Remote sensing aircraft target detection is an important task in the field of remote sensing images interpretation. The general target detection network often includes a large number of parameters, slow detection speed, and poor performance when directly applied to aircraft detection tasks. In order to solve the issues, a novel remote sensing aircraft target detection method based on lightweight YOLOv4 is proposed in this paper. Firstly, Lightweight YOLOv4 adopts the MobileNetV3 and depthwise separable convolution to greatly reduce the amount of model parameters. Then, to further enhance the feature extraction ability of the network and make the network more lightweight, this paper uses the feature enhancement module (FEM) and residual fusion module (RFM). Extensive experiments on the DOTA aircraft dataset demonstrate that the Lightweight YOLOv4 can significantly improve the detection accuracy and efficiency, as well as owns fewer model parameters.
Recently, a complex-valued convolutional neural network (CV-CNN) has been used for the classification of polarimetric synthetic aperture radar (PolSAR) images, and has shown superior performance to most traditional algorithms. However, it usually yields unreliable results for the pixels distributing within heterogeneous regions or the edge areas. To solve this problem, in this paper, an edge reassigning scheme based on Markov random field (MRF) is considered to combine with the CV-CNN. In this scheme,both the polarimetric statistical property and label context information are employed. The experiments performed on a benchmark PolSAR image of Flevoland has demonstrated the superior performance of the proposed algorithm.
This paper presents a new unsupervised classification framework based on tensor product graph (TPG) diffusion, which is generally utilized for optical image segmentation or image retrieval and for the first time used for PolSAR image classification in our work. First, the PolSAR image is divided into many superpixels by using a fast superpixel segmentation method. Second, seven features are extracted from the PolSAR image to form a feature vector based on segmented superpixels and construct a similarity matrix by using the Gaussian kernel. Third, TPG diffusion is performed on this similarity matrix to obtain a more discriminative similarity matrix by mining the higher order information between data points. Finally, spectral clustering based on diffused similarity matrix is adopted to automatically achieve the classification results. The experimental results conducted on both a simulated PolSAR image and a real-world PolSAR image demonstrate that our algorithm can effectively combine higher order neighborhood information and achieve higher classification accuracy.
KEYWORDS: Reconstruction algorithms, 3D image processing, Synthetic aperture radar, 3D image reconstruction, Data modeling, 3D acquisition, 3D modeling, Stereoscopy, Image restoration, Detection and tracking algorithms
There is an increasing interest in three-dimensional Synthetic Aperture Radar (3-D SAR) imaging from observed sparse scattering data. However, the existing 3-D sparse imaging method requires large computing times and storage capacity. In this paper, we propose a modified method for the sparse 3-D SAR imaging. The method processes the collection of noisy SAR measurements, usually collected over nonlinear flight paths, and outputs 3-D SAR imagery. Firstly, the 3-D sparse reconstruction problem is transformed into a series of 2-D slices reconstruction problem by range compression. Then the slices are reconstructed by the modified SL0 (smoothed l0 norm) reconstruction algorithm. The improved algorithm uses hyperbolic tangent function instead of the Gaussian function to approximate the l0 norm and uses the Newton direction instead of the steepest descent direction, which can speed up the convergence rate of the SL0 algorithm. Finally, numerical simulation results are given to demonstrate the effectiveness of the proposed algorithm. It is shown that our method, compared with existing 3-D sparse imaging method, performs better in reconstruction quality and the reconstruction time.
Optical clocks surpass the primary Cs microwave clocks with excellent performances. This allows new studies both in
fundamental physics and engineering. The paper presents the optical system for our space optical clock at NTSC.
Different from it in the laboratory, novel approaches and techniques were used to meet the space requirement of
compactness and reliability. The modular consisting of three robust subunits was developed, which was one laser sources
breadboard and two optical paths systems breadboards. The compact dimension of the optical system is
540mm×440mm×130mm and the total mass was approximate 28 kilogram. The deformation of two optical paths
systems was calculated under an overload test by a mechanical analysis and it could meet the requirement. It is a
advancement from lab to engineering application based on the work, which provides effective foundation for improving
the optical system.
Outliers and occlusions are important degradation in the real application of point matching. In this paper, a novel point matching algorithm based on the reference point pairs is proposed. In each iteration, it firstly eliminates the dubious matches to obtain the relatively accurate matching points (reference point pairs), and then calculates the shape contexts of the removed points with reference to them. After re-matching the removed points, the reference point pairs are combined to achieve better correspondences. Experiments on synthetic data validate the advantages of our method in comparison with some classical methods.
Detection of anomalous targets of various sizes in hyperspectral data has received a lot of attention in reconnaissance and surveillance applications. Many anomaly detectors have been proposed in literature. However, current methods are susceptible to anomalies in the processing window range and often make critical assumptions about the distribution of the background data. Motivated by the fact that anomaly pixels are often distinctive from their local background, in this letter, we proposed a novel hyperspectral anomaly detection framework for real-time remote sensing applications. The proposed framework consists of four major components, sparse feature learning, pyramid grid window selection, joint spatial-spectral collaborative coding and multi-level divergence fusion. It exploits the collaborative representation difference in the feature space to locate potential anomalies and is totally unsupervised without any prior assumptions. Experimental results on airborne recorded hyperspectral data demonstrate that the proposed methods adaptive to anomalies in a large range of sizes and is well suited for parallel processing.
This paper proposes an algorithm of simulating spatially correlated polarimetric synthetic aperture radar (PolSAR) images based on the inverse transform method (ITM). Three flexible non-Gaussian models are employed as the underlying distributions of PolSAR images, including the KummerU, W and M models. Additionally, the spatial correlation of the texture component is considered, which is described by a parametric model called the anisotropic Gaussian function. In the algorithm, PolSAR images are simulated by multiplying two independent components, the speckle and texture, that are generated separately. There are two main contributions referring to two important aspects of the ITM. First, the inverse cumulative distribution functions of all the considered texture distributions are mathematically derived, including the Fisher, Beta, and inverse Beta models. Second, considering the high computational complexities the implicitly expressed correlation transfer functions of these texture distributions have, we develop an alternative fast scheme for their computation by using piecewise linear functions. The effectiveness of the proposed simulation algorithm is demonstrated with respect to both the probability density function and spatial correlation.
Precise segmentation is crucial for the feature extraction and classification of ships in SAR imagery. To alleviate the Doppler shift and the cross ambiguity, this paper propose to segment the ship area from its background based on the radon transform. Assuming that the region of interest (ROI) of ship in SAR imagery has been extracted, the detail procedures of the proposed refined segmentation can be summarized as follows. First, the ship’s ROI image is transformed to radon domain, in which pixel intensities are cumulated along different directions. Then, the peak areas are separated to extract the ship’s orientation and the main image area of the ship that orthogonal to the principal axis. Finally, the refined segmentation is achieved in the main image area. Experiments, accomplished over measured medium and high resolution SAR ship images, show the effectiveness of the proposed approach.
As a method of representing the test sample with few training samples from an overcomplete dictionary, sparse representation has attracted much attention in target classification recently. In this paper, we develop a new SAR vehicle classification method based on sparse representation, in which the correlation between the vehicle’s aspect angle and the sparse representation coefficients is exploited. The detail procedure presented in this paper can be summarized as follows. Initially, the sparse coefficient vector of a test sample is solved by sparse representation algorithm with a pixel based dictionary. Then the coefficient vector is projected onto a sparser one with the constraint of vehicle’s aspect angle. Finally, the vehicle is classified to a certain category that minimizes the reconstruct error with the sparse coefficient vector. We present promising results of applying the proposed method to the MSTAR dataset.
Ship detection is significant especially with the increasing worldwide cooperation in commerce and military affairs.
Space-borne Synthetic Aperture Radar (SAR) is optimal for ship detection due to its high resolution over wide swaths
and all-weather working capability. Constant False Alarm Rate (CFAR) detection of ships in SAR imagery is a robust
and popular choice. K distribution has been widely accepted for homogeneous sea clutter modeling. Although localized
K-distribution based CFAR detection has been developed to solve the non-homogeneous problem, it is not satisfied
under adverse conditions, for example, interference target appears in the background window. In order to overcome its
shortcomings, this paper presents an adaptive algorithm to improve the performance. It mainly includes the homogeneity
assessment of the local background area and the automatic selection between the localized K-distribution-based CFAR
detector and the OS-CFAR detector, which has better detecting performance in non-homogeneous situation. The theory
is investigated in detail firstly, and then experiments are carried out and the results illustrate that the novel algorithm
outperforms the state-of-art methods especially under complex sea background condition.
In this paper, a novel CFAR algorithm for detecting layover and shadow areas in Interferometric synthetic aperture radar (InSAR) images is proposed. Firstly, the probability density function (PDF) of the square root amplitude of InSAR image is estimated by the kernel density estimation. Then, a CFAR algorithm combining with the morphological method for detecting both layover and shadow is presented. Finally, the proposed algorithm is evaluated on a real InSAR image obtained by TerraSAR-X system. The experimental results have validated the effectiveness of the proposed method.