检索结果-内蒙古大学图书馆

14th International Conference on Bio-inspired Computing: Theories and Applications, BIC-TA 2019

作者： Cao, Bin Li, Ruichang Chen, Shuai Hebei University of Technology Tianjin300401 China Key Laboratory of Intelligent Perception and Image Understanding of Ministry of Education Xi’an710071 China Hebei Provincial Key Laboratory of Big Data Calculation Tianjin300401 China

ISBN: (纸本)9789811534249

The capacitated arc routing problem (CARP) has important research significance in the area of vehicle routing services. We study a multi-depot green capacitated arc routing problem by minimizing economic cost, makespan and carbon emission cost, and then, present a modified memetic algorithm for multi-depot (MMAMD) that aims to solve the problem efficiently. According to the characteristics of multi-depots in this paper, a specific encoding scheme via the two-dimensional array and the dynamic adjustment strategy of boundary arcs between depots are proposed. By combining ant colony algorithm with local search, we present an ant colony local search strategy with extended neighborhood search for the memetic algorithm. The experimental results show that the proposed algorithm which is used to solve the model is effective, which can provide a reference for waste clearing decision. © 2020, Springer Nature Singapore Pte Ltd.

关键词： Vehicle routing

来源：评论

学校读者我要写书评

暂无评论

Research on Energy Saving Control of Lighting Systems in Public Buildings

Research on Energy Saving Control of Lighting Systems in Pub...

引用

第33届中国控制与决策会议

作者： Shengyan Li Hongyan Ma Yongxue Ye Jiaming Dou Jingjian Yang Beijing Key Laboratory of Intelligent Processing for Building Big Data Beijing University of Civil Engineering and Architecture National Virtual Simulation Experimental Center for Smart City Education Institute of Distributed Energy Storage Safety Big Data

Most indoor users are unwilling to adjust the lighting equipment and shutters in daily life,resulting in increasing energy consumption of *** this paper,in order to reduce the energy consumption,using fuzzy control to solve this *** using fuzzy control,it could control the opening of indoor lamps,and let the shutters could determine the opening degree of the shutters according to the difference between the set illuminance values and the actual illuminance values on the working *** results show that the fuzzy control of classroom lighting system can reasonably control the running status of lamps according to the actual situation of the environment and reduce the energy consumption of the lighting *** with the single fuzzy controller,the double fuzzy controller of the shutter can make the difference between the illuminance value of indoor working surface,and the set illuminance value smaller,control the light sources in different areas of the room,and improve the uniformity of natural light in the *** method proposed in this paper can solve the energy consumption problem.

关键词： Lighting System Shutter Double Fuzzy Control System Fuzzy Control

来源：评论

学校读者我要写书评

暂无评论

A Diabetic Retinopathy Classification Method Based on Novel Attention Mechanism 17th

A Diabetic Retinopathy Classification Method Based on Novel ...

引用

17th International Conference on intelligent Computing, ICIC 2021

作者： Zou, Jinfan Zhang, Xiaolong Lin, Xiaoli School of Computer Science and Technology Wuhan University of Science and Technology Wuhan China Institute of Big Data Science and Engineering Wuhan University of Science and Technology Wuhan China Hubei Key Laboratory of Intelligent Information Processing and Real-Time Industrial System Wuhan University of Science and Technology Wuhan China

ISBN: (纸本)9783030845216

The convolutional neural network (CNN) has been effectively used to do feature extraction in medical image analysis including fundus images for diabetic retinopathy. However, some important detail features are easily ignored, which causes low prediction accuracy. To solve this problem, a CNN feature extraction method based on a novel pixel-level attention mechanism is proposed to achieve an efficient and accurate identification of diabetic retinopathy. Firstly, data normalization and data augmentation are used to improve the quality of datasets. Then, according to the structure characteristics of the two backbone networks Inception- ResnetV2 and EfficientNet-B5, the corresponding attention modules are introduced respectively. Finally, attention networks are trained as feature extractors of fundus images, and complementary deep attention feature descriptors are formed through feature fusion, which can effectively improve the accuracy of the classification model. The experimental results on EyePACS, Messidor and OIA datasets show that the proposed method outperforms the previous ones. © 2021, Springer Nature Switzerland AG.

关键词： Image classification

来源：评论

学校读者我要写书评

暂无评论

Research on Human Pose Classification Based on Improved SVM

Research on Human Pose Classification Based on Improved SVM

引用

第33届中国控制与决策会议

作者： Yingda Zhang Hongyan Ma Yongxue Ye Jiaming Dou Jingjian Yang Beijing Key Laboratory of Intelligent Processing for Building Big Data Beijing University of Civil Engineering and Architecture National Virtual Simulation Experimental Center for Smart City Education Institute of Distributed Energy Storage Safety Big Data

With the development of intelligent buildings and aging of population,designing a high-accuracy alarm system is essential to prevent the elderly from accidentally falling *** address the problems of poor generalization and low accuracy of the SVM in the application of indoor elderly fall alarm systems,in this paper,the human keypoints detection algorithm,which increases the discrimination of the dataset,is proposed to pre-process the *** on this,the features of images are extracted from the histogram of the oriented gradient and the gray-level co-occurrence matrix,and the improved SVM algorithm is then used again for image *** simulation results show that the classification accuracy of using SVM to detect the dataset is around 91%,while the accuracy of the dataset classified as using improved SVM is around 97%.The method proposed in this paper is effective in increasing the accuracy of computer recognition of accidental falls by the elderly.

关键词： SVM Human keypoints Detection Image Classification Accidental Indoor Falls Aging of Population

来源：评论

学校读者我要写书评

暂无评论

Densely packed object detection with transformer-based head and EM-merger

引用

SERVICE ORIENTED COMPUTING AND APPLICATIONS 2023年第2期17卷 109-117页

作者： Zhong, Xiaojing Zhang, Ni Hu, Hao Li, Li Cen, Junhua Wu, Qingyao South China Univ Technol Sch Software Engn Guangzhou Peoples R China Minist Educ Key Lab Big Data & Intelligent Robot Guangzhou Peoples R China Pazhou Lab Guangzhou Peoples R China Peng Cheng Lab Guangzhou Peoples R China Zhongnan Bldg Mat Grp Co Ltd Guangzhou Guangdong Gw Peoples R China Internet Technol Co Ltd Guangzhou Guangdong Gw Peoples R China

Due to the high density of objects and their varying sizes, detecting them accurately and without repetition in such scenarios is more challenging than traditional object detection methods. In this paper, we propose a YOLOv5-based object detection approach equipped with a Transformer-based Head and EM-Merger unit specifically designed for densely packed scenes. We incorporate the transformer architecture into the prediction heads to enable a self-attention mechanism that captures long-term dependencies between the densely packed objects. Additionally, we introduce an EM-Merger unit to resolve redundant object detections. Experimental results on the RebarDSC and SKU110K datasets demonstrate that our method significantly outperforms the baseline approach, achieving new state-of-the-art detection performance.

关键词： Object detection Densely packed scenarios Attention mechanism

来源：评论

学校读者我要写书评

暂无评论

Novel traction device for endoscopic submucosal dissection: a rotatable transparent cap, an additional channel, and a traction wire

引用

ENDOSCOPY 2024年第Sup1期56卷 E70-E71页

作者： Chen, Xi Lin, Zhenqun Li, Xiangqing Yang, Xiaoning Liu, Wenjuan Huang, Zehui Cai, Shuntian Zhangzhou Tradit Chinese Med Hosp Gastroenterol Zhangzhou Peoples R China Fujian Univ Tradit Chinese Med Gastroenterol Fuzhou Peoples R China Shandong First Med Univ & Shandong Acad Med Sci Coll Artificial Intelligence & Big Data Med Sci Intelligent Biomed Robot Lab Jinan Peoples R China Xiamen Univ Zhongshan Hosp Sch Med Gastroenterol Xiamen Peoples R China Xiamen Univ Zhongshan Hosp Sch Med Xiamen Key Lab Intestinal Microbiome & Human Hlth Xiamen Peoples R China Fujian Univ Tradit Chinese Med Fuzhou Peoples R China Zhangzhou Tradit Chinese Med Hosp Endocrinol Zhangzhou Peoples R China Zhangzhou Tradit Chinese Med Hosp Zhangzhou Peoples R China Xiamen Univ Xiamen Key Lab Intestinal Microbiome & Human Hlth Zhongshan Hosp Xiamen Peoples R China Xiamen Univ Xiamen Univ Sch Med Dept Gastroenterol 201-209 Hubinnan Rd Xiamen 361004 Fujian Peoples R China

Traditionally, endoscopic procedures have been limited to a single–channel device, making it difficult for endoscopists to perform endoscopic submucosal dissection (ESD). As a result, several solutions have been developed, each with its own advantages and disadvantages, including traction devices [ 1 ] [ 2 ], double-channel endoscopes [ 3 ], laparoscopy–endoscopy combined surgery [ 4 ], and surgical endoscopic robots [ 5 ]. To address these challenges, we fabricated a rotatable transparent cap with a fixed channel that can accommodate endoscopic surgical instruments. The traction device consists of three components: a rotatable transparent cap, an additional channel, and a traction wire ([ Fig. 1 ]). The additional channel and traction wire are connected to a rotatable ring on the transparent cap. By pulling the traction wire, the additional channel can be rotated around the transparent cap ([ Fig. 1 ]). Fig. 1 Structure and functionality of the rotatable transparent cap device (Micro-Tech, Nanjing, China). a The rotatable transparent cap unit consists of a transparent cap with a rotatable ring in the middle section. A nylon wire threaded through a channel in the rotatable ring is fixed at its midpoint to the ring. b Auxiliary channel fixed to the rotatable ring. c Ends of the nylon wire that is connected to the ring and additional channel. Both ends of the wire can be used to pull the ring, rotating it. d The wire can be pulled to rotate the auxiliary channel on the ring around the transparent cap, enabling precise control. During ESD, foreign-body forceps are inserted into the additional channel to clamp and push the tumor to the distal end, acting as a traction device. During the operation, the position of the additional channel can be controlled dynamically by pulling the traction wire. To assess the efficacy of this device, we performed ESD in four directions: downward, upward, leftward, and rightward. The results demonstrated that this rotatable transparent ca

关键词：

来源：评论

学校读者我要写书评

暂无评论

MFCLIP: Multi-modal Fine-grained CLIP for Generalizable Diffusion Face Forgery Detection

arXiv

引用

arXiv 2024年

作者： Zhang, Yaning Wang, Tianyi Yu, Zitong Gao, Zan Shen, Linlin Chen, Shengyong Jinan250353 China Nanyang Technological University 50 Nanyang Ave Block N 4 639798 Singapore School of Computing and Information Technology Great Bay University Dongguan523000 China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen518060 China Jinan250014 China The Key Laboratory of Computer Vision and System Ministry of Education Tianjin University of Technology Tianjin300384 China Computer Vision Institute College of Computer Science and Software Engineering Shenzhen University Shenzhen518060 China National Engineering Laboratory for Big Data System Computing Technology Shenzhen University Shenzhen China Shenzhen Institute of Artificial Intelligence and Robotics for Society Shenzhen518129 China Guangdong Key Laboratory of Intelligent Information Processing Shenzhen University Shenzhen518060 China

The rapid development of photo-realistic face generation methods has raised significant concerns in society and academia, highlighting the urgent need for robust and generalizable face forgery detection (FFD) techniques. Although existing approaches mainly capture face forgery patterns using image modality, other modalities like fine-grained noises and texts are not fully explored, which limits the generalization capability of the model. In addition, most FFD methods tend to identify facial images generated by GAN, but struggle to detect unseen diffusion-synthesized ones. To address the limitations, we aim to leverage the cutting-edge foundation model, contrastive language-image pre-training (CLIP), to achieve generalizable diffusion face forgery detection (DFFD). In this paper, we propose a novel multi-modal fine-grained CLIP (MFCLIP) model, which mines comprehensive and fine-grained forgery traces across image-noise modalities via language-guided face forgery representation learning, to facilitate the advancement of DFFD. Specifically, we devise a fine-grained language encoder (FLE) that extracts fine global language features from hierarchical text prompts. We design a multi-modal vision encoder (MVE) to capture global image forgery embeddings as well as fine-grained noise forgery patterns extracted from the richest patch, and integrate them to mine general visual forgery traces. Moreover, we build an innovative plug-and-play sample pair attention (SPA) method to emphasize relevant negative pairs and suppress irrelevant ones, allowing cross-modality sample pairs to conduct more flexible alignment. Extensive experiments and visualizations show that our model outperforms the state of the arts on different settings like cross-generator, cross-forgery, and cross-dataset evaluations. Copyright © 2024, The Authors. All rights reserved.

关键词： Alignment

来源：评论

学校读者我要写书评

暂无评论

Refactor Business Process Models for Efficiency Improvement 1

引用

9th EAI International Conference on Cloud Computing, CloudComp 2019 and the 4th EAI International Conference on Smart Grid and Innovative Frontiers in Telecommunications, SmartGIFT 2019

作者： Dai, Fei Liu, Miao Mo, Qi Huang, Bi Li, Tong School of Big Data and Intelligent Engineering Southwest Forestry University Kunming China School of Software Yunnan University Kunming China School of Big Data Yunnan Agricultural University Kunming China Key Laboratory for Software Engineering of Yunnan Province Kunming China

ISBN: (数字)9783030485139

ISBN: (纸本)9783030485122

Since business processes describe the core value chain of enterprises, thousands of business processes are modeled in business process models. A problem is how to improve the efficiency of these models. In this paper, we propose an approach to refactor these models for efficiency improvement. More specifically, we first identify false sequence relations that affect model efficiency based on the sequence relation matrix and the dependency relation matrix. Second, we refactor a business process model by constructing and transforming a dependency graph without altering its output result. After refactoring, the concurrent execution of business tasks in the original models can be maximized such that its efficiency can be improved. Experimental results show the effectiveness of our approach. © 2020, ICST Institute for Computer Sciences, Social Informatics and Telecommunications Engineering.

关键词： Efficiency

来源：评论

学校读者我要写书评

暂无评论

Directed Acyclic Graph Factorization Machines for CTR Prediction via Knowledge Distillation

arXiv

引用

arXiv 2022年

作者： Tian, Zhen Bai, Ting Zhang, Zibin Xu, Zhiyuan Lin, Kangyi Wen, Ji-Rong Zhao, Wayne Xin Gaoling School of Artificial Intelligence Renmin University of China Beijing China Beijing University of Posts and Telecommunications Beijing China Weixin Open Platform Tencent Guangzhou China Beijing Key Laboratory of Intelligent Telecommunications Software and Multimedia China Beijing Key Laboratory of Big Data Management and Analysis Methods China

With the growth of high-dimensional sparse data in web-scale recommender systems, the computational cost to learn high-order feature interaction in CTR prediction task largely increases, which limits the use of high-order interaction models in real industrial applications. Some recent knowledge distillation based methods transfer knowledge from complex teacher models to shallow student models for accelerating the online model inference. However, they suffer from the degradation of model accuracy in knowledge distillation process. It is challenging to balance the efficiency and effectiveness of the shallow student models. To address this problem, we propose a Directed Acyclic Graph Factorization Machine (KD-DAGFM) to learn the high-order feature interactions from existing complex interaction models for CTR prediction via Knowledge Distillation. The proposed lightweight student model DAGFM can learn arbitrary explicit feature interactions from teacher networks, which achieves approximately lossless performance and is proved by a dynamic programming algorithm. Besides, an improved general model KD-DAGFM+ is shown to be effective in distilling both explicit and implicit feature interactions from any complex teacher model. Extensive experiments are conducted on four real-world datasets, including a large-scale industrial dataset from WeChat platform with billions of feature dimensions. KD-DAGFM achieves the best performance with less than 21.5% FLOPs of the state-ofthe-art method on both online and offline experiments, showing the superiority of DAGFM to deal with the industrial scale data in CTR prediction task1 © 2022, CC BY.

关键词： Recommender systems

来源：评论

学校读者我要写书评

暂无评论

Self-Supervised Multiscale Adversarial Regression Network for Stereo Disparity Estimation

引用

IEEE TRANSACTIONS ON CYBERNETICS 2021年第10期51卷 4770-4783页

作者： Wang, Chen Bai, Xiao Wang, Xiang Liu, Xianglong Zhou, Jun Wu, Xinyu Li, Hongdong Tao, Dacheng Beihang Univ Beijing Adv Innovat Ctr Big Data & Brain Comp Sch Comp Sci & Engn State Key Lab Software Dev Environm Beijing 100191 Peoples R China Beihang Univ State Key Lab Software Dev Environm Beijing 100191 Peoples R China Beihang Univ Beijing Adv Innovat Ctr Big Data Based Precis Med Beijing 100191 Peoples R China Griffith Univ Sch Informat & Commun Technol Nathan Qld 4111 Australia Chinese Acad Sci Shenzhen Inst Adv Technol Guangdong Prov Key Lab Robot & Intelligent Syst Shenzhen 518055 Peoples R China Australian Natl Univ Res Sch Engn Canberra ACT 2600 Australia Univ Sydney Fac Engn UBTECH Sydney Artificial Intelligence Ctr Darlington NSW 2008 Australia Univ Sydney Fac Engn Sch Comp Sci Darlington NSW 2008 Australia

Deep learning approaches have significantly contributed to recent progress in stereo matching. These deep stereo matching methods are usually based on supervised training, which requires a large amount of high-quality ground-truth depth map annotations that are expensive to collect. Furthermore, only a limited quantity of stereo vision training data are currently available, obtained either by active sensors (Lidar and ToF cameras) or through computer graphics simulations and not meeting requirements for deep supervised training. Here, we propose a novel deep stereo approach called the "self-supervised multiscale adversarial regression network (SMAR-Net)," which relaxes the need for ground-truth depth maps for training. Specifically, we design a two-stage network. The first stage is a disparity regressor, in which a regression network estimates disparity values from stacked stereo image pairs. Stereo image stacking method is a novel contribution as it not only contains the spatial appearances of stereo images but also implies matching correspondences with different disparity values. In the second stage, a synthetic left image is generated based on the left-right consistency assumption. Our network is trained by minimizing a hybrid loss function composed of a content loss and an adversarial loss. The content loss minimizes the average warping error between the synthetic images and the real ones. In contrast to the generative adversarial loss, our proposed adversarial loss penalizes mismatches using multiscale features. This constrains the synthetic image and real image as being pixelwise identical instead of just belonging to the same distribution. Furthermore, the combined utilization of multiscale feature extraction in both the content loss and adversarial loss further improves the adaptability of SMAR-Net in ill-posed regions. Experiments on multiple benchmark datasets show that SMAR-Net outperforms the current state-of-the-art self-supervised methods and achieves

关键词： Estimation Feature extraction Training Sensors Generative adversarial networks Generators Annotations Generative adversarial network (GAN) multiscale feature extraction self-supervised learning stereo disparity estimation

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：