Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

Обнаружение транспортных средств на изображениях загородных шоссе на основе метода Single shot multibox Detector

This chapter introduces the subject of statistical pattern recognition (SPR). It starts by considering how features are defined and emphasizes that the nearest neighbor algorithm achieves error rates comparable with those of an ideal Bayes’ classifier. The concepts of an optimal number of features, representativeness of the training data, and the need to avoid overfitting to the training data are stressed. The chapter shows that methods such as the support vector machine and artificial neural networks are subject to these same training limitations, although each has its advantages. For neural networks, the multilayer perceptron architecture and back-propagation algorithm are described. The chapter distinguishes between supervised and unsupervised learning, demonstrating the advantages of the latter and showing how methods such as clustering and principal components analysis fit into the SPR framework. The chapter also defines the receiver operating characteristic, which allows an optimum balance between false positives and false negatives to be achieved.

Statistical Pattern Recognition

Object detection in optical remote sensing images, being a fundamental but challenging problem in the field of aerial and satellite image analysis, plays an important role for a wide range of applications and is receiving significant attention in recent years. While enormous methods exist, a deep review of the literature concerning generic object detection is still lacking. This paper aims to provide a review of the recent progress in this field. Different from several previously published surveys that focus on a specific object class such as building and road, we concentrate on more generic object categories including, but are not limited to, road, building, tree, vehicle, ship, airport, urban-area. Covering about 270 publications we survey (1) template matching-based object detection methods, (2) knowledge-based object detection methods, (3) object-based image analysis (OBIA)-based object detection methods, (4) machine learning-based object detection methods, and (5) five publicly available datasets and three standard evaluation metrics. We also discuss the challenges of current studies and propose two promising research directions, namely deep learning-based feature representation and weakly supervised learning-based geospatial object detection. It is our hope that this survey will be beneficial for the researchers to have better understanding of this research field.

A survey on object detection in optical remote sensing images

In this paper, we examine the role of polarimetry in synthetic aperture radar (SAR) interferometry. We first propose a general formulation for vector wave interferometry that includes conventional scalar interferometry as a special case. Then, we show how polarimetric basis transformations can be introduced into SAR interferometry and applied to form interferograms between all possible linear combinations of polarization states. This allows us to reveal the strong polarization dependency of the interferometric coherence. We then solve the coherence optimization problem involving maximization of interferometric coherence and formulate a new coherent decomposition for polarimetric SAR interferometry that allows the separation of the effective phase centers of different scattering mechanisms. A simplified stochastic scattering model for an elevated forest canopy is introduced to demonstrate the effectiveness of the proposed algorithms. In this way, we demonstrate the importance of wave polarization for the physical interpretation of SAR interferograms. We investigate the potential of polarimetric SAR interferometry using results from the evaluation of fully polarimetric interferometric shuttle imaging radar (SIR)-C/X-SAR data collected during October 8-9, 1994, over the SE Baikal Lake Selenga delta region of Buriatia, Southeast Siberia, Russia.

Polarimetric SAR interferometry

Due to the skew data support region (DSR) of the 2-D wavenumber spectrum for high-squint synthetic aperture radar (HS-SAR), the conventional $\omega$ – $k$ algorithm cannot fully utilize the DSR and thus degrades the resolution if simply taking a rectangle region. Meanwhile, for the small-aperture data, direct azimuth imaging in the distance domain will lead to serious aliasing. A modified $\omega$ – $k$ algorithm is derived in this paper to solve these problems. The maximum usage of DSR is achieved by the coordinate rotation. As for the azimuth dependence, the method of azimuth resampling is used to get the uniform focusing. Different from the traditional $\omega$ – $k$ method, the modified $\omega$ – $k$ algorithm focuses the small-aperture data in the azimuth wavenumber domain by SPECAN processing, which avoids padding a large number of zeros when imaging in the azimuth distance domain. Simulated and real-data results show the validity and effectiveness of the presented algorithm.

A Modified $\omega$ – $k$ Algorithm for HS-SAR Small-Aperture Data Imaging

Bistatic forward-looking synthetic aperture radar (BFSAR) breaks through the limitations of the conventional monostatic SAR on the forward-looking imaging. However, the problems of range cell migration (RCM) caused by the linear range walk and 2-D spatial variability of Doppler parameters become more serious and complicated in translational variant BFSAR. In this article, a keystone transform is introduced to correct the linear RCM. Based on the characteristics of a small aperture, the nonlinear chirp scaling (NCS) is discussed in the frequency domain to equalize the azimuth-range-dependent Doppler parameters. The improved NCS in our newly proposed BFSAR imaging algorithm, especially the re-definition of range direction and the model of spatial variant phase, differentiates this article from all the existing studies in the literature on BFSAR signal processing. Simulation results and real data processing further validate the effectiveness of the proposed algorithm.

A Frequency-Domain Imaging Algorithm for Translational Variant Bistatic Forward-Looking SAR

A new algorithm for filling sparse aperture synthetic aperture radar (SAR)/inverse SAR (ISAR) data, which applies for widely gapped apertures, is proposed in this letter. An Estimating Signal Parameter via Rotational Invariance Techniques(ESPRIT)-based parametric approach is first used to estimate the power distribution with the sparse data. With the estimated power spectrum as prior information, by minimizing a weighted norm as a constraint, the full aperture data can be estimated. Although the algorithm is proposed for the sparse aperture interpolation in SAR/ISAR, it can be applied to other gapped data spectral estimation problems as well. Both numerical and experimental examples are provided to demonstrate the performance of the proposed algorithm.

A New Algorithm for Sparse Aperture Interpolation

In wideband radar imaging, it is necessary to increase the target observation angle, that is, the imaging time to obtain high azimuth resolution. For wide-angle targets, however, migration through resolution cell will occur and the Doppler information of the scatterers will be time varying, which can lead to image defocusing in both the range and azimuth domain if the range–Doppler (RD) method is applied. Aimed at this problem, this study builds up the imaging model for steadily moving targets with a constant rotation axis. Then it compares the RD algorithm, the Keystone algorithm and the polar format algorithm with theoretical analysis and deduces the maximum observation angles that fit for these methods, respectively. Additionally, this study proposes a new algorithm for imaging of targets with large observation angles based on the complex inverse Radon transform. Instead of applying the Taylor's series approximation, this method strictly follows the signal model for scatterers with a wide observation angle and utilises curvilinear integral with phase information for image formation. Therefore it is valid in wide-angle scenarios. Finally, imaging results for both the simulated and measured data are given to analyse the performance of the available algorithms and to prove the validity of the proposed algorithm.

Analysis of wide-angle radar imaging

A squinted observation geometry along with long integration time significantly aggravates the range walk and spatial variation of a medium-earth-orbit (MEO) synthetic aperture radar (SAR) signal. Variable pulse repeating frequency (PRF) is recommended to avoid the blockage in echo recording and save storage space. The existing wavenumber algorithms cannot handle the nonlinear and range–azimuth-coupled spatial variation (RACSP) over a large scene. In this paper, we propose a modified Stolt mapping method along with a modified joint time and Doppler resampling (JTDR) for highly squinted MEO SAR data processing. An azimuth timescale transformation is used to deal with the nonlinear spatial variation of the azimuth frequency-modulation (FM) rate. An extended Omega-K is used to linearize the range frequency and achieve range cell migration correction (RCMC). To address the RACSP, the Doppler is linearized in the range-Doppler domain using a range-dependent Doppler scale transformation. The computational complexity and geometry distortion correction (GDC) are also discussed. Simulation results are shown to verify the effectiveness of the developed focusing approaches.

Mengdao Xing

Papers

A Modified $\omega$ – $k$ Algorithm for HS-SAR Small-Aperture Data Imaging

A Frequency-Domain Imaging Algorithm for Translational Variant Bistatic Forward-Looking SAR

A New Algorithm for Sparse Aperture Interpolation

Analysis of wide-angle radar imaging

Highly Squinted MEO SAR Focusing Based on Extended Omega-K Algorithm and Modified Joint Time and Doppler Resampling