How to recognize targets with similar appearances from remote sensing images(RSIs) effectively and efficiently has become a big challenge. Recently, convolutional neural network(CNN) is preferred in the target classif...How to recognize targets with similar appearances from remote sensing images(RSIs) effectively and efficiently has become a big challenge. Recently, convolutional neural network(CNN) is preferred in the target classification due to the powerful feature representation ability and better performance. However,the training and testing of CNN mainly rely on single machine.Single machine has its natural limitation and bottleneck in processing RSIs due to limited hardware resources and huge time consuming. Besides, overfitting is a challenge for the CNN model due to the unbalance between RSIs data and the model structure.When a model is complex or the training data is relatively small,overfitting occurs and leads to a poor predictive performance. To address these problems, a distributed CNN architecture for RSIs target classification is proposed, which dramatically increases the training speed of CNN and system scalability. It improves the storage ability and processing efficiency of RSIs. Furthermore,Bayesian regularization approach is utilized in order to initialize the weights of the CNN extractor, which increases the robustness and flexibility of the CNN model. It helps prevent the overfitting and avoid the local optima caused by limited RSI training images or the inappropriate CNN structure. In addition, considering the efficiency of the Na¨?ve Bayes classifier, a distributed Na¨?ve Bayes classifier is designed to reduce the training cost. Compared with other algorithms, the proposed system and method perform the best and increase the recognition accuracy. The results show that the distributed system framework and the proposed algorithms are suitable for RSIs target classification tasks.展开更多
Removal of cloud cover on the satellite remote sensing image can effectively improve the availability of remote sensing images. For thin cloud cover, support vector value contourlet transform is used to achieve multi-...Removal of cloud cover on the satellite remote sensing image can effectively improve the availability of remote sensing images. For thin cloud cover, support vector value contourlet transform is used to achieve multi-scale decomposition of the area of thin cloud cover on remote sensing images. Through enhancing coefficients of high frequency and suppressing coefficients of low frequency, the thin cloud is removed. For thick cloud cover, if the areas of thick cloud cover on multi-source or multi-temporal remote sensing images do not overlap, the multi-output support vector regression learning method is used to remove this kind of thick clouds. If the thick cloud cover areas overlap, by using the multi-output learning of the surrounding areas to predict the surface features of the overlapped thick cloud cover areas, this kind of thick cloud is removed. Experimental results show that the proposed cloud removal method can effectively solve the problems of the cloud overlapping and radiation difference among multi-source images. The cloud removal image is clear and smooth.展开更多
In the field of satellite imagery, remote sensing image captioning(RSIC) is a hot topic with the challenge of overfitting and difficulty of image and text alignment. To address these issues, this paper proposes a visi...In the field of satellite imagery, remote sensing image captioning(RSIC) is a hot topic with the challenge of overfitting and difficulty of image and text alignment. To address these issues, this paper proposes a vision-language aligning paradigm for RSIC to jointly represent vision and language. First, a new RSIC dataset DIOR-Captions is built for augmenting object detection in optical remote(DIOR) sensing images dataset with manually annotated Chinese and English contents. Second, a Vision-Language aligning model with Cross-modal Attention(VLCA) is presented to generate accurate and abundant bilingual descriptions for remote sensing images. Third, a crossmodal learning network is introduced to address the problem of visual-lingual alignment. Notably, VLCA is also applied to end-toend Chinese captions generation by using the pre-training language model of Chinese. The experiments are carried out with various baselines to validate VLCA on the proposed dataset. The results demonstrate that the proposed algorithm is more descriptive and informative than existing algorithms in producing captions.展开更多
A hybrid feature selection and classification strategy was proposed based on the simulated annealing genetic algonthrn and multiple instance learning （MIL）. The band selection method was proposed from subspace decom...A hybrid feature selection and classification strategy was proposed based on the simulated annealing genetic algonthrn and multiple instance learning （MIL）. The band selection method was proposed from subspace decomposition, which combines the simulated annealing algorithm with the genetic algorithm in choosing different cross-over and mutation probabilities, as well as mutation individuals. Then MIL was combined with image segmentation, clustering and support vector machine algorithms to classify hyperspectral image. The experimental results show that this proposed method can get high classification accuracy of 93.13% at small training samples and the weaknesses of the conventional methods are overcome.展开更多
A novel image restoration scheme, which is super-resolution image restoration algorithm Poisson-maximum-afterword-probability based on Markvo constraint (MPMAP) combined with evaluating image detail parameter D, has b...A novel image restoration scheme, which is super-resolution image restoration algorithm Poisson-maximum-afterword-probability based on Markvo constraint (MPMAP) combined with evaluating image detail parameter D, has been proposed. The advantage of super-resolution algorithm MPMAP incorporated with parameter D lies in the fact that super-resolution algorithm MPMAP model is discrete, which is in accordance with remote-sensing imaging model, and the algorithm MPMAP is proved applicable to linear and non-linear imaging models with a unique solution when noise is not severe. According to simulation experiments for practical images, super-resolution algorithm MPMAP can retain image details better than most of traditional restoration methods; at the same time, the proposed parameter D can help to identify real point spread function (PSF) value of degradation process. Processing result of practical remote-sensing images by MPMAP combined with parameter D are given, it illustrates that MPMAP restoration scheme combined PSF estimation has a better restoration result than that of Photoshop processing, based on the same original images. It is proved that the proposed scheme is helpful to offset the lack of resolution of the original remote-sensing images and has its extensive application foreground.展开更多
Ⅰ. INTRODUCTION Changbai Mountain is situated between E127°54′-128°08′, N40°58′-42°06′ about 2700 meters above sea level. It is the typical area of the mountainous climate in the monsoon area ...Ⅰ. INTRODUCTION Changbai Mountain is situated between E127°54′-128°08′, N40°58′-42°06′ about 2700 meters above sea level. It is the typical area of the mountainous climate in the monsoon area of the temperate zone on the globe. The well reserved primeval vertical distribution of natural landscape belts and the Natural Conservation of Changbai Mountains adopted by the UNESCO′s MAB Program cause the worldwide attention of geographers. Beside the complexity of the climatic structure itself, the mechanical effection of the high mountain body also effect the climate in the eastern part of China. In the mountain area where short of meteorological observation data, the climatic study by remote sensing is favorable for discovery and representation of climatic law in large area.展开更多
The Landsat image information has recently been widely applied to structural geology, especially to the analysis of lineaments, owing to their macroscopic, visual and comprehensive features. The images will be more ef...The Landsat image information has recently been widely applied to structural geology, especially to the analysis of lineaments, owing to their macroscopic, visual and comprehensive features. The images will be more effective when applied to the interpretation of active faults. Active faults are widely ditributed in China. Much attention has been paid to the study of active faults both in China and abroad. There is certain controversy concerning the implication of the term "active fault". Strictly speaking, the term should refer only to the faults that are still active in the present day. However, the term also usually refers to the faults which have been active continually or intermittently from the Quaternary (or the end of Tertiary) to the present day. We propose that the tones and the configurations of features on Landsat images are the principal keys to the interpretation of active faults. The faults, which display the most prominent展开更多
Ophiolites, which have been tectonically emplaced along continental margins and island arcs, are significant to the understanding of mountain belt evolution. In the Himalayas, the ophiolitic suite of rocks occur along...Ophiolites, which have been tectonically emplaced along continental margins and island arcs, are significant to the understanding of mountain belt evolution. In the Himalayas, the ophiolitic suite of rocks occur along the Indussuture zone from Hanle in the southeast to Dras\|Kargil sector in the northwest and it represents the remnant of the compressed uplifted wedge of the oceanic crust between the two colliding continental masses, the Indian and the Asian plates.. These ophiolites are temporally and spatially correlated with the culminating phase of the Himalayan orogeny. The Indus River flows to its north separating the ophiolite from the Trans Himalayan litho\|units. Geological mapping in the hostile and inaccessible mountainous terrains of the Himalaya has always posed a great challenge to geologists. Nevertheless, a number of geologists have undertaken such arduous mapping expeditions in the past and prepared fairly good geological maps of these terrains .However there always existed disputes on the accuracy of lithological boundaries and structural details in these maps because many of these boundaries and structural features were completed through extrapolations and/or interpolations as the ruggedness and inaccessibility of a large part of the terrain forbid physical examination of every outcrop. It is in this context the potential of remote sensing, especially of satellite images, is to be appreciated.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
Considering the joint effects of various factors such as temporal baseline, spatial baseline, thermal noise, the difference of Doppler centroid frequency and the error of data processing on the interference correlatio...Considering the joint effects of various factors such as temporal baseline, spatial baseline, thermal noise, the difference of Doppler centroid frequency and the error of data processing on the interference correlation, an optimum selection method of common master images for ground deformation monitoring based on the permanent scatterer and differential SAR interferometry （PS-DInSAR） technique is proposed, in which the joint correlation coeficient is used as the evaluation function. The principle and realization method of PS-DInSAR technology is introduced, the factors affecting the DInSAR correlation are analysed, and the joint correlation function model and its solution are presented. Finally an experiment for the optimum selection of common master images is performed by using 25 SAR images over Shanghai taken by the ERS-1/2 as test data. The results indicate that the optimum selection method for PS-DInSAR common master images is effective and reliable.展开更多
针对遥感地物建筑物图像目标尺度差异大、样本空间分布不均衡、地物边界模糊、场景区域跨度大所导致的分割效果不佳问题,本文提出一种融合动态特征增强高精度遥感建筑物分割算法。首先,构建New_GhostNetV2网络,利用自适应上下文感知卷积...针对遥感地物建筑物图像目标尺度差异大、样本空间分布不均衡、地物边界模糊、场景区域跨度大所导致的分割效果不佳问题,本文提出一种融合动态特征增强高精度遥感建筑物分割算法。首先,构建New_GhostNetV2网络,利用自适应上下文感知卷积,增强算法对样本空间特征的捕捉能力。其次,采用Ghost Convolution结合跳跃连接和特征分支策略设计多层级信息增强模块,增强特征整合。随后引入级联注意力CGA(cascaded group attention),通过组内独立注意力计算,加强模型对多样化地物形态的适应性。最后,通过动态深度特征增强器构造特征融合模块,进一步加强模型捕获能力。在WHU数据集上实验结果表明:改进算法较基线模型F1-Score提高8.57%,mIoU提高12.48%,与其他主流语义分割模型相比,改进DeepLabv3+具有更好的分割精度。展开更多
遥感图像目标检测在军事侦察、智慧农业等领域意义重大,特别是小目标检测一直获得持续关注。然而,遥感图像中的小目标面临特征信息不足、检测难度大等问题,成为困扰遥感检测应用发展的最大障碍。为此,提出YOLO-HF(you only look once-hy...遥感图像目标检测在军事侦察、智慧农业等领域意义重大,特别是小目标检测一直获得持续关注。然而,遥感图像中的小目标面临特征信息不足、检测难度大等问题,成为困扰遥感检测应用发展的最大障碍。为此,提出YOLO-HF(you only look once-hybrid feature)算法,该算法在传统YOLOv7模型的网络中,引入通道注意力和自注意力的混合注意力机制提取目标深层特征,并将浅层特征和深层特征进行融合,增加局部特征的丰富性;为进一步加强对全局信息的关注,在提取特征后为小尺度目标添加全局注意力机制,实现全局特征表达能力的提升;为避免传统损失函数对小目标位置偏差敏感,导致检测效果不佳,选择使用一种新的度量方式,将其嵌入边界框损失函数的计算中,从而加快损失函数的收敛,实现小目标检测精度的提升。实验结果表明:与传统YOLOv7算法相比,所提算法在RSOD和NWPU VHR-10数据集上均表现出优越性,特别地,在RSOD数据集上均值平均精度提升了2.90%,在NWPU VHR-10数据集上均值平均精度实现了3.61%的提升。展开更多
基金supported by the National Natural Science Foundation of China(U1435220)
文摘How to recognize targets with similar appearances from remote sensing images(RSIs) effectively and efficiently has become a big challenge. Recently, convolutional neural network(CNN) is preferred in the target classification due to the powerful feature representation ability and better performance. However,the training and testing of CNN mainly rely on single machine.Single machine has its natural limitation and bottleneck in processing RSIs due to limited hardware resources and huge time consuming. Besides, overfitting is a challenge for the CNN model due to the unbalance between RSIs data and the model structure.When a model is complex or the training data is relatively small,overfitting occurs and leads to a poor predictive performance. To address these problems, a distributed CNN architecture for RSIs target classification is proposed, which dramatically increases the training speed of CNN and system scalability. It improves the storage ability and processing efficiency of RSIs. Furthermore,Bayesian regularization approach is utilized in order to initialize the weights of the CNN extractor, which increases the robustness and flexibility of the CNN model. It helps prevent the overfitting and avoid the local optima caused by limited RSI training images or the inappropriate CNN structure. In addition, considering the efficiency of the Na¨?ve Bayes classifier, a distributed Na¨?ve Bayes classifier is designed to reduce the training cost. Compared with other algorithms, the proposed system and method perform the best and increase the recognition accuracy. The results show that the distributed system framework and the proposed algorithms are suitable for RSIs target classification tasks.
基金supported by the National Natural Science Foundation of China(61172127)the Natural Science Foundation of Anhui Province(1408085MF121)
文摘Removal of cloud cover on the satellite remote sensing image can effectively improve the availability of remote sensing images. For thin cloud cover, support vector value contourlet transform is used to achieve multi-scale decomposition of the area of thin cloud cover on remote sensing images. Through enhancing coefficients of high frequency and suppressing coefficients of low frequency, the thin cloud is removed. For thick cloud cover, if the areas of thick cloud cover on multi-source or multi-temporal remote sensing images do not overlap, the multi-output support vector regression learning method is used to remove this kind of thick clouds. If the thick cloud cover areas overlap, by using the multi-output learning of the surrounding areas to predict the surface features of the overlapped thick cloud cover areas, this kind of thick cloud is removed. Experimental results show that the proposed cloud removal method can effectively solve the problems of the cloud overlapping and radiation difference among multi-source images. The cloud removal image is clear and smooth.
基金supported by the National Natural Science Foundation of China (61702528,61806212)。
文摘In the field of satellite imagery, remote sensing image captioning(RSIC) is a hot topic with the challenge of overfitting and difficulty of image and text alignment. To address these issues, this paper proposes a vision-language aligning paradigm for RSIC to jointly represent vision and language. First, a new RSIC dataset DIOR-Captions is built for augmenting object detection in optical remote(DIOR) sensing images dataset with manually annotated Chinese and English contents. Second, a Vision-Language aligning model with Cross-modal Attention(VLCA) is presented to generate accurate and abundant bilingual descriptions for remote sensing images. Third, a crossmodal learning network is introduced to address the problem of visual-lingual alignment. Notably, VLCA is also applied to end-toend Chinese captions generation by using the pre-training language model of Chinese. The experiments are carried out with various baselines to validate VLCA on the proposed dataset. The results demonstrate that the proposed algorithm is more descriptive and informative than existing algorithms in producing captions.
文摘A hybrid feature selection and classification strategy was proposed based on the simulated annealing genetic algonthrn and multiple instance learning （MIL）. The band selection method was proposed from subspace decomposition, which combines the simulated annealing algorithm with the genetic algorithm in choosing different cross-over and mutation probabilities, as well as mutation individuals. Then MIL was combined with image segmentation, clustering and support vector machine algorithms to classify hyperspectral image. The experimental results show that this proposed method can get high classification accuracy of 93.13% at small training samples and the weaknesses of the conventional methods are overcome.
文摘A novel image restoration scheme, which is super-resolution image restoration algorithm Poisson-maximum-afterword-probability based on Markvo constraint (MPMAP) combined with evaluating image detail parameter D, has been proposed. The advantage of super-resolution algorithm MPMAP incorporated with parameter D lies in the fact that super-resolution algorithm MPMAP model is discrete, which is in accordance with remote-sensing imaging model, and the algorithm MPMAP is proved applicable to linear and non-linear imaging models with a unique solution when noise is not severe. According to simulation experiments for practical images, super-resolution algorithm MPMAP can retain image details better than most of traditional restoration methods; at the same time, the proposed parameter D can help to identify real point spread function (PSF) value of degradation process. Processing result of practical remote-sensing images by MPMAP combined with parameter D are given, it illustrates that MPMAP restoration scheme combined PSF estimation has a better restoration result than that of Photoshop processing, based on the same original images. It is proved that the proposed scheme is helpful to offset the lack of resolution of the original remote-sensing images and has its extensive application foreground.
文摘Ⅰ. INTRODUCTION Changbai Mountain is situated between E127°54′-128°08′, N40°58′-42°06′ about 2700 meters above sea level. It is the typical area of the mountainous climate in the monsoon area of the temperate zone on the globe. The well reserved primeval vertical distribution of natural landscape belts and the Natural Conservation of Changbai Mountains adopted by the UNESCO′s MAB Program cause the worldwide attention of geographers. Beside the complexity of the climatic structure itself, the mechanical effection of the high mountain body also effect the climate in the eastern part of China. In the mountain area where short of meteorological observation data, the climatic study by remote sensing is favorable for discovery and representation of climatic law in large area.
文摘The Landsat image information has recently been widely applied to structural geology, especially to the analysis of lineaments, owing to their macroscopic, visual and comprehensive features. The images will be more effective when applied to the interpretation of active faults. Active faults are widely ditributed in China. Much attention has been paid to the study of active faults both in China and abroad. There is certain controversy concerning the implication of the term "active fault". Strictly speaking, the term should refer only to the faults that are still active in the present day. However, the term also usually refers to the faults which have been active continually or intermittently from the Quaternary (or the end of Tertiary) to the present day. We propose that the tones and the configurations of features on Landsat images are the principal keys to the interpretation of active faults. The faults, which display the most prominent
文摘Ophiolites, which have been tectonically emplaced along continental margins and island arcs, are significant to the understanding of mountain belt evolution. In the Himalayas, the ophiolitic suite of rocks occur along the Indussuture zone from Hanle in the southeast to Dras\|Kargil sector in the northwest and it represents the remnant of the compressed uplifted wedge of the oceanic crust between the two colliding continental masses, the Indian and the Asian plates.. These ophiolites are temporally and spatially correlated with the culminating phase of the Himalayan orogeny. The Indus River flows to its north separating the ophiolite from the Trans Himalayan litho\|units. Geological mapping in the hostile and inaccessible mountainous terrains of the Himalaya has always posed a great challenge to geologists. Nevertheless, a number of geologists have undertaken such arduous mapping expeditions in the past and prepared fairly good geological maps of these terrains .However there always existed disputes on the accuracy of lithological boundaries and structural details in these maps because many of these boundaries and structural features were completed through extrapolations and/or interpolations as the ruggedness and inaccessibility of a large part of the terrain forbid physical examination of every outcrop. It is in this context the potential of remote sensing, especially of satellite images, is to be appreciated.
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Considering the joint effects of various factors such as temporal baseline, spatial baseline, thermal noise, the difference of Doppler centroid frequency and the error of data processing on the interference correlation, an optimum selection method of common master images for ground deformation monitoring based on the permanent scatterer and differential SAR interferometry （PS-DInSAR） technique is proposed, in which the joint correlation coeficient is used as the evaluation function. The principle and realization method of PS-DInSAR technology is introduced, the factors affecting the DInSAR correlation are analysed, and the joint correlation function model and its solution are presented. Finally an experiment for the optimum selection of common master images is performed by using 25 SAR images over Shanghai taken by the ERS-1/2 as test data. The results indicate that the optimum selection method for PS-DInSAR common master images is effective and reliable.
文摘针对遥感地物建筑物图像目标尺度差异大、样本空间分布不均衡、地物边界模糊、场景区域跨度大所导致的分割效果不佳问题,本文提出一种融合动态特征增强高精度遥感建筑物分割算法。首先,构建New_GhostNetV2网络,利用自适应上下文感知卷积,增强算法对样本空间特征的捕捉能力。其次,采用Ghost Convolution结合跳跃连接和特征分支策略设计多层级信息增强模块,增强特征整合。随后引入级联注意力CGA(cascaded group attention),通过组内独立注意力计算,加强模型对多样化地物形态的适应性。最后,通过动态深度特征增强器构造特征融合模块,进一步加强模型捕获能力。在WHU数据集上实验结果表明:改进算法较基线模型F1-Score提高8.57%,mIoU提高12.48%,与其他主流语义分割模型相比,改进DeepLabv3+具有更好的分割精度。
