Cloud detection is a fundamental first step in the preprocessing of optical remote sensing images and a pivotal element for subsequent analytical tasks. High-resolution remote sensing images often require cropping into smaller patches to facilitate effective cloud detection. However, these segmented patches often lack rich features and exhibit high similarity between classes, which complicates accurate classification, particularly in intricate scenarios such as those involving both clouds and snow. To tackle this challenge, we introduce a novel cloud detection approach that employs boundary position attention, based on the CloudS26 cloud-snow dataset. This technique utilizes deformable convolution to generate boundary-focused attention, proficiently delineating the interfaces between clouds and snow in coexistent environments. Our approach has demonstrated robust detection capabilities in both the CloudS26 and CSWV datasets, showcasing its effectiveness in handling complex meteorological phenomena.
The staggered imaging camera is an important kind of remote sensing satellite camera. The staggered imaging technology can improve the spatial resolution of the camera without changing the focal length and pixel size of the optical system. However, the image resolution directly obtained by the staggered imaging camera is not enough. The existing traditional super-resolution methods have certain interpretability; the performance of deep learning super-resolution is related to the quality and quantity of training data set. There is no suitable data set for the remote sensing image generated by the staggered imaging system. So the frequency domain super-resolution technology is proposed for the image of the staggered imaging system. Low resolution remote sensing image can be regarded as the result of effective shift and sum of high-resolution image in time domain. There is a certain phase difference between low-resolution remote sensing image and high-resolution image in frequency domain. The time-shift property of two-dimensional image Fourier transform is used to find the difference between low-resolution image and high-resolution real image in frequency domain. The super resolution image is obtained by compensation coefficient. The frequency domain super-resolution algorithm is sensitive to noise. In order to suppress the noise interference, a special frequency domain filter is designed to filter the noise. It is verified by SPOT5 data that the frequency domain super-resolution can completely recover the noiseless image. Under the condition of image noise, the peak signal-to-noise ratio and the structure similarity can reach 35.754dB and 0.97 in 1.001 seconds.
The imagery vendors of the most advanced remote sensing satellites usually only provide the coefficients of rational function model (RFM) to replace the sensor model and the precise imaging parameters (orbit parameter, attitude parameter, and so on). So, the rigorous imaging model was limited to use in the geometric correction of remote sensing image. The RFM method could obtain a better correction performance in most cases. However, when the image contains few numbers or uneven distribution of ground control points (GCPs), such as infrared image, the RFM method could not obtain the expected performance. Therefore, a geometric correction method for linear pushbroom infrared imagery using compressive sampling (CS) is proposed. The core idea of the proposed method is to use the equivalent bias angles to approximate the influence of the errors (thermal distortion, optical distortion, assembly error, satellite orbit errors, attitude errors, and so on) in the imaging process and adopt the CS method to recover the equivalent bias angle signals. Most of the data are processed scene by scene with enough GCPs for each scene in conventional methods. This restriction is broken by using the sparsity of equivalent bias angle signals in the proposed method. The infrared images from the Hyperion of EO-1 are used as experiment data, and the results of experiments demonstrate the feasibility and superior performance of proposed method.
Attitude jitter occurs widely in the applications of high-resolution satellites. It is a vital factor that deteriorates the accuracy of geopositioning and mapping. However, the normal geometric correction methods cannot eliminate the influence of jitter on remote sensing images. Therefore, it is important to design a method that can handle this problem. This paper presents a geometric correction method using a rational function model (RFM) and compressive sampling called RFM-CS. This method is divided into two modules: precorrection and compensation. In the precorrection part, the original image is geometrically corrected with an RFM. However, when raw images contain distortion caused by attitude jitter, the rational polynomial coefficients (RPCs) don’t approximate the real imaging process well, and there are significant residual distortions in precorrected images. In the compensation part, we propose a new idea by which the residual distortions of images can be expressed as two-dimensional signals, which are called distortion signals. Using the new idea and CS, distortion signals, even those containing the influence of attitude jitter, are exactly reconstructed with a small set of ground control points. Based on the reconstructed distortion signals, the residual geometric error in the remote sensing images can be compensated by resampling. The experiments with images from Advanced Spaceborne Thermal Emission and Reflection Radiometer, Advanced Land Observing Satellite, and simulation demonstrate the promising performance and feasibility of the proposed method.
KEYWORDS: Error analysis, 3D modeling, Remote sensing, Monte Carlo methods, Virtual point source, Annealing, Image quality, 3D image processing, Image processing, Algorithms
The Ground Control Points (GCPs) are widely used in geometric correction for remote sensing imagery, and the distribution of them is a key factor which affects the accuracy and quality of image correction. In this paper, we propose a new sampling design method, called Smallest Singular Value-based Sampling (SSVS), to obtain the optimal distribution of the GCPs. When the geometric correction of remote sensing imagery is performed with a 2D or 3D polynomial function model, the estimation of geometric correction model parameters can be interpreted as an estimation of regression coefficients with a Multiple Linear Regression(MLR) model, whose design matrix depends on the coordinates of GCPs. From the perspective of regression model, the design matrix of MLR should be optimized to obtain the most accurate regression coefficients. In this paper, it has been proved that the Smallest Singular Value(SSV) of design matrix is inversely proportional to the upper bound of estimation errors. By choosing the optimal distribution of GCPs, the SSV of design matrix can be maximized and the upper bound of estimation errors can be minimized. Therefore, the SSV of design matrix is used as a criterion, and the objective of SSVS is to find the sample pattern that has the biggest SSV. In this paper, the simulation annealing is employed to search the optimal pattern. Two experiments were carried out to test SSVS. The results indicate that the SSVS is an effective GCPs sampling design method and can be applied to evaluate upper bound of estimation error.
The GCPs are widely used in remote sense image registration and geometric correction. Normally, the DRG and DOM are the major data source from which GCPs are extracted. But the high accuracy products of DRG and DOM are usually costly to obtain. Some of the production are free, yet without any guarantee. In order to balance the cost and the accuracy, the paper proposes a method of extracting the GCPs from SRTM data. The method consist of artificial assistance, binarization, data resample and reshape. With artificial assistance to find out which part of SRTM data could be used as GCPs, such as the islands or sharp coast line. By utilizing binarization algorithm , the shape information of the region is obtained while other information is excluded. Then the binary data is resampled to a suitable resolution required by specific application. At last, the data would be reshaped according to satellite imaging type to obtain the GCPs which could be used. There are three advantages of the method proposed in the paper. Firstly, the method is easy for implementation. Unlike the DRG data or DOM data that charges a lot, the SRTM data is totally free to access without any constricts. Secondly, the SRTM has a high accuracy about 90m that is promised by its producer, so the GCPs got from it can also obtain a high quality. Finally, given the SRTM data covers nearly all the land surface of earth between latitude -60° and latitude +60°, the GCPs which are produced by the method can cover most important regions of the world. The method which obtain GCPs from SRTM data can be used in meteorological satellite image or some situation alike, which have a relative low requirement about the accuracy. Through plenty of simulation test, the method is proved convenient and effective.
Access to the requested content is limited to institutions that have purchased or subscribe to SPIE eBooks.
You are receiving this notice because your organization may not have SPIE eBooks access.*
*Shibboleth/Open Athens users─please
sign in
to access your institution's subscriptions.
To obtain this item, you may purchase the complete book in print or electronic format on
SPIE.org.
INSTITUTIONAL Select your institution to access the SPIE Digital Library.
PERSONAL Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.