yanagido Gifu-shi Gifu501-1193 Japan

ISBN: (数字)9781510644212

ISBN: (纸本)9781510644205

Supervised learning for image segmentation requires annotated images. However, image annotation has the problem that it is time-consuming. This problem is particularly significant in the erector spinae muscle segmentation due to the large size of the muscle. Therefore, this study considers the relationship between the number of annotated images used for training and segmentation accuracy of the erector spinae muscle in torso CT images. We use Bayesian U-Net, which has shown high accuracy in thigh muscle segmentation, for the segmentation of the erector spinae muscle. In the network training, we limit the number of slices for each case and the number of cases to 100%, 50%, 25%, and 10%. In the experiment, we use 30 torso CT images, including 6 cases for the test dataset. Experimental results are evaluated by the mean Dice value of the test dataset. Using 100% of the slices per case, the segmentation accuracy with 100%, 50%, 25%, and 10% of the cases was 0.934, 0.927, 0.926, and 0.890, respectively. On the other hand, using 100% of the cases, the segmentation accuracy with 100%, 50%, 25%, and 10% of the slices per case was 0.934, 0.934, 0.933, and 0.931, respectively. Furthermore, the segmentation accuracy with 100% of the cases and 10% of the slices per case was higher than that of the previous method. We showed that it is feasible to achieve high segmentation accuracy with a limited number of annotated images by selecting several slices from a limited number of cases for training. © 2021 SPIE.

关键词： Image segmentation

Geo-Localization via Ground-to-Satellite Cross-View Image Retrieval

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Zeng, Zelong Wang, Zheng Yang, Fan Satoh, Shin'ichi The School of Computer Science National Engineering Research Center for Multimedia Software Wuhan University China The Department of Information and Communication Engineering Graduate School of Information Science and Technology The University of Tokyo Japan The Digital Content and Media Sciences Research Division National Institute of Informatics Japan

The large variation of viewpoint and irrelevant content around the target always hinder accurate image retrieval and its subsequent tasks. In this paper, we investigate an extremely challenging task: given a ground-view image of a landmark, we aim to achieve cross-view geo-localization by searching out its corresponding satellite-view images. Specifically, the challenge comes from the gap between ground-view and satellite-view, which includes not only large viewpoint changes (some parts of the landmark may be invisible from front view to top view) but also highly irrelevant background (the target landmark tend to be hidden in other surrounding buildings), making it difficult to learn a common representation or a suitable mapping. To address this issue, we take advantage of drone-view information as a bridge between ground-view and satellite-view domains. We propose a Peer Learning and Cross Diffusion (PLCD) framework. PLCD consists of three parts: 1) a peer learning across ground-view and drone-view to find visible parts to benefit ground-drone cross-view representation learning;2) a patch-based network for satellite-drone cross-view representation learning;3) a cross diffusion between ground-drone space and satellite-drone space. Extensive experiments conducted on the University-Earth and University-Google datasets show that our method outperforms state-of-the-arts significantly. Copyright © 2022, The Authors. All rights reserved.

关键词： Satellites

Estimation of PN Sequence for Spread Spectrum Pilot Signals in Grant-Free Access System

学校读者我要写书评

暂无评论

Estimation of PN Sequence for Spread Spectrum Pilot Signals ...

IEEE Conference on Vehicular Technology (VTC)

作者： Yuan Zhang Dongyang Xu Pinyi Ren James A. Ritcey Keping Yu Joel J. P. C. Rodrigues School of Information and Communications Engineering Xi’an Jiaotong University Xi’an China National Mobile Communications Research Laboratory Southeast University China Department of Electrical and Computer Engineering University of Washington Seattle WA USA Graduate School of Science and Engineering Hosei University Tokyo Japan College of Computer Science and Technology China University of Petroleum (East China) Qingdao China Instituto de Telecomunicações Covilhã Portugal

For the grant-free random access system in the Internet of Thing (IoT) scenario, the recovery of the pilot sequence and the identification of the IoT device is a crucial issue. Contrapose the problem that the existing grant-free access schemes cannot accurately recover the pilot sequence in the intensive industrial zone with impulse noise, this paper proposes to use spread spectrum signal as pilot signal and proposes an estimation algorithm based on joint k-means and M estimation accordingly. This algorithm dynamically suppresses the influence of noise with adaptive weighted function according to the estimated noise energy in the iterative process. First, the received signal is segmented to obtain samples. Second, the samples are clustered using the K-means algorithm. In the iterative process of the algorithm, cluster centers are used to estimate the energy of signal noise. According to the estimation result of the noise energy, the adaptive weighted function is used to dynamically update the cluster centers and the similarity between samples and cluster centers. Finally, assigning +1 or −1 to the samples according to the clustering results, and then the estimation of pseudo-code sequence (PN sequence) is realized while impulse noise is suppressed. Simulation results show that the proposed algorithm can greatly improve the performance of PN sequence estimation under impulse noise channel. The bit error ratio (BER) of the proposed algorithm valued 0.008 outperformed that of EVD valued 0.3 when the generalized signal-to-noise ratio (GSNR) is −4dB. In particular, the proposed algorithm has better performance when the noise distribution has heavier tails, which is different from traditional algorithms.

关键词：