A Multi-Scale U-Shaped Convolution Auto-Encoder Based on Pyramid Pooling Module for Object Recognition in Synthetic Aperture Radar Images

Abstract

Although unsupervised representation learning (RL) can tackle the performance deterioration caused by limited labeled data in synthetic aperture radar (SAR) object classification, the neglected discriminative detailed information and the ignored distinctive characteristics of SAR images can lead to performance degradation. In this paper, an unsupervised multi-scale convolution auto-encoder (MSCAE) was proposed which can simultaneously obtain the global features and local characteristics of targets with its U-shaped architecture and pyramid pooling modules (PPMs). The compact depth-wise separable convolution and the deconvolution counterpart were devised to decrease the trainable parameters. The PPM and the multi-scale feature learning scheme were designed to learn multi-scale features. Prior knowledge of SAR speckle was also embedded in the model. The reconstruction loss of the MSCAE was measured by the structural similarity index metric (SSIM) of the reconstructed data and the images filtered by the improved Lee sigma filter. A speckle suppression restriction was also added in the objective function to guarantee that the speckle suppression procedure would take place in the feature learning stage. Experimental results with the MSTAR dataset under the standard operating condition and several extended operating conditions demonstrated the effectiveness of the proposed model in SAR object classification tasks.

Locality Preserving Property Constrained Contrastive Learning for Object Classification in SAR Imagery

“…However, under EOC, images of all the serial numbers were used to test the performance of the proposed method [40]. Following the setup of the EOC experimental method in [41], we used MSTAR datasets with different configurations for evaluation. Due to insufficient training samples, we did not evaluate the performance of the proposed model at the variance of the depression angle.…”

“…However, under EOC, images of all the serial numbers were used to test the performance of the proposed method [40]. Following the setup of the EOC experimental method in [41], we used MSTAR datasets with different config-…”

Convolutional Autoencoder-Based Flaw Detection for Steel Wire Ropes

“…The convolution denoising autoencoder combines the local convolution connection with the autoencoder. Specifically, convolutional operations are used to replace fully connected layers in encoders, while deconvolution operation is used to replace fully connected layers in decoders [ 25 ]. Training deep neural networks is complicated by the fact that the distribution of each layer’s inputs changes during training, as the parameters of the previous layers change [ 26 ].…”

Non-Cooperative SAR Automatic Target Recognition Based on Scattering Centers Models

“…Regardless of the category of the algorithm, it is supported by features of three types: (1) Geometric features that describe the target by its area [ 5 , 6 , 7 , 8 , 9 ], contour [ 10 , 11 ] or shadow [ 11 , 12 ]; (2) Transformation features that reduce the dimensionality of the target data by representing it in another domain such as Discrete Cosine Transform (DCT) [ 13 ], Non-Negative Matrix Factorization (NMF) [ 14 , 15 ], Linear Discriminant Analysis (LDA) [ 16 ] and Principal Component Analysis (PCA) [ 17 ]; and (3) Scattering Centers Features which are based on the highest amplitude returns of the targets [ 18 ] and based on a statistical distance, such as Euclidean [ 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 ], Mahalanobis [ 27 , 28 , 29 , 30 ], or another statistical distance [ 31 , 32 , 33 , 34 , 35 , 36 , 37 ].…”

“…Feature-based algorithms are those with methods that run offline training supported exclusively by features extracted from the targets of interest. Among the methods employed by feature-based algorithms, we can highlight the following: Template Matching (TM) [ 5 , 6 , 7 , 11 , 30 , 37 ], Hidden Markov Model (HMM) [ 12 , 13 , 22 ], K-Nearest Neighbor (KNN) [ 27 , 28 ], Sparse Representation-based Classification (SRC) [ 8 , 29 ], Convolutional Neural Networks (CNN) [ 17 , 18 , 36 , 38 , 39 , 40 , 41 , 42 , 43 , 44 , 45 , 46 , 47 , 48 ], Support Vectors Machine (SVM) [ 9 ] and Gaussian Mixture Model (GMM) [ 10 ].…”