Urban land provides a suitable location for various economic activities which affect the development of surrounding areas. With rapid industrialization and urbanization, the contradictions in land-use become more noti...Urban land provides a suitable location for various economic activities which affect the development of surrounding areas. With rapid industrialization and urbanization, the contradictions in land-use become more noticeable. Urban administrators and decision-makers seek modern methods and technology to provide information support for urban growth. Recently, with the fast development of high-resolution sensor technology, more relevant data can be obtained, which is an advantage in studying the sustainable development of urban land-use. However, these data are only information sources and are a mixture of ＂information＂ and ＂noise＂. Processing, analysis and information extraction from remote sensing data is necessary to provide useful information. This paper extracts urban land-use information from a high-resolution image by using the multi-feature information of the image objects, and adopts an object-oriented image analysis approach and multi-scale image segmentation technology. A classification and extraction model is set up based on the multi-features of the image objects, in order to contribute to information for reasonable planning and effective management. This new image analysis approach offers a satisfactory solution for extracting information quickly and efficiently.展开更多
In the video-based surveillance application, moving shadows can affect the correct localization and detection of moving objects. This paper aims to present a method for shadow detection and suppression used for moving...In the video-based surveillance application, moving shadows can affect the correct localization and detection of moving objects. This paper aims to present a method for shadow detection and suppression used for moving visual object detection. The major novelty of the shadow suppression is the integration of several features including photometric invariant color feature, motion edge feature, and spatial feature etc. By modifying process for false shadow detected, the averaging detection rate of moving object reaches above 90% in the test of Hall-Monitor sequence.展开更多
To research the effect of the selection method of multi-objects genetic algorithm problem on optimizing result, thismethod is analyzed theoretically and discussed by using an autonomous underwater vehicle(AUV) as an o...To research the effect of the selection method of multi-objects genetic algorithm problem on optimizing result, thismethod is analyzed theoretically and discussed by using an autonomous underwater vehicle(AUV) as an object. A changingweight vtlue method is put forward and a selection formula is modified. Some experiments were implemented on an AUV.TwinBurger. The results shows that this method is effective and feasible.展开更多
Deep Learning is a powerful technique that is widely applied to Image Recognition and Natural Language Processing tasks amongst many other tasks. In this work, we propose an efficient technique to utilize pre-trained ...Deep Learning is a powerful technique that is widely applied to Image Recognition and Natural Language Processing tasks amongst many other tasks. In this work, we propose an efficient technique to utilize pre-trained Convolutional Neural Network (CNN) architectures to extract powerful features from images for object recognition purposes. We have built on the existing concept of extending the learning from pre-trained CNNs to new databases through activations by proposing to consider multiple deep layers. We have exploited the progressive learning that happens at the various intermediate layers of the CNNs to construct Deep Multi-Layer (DM-L) based Feature Extraction vectors to achieve excellent object recognition performance. Two popular pre-trained CNN architecture models i.e. the VGG_16 and VGG_19 have been used in this work to extract the feature sets from 3 deep fully connected multiple layers namely “fc6”, “fc7” and “fc8” from inside the models for object recognition purposes. Using the Principal Component Analysis (PCA) technique, the Dimensionality of the DM-L feature vectors has been reduced to form powerful feature vectors that have been fed to an external Classifier Ensemble for classification instead of the Softmax based classification layers of the two original pre-trained CNN models. The proposed DM-L technique has been applied to the Benchmark Caltech-101 object recognition database. Conventional wisdom may suggest that feature extractions based on the deepest layer i.e. “fc8” compared to “fc6” will result in the best recognition performance but our results have proved it otherwise for the two considered models. Our experiments have revealed that for the two models under consideration, the “fc6” based feature vectors have achieved the best recognition performance. State-of-the-Art recognition performances of 91.17% and 91.35% have been achieved by utilizing the “fc6” based feature vectors for the VGG_16 and VGG_19 models respectively. The recognition performance has been achieved by considering 30 sample images per class whereas the proposed system is capable of achieving improved performance by considering all sample images per class. Our research shows that for feature extraction based on CNNs, multiple layers should be considered and then the best layer can be selected that maximizes the recognition performance.展开更多
在复杂的低照度环境中获取的图像存在亮度低、噪声多和细节信息丢失等问题,直接使用通用的目标检测方法无法达到较为理想的效果.为此,提出低照度目标检测方法——Dark-YOLO.首先,使用CSPDarkNet-53骨干网络提取低照度图像特征,并提出路...在复杂的低照度环境中获取的图像存在亮度低、噪声多和细节信息丢失等问题,直接使用通用的目标检测方法无法达到较为理想的效果.为此,提出低照度目标检测方法——Dark-YOLO.首先,使用CSPDarkNet-53骨干网络提取低照度图像特征,并提出路径聚合增强模块以进一步增强特征表征能力;然后,设计金字塔平衡注意力模块捕获多尺度特征并加以有效利用,生成包含不同尺度且更具判别力的特征;最后,使用预测交并比(intersection over union,IoU)改进检测头,IoU预测分支为每个预测框预测IoU值,使得目标定位更加准确.在ExDark数据集上的实验结果表明,相较于YOLOv4,均值平均精度(mAP)提升了4.10%,Dark-YOLO方法能够有效地提高在低照度场景下目标检测的性能.展开更多
To meet the development trend of multi-bar warp knit-ting machine towards high-speed,advanced technologyand computer control and the requirements of variousproducts with small quantity,there are many researcheson the ...To meet the development trend of multi-bar warp knit-ting machine towards high-speed,advanced technologyand computer control and the requirements of variousproducts with small quantity,there are many researcheson the computer-aided pattern design of multi-barwarp knitted fabrics.In terms of the special propertiesof the computer-aided pattern design of multi-barwarp knitted fabrics,the Object Oriented Program(OOP)programming-Object Windows class Library(OWL)programming is selected.According to thecharacters of the OWL programming,various functionsare defined.Pattern design and technical parameters canbe output,which offers a great convenience for the fac-tory.展开更多
针对3D目标检测点云处理方法存在下采样目标点少,小目标特征信息丢失的问题,提出改进的PointVoxel特征提取方法.首先,以当前先进3D目标检测PV-RCNN(Point-Voxel Feature Set Abstraction for 3D Object Detection)模型为基础,就point-ba...针对3D目标检测点云处理方法存在下采样目标点少,小目标特征信息丢失的问题,提出改进的PointVoxel特征提取方法.首先,以当前先进3D目标检测PV-RCNN(Point-Voxel Feature Set Abstraction for 3D Object Detection)模型为基础,就point-based (基于纯点云)采样后目标点数量较少的问题,提出C-FPS (基于中心最远点采样)算法,即通过图像筛选点云范围,根据标签设置对X增加一个归一化乘以中心点的偏移量,优化点云分布,提高下采样目标点数量;然后,针对voxel-based (基于体素)需要划分体素大小与特征提取平衡的问题,提出体素图像特征融合方法,通过多通道卷积神经网络提取目标图像特征,将多通道特征与voxel-based提取的点云特征进行融合,弥补划分大小导致的特征信息丢失;最后,在KITTI数据集上进行验证.实验表明,与PV-RCNN模型相比,在当前困扰计算机视觉中的小目标检测上,该特征提取方法有效地提升了对小目标的检测能力,对于小目标行人和骑行者,其平均识别精度均优于PV-RCNN模型,提升幅度分别达到了1.62%,1.81%.展开更多
基金The paper is supported by the Research Foundation for OutstandingYoung Teachers , China University of Geosciences ( Wuhan) ( No .CUGQNL0616) Research Foundationfor State Key Laboratory of Geo-logical Processes and Mineral Resources ( No . MGMR2002-02)Hubei Provincial Depart ment of Education (B) .
文摘Urban land provides a suitable location for various economic activities which affect the development of surrounding areas. With rapid industrialization and urbanization, the contradictions in land-use become more noticeable. Urban administrators and decision-makers seek modern methods and technology to provide information support for urban growth. Recently, with the fast development of high-resolution sensor technology, more relevant data can be obtained, which is an advantage in studying the sustainable development of urban land-use. However, these data are only information sources and are a mixture of ＂information＂ and ＂noise＂. Processing, analysis and information extraction from remote sensing data is necessary to provide useful information. This paper extracts urban land-use information from a high-resolution image by using the multi-feature information of the image objects, and adopts an object-oriented image analysis approach and multi-scale image segmentation technology. A classification and extraction model is set up based on the multi-features of the image objects, in order to contribute to information for reasonable planning and effective management. This new image analysis approach offers a satisfactory solution for extracting information quickly and efficiently.
文摘In the video-based surveillance application, moving shadows can affect the correct localization and detection of moving objects. This paper aims to present a method for shadow detection and suppression used for moving visual object detection. The major novelty of the shadow suppression is the integration of several features including photometric invariant color feature, motion edge feature, and spatial feature etc. By modifying process for false shadow detected, the averaging detection rate of moving object reaches above 90% in the test of Hall-Monitor sequence.
文摘由于低照度图像具有对比度低、细节丢失严重、噪声大等缺点,现有的目标检测算法对低照度图像的检测效果不理想.为此,本文提出一种结合空间感知注意力机制和多尺度特征融合(Spatial-aware Attention Mechanism and Multi-Scale Feature Fusion,SAM-MSFF)的低照度目标检测方法 .该方法首先通过多尺度交互内存金字塔融合多尺度特征,增强低照度图像特征中的有效信息,并设置内存向量存储样本的特征,捕获样本之间的潜在关联性;然后,引入空间感知注意力机制获取特征在空间域的长距离上下文信息和局部信息,从而增强低照度图像中的目标特征,抑制背景信息和噪声的干扰;最后,利用多感受野增强模块扩张特征的感受野,对具有不同感受野的特征进行分组重加权计算,使检测网络根据输入的多尺度信息自适应地调整感受野的大小.在ExDark数据集上进行实验,本文方法的平均精度(mean Average Precision,mAP)达到77.04%,比现有的主流目标检测方法提高2.6%~14.34%.
文摘To research the effect of the selection method of multi-objects genetic algorithm problem on optimizing result, thismethod is analyzed theoretically and discussed by using an autonomous underwater vehicle(AUV) as an object. A changingweight vtlue method is put forward and a selection formula is modified. Some experiments were implemented on an AUV.TwinBurger. The results shows that this method is effective and feasible.
文摘Deep Learning is a powerful technique that is widely applied to Image Recognition and Natural Language Processing tasks amongst many other tasks. In this work, we propose an efficient technique to utilize pre-trained Convolutional Neural Network (CNN) architectures to extract powerful features from images for object recognition purposes. We have built on the existing concept of extending the learning from pre-trained CNNs to new databases through activations by proposing to consider multiple deep layers. We have exploited the progressive learning that happens at the various intermediate layers of the CNNs to construct Deep Multi-Layer (DM-L) based Feature Extraction vectors to achieve excellent object recognition performance. Two popular pre-trained CNN architecture models i.e. the VGG_16 and VGG_19 have been used in this work to extract the feature sets from 3 deep fully connected multiple layers namely “fc6”, “fc7” and “fc8” from inside the models for object recognition purposes. Using the Principal Component Analysis (PCA) technique, the Dimensionality of the DM-L feature vectors has been reduced to form powerful feature vectors that have been fed to an external Classifier Ensemble for classification instead of the Softmax based classification layers of the two original pre-trained CNN models. The proposed DM-L technique has been applied to the Benchmark Caltech-101 object recognition database. Conventional wisdom may suggest that feature extractions based on the deepest layer i.e. “fc8” compared to “fc6” will result in the best recognition performance but our results have proved it otherwise for the two considered models. Our experiments have revealed that for the two models under consideration, the “fc6” based feature vectors have achieved the best recognition performance. State-of-the-Art recognition performances of 91.17% and 91.35% have been achieved by utilizing the “fc6” based feature vectors for the VGG_16 and VGG_19 models respectively. The recognition performance has been achieved by considering 30 sample images per class whereas the proposed system is capable of achieving improved performance by considering all sample images per class. Our research shows that for feature extraction based on CNNs, multiple layers should be considered and then the best layer can be selected that maximizes the recognition performance.
文摘在复杂的低照度环境中获取的图像存在亮度低、噪声多和细节信息丢失等问题,直接使用通用的目标检测方法无法达到较为理想的效果.为此,提出低照度目标检测方法——Dark-YOLO.首先,使用CSPDarkNet-53骨干网络提取低照度图像特征,并提出路径聚合增强模块以进一步增强特征表征能力;然后,设计金字塔平衡注意力模块捕获多尺度特征并加以有效利用,生成包含不同尺度且更具判别力的特征;最后,使用预测交并比(intersection over union,IoU)改进检测头,IoU预测分支为每个预测框预测IoU值,使得目标定位更加准确.在ExDark数据集上的实验结果表明,相较于YOLOv4,均值平均精度(mAP)提升了4.10%,Dark-YOLO方法能够有效地提高在低照度场景下目标检测的性能.
基金安徽省高等学校科学研究项目(2022AH050834)安徽理工大学引进人才科研启动基金项目(2022yjrc61)+1 种基金安徽理工大学矿山智能技术与装备省部共建协同创新中心开放基金项目(CICJMITE202206)Open Fund of State Key Laboratory of Mining Response and Disaster Prevention and Control in Deep Coal Mines(SKLMRDPC22KF24)。
文摘To meet the development trend of multi-bar warp knit-ting machine towards high-speed,advanced technologyand computer control and the requirements of variousproducts with small quantity,there are many researcheson the computer-aided pattern design of multi-barwarp knitted fabrics.In terms of the special propertiesof the computer-aided pattern design of multi-barwarp knitted fabrics,the Object Oriented Program(OOP)programming-Object Windows class Library(OWL)programming is selected.According to thecharacters of the OWL programming,various functionsare defined.Pattern design and technical parameters canbe output,which offers a great convenience for the fac-tory.
文摘针对3D目标检测点云处理方法存在下采样目标点少,小目标特征信息丢失的问题,提出改进的PointVoxel特征提取方法.首先,以当前先进3D目标检测PV-RCNN(Point-Voxel Feature Set Abstraction for 3D Object Detection)模型为基础,就point-based (基于纯点云)采样后目标点数量较少的问题,提出C-FPS (基于中心最远点采样)算法,即通过图像筛选点云范围,根据标签设置对X增加一个归一化乘以中心点的偏移量,优化点云分布,提高下采样目标点数量;然后,针对voxel-based (基于体素)需要划分体素大小与特征提取平衡的问题,提出体素图像特征融合方法,通过多通道卷积神经网络提取目标图像特征,将多通道特征与voxel-based提取的点云特征进行融合,弥补划分大小导致的特征信息丢失;最后,在KITTI数据集上进行验证.实验表明,与PV-RCNN模型相比,在当前困扰计算机视觉中的小目标检测上,该特征提取方法有效地提升了对小目标的检测能力,对于小目标行人和骑行者,其平均识别精度均优于PV-RCNN模型,提升幅度分别达到了1.62%,1.81%.
