通讯机构：
[Cao Yuan] S;School of Mathematics and Computer Science, Wuhan Polytechnic University, Wuhan 430024, China<&wdkj&>Author to whom correspondence should be addressed.
关键词：
end to end;deep learning;point cloud completion;squeeze and excitation;trilinear interpolation
摘要：
<jats:p>We propose a conceptually simple, general framework and end-to-end approach to point cloud completion, entitled PCA-Net. This approach differs from the existing methods in that it does not require a “simple” network, such as multilayer perceptrons (MLPs), to generate a coarse point cloud and then a “complex” network, such as auto-encoders or transformers, to enhance local details. It can directly learn the mapping between missing and complete points, ensuring that the structure of the input missing point cloud remains unchanged while accurately predicting the complete points. This approach follows the minimalist design of U-Net. In the encoder, we encode the point clouds into point cloud blocks by iterative farthest point sampling (IFPS) and k-nearest neighbors and then extract the depth interaction features between the missing point cloud blocks by the attention mechanism. In the decoder, we introduce a new trilinear interpolation method to recover point cloud details, with the help of the coordinate space and feature space of low-resolution point clouds, and missing point cloud information. This paper also proposes a method to generate multi-view missing point cloud data using a 3D point cloud hidden point removal algorithm, so that each 3D point cloud model generates a missing point cloud through eight uniformly distributed camera poses. Experiments validate the effectiveness and superiority of PCA-Net in several challenging point cloud completion tasks, and PCA-Net also shows great versatility and robustness in real-world missing point cloud completion.</jats:p>
通讯机构：
[Shengyong Xu] C;College of Engineering, Huazhong Agricultural University, Wuhan 430070, China<&wdkj&>Key Laboratory of Agricultural Equipment for the Middle and Lower Reaches of the Yangtze River, Ministry of Agriculture, Wuhan 430070, China<&wdkj&>Shenzhen Branch, Guangdong Laboratory for Lingnan Modern Agriculture, Genome Analysis Laboratory of the Ministry of Agriculture, Agricultural Genomics Institute at Shenzhen, Chinese Academy of Agricultural Sciences, Shenzhen 518000, China<&wdkj&>Shenzhen Institute of Nutrition and Health, Huazhong Agricultural University, Shenzhen 518000, China<&wdkj&>Author to whom correspondence should be addressed.
关键词：
point cloud segmentation;point cloud completion;leaf area measurement;MIX-Net;seedlings;deep learning
摘要：
<jats:p>In this paper, a novel point cloud segmentation and completion framework is proposed to achieve high-quality leaf area measurement of melon seedlings. In particular, the input of our algorithm is the point cloud data collected by an Azure Kinect camera from the top view of the seedlings, and our method can enhance measurement accuracy from two aspects based on the acquired data. On the one hand, we propose a neighborhood space-constrained method to effectively filter out the hover points and outlier noise of the point cloud, which can enhance the quality of the point cloud data significantly. On the other hand, by leveraging the purely linear mixer mechanism, a new network named MIX-Net is developed to achieve segmentation and completion of the point cloud simultaneously. Different from previous methods that separate these two tasks, the proposed network can better balance these two tasks in a more definite and effective way, leading to satisfactory performance on these two tasks. The experimental results prove that our methods can outperform other competitors and provide more accurate measurement results. Specifically, for the seedling segmentation task, our method can obtain a 3.1% and 1.7% performance gain compared with PointNet++ and DGCNN, respectively. Meanwhile, the R2 of leaf area measurement improved from 0.87 to 0.93 and MSE decreased from 2.64 to 2.26 after leaf shading completion.</jats:p>
作者机构：
[Xu, Xiangrui; Li, Yaqin; Yuan, Cao] Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan 430023, Peoples R China.
通讯机构：
[Yuan, Cao] W;Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan 430023, Peoples R China.
关键词：
Deep neural network;ownership verification;security and privacy;serial number;watermarking
摘要：
The power of deep learning and the enormous effort and money required to build a deep learning model makes stealing them a hugely worthwhile and highly lucrative endeavor. Worse still, model theft requires little more than a high-school understanding of computer functions, which ensures a healthy and vibrant black market full of choice for any would-be pirate. As such, estimating how many neural network models are likely to be illegally reproduced and distributed in future is almost impossible. Therefore, we propose an embedded & x2018;identity bracelet & x2019; for deep neural networks that acts as proof of a model & x2019;s owner. Our solution is an extension to the existing trigger-set watermarking techniques that embeds a post-cryptographic-style serial number into the base deep neural network (DNN). Called a DNN-SN, this identifier works like an identity bracelet that proves a network & x2019;s rightful owner. Further, a novel training method based on non-related multitask learning ensures that embedding the DNN-SN does not compromise model performance. Experimental evaluations of the framework confirm that a DNN-SN can be embedded into a model when training from scratch or in the student network component of Net2Net.
会议名称：
11th International Symposium on Multispectral Image Processing and Pattern Recognition (MIPPR) - Automatic Target Recognition and Navigation
会议时间：
NOV 02-03, 2019
会议地点：
Wuhan, PEOPLES R CHINA
会议主办单位：
[Xu, Xiangrui;Li, Yaqin;Gao, Yunlong;Yuan, Cao] Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan, Hubei, Peoples R China.
会议论文集名称：
Proceedings of SPIE
关键词：
Deep neural network;identity number (ID);Ownership verification
摘要：
Amid the maturity of machine learning, deep neural networks are gradually applied in the business sector rather than be restricted in the laboratory. However, its intellectual property protection encounters a significant challenge. In this paper, we aim at embedding a unique identity number (ID) to the deep neural network for model ownership verification. To this end, a scheme of generating DNN ID is proposed, which is the criterion for model ownership verification. After embedding, the model can complete the original performance and own a unique ID of this model as well. DNN ID can only be generated by the owner to check the model authorship. We evaluate this method on MNIST. Experiment results demonstrate that the DNN ID can accurately verify the ownership of our trained model.
会议名称：
8th International Conference on Intelligent Human-Machine Systems and Cybernetics (IHMSC)
会议时间：
SEP 11-12, 2016
会议地点：
Zhejiang Univ, Hangzhou, PEOPLES R CHINA
会议主办单位：
Zhejiang Univ
会议论文集名称：
International Conference on Intelligent Human-Machine Systems and Cybernetics
关键词：
segmentation;active contour;smoothing parameter;edge information
摘要：
Active contour model is one of the most popular image segmentation frameworks. Conventional active contour model requires the empirical adjustment of smoothing parameter in energy functional and the smoothing parameter for active contour is a challenging problem. In this paper, we propose an automated adjustment method for the smoothing parameter in region-based active contour models and thus a full automated segmentation method is obtained. In proposed active contour model, the region term is the same as that in traditional region-based active contour model, whereas the prior term in traditional region-based active contour model is substituted by gray statistic of edge image on the contour. Thus the evolution of contour has the same iteration form as that in traditional region-based active contour model. But a driving force from the edge information takes the role of smoothing term and the adjustment of smoothing parameter is avoided. Experimental results show the proposed model can obtain segmentation quality comparable to the those obtained by traditional region-based active contour model, without the cumbersome trial of smoothing parameter.
作者机构：
[Li, Yaqin; Gong, Cheng; Yuan, Cao] Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan, Peoples R China.
会议名称：
6th International Conference on Electronic, Mechanical, Information and Management Society (EMIM)
会议时间：
APR 01-03, 2016
会议地点：
Shenyang, PEOPLES R CHINA
会议主办单位：
[Li, Yaqin;Yuan, Cao;Gong, Cheng] Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan, Peoples R China.
会议论文集名称：
ACSR-Advances in Comptuer Science Research
关键词：
Online-teaching;PCI;FPGA;DSP
摘要：
These online-teaching database systems are generated from online transactions, emails, videos, audios, images. They are stored in databases grow massively and become difficult to capture, form, store, manage, share, analyze and visualize via typical database software tools. The adventure of large amount of data and real-time requirements presents great challenges using traditional desktop computers. In this paper, to tackle the specific problem with a critical requirement of 10ms for whole data processing pipeline, we proposed a high performance embedded system as opposed to personal computer. Peripheral Component Interconnect (PCI) for data transferring from computer to Field Programmable Gate Array (FPGA) is proposed to solve the problem, and a custom-designed dual-port dual-channel RAM to realize simultaneous data exchange and data processing which was implemented using a 6-core Digital Signal Processor (DSP). The hardware system was designed and tested the performance of individual modules as well as the integration of them as a whole. We reported a total time using such pipeline of 7.5ms, meeting the critical requirement and demonstrating its feasibility in practical application.
作者：
Zhang, Xiaoqing;Qiu, Lan*;Qian, Qiongfen;Li, Yaqin
期刊：
Journal of Computational Information Systems,2015年11(14):5251-5258 ISSN：1553-9105
通讯作者：
Qiu, Lan
作者机构：
[Zhang, Xiaoqing; Li, Yaqin] School of Mathematics and Computer Science, Wuhan Polytechnic University, Wuhan, 430023, China;[Qiu, Lan] Library, Wuhan University, Wuhan, 430070, China;[Qian, Qiongfen] Department 4, Air Force Early Warning Academy, Wuhan, 430019, China
摘要：
With the development of information and the integration of media, it has great practical significance and research value to build a digital learning environment based on the complicated electronic circuit. However, the complicated electronic circuit in real-time need a complex and expensive technology. In order to overcome the high cost and technology, an approach was proposed for simplifying generation by approximating the excitations with rectangular pulses, triangular pulses and cosine waves which can be implemented with a moderate cost in analogical electronics. In this work, we improved a novel approach based on genetic programming, The differences between theoretical excitation signals and the approximation driving pulses, related to their excitation effects, were minimized by genetic programming. From these results, the accuracy of simulation can be improved by the new approach, the difference between theoretical complicated digital signals and the new approach is reduced. A trade off is obtained between the costs of implementation of digital processing in digital learning environments.
期刊：
Proceedings of SPIE - The International Society for Optical Engineering,2015年9814 ISSN：0277-786X
通讯作者：
Li, Yaqin
作者机构：
[Wang, Xuan] Wuhan Polytech Univ, Coll Hlth Sci & Nursing, Wuhan, Peoples R China.;[Li, Yaqin; Li, Shigao] Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan, Peoples R China.
通讯机构：
[Li, Yaqin] W;Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan, Peoples R China.
会议名称：
9th International Symposium on Multispectral Image Processing and Pattern Recognition (MIPPR) - Parallel Processing of Images and Optimization; and Medical Imaging Processing
会议时间：
OCT 31-NOV 01, 2015
会议地点：
Enshi, PEOPLES R CHINA
会议主办单位：
[Wang, Xuan] Wuhan Polytech Univ, Coll Hlth Sci & Nursing, Wuhan, Peoples R China.^[Li, Yaqin;Li, Shigao] Wuhan Polytech Univ, Sch Math & Comp Sci, Wuhan, Peoples R China.
