IPC분류정보
국가/구분 |
United States(US) Patent
등록
|
국제특허분류(IPC7판) |
|
출원번호 |
US-0002043
(2007-12-14)
|
등록번호 |
US-8170095
(2012-05-01)
|
발명자
/ 주소 |
|
출원인 / 주소 |
|
대리인 / 주소 |
Schwabe, Williamson & Wyatt, P.C.
|
인용정보 |
피인용 횟수 :
4 인용 특허 :
74 |
초록
▼
Methods, medium, and machines associated with image processing performance are disclosed. Image data may be copied from input memory to main memory before performing CPU intensive operations, and efficiently copied from memory thereafter. CPU intensive operations may include decryption, decompressio
Methods, medium, and machines associated with image processing performance are disclosed. Image data may be copied from input memory to main memory before performing CPU intensive operations, and efficiently copied from memory thereafter. CPU intensive operations may include decryption, decompression, image enhancement, or reformatting. Real time compression may be achieved by sub-sampling each frame. A separate plane compression aspect may provide for distinguishing between regions of an image, separating and/or masking the original image into multiple image planes, and compressing each separated image plane with a compression method that is optimal for its characteristics.
대표청구항
▼
1. A method of increasing image processing performance, comprising: copying image data of a selected subset of an entire image existing in an I/O RAM into a copy of the selected subset in a buffer in a main memory by multiple calls to a memory copy function copying each image line of the selected su
1. A method of increasing image processing performance, comprising: copying image data of a selected subset of an entire image existing in an I/O RAM into a copy of the selected subset in a buffer in a main memory by multiple calls to a memory copy function copying each image line of the selected subset to a buffer line of the buffer;performing CPU intensive operations on the copy selected subset in the main memory and not on the image data in said I/O RAM to generate a processed image and storing the processed image in the buffer; andcopying the processed image from the buffer in the main memory to the I/O RAM by multiple calls to the memory copy function copying each buffer line of the buffer to an image line to generate a copy of the processed image in the I/O RAM;wherein performing CPU intensive operations comprises bit-wise sub-sampling pixels of the copy selected subset to extract corresponding subsets of pixel bits from the pixels;counting consecutive repeated instances of various pixel values represented by the extracted subsets of pixel bits; andoutputting a series of data elements for the pixels to form the processed image, based on the extracted subsets of pixel bits;wherein a data element comprises a data field and a run-length field, wherein the data field comprises a pixel value in a range spanning from zero to a maximum value represent-able by a number of pixel bits sub-sampled from a pixel, and wherein the run-length field comprises a consecutive repeat count of the pixel value in the data field. 2. The method of claim 1, wherein the entire image comprises a television video image including a visible portion visible on a display device and a non-visible portion not visible on the display device, and the selected subset of the entire image substantially comprises only the visible portion. 3. The method of claim 1, wherein said main memory is cached in a CPU cache or an external cache. 4. The method of claim 1, wherein copying of a selected subset of the entire image from the I/O RAM or copying of the processed image into the I/O RAM comprises copying using DMA circuitry. 5. The method of claim 1, wherein said I/O RAM is a display video RAM associated with a video output device configured to display the copy of the processed image. 6. The method of claim 1, wherein a pixel comprises a plurality of color components, a color component comprises a number of color component bits, and the plurality of pixel bits of a pixel comprises the color component bits of the color components, and wherein extracting a subset of pixel bits from a pixel comprises extracting respective most significant color component bits of the color components of the pixel. 7. The method of claim 6, wherein performing CPU intensive operations further comprises encoding the various pixel values, wherein a data field in a data element comprises a pixel value in an encoded form. 8. The method of claim 6, wherein performing CPU intensive operations further comprises variably selecting one of 3, 4, 5, 8, 9, 12, 15, 16, and 24, as the number of pixel bits to sub-sample from a pixel, wherein the number of pixel bits sub-sampled from a pixel is less than a number of pixel bits of the pixel being bit-wise sub-sampled. 9. The method of claim 6, wherein performing CPU intensive operations further comprises receiving a specification of the number of pixel bits to sub-sample from a pixel. 10. The method of claim 9, wherein performing CPU intensive operations further comprises: setting a size of the data field to the specification of the number of pixel bits to sub-sample; andsubtracting the size of the data field from a supported data size to derive a size of the run-length field. 11. A machine for image processing comprising: an I/O device including an I/O RAM for storing an entire image;a buffer of a main memory for storing-a copy of a selected subset of the entire image;means for copying the selected subset of the entire image from the I/O RAM to the main memory to generate the copy of the selected subset by multiple calls to a memory copy function to copy each image line of the selected subset to a buffer line of the buffer;means for performing computing intensive processing on-the copy of the selected subset to generate a processed image and to store the processed image in the buffer; andmeans for copying the processed image from said main memory to the I/O RAM to generate a copy of the processed image by multiple calls to the memory copy function to copy each buffer line of the buffer to an image line in the I/O RAM;wherein means for performing CPU intensive operations comprises means for bit-wise sub-sampling pixels of the copy selected subset to extract corresponding subsets of pixel bits from the pixels;means for counting consecutive repeated instances of various pixel values represented by the extracted subsets of pixel bits; andmeans for outputting a series of data elements for the pixels to form the processed image, based on the extracted subsets of pixel bits;wherein a data element comprises a data field and a run-length field, wherein the data field comprises a pixel value in a range spanning from zero to a maximum value represent-able by a number of pixel bits sub-sampled from a pixel, and wherein the run-length field comprises a consecutive repeat count of the pixel value in the data field. 12. The machine of claim 11, wherein the entire image comprises a television video image including a visible portion visible on a display device and a non-visible portion not visible on the display device and the selected subset of the entire image substantially comprises only the visible portion. 13. The machine of claim 11, wherein said I/O RAM is a capture video RAM. 14. The machine of claim 11, wherein said I/O device further comprises means for outputting the copy of the processed image. 15. The machine of claim 11, wherein the main memory comprises cache memory to store the copy of selected subset and the processed image. 16. The machine of claim 11, wherein a pixel comprises a plurality of color components, a color component comprises a number of color component bits, and the plurality of pixel bits of a pixel comprises the color component bits of the color components, and wherein extracting a subset of pixel bits from a pixel comprises extracting respective most significant color component bits of the color components of the pixel. 17. The machine of claim 16, wherein means for performing CPU intensive operations further comprises means for encoding the various pixel values, wherein a data field in a data element comprises a pixel value in an encoded form. 18. The machine of claim 16, wherein means for performing CPU intensive operations further comprises means for variably selecting one of 3, 4, 5, 8, 9, 12, 15, 16, and 24, as the number of pixel bits to sub-sample from a pixel, wherein the number of pixel bits sub-sampled from a pixel is less than a number of pixel bits of the pixel being bit-wise sub-sampled. 19. The machine of claim 16, wherein means for performing CPU intensive operations further comprises means for receiving a specification of the number of pixel bits to sub-sample from a pixel. 20. The machine of claim 19, wherein means for performing CPU intensive operations further comprises: means for setting a size of the data field to the specification of the number of pixel bits to sub-sample; andmeans for subtracting the size of the data field from a supported data size to derive a size of the run-length field. 21. A machine for image processing comprising: an image input device configured to capture an entire image and including a capture video RAM to store the entire image;a buffer in a main memory to store a copy of a selected subset of the entire image, and image planes separated from the selected subset of the entire image;means for copying the selected subset from the capture video RAM to the buffer to generate a copy of the selected subset by multiple calls to a memory copy function to copy each image line of the selected subset to a buffer line of the buffer;means for separating the copy of the selected subset in the buffer into a plurality of image planes;means for compressing the image planes using different compression methods to generate a plurality of compressed image planes for storage in the main memory; andmeans for copying the compressed image planes from the main memory to the capture video RAM by multiple calls to the memory copy function copying each buffer line of the buffer to generate an image line of a copy of a compressed image plane in the capture video RAM. 22. The machine of claim 21, wherein the entire image comprises a television video image including a visible portion visible on a display device and a non-visible portion not visible on the display device, and the selected subset of the entire image substantially comprises only the visible portion. 23. The machine of claim 21, wherein the main memory comprises cache memory to store the entire image, the uncompressed image planes and the compressed image planes. 24. The machine of claim 21, wherein said means for separating comprises means for separating a first set of pixels of the copy of the selected subset having a distinguishing characteristic into a first image plane, and means for separating a second set of pixels of the copy of the selected subset not having the distinguishing characteristic into a second image plane. 25. The machine of claim 24, wherein the distinguishing characteristic comprises an effective color depth, a rate of pixel value change between adjacent pixels, a marker or an indicator indicating selection by a user, associated with the first set of pixels. 26. The machine of claim 21, wherein said means for separating comprises means for separating grayscale pixel values of the copy of the selected subset into one or more grayscale image planes, and means for separating Doppler enhanced pixel values of the copy of the selected subset into one or more Doppler enhanced image planes. 27. The machine of claim 21, wherein said means for compressing comprises first means for compressing a first image plane using a first compression method, and second means for compressing a second image plane using a second compression method, wherein the first and second compression methods are different compression methods. 28. An image processing method comprising: copying a selected subset of an entire image from a capture video RAM to a buffer of a main memory to generate a copy of the selected subset by multiple calls to a memory copy function to copy each image line of the selected subset to a buffer line of the buffer;separating the copy of the selected subset in the buffer into a plurality of image planes;compressing the image planes using different compression methods to generate a plurality of compressed image planes for storage in the main memory; andcopying the compressed image planes from the main memory by multiple calls to the memory copy function copying each buffer line of the buffer to generate an image line of a copy of a compressed image plane in the capture video RAM. 29. The method of claim 28, wherein the entire image comprises a television video image including a visible portion visible on a display device and a non-visible portion not visible on the display device, and the selected subset of the entire image substantially comprises only the visible portion. 30. The method of claim 28, wherein separating comprises separating a first set of pixels of the copy of the selected subset having a distinguishing characteristic into a first image plane, and separating a second set of pixels of the copy of the selected subset not having the distinguishing characteristic into a second image plane. 31. The method of claim 30, wherein the distinguishing characteristic comprises an effective color depth, a rate of pixel value change between adjacent pixels, a marker or an indicator indicating selection by a user, associated with the first set of pixels. 32. The method of claim 28, wherein separating comprises separating grayscale pixel values of the copy of the selected subset into one or more grayscale image planes, and separating Doppler enhanced pixel values of the copy of the selected subset into one or more Doppler enhanced image planes. 33. The method of claim 28, wherein compressing comprises compressing a first image plane using a first compression method, and compressing a second image plane using a second compression method, wherein the first and second compression methods are different compression methods.
※ AI-Helper는 부적절한 답변을 할 수 있습니다.