Yesheng Gao

Bio: Yesheng Gao is an academic researcher from Shanghai Jiao Tong University. The author has contributed to research in topics: Synthetic aperture radar & Inverse synthetic aperture radar. The author has an hindex of 7, co-authored 88 publications receiving 182 citations.

Optronic convolutional neural networks of multi-layers with different functions executed in optics for image classification.

Abstract: Although deeper convolutional neural networks (CNNs) generally obtain better performance on classification tasks, they incur higher computation costs. To address this problem, this study proposes the optronic convolutional neural network (OPCNN) in which all computation operations are executed in optics, and data transmission and control are executed in electronics. In OPCNN, we implement convolutional layers with multi input images by the lenslet 4f system, downsampling layers by optical-strided convolution and obtaining nonlinear activation by adjusting the camera's curve and fully connected layers by optical dot product. The OPCNN demonstrates good performance on the classification tasks in simulations and experiments and achieves better performance than other current optical convolutional neural networks by comparison due to the more complex architecture. The scalability of OPCNN contributes to building deeper networks when facing complicated datasets.

Improved Channel Error Calibration Algorithm for Azimuth Multichannel SAR Systems

Abstract: Multichannel synthetic aperture radar systems in azimuth can effectively suppress azimuth ambiguity and are promising in high-resolution wide-swath imaging. However, unavoidable channel errors will significantly degrade the performance of ambiguity suppression. Conventional subspace calibration methods usually estimate phase error via decomposing a Doppler-variant covariance matrix from one Doppler bin, and then average these errors estimated from several Doppler bins to improve the estimation accuracy, which will result in a large computational load. This letter presents an improved channel error calibration method, which works on the undersampled data of the individual azimuth channel. By a proposed matrix transformation method, the Doppler-variant covariance matrices will be transformed into a constant covariance matrix. Therefore, the improved calibration algorithm needs to estimate and decompose the new covariance matrix only once. The computation load could be greatly reduced. Moreover, the new covariance matrix can be estimated by training samples not only from range bins but also from Doppler bins, which will improve the estimation accuracy. Theoretical analysis and experiments based on simulations and measurements showed the high accuracy, efficiency, and robustness of the improved method, particularly in low signal-to-noise ratio.

Robust Channel Phase Error Calibration Algorithm for Multichannel High-Resolution and Wide-Swath SAR Imaging

Abstract: High-resolution and wide-swath synthetic aperture radar (SAR) imaging can be achieved by the azimuth multichannel system. The minimum variance distortionless response (MVDR) beamformer can be utilized to suppress azimuth ambiguities. However, the presence of channel phase errors significantly deteriorates the performance of the azimuth multichannel SAR system. Instead of employing subspace techniques, this letter proposes a robust channel phase error calibration algorithm via maximizing the MVDR beamformer output power. Compared with the conventional subspace-based calibration methods, there is no redundancy of channels required to estimate the subspaces in the proposed algorithm. Also, the proposed algorithm is relatively robust, because it avoids the subspace swap phenomenon, which probably takes place at low signal-to-noise ratios for the subspace techniques. Moreover, the proposed method has the advantage of estimating the channel phase errors without covariance matrix decomposition, which reduces the computation load. The simulation experiments and the real data processing validate the effectiveness of the proposed calibration method.

Imaging through scattering media using speckle pattern classification based support vector regression.

Abstract: Imaging through scattering media is a common practice in many applications of biomedical imaging. Object image would deteriorate into unrecognizable speckle pattern when scattering media is presented. Many methods have been investigated to reconstruct the object image when only speckle pattern is available. In this paper, we demonstrate a method of single-shot imaging through scattering media. This method is based on classification and support vector regression of the measured speckle pattern. We prove the possibility of speckle pattern classification and related formulas are presented. The specified and limited imaging capability without speckle pattern classification is demonstrated. Our proposed approach, that is, speckle pattern classification based support vector regression method, makes up the deficiency. Experimental results show that, with our approach, speckle patterns could be utilized for classification when object images are unavailable, and object images can be reconstructed with high fidelity. The proposed approach for imaging through scattering media is expected to be applicable to various sensing schemes.

Spotlight Synthetic Aperture Radar Signal Processing Algorithms

Abstract: Thank you very much for downloading spotlight synthetic aperture radar signal processing algorithms. Maybe you have knowledge that, people have search numerous times for their favorite books like this spotlight synthetic aperture radar signal processing algorithms, but end up in malicious downloads. Rather than reading a good book with a cup of coffee in the afternoon, instead they juggled with some harmful virus inside their laptop.

SAR Imaging via Modern 2-D Spectral Estimation Methods. Volume 1. Imaging Methods.

Abstract: : This report discusses the use of modern 2-D spectral estimation algorithms for SAR imaging, and makes two principal contributions to the field of adaptive SAR imaging. First, it is a comprehensive comparison of 2-D spectral estimation methods for SAR imaging. It provides a synopsis of the algorithms available, discusses their relative merits for SAR imaging, and illustrates their performance on simulated and collected SAR imagery. The discussion of autoregressive linear predictive techniques (ARLP), including the Tufts Kumaresan variant, is somewhat more general than appears in most of the literature, in that it allows the prediction element to be varied throughout the subaperture. This generality leads to a theoretical link between ARLP and one of Pisarenko's methods. The report also provides a theoretical analysis that predicts the impact of the adaptive sidelobe reduction (ASR) algorithm on target to clutter ratio and provides insight into order and constraint selection. Second, this work develops multi-channel variants of three related algorithms, minimum variance method (MVM), reduced rank MVM (RRMVM), and ASR to estimate both reflectivity intensity and interferometric height from polarimetric displaced-aperture interferometric data. Examples illustrate that MVM and ASR both offer significant advantages over Fourier methods for estimating both scattering intensity and interferometric height, and allow empirical comparison of the accuracies of Fourier, MVM, and- ometric height estimates.

Massively parallel amplitude-only Fourier neural network

Abstract: Machine intelligence has become a driving factor in modern society. However, its demand outpaces the underlying electronic technology due to limitations given by fundamental physics, such as capacitive charging of wires, but also by system architecture of storing and handling data, both driving recent trends toward processor heterogeneity. Task-specific accelerators based on free-space optics bear fundamental homomorphism for massively parallel and real-time information processing given the wave nature of light. However, initial results are frustrated by data handling challenges and slow optical programmability. Here we introduce a novel amplitude-only Fourier-optical processor paradigm capable of processing large-scale ∼(1000×1000) matrices in a single time step and 100 µs-short latency. Conceptually, the information flow direction is orthogonal to the two-dimensional programmable network, which leverages 106 parallel channels of display technology, and enables a prototype demonstration performing convolutions as pixelwise multiplications in the Fourier domain reaching peta operations per second throughputs. The required real-to-Fourier domain transformations are performed passively by optical lenses at zero-static power. We exemplary realize a convolutional neural network (CNN) performing classification tasks on 2 megapixel large matrices at 10 kHz rates, which latency-outperforms current graphic processing unit and phase-based display technology by 1 and 2 orders of magnitude, respectively. Training this optical convolutional layer on image classification tasks and utilizing it in a hybrid optical-electronic CNN, shows classification accuracy of 98% (Modified National Institute of Standards and Technology) and 54% (CIFAR-10). Interestingly, the amplitude-only CNN is inherently robust against coherence noise in contrast to phase-based paradigms and features a delay over 2 orders of magnitude lower than liquid-crystal-based systems. Such an amplitude-only massively parallel optical compute paradigm shows that the lack of phase information can be accounted for via training, thus opening opportunities for high-throughput accelerator technology for machine intelligence with applications in network-edge processing, in data centers, or in pre-processing information or filtering toward near-real-time decision making.

2010 ieee international geoscience and remote sensing symposium

Abstract: Volume 1 Pages 1-251