Interferometric SAR processor for elevation

• US 5,659,318 A
• Filed: 05/31/1996
• Issued: 08/19/1997
• Est. Priority Date: 05/31/1996
• Status: Expired due to Fees

• Current Assignee: California Institute of Technology
• Original Assignee: California Institute of Technology
• Inventors: Martin, Jan M., Rosen, Paul A., Hensley, Scott, Madsen, Soren N., Kim, Yunjin, Imel, David A.
• Primary Examiner(s): Sotomayor, John B.
• Application Number:
  US08/657,602
• Time in Patent Office: 445 Days
• Field of Search: 342/25, 342/190, 342/191, 342/192, 342/193, 342/194, 342/196, 342/156
• US Class Current: 342/25.00C
• CPC Class Codes: G01S 13/9023 combined with interferometric techniques

Abstract

An interferometric Synthetic Aperture Radar (SAR) system having a special-purpose processor to integrate motion compensation, interferogram coregistration, and a spectral shifting for optimal interferogram correlation together to achieve efficient, accurate, and robust three-dimensional imaging. A simple radar mapping coordinate system is implemented in the present invention to enhance the overall image processing and in particular to improve the accuracy and efficiency. The present invention substantially improves the fidelity and efficiency of phase unwrapping by incorporating phase-bootstrapping process, a correlation filter, and other processes. Absolute phase determination is implemented without known ground references by two proposed techniques employing cross-correlation of sub-patches to generate a number of estimates within a patch and weighting correlation processes.

Claims

1. An interferometric method for generating a three dimensional image of a target, comprising:

transmitting coherent electromagnetic probing pulses to said target by using at least one coherent electromagnetic pulse generator, allowing spatial overlap of two illuminated areas upon said target by two sequentially transmitted pulses;

providing a moving carriage and two signal receivers attached to said moving carriage, said receivers being spatially displaced relative to each other and operating to detect electromagnetic signals;

detecting scattered electromagnetic pulses from said target illuminated by said probing waves by using said receivers on said moving carriage;

transforming analog signals of said detected scattered pulses from said receivers to digital signals, forming a first data channel and a second data channel corresponding to said two receivers, respectively;

employing at least one positioning device to obtain attitude, position and velocity information of said moving carriage and thereby attitude, position and velocity information of said receivers;

sequentially dividing received data into a series of patches at a predetermined resolution with any two adjacent patches having partial overlap with each other;

performing a first correlation process on each of said two data channels in a range direction from said carriage to said target based on a first information indicative of a property of said pulses generated by said generator;

aligning said two data channels relative to each other and in particular to a pre-determined spacing in a reference coordinate system within which said data channels are processed;

including Doppler effect in squinted observations in said aligning process using said position and velocity information from said positioning device;

performing a motion compensation process to project both channels onto a local reference motion track in said reference coordinate system, resulting a range shift and thereby a phase shift for both said channels;

carrying out a resampling procedure to interpolate said data at said shifted range value and apply said phase shift to each data point in said first and said second channels;

applying a range migration procedure to both said first and said second channels to adjust a dependence of range on positioning of said carriage with respect to said target by applying a second correlation process to each said data channels in an azimuth direction parallel to said motion track based on a second information indicative of properties of said pulses from said receivers and said attitude information of said receivers;

performing a multiplication of said first data channel with a complex conjugate of said second data channel and thereby producing an interferogram of said two channels;

performing a phase unwrapping process to determine relative phase values of every data point in a patch;

carrying out one of an absolute phase determination process to an unwrapped patch to directly relate said phase of said patch to a range difference from said receivers to said target or a phase-bootstrapping procedure in which phase information of an overlapped region of two adjacent data patches are used for correlating phase values of said two adjacent patches; and

producing a three-dimensional image of said target by further processing said interferogram using said position and velocity information from said positioning devices.

2. A method as in claim 1, wherein said range migration procedure includes:

performing a transformation on said first data channel and said second data channel to generated a first transformed channel and a second transformed channel in frequency domain, respectively;

aligning said first transformed channel and a second transformed channel relative to each other in said frequency domain;

obtaining a plurality of range reference functions from said first information with each of said range reference functions corresponding to a range value, said property of said pulses including pulsing frequency and thereby spatial spacing of illuminated areas on said target by said pulses for a predetermined velocity of said carriage;

performing multiplication of complex conjugate of said transformed range reference functions and each of said first transformed channel and said second transformed channel to obtain first product functions and second product functions, respectively; and

performing a reverse transformation corresponding to said transformation on each of said first and second product functions to obtain cross correlation between said range reference functions and each of said first and second channels, respectively.

3. A method as in claim 2, wherein said transformation is a fast Fourier transformation.

4. A method as in claim 1, wherein said second correlation process includes:

obtaining a plurality of azimuth reference functions from said second information with each corresponding to a range value, said property of said pulses being indicative of pulsing frequency and thereby spatial spacing of illuminated areas on said target by said pulses for a predetermined velocity of said carriage, and relative positioning of said receivers with respect to said target;

performing a Fourier transformation on said first data channel, said second data channel, and said azimuth reference functions to generate a first transformed channel, a second transformed channel, and transformed azimuth reference functions, respectively;

aligning said first transformed channel and said second transformed channel relative to each other;

performing multiplication of complex conjugate of said transformed azimuth reference functions and each of said first and second transformed channels to obtain first product functions and second product functions, respectively; and

performing a reverse Fourier transformation on each of said first product functions and said second product functions to obtain a range-Doppler image for each of said channels.

5. A method as in claim 1, further comprising:

forming an interferogram power array by using power values of each pixel in said interferogram subsequent to said producing of said interferogram; and

performing an averaging process to smooth said power array and said interferogram for a predetermined mapping resolution.

6. A method as in claim 5, further comprising a correlation filtering process including:

generating a correlation filter of said interferogram by using amplitude information of each pixel in said interferogram and a mean of power values of said two data channels, said correlation filter operating to indicate a degree of correlation in said interferogram;

applying said correlation filter to determine correlation values of each pixel in said interferogram; and

identifying regions in said interferogram that have a correlation value less than a predetermined threshold value and removing said regions from said phase unwrapping and thus said phase-bootstrapping in determining said relative phase values of each data point.

7. A method as in claim 6, wherein:

said predetermined threshold value is selected from a plurality of selected values; and

said applying of said correlation filter and said removing said regions in said interferogram that have a correlation value less than said predetermined threshold value are repeated for each of said selected values.

8. A method as in claim 1, wherein said absolute phase determination process is a residual delay process comprising:

obtaining a first image from said first data channel and a second image from said second data channel subsequent to said second correlation process in said azimuth direction;

applying a range-shifting process and a phase-correcting process to said first image and said second image and thereby resampling said first image and said second image;

aligning and registering said first image and said second image with respect to each other, said aligning producing a residual phase value; and

determining said residual phase value.

9. A method as in claim 1, further comprising:

filtering said first data channel and said second data channel prior to said performing of said motion compensation process to divide each data channel in to a low frequency channel indicated by a predetermined low carrier frequency and a high frequency channel indicated by a predetermined high carrier frequency, thus resulting in four filtered channels;

a first low frequency band channel and a first high frequency band channel from said first data channel and a second low frequency band channel and a second high frequency band channel from said second data channel;

subsequently performing said motion compensation process, said resampling procedure, said range migration procedure, said second correlation process to said four filtered channels;

forming a low frequency band interferogram by using said first low frequency band channel and said second low frequency band channel;

forming a high frequency band interferogram by using said first high frequency band channel and said second high frequency band channel;

forming a full resolution interferogram with said four filtered channels;

producing a differential interferogram by using said low frequency band interferogram and said high frequency band interferogram, said differential interferogram having a differential phase value that is proportional to the absolute phase value of said full resolution interferogram;

applying said phase unwrapping process to said full resolution interferogram to obtain an unwrapped phase value thereof; and

performing an averaging operation on the difference between said differential phase value and said unwrapped phase value and thereby determining said absolute phase value.

10. An interferometric synthetic aperture radar system for three-dimensional imaging of a target area, comprising:

a carriage moving along a motion track in said target area;

an electromagnetic pulse generator disposed on said carriage, sequentially producing and transmitting coherent electromagnetic pulsed signals to said target area, said generator and carriage operating to allow spatial overlap of two illuminated areas of two sequentially transmitted pulses;

two electromagnetic signal receivers attached to said carriage and spatially displaced from each other, detecting scattered pulsed signals of said coherent transmitted pulsed signals from said target area;

an analog-to-digital converter capable of transforming analog signals into digital form;

a data storing device for retaining digital signals;

a positioning device, providing attitude, velocity and position information of said carriage and thereby said signal receivers;

an electronic processor, operating to process said signals in digital form for a three dimensional electronic image of said target area based on a coherence property of said signals;

a first signal processing mechanism in said processor, enabling said processor to encode said attitude, position and velocity information of said receivers to said signals, forming a first data channel and a second data channel corresponding to said two receivers, respectively;

a second signal processing mechanism in said processor, enabling said processor to obtain spatial frequency information of said signals in both an azimuth direction along said motion track and in a range direction from said carriage to said target area, to sequentially divide received data into a series of patches at a predetermined resolution with any two adjacent patches having partial overlap with each other, and to perform a first correlation process on each of said two data channels in a range direction from said carriage to said target based on a first information indicative of a property of said pulses detected by said receivers;

a third signal processing mechanism in said processor, enabling said processor to align said first data channel and said second data channel with respect to a common reference, to perform a motion compensation process to project both channels onto said reference, resulting a range shift and a phase shift for both said channels;

said third signal processing mechanism operating to enable said processor to carry out a resampling procedure to interpolate said data at said shifted range value and apply said phase shift to signal values in said first and said second channels and to apply a range migration procedure to both said first and said second channels to adjust a dependence of range on positioning of said carriage with respect to said target;

a fourth signal processing mechanism in said processor, enabling said processor to apply a second correlation process to each said data channels in said azimuth direction relative to said reference based on a second information indicative of a property of said pulses, said velocity information of said carriage, and relative positioning of said receivers with respect to said target;

said fourth signal processing mechanism further enabling said processor to produce an interferogram by correlating said first data channel with said second data channel;

a phase unwrapping mechanism in said processor, enabling said processor to perform a phase-bootstrapping procedure in which phase information of a overlapped region of two adjacent data patches are used for correlating phase values of said two adjacent patches, and to determine relative phase values of every data point in a patch that is bootstrapped by said phase-bootstrapping process;

a phase resolving mechanism in said processor, enabling said processor to cross correlate a first subset of data of said interferogram with a second sub set of data thereof and thereby to determine a absolute phase value of said interferogram directly relating to a difference between a first range from said target area to said first receiver and a second range from said target area to said second receiver; and

an imaging mechanism in said processor, producing said three dimensional image by having said processor to correlate each pixel in said interferogram with said position information from said positioning device.

11. A system as in claim 10, wherein said first correlation process includes:

performing a transformation on said first data channel and said second data channel to generated a first transformed channel and a second transformed channel in frequency domain, respectively;

aligning said first transformed channel and a second transformed channel relative to each other in said frequency domain;

obtaining a plurality of range reference functions from said first information with each of said range reference functions corresponding to a range value, said property of said pulses including pulsing frequency and thereby spatial spacing of illuminated areas on said target by said pulses for a predetermined velocity of said carriage;

performing multiplication of complex conjugate of said transformed range reference functions and each of said first transformed channel and said second transformed channel to obtain first product functions and second product functions, respectively; and

performing a reverse transformation corresponding to said transformation on each of said first and second product functions to obtain cross correlation between said range reference functions and each of said first and second channels, respectively.

12. A system as in claim 10, wherein said second correlation process includes:

obtaining a plurality of azimuth reference functions from said second information with each corresponding to a range value, said property of said pulses being indicative of pulsing frequency and thereby spatial spacing of illuminated areas on said target by said pulses for a predetermined velocity of said carriage, and relative positioning of said receivers with respect to said target;

performing a Fourier transformation on said first data channel, said second data channel, and said azimuth reference functions to generate a first transformed channel, a second transformed channel, and transformed azimuth reference functions, respectively;

aligning said first transformed channel and said second transformed channel relative to each other;

performing multiplication of complex conjugate of said transformed azimuth reference functions and each of said first and second transformed channels to obtain first product functions and second product functions, respectively; and

performing a reverse Fourier transformation on each of said first product functions and said second product functions to obtain a range-Doppler image for each of said channels.

13. A system as in claim 10, wherein said fourth signal processing can further enable said processor to:

form an interferogram power array by using power values of each pixel in said interferogram subsequent to said producing of said interferogram; and

perform an averaging process to smooth said power array and said interferogram for a predetermined mapping resolution.

14. A system as in claim 13, further comprising a correlation filtering mechanism in said processor, operating to make said processor capable of:

generating a correlation filter of said interferogram by using amplitude information of each pixel in said interferogram and a mean of power values of said two data channels, said correlation filter operating to indicate a degree of correlation in said interferogram;

applying said correlation filter to determine correlation values of each pixel in said interferogram; and

identifying regions in said interferogram that have a correlation value that is less than a predetermined threshold value and removing said regions from said phase unwrapping and thus said phase-bootstrapping in determining said relative phase values of each data point.