Focused on the task of fast and accurate armored target detection in ground battlefield,a detection method based on multi-scale representation network(MS-RN) and shape-fixed Guided Anchor(SF-GA)scheme is proposed.Firs...Focused on the task of fast and accurate armored target detection in ground battlefield,a detection method based on multi-scale representation network(MS-RN) and shape-fixed Guided Anchor(SF-GA)scheme is proposed.Firstly,considering the large-scale variation and camouflage of armored target,a new MS-RN integrating contextual information in battlefield environment is designed.The MS-RN extracts deep features from templates with different scales and strengthens the detection ability of small targets.Armored targets of different sizes are detected on different representation features.Secondly,aiming at the accuracy and real-time detection requirements,improved shape-fixed Guided Anchor is used on feature maps of different scales to recommend regions of interests(ROIs).Different from sliding or random anchor,the SF-GA can filter out 80% of the regions while still improving the recall.A special detection dataset for armored target,named Armored Target Dataset(ARTD),is constructed,based on which the comparable experiments with state-of-art detection methods are conducted.Experimental results show that the proposed method achieves outstanding performance in detection accuracy and efficiency,especially when small armored targets are involved.展开更多
This paper proposes a multi-scale self-recovery(MSSR)approach to protect images against content forgery.The main idea is to provide more resistance against image tampering while enabling the recovery process in a mult...This paper proposes a multi-scale self-recovery(MSSR)approach to protect images against content forgery.The main idea is to provide more resistance against image tampering while enabling the recovery process in a multi-scale quality manner.In the proposed approach,the reference data composed of several parts and each part is protected by a channel coding rate according to its importance.The first part,which is used to reconstruct a rough approximation of the original image,is highly protected in order to resist against higher tampering rates.Other parts are protected with lower rates according to their importance leading to lower tolerable tampering rate(TTR),but the higher quality of the recovered images.The proposed MSSR approach is an efficient solution for the main disadvantage of the current methods,which either recover a tampered image in low tampering rates or fails when tampering rate is above the TTR value.The simulation results on 10000 test images represent the efficiency of the multi-scale self-recovery feature of the proposed approach in comparison with the existing methods.展开更多
This paper proposes a universal framework,termed as Multi-Task Hybrid Convolutional Neural Network(MHCNN),for joint face detection,facial landmark detection,facial quality,and facial attribute analysis.MHCNN consists ...This paper proposes a universal framework,termed as Multi-Task Hybrid Convolutional Neural Network(MHCNN),for joint face detection,facial landmark detection,facial quality,and facial attribute analysis.MHCNN consists of a high-accuracy single stage detector(SSD)and an efficient tiny convolutional neural network(T-CNN)for joint face detection refinement,alignment and attribute analysis.Though the SSD face detectors achieve promising results,we find that applying a tiny CNN on detections further boosts the detected face scores and bounding boxes.By multi-task training,our T-CNN aims to provide five facial landmarks,facial quality scores,and facial attributes like wearing sunglasses and wearing masks.Since there is no public facial quality data and facial attribute data as we need,we contribute two datasets,namely FaceQ and FaceA,which are collected from the Internet.Experiments show that our MHCNN achieves face detection performance comparable to the state of the art in face detection data set and benchmark(FDDB),and gets reasonable results on AFLW,FaceQ and FaceA.展开更多
For face detection under complex background and illumination, a detection method that combines the skin color segmentation and cost-sensitive Adaboost algorithm is proposed in this paper. First, by using the character...For face detection under complex background and illumination, a detection method that combines the skin color segmentation and cost-sensitive Adaboost algorithm is proposed in this paper. First, by using the characteristic of human skin color clustering in the color space, the skin color area in YC b C r color space is extracted and a large number of irrelevant backgrounds are excluded; then for remedying the deficiencies of Adaboost algorithm, the cost-sensitive function is introduced into the Adaboost algorithm; finally the skin color segmentation and cost-sensitive Adaboost algorithm are combined for the face detection. Experimental results show that the proposed detection method has a higher detection rate and detection speed, which can more adapt to the actual field environment.展开更多
Since the coal mine in-pit personnel positioning system neither can effectively achieve the function to detect the uniqueness of in-pit coal-mine personnel nor can identify and eliminate violations in attendance manag...Since the coal mine in-pit personnel positioning system neither can effectively achieve the function to detect the uniqueness of in-pit coal-mine personnel nor can identify and eliminate violations in attendance management such as multiple cards for one person, and swiping one's cards by others in China at present. Therefore, the research introduces a uniqueness detection system and method for in-pit coal-mine personnel integrated into the in-pit coal mine personnel positioning system, establishing a system mode based on face recognition + recognition of personnel positioning card + release by automatic detection. Aiming at the facts that the in-pit personnel are wearing helmets and faces are prone to be stained during the face recognition, the study proposes the ideas that pre-process face images using the 2D-wavelet-transformation-based Mallat algorithm and extracts three face features: miner light, eyes and mouths, using the generalized symmetry transformation-based algorithm. This research carried out test with 40 clean face images with no helmets and 40 lightly-stained face images, and then compared with results with the one using the face feature extraction method based on grey-scale transformation and edge detection. The results show that the method described in the paper can detect accurately face features in the above-mentioned two cases, and the accuracy to detect face features is 97.5% in the case of wearing helmets and lightly-stained faces.展开更多
The skewed symmetry detection plays an improtant role in three-dimensional（3-D） reconstruction. The skewed symmetry depicts a real symmetry viewed from some unknown viewing directions. And the skewed symmetry detect...The skewed symmetry detection plays an improtant role in three-dimensional（3-D） reconstruction. The skewed symmetry depicts a real symmetry viewed from some unknown viewing directions. And the skewed symmetry detection can decrease the geometric constrains and the complexity of 3-D reconstruction. The detection technique for the quadric curve ellipse proposed by Sugimoto is improved to further cover quadric curves including hyperbola and parabola. With the parametric detection, the 3-D quadric curve projection matching is automatical- ly accomplished. Finally, the skewed symmetry surface of the quadric surface solid is obtained. Several examples are used to verify the feasibility of the algorithm and satisfying results can be obtained.展开更多
In response to the challenge of low detection accuracy and susceptibility to missed and false detections of small targets in unmanned aerial vehicles(UAVs)aerial images,an improved UAV image target detection algorithm...In response to the challenge of low detection accuracy and susceptibility to missed and false detections of small targets in unmanned aerial vehicles(UAVs)aerial images,an improved UAV image target detection algorithm based on YOLOv8 was proposed in this study.To begin with,the CoordAtt attention mechanism was employed to enhance the feature extraction capability of the backbone network,thereby reducing interference from backgrounds.Additionally,the BiFPN feature fusion network with an added small object detection layer was used to enhance the model's ability to perceive for small objects.Furthermore,a multi-level fusion module was designed and proposed to effectively integrate shallow and deep information.The use of an enhanced MPDIoU loss function further improved detection performance.The experimental results based on the publicly available VisDrone2019 dataset showed that the improved model outperformed the YOLOv8 baseline model,mAP@0.5 improved by 20%,and the improved method improved the detection accuracy of the model for small targets.展开更多
Detection of floating garbage in inland rivers is crucial for water environmental protection,as it effectively reduces ecological damage and ensures the safety of water resources.To address the inefficiency of traditi...Detection of floating garbage in inland rivers is crucial for water environmental protection,as it effectively reduces ecological damage and ensures the safety of water resources.To address the inefficiency of traditional cleanup methods and the challenges in detecting small targets,an improved YOLOv5 object detection model was proposed in this study.In order to enhance the model’s sensitivity to small targets and mitigate the impact of redundant information on detection performance,a bi-level routing attention mechanism was introduced and embedded into the backbone network.Additionally,a multi-scale detection head was incorporated into the model,allowing for more comprehensive coverage of floating garbage of various sizes through multi-scale feature extraction and detection.The Focal-EIoU loss function was also employed to optimize the model parameters,improving localization accuracy.Experimental results on the publicly available FloW_Img dataset demonstrated that the improved YOLOv5 model outperforms the original YOLOv5 model in terms of precision and recall,achieving a mAP(mean average precision)of 86.12%,with significant improvements and faster convergence.展开更多
The automatic detection of faces is a very important problem. The effectiveness of biometric authentication based on face mainly depends on the method used to locate the face in the image. This paper presents a hybrid...The automatic detection of faces is a very important problem. The effectiveness of biometric authentication based on face mainly depends on the method used to locate the face in the image. This paper presents a hybrid system for faces detection in unconstrained cases in which the illumination, pose, occlusion, and size of the face are uncontrolled. To do this, the new method of detection proposed in this paper is based primarily on a technique of automatic learning by using the decision of three neural networks, a technique of energy compaction by using the discrete cosine transform, and a technique of segmentation by the color of human skin. A whole of pictures （faces and no faces） are transformed to vectors of data which will be used for learning the neural networks to separate between the two classes. Discrete cosine transform is used to reduce the dimension of the vectors, to eliminate the redundancies of information, and to store only the useful information in a minimum number of coefficients while the segmentation is used to reduce the space of research in the image. The experimental results have shown that this hybridization of methods will give a very significant improvement of the rate of the recognition, quality of detection, and the time of execution.展开更多
人脸识别技术广泛应用于考勤管理、移动支付等智慧建设中。伴随着常态化的口罩干扰,传统人脸识别算法已无法满足实际应用需求,为此,本文利用深度学习模型SSD以及FaceNet模型对人脸识别系统展开设计。首先,为消除现有数据集中亚洲人脸占...人脸识别技术广泛应用于考勤管理、移动支付等智慧建设中。伴随着常态化的口罩干扰,传统人脸识别算法已无法满足实际应用需求,为此,本文利用深度学习模型SSD以及FaceNet模型对人脸识别系统展开设计。首先,为消除现有数据集中亚洲人脸占比小造成的类内间距变化差距不明显的问题,在CAS-IA Web Face公开数据集的基础上对亚洲人脸数据进行扩充;其次,为解决不同口罩样式对特征提取的干扰,使用SSD人脸检测模型与DLIB人脸关键点检测模型提取人脸关键点,并利用人脸关键点与口罩的空间位置关系,额外随机生成不同的口罩人脸,组成混合数据集;最后,在混合数据集上进行模型训练并将训练好的模型移植到人脸识别系统中,进行检测速度与识别精度验证。实验结果表明,系统的实时识别速度达20 fps以上,人脸识别模型准确率在构建的混合数据集中达到97.1%,在随机抽取的部分LFW数据集验证的准确率达99.7%,故而该系统可满足实际应用需求,在一定程度上提高人脸识别的鲁棒性与准确性。展开更多
Tunneling machines, or excavators, are large and good conductors and affect the reliability of data gathering and interpretation in advanced detection using transient electromagnetic methods. In our experiment, we use...Tunneling machines, or excavators, are large and good conductors and affect the reliability of data gathering and interpretation in advanced detection using transient electromagnetic methods. In our experiment, we used a coincident-loop and central loop type of configuration, where the coil plane l） vertical to and 2） parallel to the working face. A SIROTEM instrument at different locations was used to observe the transient electromagnetic responses of the excavator and to analyze the response amplitudes. The result shows that the tunneling machine affects the advanced detection data and is related to the way the coil is coupled. When the excavator is 6 m from the observatory, the interference of tunneling machine can be ignored.展开更多
In this paper,based on a bidirectional parallel multi-branch feature pyramid network(BPMFPN),a novel one-stage object detector called BPMFPN Det is proposed for real-time detection of ground multi-scale targets by swa...In this paper,based on a bidirectional parallel multi-branch feature pyramid network(BPMFPN),a novel one-stage object detector called BPMFPN Det is proposed for real-time detection of ground multi-scale targets by swarm unmanned aerial vehicles(UAVs).First,the bidirectional parallel multi-branch convolution modules are used to construct the feature pyramid to enhance the feature expression abilities of different scale feature layers.Next,the feature pyramid is integrated into the single-stage object detection framework to ensure real-time performance.In order to validate the effectiveness of the proposed algorithm,experiments are conducted on four datasets.For the PASCAL VOC dataset,the proposed algorithm achieves the mean average precision(mAP)of 85.4 on the VOC 2007 test set.With regard to the detection in optical remote sensing(DIOR)dataset,the proposed algorithm achieves 73.9 mAP.For vehicle detection in aerial imagery(VEDAI)dataset,the detection accuracy of small land vehicle(slv)targets reaches 97.4 mAP.For unmanned aerial vehicle detection and tracking(UAVDT)dataset,the proposed BPMFPN Det achieves the mAP of 48.75.Compared with the previous state-of-the-art methods,the results obtained by the proposed algorithm are more competitive.The experimental results demonstrate that the proposed algorithm can effectively solve the problem of real-time detection of ground multi-scale targets in aerial images of swarm UAVs.展开更多
针对在现有人脸静态识别过程中被识别人需等待配合的问题,文中提出了一种动态人脸识别系统。该系统采用了基于RetinaFace与FaceNet算法的动态人脸检测和识别方法,并进行了优化,以达到高识别精度和实时性的目标。其中,RetinaFace检测采用...针对在现有人脸静态识别过程中被识别人需等待配合的问题,文中提出了一种动态人脸识别系统。该系统采用了基于RetinaFace与FaceNet算法的动态人脸检测和识别方法,并进行了优化,以达到高识别精度和实时性的目标。其中,RetinaFace检测采用GhostNet作为骨干网络,使用Adaptive-NMS(Non Max Suppression)非极大值抑制用于人脸框的回归,FaceNet识别采用MobileNetV1作为骨干网络,使用Triplet损失与交叉熵损失结合的联合损失函数用以人脸分类。优化后的算法在检测与识别上具有良好表现,改进RetinaFace算法在WiderFace数据集下检测精度为93.35%、90.84%和80.43%,FPS(Frames Per Second)可达53 frame·s^(-1)。动态人脸检测平均检测精度为96%,FPS为21 frame·s^(-1)。当FaceNet阈值设为1.15时,识别率最高达到98.23%。动态识别系统平均识别精度98%,FPS可达20 frame·s^(-1)。实验结果表明,该系统解决了人脸静态识别中需等待配合的问题,具有较高的识别精度与实时性。展开更多
Technological advances in computer science and their application in our daily life allow us to improve our understanding of problems and solve them effectively.A system design to detect people with fever and determine...Technological advances in computer science and their application in our daily life allow us to improve our understanding of problems and solve them effectively.A system design to detect people with fever and determine highrisk areas using infrared thermography and big data is presented.In order to detect people with fever,face detection algorithms of Viola-Jones and Kanade-Lucas are investigated,and comparison between them is presented using a training set of 406 thermal images and a test set of 2072 thermal images.Thermography analysis is performed on detected faces to obtain the temperature level on Celsius scale.With this information a sample database is created.To perform big data experimental analysis,Power Bi tool is used to determine the high-risk area.The experimental results show that Viola-Jones algorithm has a higher performance recognizing faces of thermal images than KanadeLucas,having a high detection rate,less false-positives rate and false-negatives rate.展开更多
With the development of sensors,the application of multi-source remote sensing data has been widely concerned.Since hyperspectral image(HSI)contains rich spectral information while light detection and ranging(LiDAR)da...With the development of sensors,the application of multi-source remote sensing data has been widely concerned.Since hyperspectral image(HSI)contains rich spectral information while light detection and ranging(LiDAR)data contains elevation information,joint use of them for ground object classification can yield positive results,especially by building deep networks.Fortu-nately,multi-scale deep networks allow to expand the receptive fields of convolution without causing the computational and training problems associated with simply adding more network layers.In this work,a multi-scale feature fusion network is proposed for the joint classification of HSI and LiDAR data.First,we design a multi-scale spatial feature extraction module with cross-channel connections,by which spatial information of HSI data and elevation information of LiDAR data are extracted and fused.In addition,a multi-scale spectral feature extraction module is employed to extract the multi-scale spectral features of HSI data.Finally,joint multi-scale features are obtained by weighting and concatenation operations and then fed into the classifier.To verify the effective-ness of the proposed network,experiments are carried out on the MUUFL Gulfport and Trento datasets.The experimental results demonstrate that the classification performance of the proposed method is superior to that of other state-of-the-art methods.展开更多
基金supported by the National Key Research and Development Program of China under grant 2016YFC0802904National Natural Science Foundation of China under grant61671470the Postdoctoral Science Foundation Funded Project of China under grant 2017M623423。
文摘Focused on the task of fast and accurate armored target detection in ground battlefield,a detection method based on multi-scale representation network(MS-RN) and shape-fixed Guided Anchor(SF-GA)scheme is proposed.Firstly,considering the large-scale variation and camouflage of armored target,a new MS-RN integrating contextual information in battlefield environment is designed.The MS-RN extracts deep features from templates with different scales and strengthens the detection ability of small targets.Armored targets of different sizes are detected on different representation features.Secondly,aiming at the accuracy and real-time detection requirements,improved shape-fixed Guided Anchor is used on feature maps of different scales to recommend regions of interests(ROIs).Different from sliding or random anchor,the SF-GA can filter out 80% of the regions while still improving the recall.A special detection dataset for armored target,named Armored Target Dataset(ARTD),is constructed,based on which the comparable experiments with state-of-art detection methods are conducted.Experimental results show that the proposed method achieves outstanding performance in detection accuracy and efficiency,especially when small armored targets are involved.
文摘This paper proposes a multi-scale self-recovery(MSSR)approach to protect images against content forgery.The main idea is to provide more resistance against image tampering while enabling the recovery process in a multi-scale quality manner.In the proposed approach,the reference data composed of several parts and each part is protected by a channel coding rate according to its importance.The first part,which is used to reconstruct a rough approximation of the original image,is highly protected in order to resist against higher tampering rates.Other parts are protected with lower rates according to their importance leading to lower tolerable tampering rate(TTR),but the higher quality of the recovered images.The proposed MSSR approach is an efficient solution for the main disadvantage of the current methods,which either recover a tampered image in low tampering rates or fails when tampering rate is above the TTR value.The simulation results on 10000 test images represent the efficiency of the multi-scale self-recovery feature of the proposed approach in comparison with the existing methods.
基金supported by ZTE Corporation and State Key Laboratory of Mobile Network and Mobile Multimedia Technology
文摘This paper proposes a universal framework,termed as Multi-Task Hybrid Convolutional Neural Network(MHCNN),for joint face detection,facial landmark detection,facial quality,and facial attribute analysis.MHCNN consists of a high-accuracy single stage detector(SSD)and an efficient tiny convolutional neural network(T-CNN)for joint face detection refinement,alignment and attribute analysis.Though the SSD face detectors achieve promising results,we find that applying a tiny CNN on detections further boosts the detected face scores and bounding boxes.By multi-task training,our T-CNN aims to provide five facial landmarks,facial quality scores,and facial attributes like wearing sunglasses and wearing masks.Since there is no public facial quality data and facial attribute data as we need,we contribute two datasets,namely FaceQ and FaceA,which are collected from the Internet.Experiments show that our MHCNN achieves face detection performance comparable to the state of the art in face detection data set and benchmark(FDDB),and gets reasonable results on AFLW,FaceQ and FaceA.
基金supported by the National Basic Research Program of China(973 Program)under Grant No.2012CB215202the National Natural Science Foundation of China under Grant No.51205046
文摘For face detection under complex background and illumination, a detection method that combines the skin color segmentation and cost-sensitive Adaboost algorithm is proposed in this paper. First, by using the characteristic of human skin color clustering in the color space, the skin color area in YC b C r color space is extracted and a large number of irrelevant backgrounds are excluded; then for remedying the deficiencies of Adaboost algorithm, the cost-sensitive function is introduced into the Adaboost algorithm; finally the skin color segmentation and cost-sensitive Adaboost algorithm are combined for the face detection. Experimental results show that the proposed detection method has a higher detection rate and detection speed, which can more adapt to the actual field environment.
基金financial supports from the National Natural Science Foundation of China (No. 51134024)the National High Technology Research and Development Program of China (No. 2012AA062203)are gratefully acknowledged
文摘Since the coal mine in-pit personnel positioning system neither can effectively achieve the function to detect the uniqueness of in-pit coal-mine personnel nor can identify and eliminate violations in attendance management such as multiple cards for one person, and swiping one's cards by others in China at present. Therefore, the research introduces a uniqueness detection system and method for in-pit coal-mine personnel integrated into the in-pit coal mine personnel positioning system, establishing a system mode based on face recognition + recognition of personnel positioning card + release by automatic detection. Aiming at the facts that the in-pit personnel are wearing helmets and faces are prone to be stained during the face recognition, the study proposes the ideas that pre-process face images using the 2D-wavelet-transformation-based Mallat algorithm and extracts three face features: miner light, eyes and mouths, using the generalized symmetry transformation-based algorithm. This research carried out test with 40 clean face images with no helmets and 40 lightly-stained face images, and then compared with results with the one using the face feature extraction method based on grey-scale transformation and edge detection. The results show that the method described in the paper can detect accurately face features in the above-mentioned two cases, and the accuracy to detect face features is 97.5% in the case of wearing helmets and lightly-stained faces.
基金Supported by the National Natural Science Foundation of China(10377007)~~
文摘The skewed symmetry detection plays an improtant role in three-dimensional（3-D） reconstruction. The skewed symmetry depicts a real symmetry viewed from some unknown viewing directions. And the skewed symmetry detection can decrease the geometric constrains and the complexity of 3-D reconstruction. The detection technique for the quadric curve ellipse proposed by Sugimoto is improved to further cover quadric curves including hyperbola and parabola. With the parametric detection, the 3-D quadric curve projection matching is automatical- ly accomplished. Finally, the skewed symmetry surface of the quadric surface solid is obtained. Several examples are used to verify the feasibility of the algorithm and satisfying results can be obtained.
文摘In response to the challenge of low detection accuracy and susceptibility to missed and false detections of small targets in unmanned aerial vehicles(UAVs)aerial images,an improved UAV image target detection algorithm based on YOLOv8 was proposed in this study.To begin with,the CoordAtt attention mechanism was employed to enhance the feature extraction capability of the backbone network,thereby reducing interference from backgrounds.Additionally,the BiFPN feature fusion network with an added small object detection layer was used to enhance the model's ability to perceive for small objects.Furthermore,a multi-level fusion module was designed and proposed to effectively integrate shallow and deep information.The use of an enhanced MPDIoU loss function further improved detection performance.The experimental results based on the publicly available VisDrone2019 dataset showed that the improved model outperformed the YOLOv8 baseline model,mAP@0.5 improved by 20%,and the improved method improved the detection accuracy of the model for small targets.
文摘Detection of floating garbage in inland rivers is crucial for water environmental protection,as it effectively reduces ecological damage and ensures the safety of water resources.To address the inefficiency of traditional cleanup methods and the challenges in detecting small targets,an improved YOLOv5 object detection model was proposed in this study.In order to enhance the model’s sensitivity to small targets and mitigate the impact of redundant information on detection performance,a bi-level routing attention mechanism was introduced and embedded into the backbone network.Additionally,a multi-scale detection head was incorporated into the model,allowing for more comprehensive coverage of floating garbage of various sizes through multi-scale feature extraction and detection.The Focal-EIoU loss function was also employed to optimize the model parameters,improving localization accuracy.Experimental results on the publicly available FloW_Img dataset demonstrated that the improved YOLOv5 model outperforms the original YOLOv5 model in terms of precision and recall,achieving a mAP(mean average precision)of 86.12%,with significant improvements and faster convergence.
基金supported by the Laboratory of Inverses Problems, Modeling, Information and Systems (PI:MIS), Department of Electronic and Telecommunication, University of 08 Mai 1945, Guelma, Algériathe Laboratory of Computer Research (LRI), Department of Computer Sciences, University of Badji Mokhtar, Annaba, Algéria
文摘The automatic detection of faces is a very important problem. The effectiveness of biometric authentication based on face mainly depends on the method used to locate the face in the image. This paper presents a hybrid system for faces detection in unconstrained cases in which the illumination, pose, occlusion, and size of the face are uncontrolled. To do this, the new method of detection proposed in this paper is based primarily on a technique of automatic learning by using the decision of three neural networks, a technique of energy compaction by using the discrete cosine transform, and a technique of segmentation by the color of human skin. A whole of pictures （faces and no faces） are transformed to vectors of data which will be used for learning the neural networks to separate between the two classes. Discrete cosine transform is used to reduce the dimension of the vectors, to eliminate the redundancies of information, and to store only the useful information in a minimum number of coefficients while the segmentation is used to reduce the space of research in the image. The experimental results have shown that this hybridization of methods will give a very significant improvement of the rate of the recognition, quality of detection, and the time of execution.
文摘人脸识别技术广泛应用于考勤管理、移动支付等智慧建设中。伴随着常态化的口罩干扰,传统人脸识别算法已无法满足实际应用需求,为此,本文利用深度学习模型SSD以及FaceNet模型对人脸识别系统展开设计。首先,为消除现有数据集中亚洲人脸占比小造成的类内间距变化差距不明显的问题,在CAS-IA Web Face公开数据集的基础上对亚洲人脸数据进行扩充;其次,为解决不同口罩样式对特征提取的干扰,使用SSD人脸检测模型与DLIB人脸关键点检测模型提取人脸关键点,并利用人脸关键点与口罩的空间位置关系,额外随机生成不同的口罩人脸,组成混合数据集;最后,在混合数据集上进行模型训练并将训练好的模型移植到人脸识别系统中,进行检测速度与识别精度验证。实验结果表明,系统的实时识别速度达20 fps以上,人脸识别模型准确率在构建的混合数据集中达到97.1%,在随机抽取的部分LFW数据集验证的准确率达99.7%,故而该系统可满足实际应用需求,在一定程度上提高人脸识别的鲁棒性与准确性。
基金support received from the National Basic Research Program of China (No2007CB209400)the National Natural Science Foundation of China (No50774085)the Young Scientists Fund of the School Science Foundation of CUMT (No2008A046)
文摘Tunneling machines, or excavators, are large and good conductors and affect the reliability of data gathering and interpretation in advanced detection using transient electromagnetic methods. In our experiment, we used a coincident-loop and central loop type of configuration, where the coil plane l） vertical to and 2） parallel to the working face. A SIROTEM instrument at different locations was used to observe the transient electromagnetic responses of the excavator and to analyze the response amplitudes. The result shows that the tunneling machine affects the advanced detection data and is related to the way the coil is coupled. When the excavator is 6 m from the observatory, the interference of tunneling machine can be ignored.
文摘In this paper,based on a bidirectional parallel multi-branch feature pyramid network(BPMFPN),a novel one-stage object detector called BPMFPN Det is proposed for real-time detection of ground multi-scale targets by swarm unmanned aerial vehicles(UAVs).First,the bidirectional parallel multi-branch convolution modules are used to construct the feature pyramid to enhance the feature expression abilities of different scale feature layers.Next,the feature pyramid is integrated into the single-stage object detection framework to ensure real-time performance.In order to validate the effectiveness of the proposed algorithm,experiments are conducted on four datasets.For the PASCAL VOC dataset,the proposed algorithm achieves the mean average precision(mAP)of 85.4 on the VOC 2007 test set.With regard to the detection in optical remote sensing(DIOR)dataset,the proposed algorithm achieves 73.9 mAP.For vehicle detection in aerial imagery(VEDAI)dataset,the detection accuracy of small land vehicle(slv)targets reaches 97.4 mAP.For unmanned aerial vehicle detection and tracking(UAVDT)dataset,the proposed BPMFPN Det achieves the mAP of 48.75.Compared with the previous state-of-the-art methods,the results obtained by the proposed algorithm are more competitive.The experimental results demonstrate that the proposed algorithm can effectively solve the problem of real-time detection of ground multi-scale targets in aerial images of swarm UAVs.
文摘针对在现有人脸静态识别过程中被识别人需等待配合的问题,文中提出了一种动态人脸识别系统。该系统采用了基于RetinaFace与FaceNet算法的动态人脸检测和识别方法,并进行了优化,以达到高识别精度和实时性的目标。其中,RetinaFace检测采用GhostNet作为骨干网络,使用Adaptive-NMS(Non Max Suppression)非极大值抑制用于人脸框的回归,FaceNet识别采用MobileNetV1作为骨干网络,使用Triplet损失与交叉熵损失结合的联合损失函数用以人脸分类。优化后的算法在检测与识别上具有良好表现,改进RetinaFace算法在WiderFace数据集下检测精度为93.35%、90.84%和80.43%,FPS(Frames Per Second)可达53 frame·s^(-1)。动态人脸检测平均检测精度为96%,FPS为21 frame·s^(-1)。当FaceNet阈值设为1.15时,识别率最高达到98.23%。动态识别系统平均识别精度98%,FPS可达20 frame·s^(-1)。实验结果表明,该系统解决了人脸静态识别中需等待配合的问题,具有较高的识别精度与实时性。
文摘Technological advances in computer science and their application in our daily life allow us to improve our understanding of problems and solve them effectively.A system design to detect people with fever and determine highrisk areas using infrared thermography and big data is presented.In order to detect people with fever,face detection algorithms of Viola-Jones and Kanade-Lucas are investigated,and comparison between them is presented using a training set of 406 thermal images and a test set of 2072 thermal images.Thermography analysis is performed on detected faces to obtain the temperature level on Celsius scale.With this information a sample database is created.To perform big data experimental analysis,Power Bi tool is used to determine the high-risk area.The experimental results show that Viola-Jones algorithm has a higher performance recognizing faces of thermal images than KanadeLucas,having a high detection rate,less false-positives rate and false-negatives rate.
基金supported by the National Key Research and Development Project(No.2020YFC1512000)the General Projects of Key R&D Programs in Shaanxi Province(No.2020GY-060)Xi’an Science&Technology Project(No.2020KJRC 0126)。
文摘With the development of sensors,the application of multi-source remote sensing data has been widely concerned.Since hyperspectral image(HSI)contains rich spectral information while light detection and ranging(LiDAR)data contains elevation information,joint use of them for ground object classification can yield positive results,especially by building deep networks.Fortu-nately,multi-scale deep networks allow to expand the receptive fields of convolution without causing the computational and training problems associated with simply adding more network layers.In this work,a multi-scale feature fusion network is proposed for the joint classification of HSI and LiDAR data.First,we design a multi-scale spatial feature extraction module with cross-channel connections,by which spatial information of HSI data and elevation information of LiDAR data are extracted and fused.In addition,a multi-scale spectral feature extraction module is employed to extract the multi-scale spectral features of HSI data.Finally,joint multi-scale features are obtained by weighting and concatenation operations and then fed into the classifier.To verify the effective-ness of the proposed network,experiments are carried out on the MUUFL Gulfport and Trento datasets.The experimental results demonstrate that the classification performance of the proposed method is superior to that of other state-of-the-art methods.
