# US 10,455,250 B2 · Method and system for distributed coding of images > Agents: this data is available as JSON from the patentgrep API and MCP server — see https://patentgrep.com/llms.txt - Number: US 10,455,250 B2 - Document ID: US10455250B2 - Status: Granted - Kind: B2 - Application number: 15/608,000 - Filed: 2017-05-30 - Published: 2018-12-06 - Granted: 2019-10-22 - Earliest claimed: 2017-05-30 - Assignees: Mitsubishi Electric Research Laboratories, Inc. - Applicants: Mitsubishi Electric Research Laboratories, Inc. - Inventors: Petros Boufounos; Maxim Goukhshtein; Toshiaki Koike-Akino; Stark C Draper - Web page: https://patentgrep.com/patent/US10455250B2 ## CPC - H04N 19/149: by estimating the code amount by means of a model, e.g. mathematical model or statistical model - G06F 17/18: for evaluating statistical data {, e.g. average values, frequency distributions, probability functions, regression analysis (forecasting specially adapted for a specific administrative, business or logistic context G06Q10/04)} - H03M 13/25: Error detection or forward error correction by signal space coding, i.e. adding redundancy in the signal constellation, e.g. Trellis Coded Modulation [TCM] {(modulation codes H03M13/31)} - H04N 19/103: Selection of coding mode or of prediction mode - H04N 19/184: the unit being bits, e.g. of the compressed video stream - H04N 19/192: the adaptation method, adaptation tool or adaptation type being iterative or recursive - H04N 19/395: {involving distributed video coding [DVC], e.g. Wyner-Ziv video coding or Slepian-Wolf video coding} - H04N 19/44: Decoders specially adapted therefor, e.g. video decoders which are asymmetric with respect to the encoder - H04N 19/593: involving spatial prediction techniques - G06V 20/194: {using hyperspectral data, i.e. more or other wavelengths than RGB} - H04N: PICTORIAL COMMUNICATION, e.g. TELEVISION - G06F: ELECTRIC DIGITAL DATA PROCESSING (computer systems based on specific computational models G06N) - H03M: CODING; DECODING; CODE CONVERSION IN GENERAL (using fluidic means F15C4/00; optical analogue/digital converters G02F7/00; coding, decoding or code conversion, specially adapted for particular applications, see the relevant subclasses, e.g. G01D, G01R, G06F, G06T, G09G, G10L, G11B, G11C, H04B, H04L, H04M, H04N; ciphering or deciphering for cryptography or other purposes involving the need for secrecy G09C) - G06V: IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING ## Abstract A method for reconstructing an uncompressed signal. The method includes obtaining an encoded signal corresponding to the signal. Obtaining side information about the signal and using the side information to obtain a prediction of dithered linear measurements of the signal. Using the prediction of the dithered linear measurements and encoded quantized dithered linear measurements to obtain quantized linear measurements of the signal based on processing each bitplane iteratively, starting from a least significant level bitplane to a most significant level bitplane. At each iteration, a prediction of each bitplane is formed using the prediction of the dithered linear measurements and the bitplanes processed in the previous iterations. Wherein each code for each bitplane is used to correct each bitplane prediction. Reconstructing the signal as a reconstructed signal using the recovered quantized dithered linear measurements, wherein the steps are performed in a processor of a decoder. ## Claims ### Claim 1 A method for reconstructing a signal, wherein the signal is uncompressed, comprising steps of: obtaining an encoded signal corresponding to the signal, wherein the encoded signal includes encoded quantized dithered linear measurements of the signal, wherein the encoded quantized dithered linear measurements of the signal correspond to quantized dithered linear measurements of the signal, such that the quantized dithered linear measurements are separated into bitplanes, and each bitplane is encoded with a code at a predetermined rate that ensures a reliable correction of a prediction of the quantized dithered linear measurements of the signal; obtaining side information about the signal; using the side information to obtain a prediction of the dithered linear measurements of the signal; using the prediction of the dithered linear measurements of the signal and the encoded quantized dithered linear measurements to obtain the quantized linear measurements of the signal based on processing each bitplane iteratively, starting from a least significant level bitplane to a most significant level bitplane, such that at each iteration, a prediction of each bitplane is formed using the prediction of the dithered linear measurements of the signal and the bitplanes processed in the previous iterations, wherein each code for each bitplane is used to correct each bitplane prediction, so as to recover the quantized dithered linear measurements; and reconstructing the signal as a reconstructed signal using the recovered quantized dithered linear measurements, wherein the steps are performed in a processor of a decoder. ### Claim 2 (depends on 1) The method of claim 1, wherein the side information is encoded as encoded side information, and further comprising: decoding the encoded side information to obtain decoded side information. ### Claim 3 (depends on 1) The method of claim 1, wherein the prediction of the signal is obtained from the side information using linear prediction. ### Claim 4 (depends on 1) The method of claim 1, further comprising: updating the prediction of the signal and the side information using the reconstructed signal; and repeating the determining, the combining, and the reconstructing steps using the updated prediction and side information. ### Claim 5 (depends on 1) The method of claim 1, wherein the side information includes a signal similar to the encoded signal. ### Claim 6 (depends on 1) The method of claim 1, wherein the side information includes statistical similarity measures of the encoded signal with a signal similar to the encoded signal. ### Claim 7 (depends on 6) The method of claim 6, wherein the statistical similarity measures includes a mean, a covariance, and a correlation of the signal. ### Claim 8 (depends on 1) The method of claim 1, wherein the side information includes quantized measurements of the signal. ### Claim 9 (depends on 1) The method of claim 1, wherein the signal is acquired by a remote sensing system. ### Claim 10 (depends on 9) The method of claim 9, wherein the remote sensing system is mounted on a satellite system. ### Claim 11 (depends on 9) The method of claim 9, wherein the remote sensing system is a radar system. ### Claim 12 (depends on 9) The method of claim 9, wherein the remote sensing system is a Synthetic Aperture Radar (SAR) system. ### Claim 13 (depends on 1) The method of claim 1, wherein the code is a distributed source code. ### Claim 14 (depends on 13) The method of claim 13, wherein the distributed source code is a low-density parity check code (LDPC). ### Claim 15 (depends on 13) The method of claim 13, wherein the distributed source code is a polar code. ### Claim 16 (depends on 13) The method of claim 13, wherein the distributed source code is a fountain code. ### Claim 17 (depends on 1) The method of claim 1, wherein the signal includes a multispectal image or a hyperspectral image. ### Claim 18 (depends on 1) The method of claim 1, wherein the obtained side information includes statistical data for the prediction of spectral bands of the signal. ### Claim 19 (depends on 1) The method of claim 1 wherein the reconstructing solves an optimization problem. ### Claim 20 (depends on 19) The method of claim 19, wherein the optimization enforces a sparsity model. ### Claim 21 (depends on 19) The method of claim 19, wherein the optimization enforces low signal total variation. ### Claim 22 (depends on 19) The method of claim 19, wherein the optimization uses weights derived from the side information. ### Claim 23 (depends on 19) The method of claim 19 wherein the optimization uses weights derived from the signal prediction, the side information, or both. ### Claim 24 (depends on 19) The method of claim 19, wherein the optimization is performed using a greedy algorithm. ### Claim 25 (depends on 19) The method of claim 19, wherein the optimization is performed using a convex relaxation. ### Claim 26 (depends on 19) The method of claim 19, wherein the optimization is using a dictionary. ### Claim 27 (depends on 26) The method of claim 26, wherein the dictionary is learned from the prediction. ### Claim 28 (depends on 26) The method of claim 26, wherein the dictionary is learned from a reconstructed signal. ### Claim 29 (depends on 26) The method of claim 26, wherein the decoder accesses the side information via a database of similar images obtained at different times, such that the database is stored in a memory in communication with the processor of the decoder. ### Claim 30 (depends on 1) The method of claim 1, wherein the predetermined rate is determined as a function of the predictive quallity of the side information. ### Claim 31 (depends on 30) The method of claim 30, wherein the predictive quality of the side information is measured using the prediction error. ### Claim 32 A non-transitory computer readable storage medium embodied thereon a program executable by a computer for performing a method, the method is for reconstructing a signal, wherein the signal is uncompressed, comprising steps of: obtaining an encoded signal corresponding to the signal, wherein the encoded signal includes encoded quantized dithered linear measurements of the signal, wherein the encoded quantized dithered linear measurements of the signal correspond to quantized dithered linear measurements of the signal, such that the quantized dithered linear measurements are separated into bitplanes, and each bitplane is encoded with a code at a predetermined rate that ensures a reliable correction of a prediction of the quantized dithered linear measurements of the signal; obtaining side information about the signal; using the side information to obtain a prediction of the dithered linear measurements of the signal; using the prediction of the dithered linear measurements of the signal and the encoded quantized dithered linear measurements to obtain the quantized linear measurements of the signal based on processing each bitplane iteratively, starting from a least significant level bitplane to a most significant level bitplane, such that at each iteration, a prediction of each bitplane is formed using the prediction of the dithered linear measurements of the signal and the bitplanes processed in the previous iterations, wherein each code for each bitplane is used to correct each bitplane prediction, so as to recover the quantized dithered linear measurements; and reconstructing the signal as a reconstructed signal using the recovered quantized dithered linear measurements, wherein the steps are performed in a decoder. ### Claim 33 A decoder for reconstructing a signal, wherein the signal is uncompressed, comprising: a receiver to receive from an encoder an encoded signal corresponding to the signal, wherein the encoded signal includes encoded quantized dithered linear measurements of the signal, such that the encoded quantized dithered linear measurements of the signal correspond to quantized dithered linear measurements of the signal, wherein the quantized dithered linear measurements are separated into bitplanes, and each bitplane is encoded with a code at a predetermined rate that ensures a reliable correction of a prediction of the quantized dithered linear measurements of the signal; a processor is configured to obtain side information about the signal, use the side information to obtain a prediction of the dithered linear measurements of the signal, and uses the prediction of the dithered linear measurements of the signal and the encoded quantized dithered linear measurements to obtain the quantized linear measurements of the signal based on processing each bitplane iteratively, starting from a least significant level bitplane to a most significant level bitplane, such that at each iteration, a prediction of each bitplane is formed using the prediction of the dithered linear measurements of the signal and the bitplanes processed in the previous iterations, wherein each code for each bitplane is used to correct each bitplane prediction, so as to recover the quantized dithered linear measurements; and reconstruct the signal as a reconstructed signal using the recovered quantized dithered linear measurements, and stores the reconstructed signal in a memory in communication with the processor. ### Claim 34 An encoder for encoding a signal, comprising: a receiver to receive the signal; a processor in communication with the receiver is configured to: obtain quantized dithered linear measurements of the signal; separate the quantized dithered linear measurements into bitplanes; encode each bitplane with a code at a predetermined rate that ensures a reliable correction of a prediction of the bitplane of the quantized dithered linear measurements of the signal; and a transmitter to transmit to a decoder the encoded signal corresponding to the signal; a receiver of the decoder to receive from the encoder the encoded signal corresponding to the signal, wherein the encoded signal includes encoded quantized dithered linear measurements of the signal, such that the encoded quantized dithered linear measurements of the signal correspond to quantized dithered linear measurements of the signal, wherein the quantized dithered linear measurements are separated into bitplanes, and each bitplane is encoded with a code at a predetermined rate that ensures a reliable correction of a prediction of the quantized dithered linear measurements of the signal; a processor of the decoder in communication with the receiver of the decoder obtains side information about the signal, and using the side information obtains a prediction of the dithered linear measurements of the signal, the decoder processor is configured to: obtain the quantized linear measurements of the signal, using the prediction of the dithered linear measurements of the signal and the encoded quantized dithered linear measurements, based on processing each bitplane iteratively, starting from a least significant level bitplane to a most significant level bitplane, such that at each iteration, a prediction of each bitplane is formed using the prediction of the dithered linear measurements of the signal and the bitplanes processed in the previous iterations, wherein each code for each bitplane is used to correct each bitplane prediction, so as to recover the quantized dithered linear measurements; and reconstruct the signal as a reconstructed signal using the recovered quantized dithered linear measurements.