As a complex engineering problem,the satellite module layout design (SMLD) is difficult to resolve by using conventional computation-based approaches. The challenges stem from three aspects:computational complexity,en...As a complex engineering problem,the satellite module layout design (SMLD) is difficult to resolve by using conventional computation-based approaches. The challenges stem from three aspects:computational complexity,engineering complexity,and engineering practicability. Engineers often finish successful satellite designs by way of their plenty of experience and wisdom,lessons learnt from the past practices,as well as the assistance of the advanced computational techniques. Enlightened by the ripe patterns,th...展开更多
The projectile penetration process into concrete target is a nonlinear complex problem.With the increase ofexperiment data,the data-driven paradigm has exhibited a new feasible method to solve such complex prob-lem.Ho...The projectile penetration process into concrete target is a nonlinear complex problem.With the increase ofexperiment data,the data-driven paradigm has exhibited a new feasible method to solve such complex prob-lem.However,due to poor quality of experimental data,the traditional machine learning(ML)methods,whichare driven only by experimental data,have poor generalization capabilities and limited prediction accuracy.Therefore,this study intends to exhibit a ML method fusing the prior knowledge with experiment data.The newML method can constrain the fitting to experimental data,improve the generalization ability and the predic-tion accuracy.Experimental results show that integrating domain prior knowledge can effectively improve theperformance of the prediction model for penetration depth into concrete targets.展开更多
Pancreatic diseases, including mass-forming chronic pancreatitis (MFCP) and pancreatic ductal adenocarcinoma(PDAC), present with similar imaging features, leading to diagnostic complexities. Deep Learning (DL) methods...Pancreatic diseases, including mass-forming chronic pancreatitis (MFCP) and pancreatic ductal adenocarcinoma(PDAC), present with similar imaging features, leading to diagnostic complexities. Deep Learning (DL) methodshave been shown to perform well on diagnostic tasks. Existing DL pancreatic lesion diagnosis studies basedon Magnetic Resonance Imaging (MRI) utilize the prior information to guide models to focus on the lesionregion. However, over-reliance on prior information may ignore the background information that is helpful fordiagnosis. This study verifies the diagnostic significance of the background information using a clinical dataset.Consequently, the Prior Difference Guidance Network (PDGNet) is proposed, merging decoupled lesion andbackground information via the Prior Normalization Fusion (PNF) strategy and the Feature Difference Guidance(FDG) module, to direct the model to concentrate on beneficial regions for diagnosis. Extensive experiments inthe clinical dataset demonstrate that the proposed method achieves promising diagnosis performance: PDGNetsbased on conventional networks record an ACC (Accuracy) and AUC (Area Under the Curve) of 87.50% and89.98%, marking improvements of 8.19% and 7.64% over the prior-free benchmark. Compared to lesion-focusedbenchmarks, the uplift is 6.14% and 6.02%. PDGNets based on advanced networks reach an ACC and AUC of89.77% and 92.80%. The study underscores the potential of harnessing background information in medical imagediagnosis, suggesting a more holistic view for future research.展开更多
Recently,deep learning-based image inpainting methods have made great strides in reconstructing damaged regions.However,these methods often struggle to produce satisfactory results when dealing with missing images wit...Recently,deep learning-based image inpainting methods have made great strides in reconstructing damaged regions.However,these methods often struggle to produce satisfactory results when dealing with missing images with large holes,leading to distortions in the structure and blurring of textures.To address these problems,we combine the advantages of transformers and convolutions to propose an image inpainting method that incorporates edge priors and attention mechanisms.The proposed method aims to improve the results of inpainting large holes in images by enhancing the accuracy of structure restoration and the ability to recover texture details.This method divides the inpainting task into two phases:edge prediction and image inpainting.Specifically,in the edge prediction phase,a transformer architecture is designed to combine axial attention with standard self-attention.This design enhances the extraction capability of global structural features and location awareness.It also balances the complexity of self-attention operations,resulting in accurate prediction of the edge structure in the defective region.In the image inpainting phase,a multi-scale fusion attention module is introduced.This module makes full use of multi-level distant features and enhances local pixel continuity,thereby significantly improving the quality of image inpainting.To evaluate the performance of our method.comparative experiments are conducted on several datasets,including CelebA,Places2,and Facade.Quantitative experiments show that our method outperforms the other mainstream methods.Specifically,it improves Peak Signal-to-Noise Ratio(PSNR)and Structure Similarity Index Measure(SSIM)by 1.141~3.234 db and 0.083~0.235,respectively.Moreover,it reduces Learning Perceptual Image Patch Similarity(LPIPS)and Mean Absolute Error(MAE)by 0.0347~0.1753 and 0.0104~0.0402,respectively.Qualitative experiments reveal that our method excels at reconstructing images with complete structural information and clear texture details.Furthermore,our model exhib展开更多
已有跌倒检测工作主要关注室内场景,且大多偏重对人员身体姿态特征进行建模,而忽略了场景背景信息以及人员与地面的交互信息。针对这个问题,从实际电梯场景应用入手,提出一种基于场景先验及注意力引导的跌倒检测算法。首先,利用电梯历...已有跌倒检测工作主要关注室内场景,且大多偏重对人员身体姿态特征进行建模,而忽略了场景背景信息以及人员与地面的交互信息。针对这个问题,从实际电梯场景应用入手,提出一种基于场景先验及注意力引导的跌倒检测算法。首先,利用电梯历史数据,以高斯概率分布建模的方式从人员的活动轨迹中自动化地学习场景先验信息;随后,把场景先验信息作为空间注意力掩膜与神经网络的全局特征融合,以此聚焦地面区域的局部信息;然后,将融合后的局部特征与全局特征采用自适应加权的方式进一步聚合,从而形成更具鲁棒性和判别力的特征;最后,将特征送入由全局平均池化层和全连接层构成的分类模块中进行跌倒类别预测。在自构建的电梯场景Elevator Fall Detection和公开的UR Fall Detection数据集上的实验结果表明,所提算法的检测准确率分别达到了95.36%和99.01%,相较于网络结构复杂的ResNet50算法,分别提高了3.52个百分点和0.61个百分点。可见所构建的高斯场景先验引导的注意力机制可使网络关注地面区域的特征,更有利于对跌倒的识别,由此得到的检测模型准确率高且算法满足实时性应用要求。展开更多
The numerous photos captured by low-price Internet of Things(IoT)sensors are frequently affected by meteorological factors,especially rainfall.It causes varying sizes of white streaks on the image,destroying the image...The numerous photos captured by low-price Internet of Things(IoT)sensors are frequently affected by meteorological factors,especially rainfall.It causes varying sizes of white streaks on the image,destroying the image texture and ruining the performance of the outdoor computer vision system.Existing methods utilise training with pairs of images,which is difficult to cover all scenes and leads to domain gaps.In addition,the network structures adopt deep learning to map rain images to rain-free images,failing to use prior knowledge effectively.To solve these problems,we introduce a single image derain model in edge computing that combines prior knowledge of rain patterns with the learning capability of the neural network.Specifically,the algorithm first uses Residue Channel Prior to filter out the rainfall textural features then it uses the Feature Fusion Module to fuse the original image with the background feature information.This results in a pre-processed image which is fed into Half Instance Net(HINet)to recover a high-quality rain-free image with a clear and accurate structure,and the model does not rely on any rainfall assumptions.Experimental results on synthetic and real-world datasets show that the average peak signal-to-noise ratio of the model decreases by 0.37 dB on the synthetic dataset and increases by 0.43 dB on the real-world dataset,demonstrating that a combined model reduces the gap between synthetic data and natural rain scenes,improves the generalization ability of the derain network,and alleviates the overfitting problem.展开更多
In foggy weather, images of outdoor scene are usually characterized with poor visibility as well as faint color saturation. The degraded hazy images may have substantial negative impact on most computer vision systems...In foggy weather, images of outdoor scene are usually characterized with poor visibility as well as faint color saturation. The degraded hazy images may have substantial negative impact on most computer vision systems. Thus image haze removal is of the practical significance in engineering. This paper proposes a fast and effective single image haze removal algorithm on the basis of the physics imaging model. To extract the global atmospheric light accurately, we exploit multiple prior rules underlying hazy images, and put forward a novel measurement to judge the likelihood that a pixel is regarded as the global atmospheric light. In addition, the rough transmission map is estimated through a multiscale fusion process based on the Laplace pyramid transform, and refined by a total variation model. Experimental results demonstrate the proposed method outperforms most of the state-of-the-art algorithms in terms of the dehazing quality, and achieves a trade-off between the computational efficiency and haze removal capability.展开更多
基金National Natural Science Foundation of China (50575031, 50275019)National High-tech Research and Development Program (2006AA04Z109)
文摘As a complex engineering problem,the satellite module layout design (SMLD) is difficult to resolve by using conventional computation-based approaches. The challenges stem from three aspects:computational complexity,engineering complexity,and engineering practicability. Engineers often finish successful satellite designs by way of their plenty of experience and wisdom,lessons learnt from the past practices,as well as the assistance of the advanced computational techniques. Enlightened by the ripe patterns,th...
基金supported by the National Natural Science Founda-tion of China(Grant No.12172381)Leading Talents of Science and Technology in the Central Plain of China(Grant No.234200510016).
文摘The projectile penetration process into concrete target is a nonlinear complex problem.With the increase ofexperiment data,the data-driven paradigm has exhibited a new feasible method to solve such complex prob-lem.However,due to poor quality of experimental data,the traditional machine learning(ML)methods,whichare driven only by experimental data,have poor generalization capabilities and limited prediction accuracy.Therefore,this study intends to exhibit a ML method fusing the prior knowledge with experiment data.The newML method can constrain the fitting to experimental data,improve the generalization ability and the predic-tion accuracy.Experimental results show that integrating domain prior knowledge can effectively improve theperformance of the prediction model for penetration depth into concrete targets.
基金the National Natural Science Foundation of China(No.82160347)Yunnan Key Laboratory of Smart City in Cyberspace Security(No.202105AG070010)Project of Medical Discipline Leader of Yunnan Province(D-2018012).
文摘Pancreatic diseases, including mass-forming chronic pancreatitis (MFCP) and pancreatic ductal adenocarcinoma(PDAC), present with similar imaging features, leading to diagnostic complexities. Deep Learning (DL) methodshave been shown to perform well on diagnostic tasks. Existing DL pancreatic lesion diagnosis studies basedon Magnetic Resonance Imaging (MRI) utilize the prior information to guide models to focus on the lesionregion. However, over-reliance on prior information may ignore the background information that is helpful fordiagnosis. This study verifies the diagnostic significance of the background information using a clinical dataset.Consequently, the Prior Difference Guidance Network (PDGNet) is proposed, merging decoupled lesion andbackground information via the Prior Normalization Fusion (PNF) strategy and the Feature Difference Guidance(FDG) module, to direct the model to concentrate on beneficial regions for diagnosis. Extensive experiments inthe clinical dataset demonstrate that the proposed method achieves promising diagnosis performance: PDGNetsbased on conventional networks record an ACC (Accuracy) and AUC (Area Under the Curve) of 87.50% and89.98%, marking improvements of 8.19% and 7.64% over the prior-free benchmark. Compared to lesion-focusedbenchmarks, the uplift is 6.14% and 6.02%. PDGNets based on advanced networks reach an ACC and AUC of89.77% and 92.80%. The study underscores the potential of harnessing background information in medical imagediagnosis, suggesting a more holistic view for future research.
基金supported in part by the National Natural Science Foundation of China under Grant 62062061/in part by the Major Project Cultivation Fund of Xizang Minzu University under Grant 324112300447.
文摘Recently,deep learning-based image inpainting methods have made great strides in reconstructing damaged regions.However,these methods often struggle to produce satisfactory results when dealing with missing images with large holes,leading to distortions in the structure and blurring of textures.To address these problems,we combine the advantages of transformers and convolutions to propose an image inpainting method that incorporates edge priors and attention mechanisms.The proposed method aims to improve the results of inpainting large holes in images by enhancing the accuracy of structure restoration and the ability to recover texture details.This method divides the inpainting task into two phases:edge prediction and image inpainting.Specifically,in the edge prediction phase,a transformer architecture is designed to combine axial attention with standard self-attention.This design enhances the extraction capability of global structural features and location awareness.It also balances the complexity of self-attention operations,resulting in accurate prediction of the edge structure in the defective region.In the image inpainting phase,a multi-scale fusion attention module is introduced.This module makes full use of multi-level distant features and enhances local pixel continuity,thereby significantly improving the quality of image inpainting.To evaluate the performance of our method.comparative experiments are conducted on several datasets,including CelebA,Places2,and Facade.Quantitative experiments show that our method outperforms the other mainstream methods.Specifically,it improves Peak Signal-to-Noise Ratio(PSNR)and Structure Similarity Index Measure(SSIM)by 1.141~3.234 db and 0.083~0.235,respectively.Moreover,it reduces Learning Perceptual Image Patch Similarity(LPIPS)and Mean Absolute Error(MAE)by 0.0347~0.1753 and 0.0104~0.0402,respectively.Qualitative experiments reveal that our method excels at reconstructing images with complete structural information and clear texture details.Furthermore,our model exhib
文摘已有跌倒检测工作主要关注室内场景,且大多偏重对人员身体姿态特征进行建模,而忽略了场景背景信息以及人员与地面的交互信息。针对这个问题,从实际电梯场景应用入手,提出一种基于场景先验及注意力引导的跌倒检测算法。首先,利用电梯历史数据,以高斯概率分布建模的方式从人员的活动轨迹中自动化地学习场景先验信息;随后,把场景先验信息作为空间注意力掩膜与神经网络的全局特征融合,以此聚焦地面区域的局部信息;然后,将融合后的局部特征与全局特征采用自适应加权的方式进一步聚合,从而形成更具鲁棒性和判别力的特征;最后,将特征送入由全局平均池化层和全连接层构成的分类模块中进行跌倒类别预测。在自构建的电梯场景Elevator Fall Detection和公开的UR Fall Detection数据集上的实验结果表明,所提算法的检测准确率分别达到了95.36%和99.01%,相较于网络结构复杂的ResNet50算法,分别提高了3.52个百分点和0.61个百分点。可见所构建的高斯场景先验引导的注意力机制可使网络关注地面区域的特征,更有利于对跌倒的识别,由此得到的检测模型准确率高且算法满足实时性应用要求。
基金supported by the National Natural Science Foundation of China under Grant no.41975183,and Grant no.41875184 and Supported by a grant from State Key Laboratory of Resources and Environmental Information System.
文摘The numerous photos captured by low-price Internet of Things(IoT)sensors are frequently affected by meteorological factors,especially rainfall.It causes varying sizes of white streaks on the image,destroying the image texture and ruining the performance of the outdoor computer vision system.Existing methods utilise training with pairs of images,which is difficult to cover all scenes and leads to domain gaps.In addition,the network structures adopt deep learning to map rain images to rain-free images,failing to use prior knowledge effectively.To solve these problems,we introduce a single image derain model in edge computing that combines prior knowledge of rain patterns with the learning capability of the neural network.Specifically,the algorithm first uses Residue Channel Prior to filter out the rainfall textural features then it uses the Feature Fusion Module to fuse the original image with the background feature information.This results in a pre-processed image which is fed into Half Instance Net(HINet)to recover a high-quality rain-free image with a clear and accurate structure,and the model does not rely on any rainfall assumptions.Experimental results on synthetic and real-world datasets show that the average peak signal-to-noise ratio of the model decreases by 0.37 dB on the synthetic dataset and increases by 0.43 dB on the real-world dataset,demonstrating that a combined model reduces the gap between synthetic data and natural rain scenes,improves the generalization ability of the derain network,and alleviates the overfitting problem.
基金supported by the National Natural Science Foundation of China(61571241)the Industry-University-research Prospective Joint Project of Jiangsu Province(BY2014014)+2 种基金the Major Projects of Jiangsu Province University Natural Science Research(15KJA510002)the Jiangsu Province Graduate Research and Innovation Project(CXZZ130476)the Science Research Fund of NUPT(NY215169)
文摘In foggy weather, images of outdoor scene are usually characterized with poor visibility as well as faint color saturation. The degraded hazy images may have substantial negative impact on most computer vision systems. Thus image haze removal is of the practical significance in engineering. This paper proposes a fast and effective single image haze removal algorithm on the basis of the physics imaging model. To extract the global atmospheric light accurately, we exploit multiple prior rules underlying hazy images, and put forward a novel measurement to judge the likelihood that a pixel is regarded as the global atmospheric light. In addition, the rough transmission map is estimated through a multiscale fusion process based on the Laplace pyramid transform, and refined by a total variation model. Experimental results demonstrate the proposed method outperforms most of the state-of-the-art algorithms in terms of the dehazing quality, and achieves a trade-off between the computational efficiency and haze removal capability.
