Detecting highly-overlapped objects in crowded scenes remains a challenging problem,especially for one-stage detector.In this paper,we extricate YOLOv4 from the dilemma in a crowd by fine-tuning its detection scheme,n...Detecting highly-overlapped objects in crowded scenes remains a challenging problem,especially for one-stage detector.In this paper,we extricate YOLOv4 from the dilemma in a crowd by fine-tuning its detection scheme,named YOLO-CS.Specifically,we give YOLOv4 the power to detect multiple objects in one cell.Center to our method is the carefully designed joint prediction scheme,which is executed through an assignment of bounding boxes and a joint loss.Equipped with the derived joint-object augmentation(DJA),refined regression loss(RL)and Score-NMS(SN),YOLO-CS achieves competitive detection performance on CrowdHuman and CityPersons benchmarks compared with state-of-the-art detectors at the cost of little time.Furthermore,on the widely used general benchmark COCO,YOLOCS still has a good performance,indicating its robustness to various scenes.展开更多
Camouflaged people are extremely expert in actively concealing themselves by effectively utilizing cover and the surrounding environment. Despite advancements in optical detection capabilities through imaging systems,...Camouflaged people are extremely expert in actively concealing themselves by effectively utilizing cover and the surrounding environment. Despite advancements in optical detection capabilities through imaging systems, including spectral, polarization, and infrared technologies, there is still a lack of effective real-time method for accurately detecting small-size and high-efficient camouflaged people in complex real-world scenes. Here, this study proposes a snapshot multispectral image-based camouflaged detection model, multispectral YOLO(MS-YOLO), which utilizes the SPD-Conv and Sim AM modules to effectively represent targets and suppress background interference by exploiting the spatial-spectral target information. Besides, the study constructs the first real-shot multispectral camouflaged people dataset(MSCPD), which encompasses diverse scenes, target scales, and attitudes. To minimize information redundancy, MS-YOLO selects an optimal subset of 12 bands with strong feature representation and minimal inter-band correlation as input. Through experiments on the MSCPD, MS-YOLO achieves a mean Average Precision of 94.31% and real-time detection at 65 frames per second, which confirms the effectiveness and efficiency of our method in detecting camouflaged people in various typical desert and forest scenes. Our approach offers valuable support to improve the perception capabilities of unmanned aerial vehicles in detecting enemy forces and rescuing personnel in battlefield.展开更多
Low-light image enhancement is one of the most active research areas in the field of computer vision in recent years.In the low-light image enhancement process,loss of image details and increase in noise occur inevita...Low-light image enhancement is one of the most active research areas in the field of computer vision in recent years.In the low-light image enhancement process,loss of image details and increase in noise occur inevitably,influencing the quality of enhanced images.To alleviate this problem,a low-light image enhancement model called RetinexNet model based on Retinex theory was proposed in this study.The model was composed of an image decomposition module and a brightness enhancement module.In the decomposition module,a convolutional block attention module(CBAM)was incorporated to enhance feature representation capacity of the network,focusing on crucial features and suppressing irrelevant ones.A multifeature fusion denoising module was designed within the brightness enhancement module,circumventing the issue of feature loss during downsampling.The proposed model outperforms the existing algorithms in terms of PSNR and SSIM metrics on the publicly available datasets LOL and MIT-Adobe FiveK,as well as gives superior results in terms of NIQE metrics on the publicly available dataset LIME.展开更多
Although compressed sensing inverse synthetic aperture radar(ISAR) imaging methods are widely used in radar signal processing, its reconstructing time and memory storage space requirements are very high. The main reas...Although compressed sensing inverse synthetic aperture radar(ISAR) imaging methods are widely used in radar signal processing, its reconstructing time and memory storage space requirements are very high. The main reason is that large scene reconstruction needs a higher dimension of the sensing matrix. To reduce this limitation, a fast high resolution ISAR imaging method,which is based on scene segmentation for random chirp frequencystepped signals, is proposed. The idea of scene segmentation is used to solve the problems aforementioned. In the method,firstly, the observed scene is divided into multiple sub-scenes and then the sub-scenes are reconstructed respectively. Secondly, the whole image scene can be obtained through the stitching of the sub-scenes. Due to the reduction of the dimension of the sensing matrix, the requirement of the memory storage space is reduced substantially. In addition, due to the nonlinear superposition of the reconstructed time of the segmented sub-scenes, the reconstruction time is reduced, and the purpose of fast imaging is achieved.Meanwhile, the feasibility and the related factors which affect the performance of the proposed method are also analyzed, and the selection criterion of the scene segmentation is afforded. Finally,theoretical analysis and simulation results demonstrate the feasibility and effectiveness of the proposed method.展开更多
In this paper, we study autonomous landing scene recognition with knowledge transfer for drones. Considering the difficulties in aerial remote sensing, especially that some scenes are extremely similar, or the same sc...In this paper, we study autonomous landing scene recognition with knowledge transfer for drones. Considering the difficulties in aerial remote sensing, especially that some scenes are extremely similar, or the same scene has different representations in different altitudes, we employ a deep convolutional neural network(CNN) based on knowledge transfer and fine-tuning to solve the problem. Then, LandingScenes-7 dataset is established and divided into seven classes. Moreover, there is still a novelty detection problem in the classifier, and we address this by excluding other landing scenes using the approach of thresholding in the prediction stage. We employ the transfer learning method based on ResNeXt-50 backbone with the adaptive momentum(ADAM) optimization algorithm. We also compare ResNet-50 backbone and the momentum stochastic gradient descent(SGD) optimizer. Experiment results show that ResNeXt-50 based on the ADAM optimization algorithm has better performance. With a pre-trained model and fine-tuning, it can achieve 97.845 0% top-1 accuracy on the LandingScenes-7dataset, paving the way for drones to autonomously learn landing scenes.展开更多
To solve the heterogeneous image scene matching problem, a non-linear pre-processing method for the original images before intensity-based correlation is proposed. The result shows that the proper matching probability...To solve the heterogeneous image scene matching problem, a non-linear pre-processing method for the original images before intensity-based correlation is proposed. The result shows that the proper matching probability is raised greatly. Especially for the low S/N image pairs, the effect is more remarkable.展开更多
The improved scene-based adaptive nonuniformity correction (NUC) algorithms using a neural network (NNT) approach for infrared image sequences are presented and analyzed. The retina-like neural networks using steepest...The improved scene-based adaptive nonuniformity correction (NUC) algorithms using a neural network (NNT) approach for infrared image sequences are presented and analyzed. The retina-like neural networks using steepest descent model was the first proposed infrared focal plane arrays (IRFPA) nonuniformity compensation method,which can perform parameter estimation of the sensors over time on a frame by frame basis. To increase the strength and the robustness of the NNT algorithm and to avoid the presence of ghosting artifacts,some optimization techniques,including momentum term,regularization factor and adaptive learning rate,were executed in the parameter learning process. In this paper,the local median filtering result of AX^U_ ij (n) is proposed as an alternative value of desired network output of neuron X_ ij (n),denoted as T_ ij (n),which is the local spatial average of AX^U_ ij (n) in traditional NNT methods. Noticeably,the NUC algorithm is inter-frame adaptive in nature and does not rely on any statistical assumptions on the scene data in the image sequence. Applications of this algorithm to the simulated video sequences and real infrared data taken with PV320 show that the correction results of image sequence are better than that of using original NNT approach,especially for the short-time image sequences (several hundred frames) subjected to the dense impulse noises with a number of dead or saturated pixels.展开更多
The process of human natural scene categorization consists of two correlated stages: visual perception and visual cognition of natural scenes.Inspired by this fact,we propose a biologically plausible approach for natu...The process of human natural scene categorization consists of two correlated stages: visual perception and visual cognition of natural scenes.Inspired by this fact,we propose a biologically plausible approach for natural scene image classification.This approach consists of one visual perception model and two visual cognition models.The visual perception model,composed of two steps,is used to extract discriminative features from natural scene images.In the first step,we mimic the oriented and bandpass properties of human primary visual cortex by a special complex wavelets transform,which can decompose a natural scene image into a series of 2D spatial structure signals.In the second step,a hybrid statistical feature extraction method is used to generate gist features from those 2D spatial structure signals.Then we design a cognitive feedback model to realize adaptive optimization for the visual perception model.At last,we build a multiple semantics based cognition model to imitate human cognitive mode in rapid natural scene categorization.Experiments on natural scene datasets show that the proposed method achieves high efficiency and accuracy for natural scene classification.展开更多
In this paper,we study scene image recognition with knowledge transfer for drone navigation.We divide navigation scenes into three macro-classes,namely outdoor special scenes(OSSs),the space from indoors to outdoors o...In this paper,we study scene image recognition with knowledge transfer for drone navigation.We divide navigation scenes into three macro-classes,namely outdoor special scenes(OSSs),the space from indoors to outdoors or from outdoors to indoors transitional scenes(TSs),and others.However,there are difficulties in how to recognize the TSs,to this end,we employ deep convolutional neural network(CNN)based on knowledge transfer,techniques for image augmentation,and fine tuning to solve the issue.Moreover,there is still a novelty detection prob-lem in the classifier,and we use global navigation satellite sys-tems(GNSS)to solve it in the prediction stage.Experiment results show our method,with a pre-trained model and fine tun-ing,can achieve 91.3196%top-1 accuracy on Scenes21 dataset,paving the way for drones to learn to understand the scenes around them autonomously.展开更多
基金the China National Key Research and Development Program(No.2016YFC0802904)National Natural Science Foundation of China(61671470)62nd batch of funded projects of China Postdoctoral Science Foundation(No.2017M623423).
文摘Detecting highly-overlapped objects in crowded scenes remains a challenging problem,especially for one-stage detector.In this paper,we extricate YOLOv4 from the dilemma in a crowd by fine-tuning its detection scheme,named YOLO-CS.Specifically,we give YOLOv4 the power to detect multiple objects in one cell.Center to our method is the carefully designed joint prediction scheme,which is executed through an assignment of bounding boxes and a joint loss.Equipped with the derived joint-object augmentation(DJA),refined regression loss(RL)and Score-NMS(SN),YOLO-CS achieves competitive detection performance on CrowdHuman and CityPersons benchmarks compared with state-of-the-art detectors at the cost of little time.Furthermore,on the widely used general benchmark COCO,YOLOCS still has a good performance,indicating its robustness to various scenes.
基金support by the National Natural Science Foundation of China (Grant No. 62005049)Natural Science Foundation of Fujian Province (Grant Nos. 2020J01451, 2022J05113)Education and Scientific Research Program for Young and Middleaged Teachers in Fujian Province (Grant No. JAT210035)。
文摘Camouflaged people are extremely expert in actively concealing themselves by effectively utilizing cover and the surrounding environment. Despite advancements in optical detection capabilities through imaging systems, including spectral, polarization, and infrared technologies, there is still a lack of effective real-time method for accurately detecting small-size and high-efficient camouflaged people in complex real-world scenes. Here, this study proposes a snapshot multispectral image-based camouflaged detection model, multispectral YOLO(MS-YOLO), which utilizes the SPD-Conv and Sim AM modules to effectively represent targets and suppress background interference by exploiting the spatial-spectral target information. Besides, the study constructs the first real-shot multispectral camouflaged people dataset(MSCPD), which encompasses diverse scenes, target scales, and attitudes. To minimize information redundancy, MS-YOLO selects an optimal subset of 12 bands with strong feature representation and minimal inter-band correlation as input. Through experiments on the MSCPD, MS-YOLO achieves a mean Average Precision of 94.31% and real-time detection at 65 frames per second, which confirms the effectiveness and efficiency of our method in detecting camouflaged people in various typical desert and forest scenes. Our approach offers valuable support to improve the perception capabilities of unmanned aerial vehicles in detecting enemy forces and rescuing personnel in battlefield.
文摘Low-light image enhancement is one of the most active research areas in the field of computer vision in recent years.In the low-light image enhancement process,loss of image details and increase in noise occur inevitably,influencing the quality of enhanced images.To alleviate this problem,a low-light image enhancement model called RetinexNet model based on Retinex theory was proposed in this study.The model was composed of an image decomposition module and a brightness enhancement module.In the decomposition module,a convolutional block attention module(CBAM)was incorporated to enhance feature representation capacity of the network,focusing on crucial features and suppressing irrelevant ones.A multifeature fusion denoising module was designed within the brightness enhancement module,circumventing the issue of feature loss during downsampling.The proposed model outperforms the existing algorithms in terms of PSNR and SSIM metrics on the publicly available datasets LOL and MIT-Adobe FiveK,as well as gives superior results in terms of NIQE metrics on the publicly available dataset LIME.
基金supported by the National Natural Science Foundation of China(61671469)
文摘Although compressed sensing inverse synthetic aperture radar(ISAR) imaging methods are widely used in radar signal processing, its reconstructing time and memory storage space requirements are very high. The main reason is that large scene reconstruction needs a higher dimension of the sensing matrix. To reduce this limitation, a fast high resolution ISAR imaging method,which is based on scene segmentation for random chirp frequencystepped signals, is proposed. The idea of scene segmentation is used to solve the problems aforementioned. In the method,firstly, the observed scene is divided into multiple sub-scenes and then the sub-scenes are reconstructed respectively. Secondly, the whole image scene can be obtained through the stitching of the sub-scenes. Due to the reduction of the dimension of the sensing matrix, the requirement of the memory storage space is reduced substantially. In addition, due to the nonlinear superposition of the reconstructed time of the segmented sub-scenes, the reconstruction time is reduced, and the purpose of fast imaging is achieved.Meanwhile, the feasibility and the related factors which affect the performance of the proposed method are also analyzed, and the selection criterion of the scene segmentation is afforded. Finally,theoretical analysis and simulation results demonstrate the feasibility and effectiveness of the proposed method.
基金supported by the National Natural Science Foundation of China (62103104)the China Postdoctoral Science Foundation(2021M690615)。
文摘In this paper, we study autonomous landing scene recognition with knowledge transfer for drones. Considering the difficulties in aerial remote sensing, especially that some scenes are extremely similar, or the same scene has different representations in different altitudes, we employ a deep convolutional neural network(CNN) based on knowledge transfer and fine-tuning to solve the problem. Then, LandingScenes-7 dataset is established and divided into seven classes. Moreover, there is still a novelty detection problem in the classifier, and we address this by excluding other landing scenes using the approach of thresholding in the prediction stage. We employ the transfer learning method based on ResNeXt-50 backbone with the adaptive momentum(ADAM) optimization algorithm. We also compare ResNet-50 backbone and the momentum stochastic gradient descent(SGD) optimizer. Experiment results show that ResNeXt-50 based on the ADAM optimization algorithm has better performance. With a pre-trained model and fine-tuning, it can achieve 97.845 0% top-1 accuracy on the LandingScenes-7dataset, paving the way for drones to autonomously learn landing scenes.
文摘To solve the heterogeneous image scene matching problem, a non-linear pre-processing method for the original images before intensity-based correlation is proposed. The result shows that the proper matching probability is raised greatly. Especially for the low S/N image pairs, the effect is more remarkable.
文摘The improved scene-based adaptive nonuniformity correction (NUC) algorithms using a neural network (NNT) approach for infrared image sequences are presented and analyzed. The retina-like neural networks using steepest descent model was the first proposed infrared focal plane arrays (IRFPA) nonuniformity compensation method,which can perform parameter estimation of the sensors over time on a frame by frame basis. To increase the strength and the robustness of the NNT algorithm and to avoid the presence of ghosting artifacts,some optimization techniques,including momentum term,regularization factor and adaptive learning rate,were executed in the parameter learning process. In this paper,the local median filtering result of AX^U_ ij (n) is proposed as an alternative value of desired network output of neuron X_ ij (n),denoted as T_ ij (n),which is the local spatial average of AX^U_ ij (n) in traditional NNT methods. Noticeably,the NUC algorithm is inter-frame adaptive in nature and does not rely on any statistical assumptions on the scene data in the image sequence. Applications of this algorithm to the simulated video sequences and real infrared data taken with PV320 show that the correction results of image sequence are better than that of using original NNT approach,especially for the short-time image sequences (several hundred frames) subjected to the dense impulse noises with a number of dead or saturated pixels.
文摘The process of human natural scene categorization consists of two correlated stages: visual perception and visual cognition of natural scenes.Inspired by this fact,we propose a biologically plausible approach for natural scene image classification.This approach consists of one visual perception model and two visual cognition models.The visual perception model,composed of two steps,is used to extract discriminative features from natural scene images.In the first step,we mimic the oriented and bandpass properties of human primary visual cortex by a special complex wavelets transform,which can decompose a natural scene image into a series of 2D spatial structure signals.In the second step,a hybrid statistical feature extraction method is used to generate gist features from those 2D spatial structure signals.Then we design a cognitive feedback model to realize adaptive optimization for the visual perception model.At last,we build a multiple semantics based cognition model to imitate human cognitive mode in rapid natural scene categorization.Experiments on natural scene datasets show that the proposed method achieves high efficiency and accuracy for natural scene classification.
基金supported by the National Natural Science Foundation of China(62103104)the Natural Science Foundation of Jiangsu Province(BK20210215)the China Postdoctoral Science Foundation(2021M690615).
文摘In this paper,we study scene image recognition with knowledge transfer for drone navigation.We divide navigation scenes into three macro-classes,namely outdoor special scenes(OSSs),the space from indoors to outdoors or from outdoors to indoors transitional scenes(TSs),and others.However,there are difficulties in how to recognize the TSs,to this end,we employ deep convolutional neural network(CNN)based on knowledge transfer,techniques for image augmentation,and fine tuning to solve the issue.Moreover,there is still a novelty detection prob-lem in the classifier,and we use global navigation satellite sys-tems(GNSS)to solve it in the prediction stage.Experiment results show our method,with a pre-trained model and fine tun-ing,can achieve 91.3196%top-1 accuracy on Scenes21 dataset,paving the way for drones to learn to understand the scenes around them autonomously.
