Image processing device, method and recording medium for compressing image data

• US 7,477,791 B2
• Filed: 11/27/2006
• Issued: 01/13/2009
• Est. Priority Date: 05/23/2002
• Status: Expired due to Fees

• Current Assignee: Fuji Xerox Company Limited (Fujifilm Holdings Corporation)
• Original Assignee: Fuji Xerox Company Limited (Fujifilm Holdings Corporation)
• Inventors: Yokose, Taro
• Primary Examiner(s): Alavi; Amir
• Application Number:
  US11/604,200
• Publication Number:
  US 20070065031A1
• Time in Patent Office: 778 Days
• Field of Search: 382/173, 382/176, 382/232, 382/233, 382/234, 382/238, 382/239, 382/243, 382/250, 382/251, 341/51, 375/240.03, 375/240.04, 375/240.1, 375/240.2, 375/240.24
• US Class Current: 382/238
• CPC Class Codes: G06T 9/004 Predictors, e.g. intraframe, interframe coding
• CPC Class Codes: H04N 19/12 Selection from among a plurality of transforms or standards, e.g. selection between discrete cosine transform [DCT] and sub-band transform or selection between H.263 and H.264
• CPC Class Codes: H04N 19/132 Sampling, masking or truncation of coding units, e.g. adaptive resampling, frame skipping, frame interpolation or high-frequency transform coefficient masking
• CPC Class Codes: H04N 19/176 the region being a block, e.g. a macroblock
• CPC Class Codes: H04N 19/182 the unit being a pixel
• CPC Class Codes: H04N 19/60 using transform coding
• CPC Class Codes: H04N 19/85 using pre-processing or post-processing specially adapted for video compression

Abstract

An image processing device for compressing image data using conversion to spatial frequency components may include a dividing section that divides the image data into a plurality of pixel blocks and computes spatial frequency components of each pixel block; a segmenting section that computes an intensity of predetermined high frequency components from information of the spatial frequency components and segments the image data into a first plane formed by including a pixel block having an intensity which is less than a predetermined threshold value and into a second plane formed by including a pixel block having an intensity which is equal to or greater than the threshold value; a compression section that executes compression for the first plane image data by applying quantization and entropy coding using the spatial frequency components information; and a run length compression section that executes run length compression of the second plane image data.

Claims

1. An image processing device which compresses image data using prediction encoding, comprising:

a prediction unit that sequentially selects a noted pixel from image data subjected to compression, and determines whether or not the noted pixel can be predicted from information for other pixels;

a segmenting unit that segments the image data into a first plane including a set of pixels determined as being predictable, and a second plane including a set of pixels determined as being not predictable; and

a compression processor that carries out compression using prediction encoding to image data of the first plane.

2. An image processing device comprising:

a pixel value prediction unit that sequentially selects a noted pixel from image data subjected to compression, and determines whether or not the noted pixel can be predicted from other pixels;

an enlargement/reduction interpolation unit that converts resolution by applying enlargement or reduction with respect to the image data subjected to compression, and interpolates a value of each pixel of image data after the resolution has been converted;

an image segmenting unit that segments the image data subjected to compression into image data of a first plane and image data of a second plane based on the determined information of whether or not the noted pixel is predictable and the interpolated value of each pixel; and

a compression processor that carries out compression using prediction encoding to at least image data of the first plane.

3. An image processing device comprising:

a pixel value prediction unit that sequentially selects a noted pixel from image data subjected to compression, and executes a forward prediction to determine whether or not the noted pixel can be predicted from pixels which have already been selected as noted pixels and a backward prediction to determine whether or not the noted pixel can be predicted from pixels that have not yet been selected as noted pixels;

an enlargement/reduction interpolation unit that converts resolution by applying enlargement or reduction with respect to the image data subjected to compression, interpolates, based on an image size of the image data subjected to compression, a value of each pixel of image data after the resolution has been converted, and outputs the result as interpolated image data;

an image segmenting unit that sequentially selects a noted pixel from the image data subjected to compression, determines whether or not the noted pixel is an edge portion by comparing a pixel value of the noted pixel and a pixel value of a pixel of the interpolated image data corresponding to the noted pixel, determines whether each pixel of the image data subjected to compression belongs to a first plane or a second plane based on at least one of the edge determination result, a determination result of the forward prediction, and a determination result of the backward prediction, and segments, based on the determination result, the image data subjected to compression into image data of the first plane and image data of the second plane; and

a compression processor that carries out compression using prediction encoding to at least image data of the first plane.

4. The image processing device according to claim 3, wherein the image segmenting unit:

determines that the present noted pixel belongs to the first plane when a previous noted pixel belongs to the second plane, the present noted pixel is the edge portion, and the determination result of the forward prediction or the backward prediction with respect to the present noted pixel is predictable;

determines that the present noted pixel belongs to the second plane regardless of whether or not the present noted pixel is the edge portion when the previous noted pixel belongs to the second plane and the determination result of the forward prediction and the backward prediction with respect to the present noted pixel is not predictable;

determines that the present noted pixel belongs to the first plane regardless of whether or not the present noted pixel is the edge portion when the previous noted pixel belongs to the first plane and the determination result of the forward prediction with respect to the present noted pixel is predictable; and

determines that the present noted pixel belongs to the first plane when the previous noted pixel belongs to the first plane, the present noted pixel is the edge portion, and the determination result of the backward prediction with respect to the present noted pixel is predictable.

5. The image processing device according to claim 4, wherein

the image segmenting unit determines that a noted pixel is the edge portion when a difference between a pixel value of the noted pixel and a pixel value of a pixel of the interpolated image data corresponding to the noted pixel is greater than or equal to a predetermined threshold value.

6. The image processing device according to claim 3, wherein

the compression processor applies compression using JPEG encoding to the image data of the second plane.

7. The image processing device according to claim 3, wherein

the compression processor applies compression using JPEG encoding to the image data of the second plane after applying a reduction process to the image data of the second plane.

8. A computer readable recording medium storing a program causing a computer to execute a process for compressing image data using prediction encoding, the process comprising:

sequentially selecting a noted pixel from image data subjected to compression, and determining whether or not the noted pixel can be predicted from information for other pixels;

segmenting the image data into a first plane including a set of pixels determined as being predictable, and a second plane including a set of pixels determined as being not predictable; and

carrying out compression using prediction encoding to image data of the first plane.

9. A method for compressing image data using prediction encoding, the method comprising:

sequentially selecting a noted pixel from image data subjected to compression, and determining whether or not the noted pixel can be predicted from information of other pixels;

segmenting the image data into a first plane including a set of pixels determined as being predictable, and a second plane including a set of pixels determined as being not predictable; and

carrying out compression using prediction encoding to image data of the first plane.

10. A computer readable recording medium storing a program causing a computer to execute a process for compressing image data using prediction encoding, the process comprising:

sequentially selecting a noted pixel from image data subjected to compression, and determining whether or not the noted pixel can be predicted from other pixels;

converting resolution by applying enlargement or reduction with respect to the image data subjected to compression and interpolating a value of each pixel of image data after the resolution has been converted;

segmenting the image data subjected to compression into image data of a first plane and image data of a second plane based on the determined information of whether or not the noted pixel is predictable and the interpolated value of each pixel; and

carrying out compression using prediction encoding to at least image data of the first plane.

11. A method for compressing image data using prediction encoding, the method comprising:

sequentially selecting a noted pixel from image data subjected to compression, and determining whether or not the noted pixel can be predicted from other pixels;

converting resolution by applying enlargement or reduction with respect to the image data subjected to compression and interpolating a value of each pixel of image data after the resolution has been converted;

segmenting the image data subjected to compression into image data of a first plane and image data of a second plane based on the determined information of whether or not the noted pixel is predictable and the interpolated value of each pixel; and

carrying out compression using prediction encoding to at least image data of the first plane.

12. A computer readable recording medium storing a program causing a computer to execute a process for compressing image data using prediction encoding, the process comprising:

sequentially selecting a noted pixel from image data subjected to compression, and executing a forward prediction to determine whether or not the noted pixel can be predicted from pixels which have already been selected as noted pixels and a backward prediction to determine whether or not the noted pixel can be predicted from pixels which have not yet been selected as noted pixels;

converting resolution by applying enlargement or reduction with respect to the image data subjected to compression, interpolating, based on an image size of the image data subjected to compression, a value of each pixel of the image data after the resolution has been converted, and outputting the result as interpolated image data;

determining whether or not a noted pixel is an edge portion by sequentially selecting the noted pixel from the image data subjected to compression and comparing a pixel value of the noted pixel and a pixel value of a pixel of the interpolated image data corresponding to the noted pixel, determining whether each pixel of the image data subjected to compression belongs to a first plane or a second plane based on at least one of the edge determination result, a determination result of the forward prediction, and a determination result of the backward prediction, and segmenting the image data subjected to compression into image data of the first plane and image data of the second plane based on the determination result; and

carrying out compression using prediction encoding to at least the image data of the first plane.

13. A method for compressing image data using prediction encoding, the method comprising:

sequentially selecting a noted pixel from image data subjected to compression, and executing a forward prediction to determine whether or not the noted pixel can be predicted from pixels which have already been selected as noted pixels and a backward prediction to determine whether or not the noted pixel can be predicted from pixels which have not yet been selected as noted pixels;

converting resolution by applying enlargement or reduction with respect to the image data subjected to compression, interpolating, based on an image size of the image data subjected to compression, a value of each pixel of the image data after the resolution has been converted, and outputting the result as interpolated image data;

determining whether or not a noted pixel is an edge portion by sequentially selecting the noted pixel from the image data subjected to compression and comparing a pixel value of the noted pixel and a pixel value of a pixel of the interpolated image data corresponding to the noted pixel, determining whether each pixel of the image data subjected to compression belongs to a first plane or a second plane based on at least one of the edge determination result, a determination result of the forward prediction, and a determination result of the backward prediction, and segmenting the image data subjected to compression into image data of the first plane and image data of the second plane based on the determination result; and

carrying out compression using prediction encoding to at least the image data of the first plane.