Fei Xia

Urban Land Use and Land Cover Classification Using Remotely Sensed SAR Data through Deep Belief Networks

Land use and land cover (LULC) mapping in urban areas is one of the core applications in remote sensing, and it plays an important role in modern urban planning and management. Deep learning is springing up in the field of machine learning recently. By mimicking the hierarchical structure of the human brain, deep learning can gradually extract features from lower level to higher level. The Deep Belief Networks (DBN) model is a widely investigated and deployed deep learning architecture. It combines the advantages of unsupervised and supervised learning and can archive good classification performance. This study proposes a classification approach based on the DBN model for detailed urban mapping using polarimetric synthetic aperture radar (PolSAR) data. Through the DBN model, effective contextual mapping features can be automatically extracted from the PolSAR data to improve the classification performance. Two-date high-resolution RADARSAT-2 PolSAR data over the Great Toronto Area were used for evaluation. Comparisons with the support vector machine (SVM), conventional neural networks (NN), and stochastic Expectation-Maximization (SEM) were conducted to assess the potential of the DBN-based classification approach. Experimental results show that the DBN-based method outperforms three other approaches and produces homogenous mapping results with preserved shape details.

Airport Detection on Optical Satellite Images Using Deep Convolutional Neural Networks

This letter proposes a method using convolutional neural networks (CNNs) for airport detection on optical satellite images. To efficiently build a deep CNN with limited satellite image samples, a transfer learning approach had been employed by sharing the common image features of the natural images. To decrease the computing cost, an efficient region proposal method had been proposed based on the prior knowledge of the line segments distribution in an airport. The transfer learning ability on deep CNN for airport detection on satellite images had been first evaluated in this letter. The proposed method was tested on an image data set, including 170 different airports and 30 nonairports. The detection rate could reach 88.8% in experiments with seconds’ computation time, which showed a great improvement over other the state-of-the-art methods.

FPGA implementation of an exact dot product and its application in variable-precision floating-point arithmetic

The current paper explores the capability and flexibility of field programmable gate-arrays (FPGAs) to implement variable-precision floating-point (VP) arithmetic. First, the VP exact dot product algorithm, which uses exact fixed-point operations to obtain an exact result, is presented. A VP multiplication and accumulation unit (VPMAC) on FPGA is then proposed. In the proposed design, the parallel multipliers generate the partial products of mantissa multiplication in parallel, which is the most time-consuming part in the VP multiplication and accumulation operation. This method fully utilizes DSP performance on FPGAs to enhance the performance of the VPMAC unit. Several other schemes, such as two-level RAM bank, carry-save accumulation, and partial summation, are used to achieve high frequency and pipeline throughput in the product accumulation stage. The typical algorithms in Basic Linear Algorithm Subprograms (i.e., vector dot product, general matrix vector product, and general matrix multiply product), LU decomposition, and Modified Gram–Schmidt QR decomposition, are used to evaluate the performance of the VPMAC unit. Two schemes, called the VPMAC coprocessor and matrix accelerator, are presented to implement these applications. Finally, prototypes of the VPMAC unit and the matrix accelerator based on the VPMAC unit are created on a Xilinx XC6VLX760 FPGA chip.Compared with a parallel software implementation based on OpenMP running on an Intel Xeon Quad-core E5620 CPU, the VPMAC coprocessor, equipped with one VPMAC unit, achieves a maximum acceleration factor of 18X. Moreover, the matrix accelerator, which mainly consists of a linear array of eight processing elements, achieves 12X–65X better performance.

Airport detection from remote sensing images using transferable convolutional neural networks

This paper presents a method for airport detection from optical satellite images using deep convolutional neural networks (CNN). To achieve fast detection with high accuracy, region proposal by searching adjacent parallel line segments has been applied to select candidate fields with potential runways. These proposals were further classified by a CNN model transfer learned from AlexNet to identify the final airport regions from other confusing classes. The proposed method has been tested on a remote sensing dataset consisting of 120 airports. Experiments showed that the proposed method could recognize airports from a large complex area in seconds with an accuracy of 84.1%.

A Fine-Grained Pipelined Implementation for Large-Scale Matrix Inversion on FPGA

Large-scale matrix inversion play an important role in many applications. However to the best of our knowledge, there is no FPGA-based implementation. In this paper, we explore the possibility of accelerating large-scale matrix inversion on FPGA. To exploit the computational potential of FPGA, we introduce a fine-grained parallel algorithm for matrix inversion. A scalable linear array processing elements (PEs), which is the core component of the FPGA accelerator, is proposed to implement this algorithm. A total of 12 PEs can be integrated into an Altera StratixII EP2S130F1020C5 FPGA on our self-designed board. Experimental results show that a factor of 2.6 speedup and the maximum power-performance of 41 can be achieved compare to Pentium Dual CPU with double SSE threads.

Local kernel alignment based multi-view clustering using extreme learning machine

Abstract A similarity or dissimilarity measure, such as the Euclidean distance, is crucial to discriminative clustering algorithms. These measures used to calculate pairwise similarities between samples rely on data representations in a feature space. However, discriminative clustering fails if the samples in a feature space are linearly inseparable. This problem can be solved by performing a nonlinear data transformation into a high dimensional feature space, which can increase the probability of the linear separability of the samples within the transformed feature space and simplify the associated data structure. Mercer kernels, which are constructed using such a nonlinear data transformation, have been widely used in clustering tasks. Extreme learning machine (ELM) is a new method that exhibits promising clustering performance owing to its universal approximation capability, easy parameter selection, explicit feature mapping process, and excellent feature representation capability. This study proposes an ELM based multi-view learning approach with different views generated by ELM random feature mapping with respect to different hidden-layer nodes, and exploits the properties of these views. Experiments show that better clustering results can be obtained by combining these views together compared with the corresponding ELM-based single-view clustering methods and the traditional algorithms which are performed in the feature space of the original data. Moreover, local kernel alignment property is widespread in these views. This alignment helps the clustering algorithm focus on closer sample pairs. This study also proposes an ELM based multiple kernel clustering algorithm with local kernel alignment maximization. The proposed algorithm is experimentally demonstrated on 10 single-view benchmark datasets and yields superior clustering performance when compared with the state-of-the-art multi-view clustering methods in recent literatures. Thus, the effectiveness and superiority of maximizing local kernel alignment on those views constructed by the proposed method are verified.

Hyperspectral image classification via local receptive fields based random weights networks

This paper proposes a classification approach for hyperspectral image using the local receptive fields based random weights networks. The local receptive field concept originates from research in neuroscience. Considering the local correlations of spectral features, it is promising to improve the performance of HSI classification by introducing the local receptive fields. The proposed classification framework consists of four layers, i.e., input layer, convolution layer, pooling layer, and output layer. The convolution and pooling layer are used for feature extracting and the last layer is used as the classifier. Experimental results on the ROSIS Pavia University dataset confirm the effectiveness of the proposed HSI classification method.