检索结果-南通市图书馆

Computational Visual Media 2024年第3期10卷 593-608页

作者： Fan zhang Gongguan Chen Hua Wang caiming zhang Shandong technology and Business University Shandong 264005China school of Information and Electrical Engineering Ludong UniversityYantai 264025China Shangdong University Shandong 250100China

Recently, facial-expression recognition (FER)has primarily focused on images in the wild, includingfactors such as face occlusion and image blurring, ratherthan laboratory images. Complex field environmentshave introduced new challenges to FER. To addressthese challenges, this study proposes a cross-fusion dualattention network. The network comprises three parts:(1) a cross-fusion grouped dual-attention mechanism torefine local features and obtain global information;(2) aproposed C2 activation function construction method,which is a piecewise cubic polynomial with threedegrees of freedom, requiring less computation withimproved ﬂexibility and recognition abilities, whichcan better address slow running speeds and neuroninactivation problems;and (3) a closed-loop operationbetween the self-attention distillation process andresidual connections to suppress redundant informationand improve the generalization ability of the *** recognition accuracies on the RAF-DB, FERPlus,and AffectNet datasets were 92.78%, 92.02%, and63.58%, respectively. Experiments show that this modelcan provide more effective solutions for FER tasks.

关键词： facial-expression recognition(FER) cubic polynomial activation function dualattention mechanism interactive learning self-attention distillation

Adaptive bidirectional diffusion for image restoration

维普期刊数据库评论

在线全文

维普期刊数据库

学校读者我要写书评

暂无评论

science china(Information sciences) 2010年第12期53卷 2452-2460页

作者： FU ShuJun 1,2 & zhang caiming 1 1 school of computer science and technology, Shandong University, Jinan 250101, china 2 school of Mathematics, Shandong University, Jinan 250100, china school of computer science and technology Shandong University Jinan China school of Mathematics Shandong University Jinan China

A large number of applications in image processing and computer vision depend on image quality. In this paper, combining the forward diffusion with the backward diffusion by different weights, we present an adaptive bidirectional diffusion method for image denoising and deblurring simultaneously. Further, we introduce a gradient factor into the data fidelity term, which forms a spatially varying constraint and allows a better restoration of image edges and fine details. In order to obtain a stable solution, we develop a numerically stable scheme and give its theoretical analysis. Finally, we show advantages of this method in deblurring edges, denoising and restoring fine details of image compared with other related methods in experiments.

关键词： image restoration bidirectional diffusion inverse diffusion anisotropic diffusion stabilizing scheme spatially varying constraint

Cubic surface fitting to image by combination

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

science china(Information sciences) 2010年第7期53卷 1287-1295页

作者： LI XueMei1, zhang caiming1,2, YUE YiZhen1 & WANG KunPeng1 1school of computer science and technology, Shandong University, Jinan 250101, china 2school of computer science and technology, Shandong Economics University, Jinan 250014, china school of computer science and technology Shandong University Jinan China school of computer science and technology Shandong Economics University Jinan China

We present a new method for constructing a fitting surface to image data. The new method is based on a supposition that the given image data are sampled from an original scene that can be represented by a surface defined by piecewise quadratic polynomials. The surface representing the original scene is known as the original surface in this paper. Unlike existing methods, which generally construct the fitting surface to the original surface using image data as interpolation data, the new method constructs the fitting surface using the image data as constraints to reverse the sampling process, which improves the approximation precision of the fitting surface. Associated with each data point and its near region, the new method constructs a quadratic polynomial patch locally using the sampling formula as constraint. The quadratic patch approximates the original surface with a quadratic polynomial precision. The fitting surface which approximates the original surface is formed by the combination of all the quadratic polynomial patches. The experiments demonstrate that compared with Bi-cubic and Separable PCC methods, the new method produced resized images with high precision and good quality.

关键词： surface fitting quadratic polynomial image resizing reversing

Medical image segmentation using improved FCM

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

science china(Information sciences) 2012年第5期55卷 1052-1061页

作者： zhang XiaoFeng 1,2,zhang caiming 1,3,4,TANG WenJing 1,2 & WEI ZhenWen 1 1 school of computer science and technology,Shandong University,Jinan 250101,china 2 school of Information and Electrical Engineering,Ludong University,Yantai 264025,china 3 school of computer science and technology,Shandong University of Finance and Economics,Jinan 250014,china 4 Shandong Province Key Lab of Digital Media technology,Jinan 250014,china school of computer science and technology Shandong University Jinan China school of Information and Electrical Engineering Ludong University Yantai China school of computer science and technology Shandong University of Finance and Economics Jinan China Shandong Province Key Lab of Digital Media technology Jinan China

Image segmentation is one of the most important problems in medical image processing,and the existence of partial volume effect and other phenomena makes the problem much more *** C-means,as an effective tool to deal with PVE,however,is faced with great challenges in *** at this,this paper proposes one improved FCM algorithm based on the histogram of the given image,which will be denoted as HisFCM and divided into two *** first phase will retrieve several intervals on which to compute cluster centroids,and the second one will perform image segmentation based on improved FCM *** with FCM and other improved algorithms,HisFCM is of much higher efficiency with satisfying *** on medical images show that HisFCM can achieve good segmentation results in less than 0.1 second,and can satisfy real-time requirements of medical image processing.

关键词： FCM histogram image segmentation medical image processing

基于雾浓度分类与暗-亮通道先验的多分支去雾网络

在线全文

同方期刊数据库

学校读者我要写书评

暂无评论

计算机研究与发展 2024年第3期61卷 762-779页

作者：张琪东迟静陈玉妍张彩明山东财经大学计算机科学与技术学院济南250014 山东省数字媒体技术重点实验室(山东财经大学) 济南250014 山东大学软件学院济南250101

在图像去雾领域中,目前多数去雾模型难以维持精度与效率的平衡,高精度的模型往往伴随着复杂的网络结构,而简单的网络结构又往往会导致低质量的结果.针对该问题提出一个基于雾浓度分类与暗-亮通道先验的多分支去雾模型,通过对带雾图像分... 详细信息

在图像去雾领域中,目前多数去雾模型难以维持精度与效率的平衡,高精度的模型往往伴随着复杂的网络结构,而简单的网络结构又往往会导致低质量的结果.针对该问题提出一个基于雾浓度分类与暗-亮通道先验的多分支去雾模型,通过对带雾图像分类,使用复杂度不同的网络来处理不同雾浓度的图像,可在保证精度的同时提高计算效率.模型由轻量级雾图像分类器和基于暗-亮通道先验的多分支去雾网络2部分构成:前者将带雾图像分为轻雾、中雾、浓雾3类,输出雾浓度标签;后者包含3个结构相同、宽度不同的分支网络,根据雾浓度标签选择不同的分支网络处理不同雾浓度图像,恢复至无雾图像.提出一个新的雾浓度分类方法以及基于该方法的雾浓度分类损失函数,可根据带雾图像的暗通道特征和恢复难度,结合生成图像质量和模型计算效率,得到对带雾图像合理准确的分类结果,达到去雾效果和算力需求的良好平衡.提出新的暗通道与亮通道先验损失函数,用于约束分支去雾网络,可有效提高去雾精度.实验结果表明,模型能够以更低的网络参数量和复杂度得到更优的去雾结果.

关键词：图像去雾雾浓度分类暗通道先验亮通道先验卷积神经网络

Genetic Algorithm for the Thermal Stresses Optimum Design ofFunctionally Gradient Material Plate

在线全文

学校读者我要写书评

暂无评论

International Journal of Minerals,Metallurgy and Materials 1999年第3期13卷 224-227页

作者： Xiaodan zhang Zhengbin Tang Changchun Ge(Applied science school, University of science and technology Beijing, Beijing 100083, china) Applied science school University of Science and Technology Beijing Beijing 100083 China

Based on the thermal stress distribution for functionally gradient material (FGM) plates, a Genetic Algorithm (GA) method for the thermal stresses optimum design of FGM plate with computer technologies is given. The m... 详细信息

关键词： functionally gradient material (FGM) thermal stress Genetic Algorithm (GA) crossover mutation

One-Dimensional Fractal Curves and Fractal Dimension Determination

在线全文

学校读者我要写书评

暂无评论

International Journal of Minerals,Metallurgy and Materials 1998年第3期12卷 165-168页

作者： Qikai Li Yue zhang Wuyang Chu(Applied science school, University of science and technology Beijing, Beijing 100083, china) Applied science school University of Science and Technology Beijing Beijing 100083 China

Various methods for evaluating the fractal curves were reviewed and simulated on computer. It is shown box-counting and power spectrum methods generally give poor measuring results, while yard and variation methods could obtain good results. However,owing to multiple influencing factors, further study needs to be done before widespread application of variation method. In order to improve the measuring accuracy of yard method, a new method has been propoed to measure the fractal dimension by changing the instrumental resolutions.

关键词： fractal fractal dimension computer simulation

在线全文

学校读者我要写书评

暂无评论

区域增强型注意力网络下的人脸表情识别

计算机辅助设计与图形学学报 2024年第1期36卷 152-160页

作者：陈公冠张帆王桦范辉张彩明山东工商学院计算机科学与技术学院烟台264005 山东省未来智能金融工程实验室烟台264005 鲁东大学信息与电气工程学院烟台264011 山东大学软件学院济南250101

为了识别人脸表情中包含复杂背景、面部遮挡等因素的真实环境下的图像,提出基于区域增强型注意力网络的人脸表情识别方法.首先提出基于注意力的区域增强网络,减弱外部因素的影响以及增强表情识别在真实环境下的鲁棒性;然后提出通道-空... 详细信息

为了识别人脸表情中包含复杂背景、面部遮挡等因素的真实环境下的图像,提出基于区域增强型注意力网络的人脸表情识别方法.首先提出基于注意力的区域增强网络,减弱外部因素的影响以及增强表情识别在真实环境下的鲁棒性;然后提出通道-空间注意力融合网络,作用于全局的特征提取;最后通过分区损失和交叉熵损失相结合的方式提升表情图像的辨识度,从而提升识别准确率.在公开数据集RAF-DB,FERPlus和AffectNet上的实验结果表明,表情识别准确率分别达到88.81%,89.32%和60.45%;所提方法具有更高的准确率和鲁棒性.

关键词：人脸表情识别区域增强注意力融合分区损失

Numerical Simulation and Water Analog of Mold Filling Processes

在线全文

学校读者我要写书评

暂无评论

Journal of Materials science & technology 2001年第1期17卷 109-110页

作者： Yuebing zhang Xiang XUE. and Bide ZHOU school of Materials science and Engineering, Harbin institute of technology, Harbin 150001, china Sch. of Mater. Sci. Harbin Inst. of Technol. Harbin 150001 China

This paper demonstrates the potential of a computer code, developed by the authors, in shaping gating systems by modeling the fluid flow phenomena through a complex gating system during mold filling. A plate casting with dimension 200 mm×200 mm×50 mm was chosen as the verifying problem. Water analog studies were carried out on this casting. The comparison indicates that computer simulation could be a powerful tool in shaping gating systems.

关键词： computer simulation flow phenomena computer code mold filling