ATOMBEAM TECHNOLOGIES INC.

A system and method for processing Synthetic Aperture Radar (SAR) data using cross polarization prediction. The system processes complex SAR data from two different polarizations, typically HH and VV, and stores their representations. A specialized reconstruction component then reconstructs data from the first polarization while utilizing the second polarization data as auxiliary information. This cross-polarization approach enhances the reconstruction process, leveraging complementary information from both polarizations to produce improved SAR imagery. The method can employ neural network-based compression and reconstruction techniques, adaptive processing based on SAR data characteristics, and sophisticated analysis of cross-polarization relationships. Advanced implementations may include interpolation between polarizations, extrapolation of polarization effects, and arithmetic operations on processed representations to further enhance the SAR data. This innovation enables the generation of high-quality, information-rich SAR images with potential applications in environmental monitoring, urban planning, agriculture, and disaster response.

A system and method are disclosed for processing hyperspectral image information. A set of hyperspectral images is obtained, and a spectrum compact representation is derived for each hyperspectral image in the set of hyperspectral images. The spectrum compact representation for each hyperspectral image in the set of hyperspectral images is encoded with a video encoder, such as an H.266 encoder, to generate an encoded spectrum compact representation for each hyperspectral image in the set of hyperspectral images. The encoded spectrum compact representation is a compressed representation of the original hyperspectral image information that can be transmitted via a communication channel. On a receiving side of the communication channel, the compressed representation can be decompressed utilizing a corresponding video decoder, in conjunction with a joint spatial-frequency (JSF) network.

For compressing synthetic aperture radar (SAR) images, preprocessing operations are performed on an input SAR image. A discrete cosine transform is performed on the image, and multiple subbands are created, where each subband represents a particular range of frequencies. The subbands are organized into multiple groups, where the multiple groups comprise a first low frequency group, a second low frequency group, and a high frequency group. A latent space representation is generated corresponding to each of the multiple groups of subbands. A first bitstream is created based on the latent space representation, and an alternate representation of the latent space is used for creating a second bitstream, enabling multiple-pass techniques for SAR image data compression, including phase unwrapping for supporting interferometric SAR (InSAR) applications.

A system and method for compressing synthetic aperture radar (SAR) images with enhanced phase recovery and unwrapping capabilities is disclosed. The system performs preprocessing on input SAR images, applies discrete cosine transform (DCT) to create subbands, and utilizes a multi-pass amplitude compression technique. A specialized neural network performs phase unwrapping using compressed amplitude information and interferogram wrapped phase data. The system employs a channel-wise transformer fusion block (CTFB) for feature fusion and a multi-stage context recovery subsystem with optimized loss functions for both amplitude and phase recovery. The method achieves improved compression efficiency and phase recovery accuracy, particularly beneficial for Interferometric SAR (InSAR) applications.

For compressing synthetic aperture radar (SAR) images, preprocessing operations are performed on an input SAR image. A discrete cosine transform is performed on the image, and multiple subbands are created, where each subband represents a particular range of frequencies. The subbands are organized into multiple groups, where the multiple groups comprise a first low frequency group, a second low frequency group, and a high frequency group. A latent space representation is generated corresponding to each of the multiple groups of subbands. A first bitstream is created based on the latent space representation, and an alternate representation of the latent space is used for creating a second bitstream, enabling multiple-pass techniques for SAR image data compression, including phase unwrapping for supporting interferometric SAR (InSAR) applications.

A system and method for compressing temporal stacks of synthetic aperture radar (SAR) images while preserving interferometric properties. The system receives multiple SAR images acquired over time, aligns them through coregistration, and maintains phase continuity across the temporal sequence. A three-dimensional discrete cosine transform processes both spatial and temporal dimensions, creating hybrid subbands organized by frequency content and temporal change characteristics. The system employs a change-aware encoder that selectively uses differential encoding for small changes between frames and full encoding at adaptive keyframe intervals. A temporal coherence network with separate pathways for amplitude and phase information ensures consistency across the image stack. The compressed output preserves interferometric coherence properties essential for applications such as ground deformation monitoring and change detection. The system achieves compression ratios from 10:1 to 50:1 for static content while maintaining higher quality for rapidly changing features.

A system and method are disclosed for processing hyperspectral image information. A set of hyperspectral images is obtained, and a spectrum compact representation is derived for each hyperspectral image in the set of hyperspectral images. The spectrum compact representation for each hyperspectral image in the set of hyperspectral images is encoded with a video encoder, such as an H.266 encoder, to generate an encoded spectrum compact representation for each hyperspectral image in the set of hyperspectral images. The encoded spectrum compact representation is a compressed representation of the original hyperspectral image information that can be transmitted via a communication channel. On a receiving side of the communication channel, the compressed representation can be decompressed utilizing a corresponding video decoder, in conjunction with a joint spatial-frequency (JSF) network.

Systems and methods are disclosed for classifying objects using electro-optical and synthetic aperture radar images through multi-modal feature alignment and fusion. A computing system acquires and preprocesses image data, then aligns features across modalities using a multi-modal alignment engine. A cross-modal attention fusion network extracts and integrates complementary information using transformer-based attention mechanisms. A modality-specific feature extraction framework processes EO and SAR images through specialized branches, ensuring optimal feature representation. An adaptive fusion decision system dynamically determines the best fusion strategy based on image quality and confidence scores. A self-supervised consistency controller enforces alignment between EO and SAR features using contrastive learning. The fused representations are processed by a neural network to generate object classifications. This system improves accuracy and robustness in environments where one modality may be degraded or missing, enhancing applications such as remote sensing, surveillance, and autonomous navigation.

A system and method are disclosed for generating hyperspectral images from multi-modal sensor data including RGB, LiDAR, thermal, and near-infrared inputs. Training data includes hyperspectral images and corresponding multi-modal measurements. Spectral band grouping is performed based on correlation coefficients. A multi-modal decomposition network with cross-modal attention mechanisms generate reconstructed hyperspectral images by fusing complementary sensor information. A fine-tuning network creates reconstructed RGB images. A comprehensive quality assurance system analyzes spectral consistency, cross-modal coherence, and fusion artifacts to generate quality metrics. Missing data compensation strategies handle corrupted sensor inputs using information from other modalities. The system includes temporal integration for video sequences and multi-resolution processing for different sensor resolutions. Quality metrics guide network weight adjustments to improve reconstruction accuracy while maintaining robustness to sensor failures and environmental variations.

For compressing synthetic aperture radar (SAR) images, preprocessing operations are performed on an input SAR image. A discrete cosine transform is performed on the image, and multiple subbands are created, where each subband represents a particular range of frequencies. The subbands are organized into multiple groups, where the multiple groups comprise a first low frequency group, a second low frequency group, and a high frequency group. A latent space representation is generated corresponding to each of the multiple groups of subbands. Compression is performed on one or more of the subbands, thereby creating a compressed bitstream, wherein the compressed bitstream is a compressed version of the input SAR image.