System and methods for upsampling of decompressed data after lossy compression using a neural network
A system and method for complex-valued Synthetic Aperture Radar (SAR) image compression integrates AI-based techniques to enhance compression quality. It incorporates a novel AI deblocking network composed of convolutional layers for feature extraction and a channel-wise transformer with attention to capture complex inter-channel dependencies. The convolutional layers extract multi-dimensional features from the complex-valued SAR image, while the channel-wise transformer learns global inter-channel relationships. This hybrid approach addresses both local and global features, mitigating compression artifacts and improving image quality. The model's outputs enable effective SAR image reconstruction, achieving advanced compression while preserving crucial information for accurate analysis.
1. A system for upsampling of decompressed data after lossy compression using a neural network, comprising:
a computing device comprising at least a memory and a processor;
a trained deep learning algorithm configured to recover phase and amplitude information associated with a Synthetic Aperture Radar (SAR) image; and
a decoder comprising a first plurality of programming instructions stored in the memory and operable on the processor, wherein the first plurality of programming instructions, when operating on the processor, cause the computing device to:
receive a compressed bit stream, the compressed bit stream comprising complex-valued SAR image data;
decode the compressed bit stream to predict both an In-phase (I) channel and a Quadrature (Q) channel; and
use both the I channel and the Q channel as inputs into the trained deep learning algorithm to recover both phase and amplitude information associated with the SAR image data.
2. The system of claim 1 , wherein the trained deep learning algorithm is a neural network that can recover signals from a compressed bitstream.
3. The system of claim 2 , wherein the trained deep learning algorithm further comprises a multi-channel transformer with attention.
4. The system of claim 1 , wherein training the deep learning algorithm comprises two stages, wherein each stage comprises a specific loss function.
5. The system of claim 4 , wherein the loss function is mean squared error.
6. The system of claim 4 , wherein one stage of the two stages is associated with a loss function for the I and Q channels.
7. The system of claim 4 , wherein one stage of the two stages is associated with the amplitude loss.
8. A method for upsampling of decompressed data after lossy compression using a neural network, comprising the steps of:
training a deep learning algorithm configured to recover phase and amplitude information associated with a Synthetic Aperture Radar (SAR) image;
receiving a compressed bit stream, the compressed bit stream comprising complex-valued SAR image data;
decoding the compressed bit stream to predict both an In-phase (I) channel and a Quadrature (Q) channel; and
using both the I channel and the Q channel as inputs into the trained deep learning algorithm to recover both phase and amplitude information associated with the SAR image data.
9. The method of claim 8 , wherein the trained deep learning algorithm is a neural network that can recover signals from a compressed bitstream.
10. The method of claim 9 , wherein the trained deep learning algorithm further comprises a multi-channel transformer with attention.
11. The method of claim 8 , wherein training the deep learning algorithm comprises two stages, wherein each stage comprises a specific loss function.
12. The method of claim 11 , wherein the loss function is mean squared error.
13. The method of claim 11 , wherein one stage of the two stages is associated with a loss function for the I and Q channels.
14. The method of claim 11 , wherein one stage of the two stages is associated with the amplitude loss.