Image processing using multiprocessor discrete wavelet transform
원문보기
IPC분류정보
국가/구분
United States(US) Patent
등록
국제특허분류(IPC7판)
G06K-009/54
G06F-017/14
G06T-007/30
G06T-001/20
G06T-005/00
출원번호
US-0437260
(2012-04-02)
등록번호
US-9836433
(2017-12-05)
발명자
/ 주소
Hoke, Jaclyn A.
Jensen, David W.
출원인 / 주소
Rockwell Collins, Inc.
대리인 / 주소
Gerdzhikov, Angel N.
인용정보
피인용 횟수 :
0인용 특허 :
19
초록
The present invention relates to improved systems and methods of image processing and more particularly to improved systems and method of image processing using modified image data to produce enhanced data and images using fewer processing cycles and lower system power.
대표청구항▼
1. An image processing system, the system comprising: a decomposition processor that utilizes a discrete wavelet transform to decompose data from one or more images into one or more sets of wavelet coefficients, the one or more sets of wavelet coefficients comprising a low frequency data set of wave
1. An image processing system, the system comprising: a decomposition processor that utilizes a discrete wavelet transform to decompose data from one or more images into one or more sets of wavelet coefficients, the one or more sets of wavelet coefficients comprising a low frequency data set of wavelet coefficients that retains essential image information of the one or more images and one or more high frequency data sets of wavelet coefficients that represent gradient between adjacent pixels of the one or more images;a memory that stores the one or more sets of wavelet coefficients; andat least one processor configured to: apply two or more image enhancement processing techniques to provide processing enhancements to at least one of the one or more images, the two or more image enhancement processing techniques including one frequency-based image enhancement processing technique configured to transform the one or more sets of wavelet coefficients into a set of modified wavelet coefficients representing one or more enhanced images; andtransform the set of modified wavelet coefficients into one or more enhanced images,wherein the two or more image enhancement processing techniques include a parallax-correcting image registration technique in conjunction with two or more images, the parallax-correcting image registration technique performed by applying one or more registration algorithms to low frequency coefficients of the two or more images to provide a coarse registration and using high frequency coefficients of the two or more images to refine the coarse registration, and leveraging high frequency coefficients of the two or more images as weights to guide and select appropriate coefficients to construct an output image, wherein larger coefficients of the two or more images are chosen to ensure more detailed image data is captured in the constructed output image. 2. The system of claim 1 where the decomposition processor is encompassed in the at least one processor. 3. The system of claim 1 where the at least one processor comprises a single general purpose processor, a multicore processor, or a hardware implementation. 4. The system of claim 1 where the at least one processor is a single processor that uses two or more temporal phases to perform the transforms on the one or more sets of wavelet coefficients. 5. The system of claim 1 where the at least one processor includes multiple processors that simultaneously perform the transforms on the one or more sets of wavelet coefficients. 6. The system of claim 1 wherein the two or more image enhancement processing techniques are used to process image frames collected from sensors include visible, low light, and infrared sensors. 7. The system of claim 1, wherein the decomposition processor further decomposes the low frequency data set recursively until only a single pixel remains in the low frequency data set. 8. The system of claim 1, wherein the two or more image enhancement processing techniques further provide processing techniques selected from a group comprising noise reduction, distortion correction, feature recognition, contrast enhancement, multi-spectral fusion, multi-focus fusion (hands free focus), super-resolution, dynamic range compression, and deblurring algorithm. 9. An image enhancement method, the enhancement method comprising: decomposing, by at least one processor, data from one or more images using a discrete wavelet transform, the data from the one or more images being decomposed into one or more sets of wavelet coefficients, the one or more sets of wavelet coefficients comprising a low frequency data set of wavelet coefficients that retains essential image information of the one or more images and one or more high frequency data sets of wavelet coefficients that represent gradient between adjacent pixels of the one or more images;storing, by the at least one processor, the one or more sets of wavelet coefficients in a memory;applying, by the at least one processor, two or more image enhancement processing techniques to provide processing enhancements to at least one of the one or more images, the two or more image enhancement processing techniques including one frequency-based image enhancement processing technique configured to transform the one or more sets of wavelet coefficients into a set of modified wavelet coefficients representing one or more enhanced images; andtransforming, by the at least one processor, the set of modified wavelet coefficients into one or more enhanced images,wherein the two or more image enhancement processing techniques include a parallax-correcting image registration technique in conjunction with two or more images, the parallax-correcting image registration technique performed by applying one or more registration algorithms to low frequency coefficients of the two or more images to provide a coarse registration and using high frequency coefficients of the two or more images to refine the coarse registration, and leveraging high frequency coefficients of the two or more images as weights to guide and select appropriate coefficients to construct an output image, wherein larger coefficients of the two or more images are chosen to ensure more detailed image data is captured in the constructed output image. 10. The method of claim 9 where the memory is a random access memory. 11. The method of claim 9 where the two or more image enhancement processing techniques are performed on a single general purpose processor, a multicore processor, or a hardware implementation. 12. The method of claim 9 where the two or more image enhancement processing techniques are performed on a single processor or hardware unit and use two or more temporal phases to transform the one or more sets of wavelet coefficients. 13. The method of claim 9 where the two or more image enhancement processing techniques are performed on multiple processors or hardware units to simultaneously transform the one or more sets of wavelet coefficients. 14. The method of claim 9 where the two or more image enhancement processing techniques process image frames collected from sensors include visible, low light, and infrared sensors. 15. The method of claim 9, wherein the decomposing step further decomposes the low frequency data set recursively until only a single pixel remains in the low frequency data set. 16. The method of claim 9, wherein the two or more image enhancement processing techniques further provide processing techniques selected from a group comprising noise reduction, distortion correction, feature recognition, contrast enhancement, multi-spectral fusion, multi-focus fusion (hands free focus), super-resolution, dynamic range compression, and deblurring algorithm. 17. An image processing system, the system comprising: a processor; anda memory coupled to the processor,the processor configured to:decompose data from one or more images into one or more sets of wavelet coefficients using a discrete wavelet transform, the one or more sets of wavelet coefficients comprising a low frequency data set of wavelet coefficients that retains essential image information of the one or more images and one or more high frequency data sets of wavelet coefficients that represent gradient between adjacent pixels of the one or more images;store the one or more sets of wavelet coefficients in the memory;apply two or more image enhancement processing techniques to provide processing enhancements to at least one of the one or more images, the two or more image enhancement processing techniques including one frequency-based image enhancement processing technique configured to transform the one or more sets of wavelet coefficients into a set of modified wavelet coefficients representing one or more enhanced images; andtransform the set of modified wavelet coefficients into one or more enhanced images,wherein the two or more image enhancement processing techniques include a parallax-correcting image registration technique in conjunction with two or more images, the parallax-correcting image registration technique performed by applying one or more registration algorithms to low frequency coefficients of the two or more images to provide a coarse registration and using high frequency coefficients of the two or more images to refine the coarse registration, and leveraging high frequency coefficients of the two or more images as weights to guide and select appropriate coefficients to construct an output image, wherein larger coefficients of the two or more images are chosen to ensure more detailed image data is captured in the constructed output image. 18. The system of claim 17 where the processor comprises a single general purpose processor, a multicore processor, or a hardware implementation and the processor uses two or more temporal phases to perform the transforms on the one or more sets of wavelet coefficients to process image frames collected from sensors including visible, low light, and infrared sensors.
연구과제 타임라인
LOADING...
LOADING...
LOADING...
LOADING...
LOADING...
이 특허에 인용된 특허 (19)
Master, Paul L.; Watson, John, Configurable hardware based digital imaging apparatus.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.