作者：
Gao, JingyuGeng, XiuruiAerospace Information Research Institute
Chinese Academy of Sciences School of Electronic Electrical and Communication Engineering University of Chinese Academy of Sciences Key Laboratory of Technology in Geo-Spatial Information Processing and Application System Chinese Academy of Sciences China
The proximal point algorithm (PPA) has been developed to solve the monotone variational inequality problem. It provides a theoretical foundation for some methods, such as the augmented Lagrangian method (ALM) and the ...
详细信息
Continual semantic segmentation (CSS) has risen as a popular field, which aims to acquire new skills constantly without forgetting past knowledge catastrophically. In CSS, we identify that there is a severe imbalance ...
详细信息
ISBN:
(数字)9798350390155
ISBN:
(纸本)9798350390162
Continual semantic segmentation (CSS) has risen as a popular field, which aims to acquire new skills constantly without forgetting past knowledge catastrophically. In CSS, we identify that there is a severe imbalance between new classes and old classes, leading to the classifier weight toward new classes. In this paper, we deal with the continual semantic segmentation problem from the class imbalance perspective via mask-based class rebalancing, avoiding the model suffering from catastrophic forgetting. More specifically, the mask-based class rebalancing depends on a mask to combine resampling with reweighting ingenuously, which mitigates the classifier bias toward new classes. Besides, we also propose a frequency knowledge distillation, leveraging multiple frequency components information to maintain the feature representation space for old classes. We demonstrate the effectiveness of our approach with an extensive evaluation of the Pascal-VOC 2012 and ADE20K datasets, significantly outperforming the state-of-the-art method.
The compensation of channel imbalances plays a vital role in signal processing of the azimuth multi-channel (AMC) synthetic aperture radar (SAR). In the operational AMC SAR system, the channel imbalance is usually con...
详细信息
ISBN:
(纸本)9781665468893
The compensation of channel imbalances plays a vital role in signal processing of the azimuth multi-channel (AMC) synthetic aperture radar (SAR). In the operational AMC SAR system, the channel imbalance is usually considered to be a constant in both the time and frequency domain for one image. However, since the antenna patterns of all channels cannot be strictly consistent, the channel imbalances may vary with the Doppler frequency. This paper establishes the signal model of AMC SAR with channel imbalances changing with Doppler frequency and proposes a novel compensation method during the reconstruction for the echo signals received by all channels. This method has been verified by simulations and real data acquired by Gaofen-3 (GF-3). The result shows that this method has excellent potential for improving the image quality for AMC SAR.
The interferometric coherence map is derived from the cross-correlation of two registered synthetic aperture radar (SAR) images. It can give additional information complementary to the intensity image, or act as an in...
详细信息
ISBN:
(纸本)9781509033331
The interferometric coherence map is derived from the cross-correlation of two registered synthetic aperture radar (SAR) images. It can give additional information complementary to the intensity image, or act as an independent information source in many applications. Compared to the plenty of work on SAR intensity statistics, there are quite fewer researches on the statistical characters of interferometric SAR (InSAR) coherence. And to our knowledge, all of the existing work that related to InSAR coherence statistics, models the coherence with Gaussian distribution with no discrimination on data resolutions or scene types. Our main contribution is the investigation on the accuracies of several typical models for high resolution coherence statistics over urban areas. We select three typical land classes including trees, buildings, and shadow, as the representatives of urban areas. And different models including Gaussian, Weibull, Rayleigh, Nakagami and Beta are evaluated. Experiment results on TanDEM-X data illustrate that the Beta model reveals a better performance than other distributions. Finally, the Beta model is used in the detection of buildings.
The accuracy of attitude observation is always the main contributor to geometric performance of earth observation satellites (EOSs). Given the ground process requirement, linear pushbroom and asynchronous imaging sens...
详细信息
The accuracy of attitude observation is always the main contributor to geometric performance of earth observation satellites (EOSs). Given the ground process requirement, linear pushbroom and asynchronous imaging sensors are widely used in EOSs, such as multispectral sensor with sequential line arrays, three-line array sensor in stereo mapping satellite, panchromatic sensor with multiple non-collinear CCD chip in some high-resolution optical satellite. By using the images of those sensors, this paper proposes a method, which is based on image registration approach, rigorous forward intersection and bundle adjustment technology, to refine attitude data of satellite for improving geometric performance of images. Preliminary experiments, which used multi-sensors asynchronous images of Chinese Mapping Satellite-1-02, demonstrate that the proposed method is capable of improving internally coincident precision of attitude data without ground control points. In particular, relative positioning accuracy of images can be directly improved, and absolute positioning accuracy can consequently be improved via additionally using a few GCPs in the stripe image data.
The slant range errors caused by traditional hyperbolic range equation (THRE) with stop-and-go assumption will lead to image defocusing in high resolution spaceborne sliding-spotlight Bistatic SAR system (ST-BiSAR). I...
详细信息
ISBN:
(数字)9781728163741
ISBN:
(纸本)9781728163758
The slant range errors caused by traditional hyperbolic range equation (THRE) with stop-and-go assumption will lead to image defocusing in high resolution spaceborne sliding-spotlight Bistatic SAR system (ST-BiSAR). In this paper, an accurate bistatic slant range model based on uniform acceleration curve motion (UARM) is proposed, which is precisely fitted with the actual range history. Then, a two-step imaging algorithm based on UARM and method of reversion series (MSR) is introduced to eliminate aliasing phenomenon and realize focus. Finally, simulation results verify the correctness and effectives of the proposed range model and imaging algorithms.
Stereoscopic image quality assessment (SIQA) has encountered non-trivial challenges due to the fast proliferation of 3D contents. In the past years, deep learning oriented SIQA methods have emerged and achieved specta...
详细信息
Deeper architectures are proven to be beneficial for the classification performance obviously in computer vision field. Inspired by this, deep CNNs are expected to make progress in the SAR target classification proble...
详细信息
Deeper architectures are proven to be beneficial for the classification performance obviously in computer vision field. Inspired by this, deep CNNs are expected to make progress in the SAR target classification problem as well. However, it is hard to train deeper CNNs for SAR images. Such CNNs have millions of parameters to be determined in the network (for example the VGGNet has more than 130 million parameters), hence large-scale dataset is indispensable when training a deep CNN. But there is no large-scale annotated SAR target dataset, and data acquisition and annotation is much more costly for SAR images. With inadequate data, the network is easy to be overfitting. Several methods based on deep learning have been proposed for SAR image classifications, but they cannot get rid of the aforementioned data limitation of labelled SAR images. To solve this problem, this paper proposes a microarchitecture called CompressUnit (CU). With CU, we design a deeper CNN. Compared with the network with the fewest parameters for SAR image classification in literature so far, our network is 2X deeper with only about 10% of parameters. In this way, we get a deeper network with much fewer parameters. This network is easier to be trained with limited SAR data and is more likely to get rid of overfitting.
Few-shot image generation aims to train generative models using a small number of training images. When there are few images available for training (e.g. 10 images), Learning From Scratch (LFS) methods often generate ...
详细信息
Polarimetric Interferometric SAR (PolInSAR) can improve the coherence of images and it plays an important role in urban remote sensing. The explanation of its scattering mechanism is concerned by many researchers. It ...
详细信息
ISBN:
(纸本)9781665468893
Polarimetric Interferometric SAR (PolInSAR) can improve the coherence of images and it plays an important role in urban remote sensing. The explanation of its scattering mechanism is concerned by many researchers. It is always explained from the aspect of different polarization decompositions but seldom from the height of scattering centers. In this paper, a Ku-band UAV-borne PolInSAR system is used to compare the relationship of scattering centers between PolInSAR and Pauli decomposition on different artificial targets. The physical meaning of PolInSAR is preliminary analyzed and our conclusion is also verified through a C-band manned airborne PolInSAR system.
暂无评论