Disclosed are a method and an apparatus for multi-scale SAR image recognition based on attention mechanism. According to the method, a whole image recognition network is adjusted by training a SAR training image by an attention prediction subnet, a region-of-interest positioning subnet and an image classification subnet in combination with a network loss, which greatly improves a network performance; and in addition, an attention prediction map is generated by attention mechanism to position a most prominent feature part in the SAR image, which greatly eliminates a redundancy of image features in a machine vision, effectively determines a region-of-interest, reduces interference of image noises, greatly reduces an image processing time, improves a target recognition accuracy, is beneficial to next target positioning, and has a significant improvement on a network recognition speed integrally.

A system and method to generate flood inundation maps in near real time. The system includes a plurality of computer processing modules: a flood trigger system, a SAR data query system, and a RAPID kernel algorithm system, running in real time, to identify the potential flood zones, query SAR data, and finally compute the inundation maps, respectively. As disclosed herein, the RAPID kernel algorithm is extended to a fully automated flood mapping system that requires no human interference from the initial flood events discovery to the final flood map production.

A method, a device and a storage medium for extracting height and deformation information of a high voltage transmission tower by using SAR tomography are disclosed. A Spaceborne satellite is used to obtain a plurality of high resolution SAR images of a region having a high voltage transmission tower. The images are then pre-processed via registration, dechirp and phase compensation to obtain observation data. The observation data is then discretized. Then the discretized observation data is resampled by using singular value decomposition, and Akaike information criterion is used to estimate the number of scattering points in the resampled observation data. The obtained number of the scattering points is used to eliminate the singular values in the reconstructed signal. Finally, the sparsity of the observation data is used together with compression sensing to realize signal reconstruction, thereby extracting the height and deformation rate at a position on the high voltage transmission tower.

A synthetic aperture radar image analysis system 20 includes: a phase correlation determination means 21 which determines a strength of the phase correlation between a plurality of pixels in an image selected from among a plurality of images on the basis of the plurality of images that have been photographed by a synthetic aperture radar and show the same point; a shape determination means 22 which determines a degree of similarity between the shape of the distribution of the plurality of pixels and an object shape indicated by geospatial information; and an association means 23 which associates the plurality of pixels with the object on the basis of the determined strength of the phase correlation and the determined degree of similarity.

A system is configured to receive synthetic aperture radar (SAR) backscatter signatures of a geographical area including the object of interest from a SAR device. The system also extracts feature vectors from the SAR backscatter signature based on the intensity values of the SAR backscatter signature. The system inputs the one or more feature vectors into a neural network model. The system receives, as output from the neural network model, coordinate values indicating one or more visual features of the object of interest. Using these coordinate values, the system determines one or more measurements of the object of interest.

An aerial vehicle system for mapping an object buried in a subterranean surface, the aerial vehicle system including an aerial vehicle, an electronic sensor, a processor, and a memory. The memory includes instructions, which when executed by the processor, cause the system to receive a first input data set by the electronic sensor, the first input data set based on an electromagnetic signal and geographic location data, generate a raw image based on the first input data set, and compare the raw image to a calibration data set, the calibration data set based on material calibration data. The material calibration data is based on unique spectral reflection patterns of an object in a controlled environment at predefined heights and subterranean conditions.

A system configured to identify a target in a synthetic aperture radar signal includes: a feature extractor configured to extract a plurality of features from the synthetic aperture radar signal; a spiking neural network configured to encode the features as a plurality of spiking signals; a readout neural layer configured to compute a signal identifier based on the spiking signals; and an output configured to output the signal identifier, the signal identifier identifying the target.

Disclosed is an electronic device including a display, a camera, a wireless communication circuit connected to an antenna array including a plurality of antenna elements and configured to perform beamforming using the antenna array, a processor, and a memory. The processor may be configured to obtain an image including a face image via the camera, and obtain face data of a face corresponding to the face image via the wireless communication circuit.

Disclosed are method and apparatus for SAR image recognition based on multi-scale features and broad learning. A region of interest of an original SAR image is extracted by centroid localization, the image is rotated and added with noise for enhancing the data volume, the image is downsampled, LBP features and PPQ features are extracted, an LBP feature vector X.sub.LBP and an LPQ feature vector X.sub.LPQ are cascaded to achieve dimension reduction by principal component analysis to obtain a fusion feature data X.sub.m, the fusion feature data X.sub.m is input to a broad learning network for image recognition and a recognition result is output. By fusing the LBP features and the LPQ features, complementary information is fully utilized and redundant information is reduced. The broad learning network is used to improve the training speed and reduce the time cost. As a result, the recognition effect is more stable, robust and reliable.

Optical images in remote sensing are contaminated by cloud cover and bad weather conditions and are only available during the daytime. Whereas SAR images are completely cloud free, independent of weather conditions and can be acquired both during the day and at night. However, due to the speckle effect and side looking imaging mechanism of SAR images, they are not easily interpretable by untrained people. To address this issue, the present disclosure provides a method and system for LULC guided SAR visualization, wherein a GAN is trained to translate SAR images to optical images for visualization. A given SAR image is fed into a first generator of the GAN to obtain LULC map which is then concatenated with the SAR image and fed into a second generator of the GAN to generate an optical image. The LULC map provides semantic information required for generation of more realistic optical image.