Back to results
US 2001/0043752 A1Published application
Image processing system for compressing image data including binary image data and continuous tone image data by a sub-band transform method with a high-compression rate
AssigneeRicoh Company, Ltd.
InventorsNEKKA MATSUURA, YUKIKO YAMAZAKI, TAKAHIRO YAGISHITA
Life of this patent
Estimating the term…
Abstract
An image processing system which compresses an image including both a binary image and a continuous tone image by a sub-band transform method with a high compression rate. A 2×2 pixel matrix block is extracted from image data. A transform factor having a plurality of frequency components is obtained from the 2×2 pixel matrix block data. The transform factor is quantized by a fixed-length quantizing method by deleting a predetermined number of lower order bits of each of the frequency components.
Claims20
An image processing system comprising:
- a buffer unit extracting n×m pixel matrix block data from image data, where n and m are integers;
- a sub-band transform unit transforming the n×m pixel matrix block data by a sub-band transform method so as to obtain a transform factor having a plurality of frequency components;
- and
- a quantizing unit quantizing the transform factor by a fixed-length quantizing method by deleting a predetermined number of lower order bits of each of the frequency components.
1 more element
An image processing system comprising:
- a buffer unit extracting n×m pixel matrix block data from image data, where n and m are integers;
- a sub-band transform unit transforming the n×m pixel matrix block data by a sub-band transform method so as to obtain a transform factor having a plurality of frequency components including a low-frequency component and a high-frequency component;
- an area discriminating unit discriminating a type of an image area corresponding to the n×m pixel matrix block data being processed so that the image area is determined as one of an edge area and a non-edge area, a discrimination being made based on whether or not an absolute value of each of the components of the transform factor exceeds a threshold value;
- and
- a quantizing unit quantizing the transform factor by a fixed-length quantizing method by deleting a predetermined number of lower order bits of each of the frequency components, the number of deleted lower order bits of each of the frequency components of the transform factor being changed in accordance with a type of image area being processed so that image data including the transform factor and flag information indicating a type of image area has a predetermined fixed length.
2 more elements
An image processing method comprising the steps of:
- extracting n×m pixel matrix block data from image data, where n and m are integers;
- transforming the n×m pixel matrix block data by a sub-band transform method so as to obtain a transform factor having a plurality of frequency components;
- and
- quantizing the transform factor by a fixed-length quantizing method by deleting a predetermined number of lower order bits of each of the frequency components.
1 more element
An image processing method comprising the steps of:
- extracting n×m pixel matrix block data from image data, where n and m are integers;
- transforming the n×m pixel matrix block data by a sub-band transform method so as to obtain a transform factor having a plurality of frequency components including a low-frequency component and a high-frequency component;
- discriminating a type of an image area corresponding to the n×m pixel matrix block data being processed so that the image area is determined as one of an edge area and a non-edge area, a discrimination being made based on whether or not an absolute value of each of the components of the transform factor exceeds a threshold value;
- and
- quantizing the transform factor by a fixed-length quantizing method by deleting a predetermined number of lower order bits of each of the frequency components, the number of deleted lower order bits of each of the frequency components of the transform factor being changed in accordance with a type of image area being processed so that image data including the transform factor and flag information indicating a type of image area has a predetermined fixed length.
2 more elements
An image processing system comprising:
- a buffer unit extracting n×m pixel matrix block data from image data, where n and m are integers;
- a binarizing unit transforming the n×m pixel matrix block data into binary data represented by a maximum value and a minimum value;
- a differential data calculating unit calculating differential data which is a difference between a value of each pixel in the n×m pixel matrix block data and one of the maximum value and the minimum value of the binary data;
- a sub-band transform unit transforming the differential data by a sub-band transform method so as to obtain a transform factor having a plurality of frequency components;
- and
- an encoding unit encoding the binary data and the sub-band transform factor so as to obtain a code representing the image data.
3 more elements
An image processing method comprising the steps of:
- extracting n×m pixel matrix block data from image data, where n and m are integers;
- transforming the n×m pixel matrix block data into binary data represented by a maximum value and a minimum value;
- calculating differential data which is a difference between a value of each pixel in the n×m pixel matrix block data and one of the maximum value and the minimum value of the binary data;
- transforming the differential data by a sub-band transform method so as to obtain a transform factor having a plurality of frequency components;
- and
- encoding the binary data and the sub-band transform factor so as to obtain a code representing the image data.
3 more elements
An image processing system comprising:
- a dividing unit dividing image data into a plurality of n×m pixel matrix block data, where n and m are integers;
- a transform unit transforming each pixel in the n×m pixel matrix block data by a frequency transform method so as to produce a transform factor including a high-frequency component and a low-frequency component;
- an image area discriminating unit for determining whether the block being processed corresponds to an edge area or a non-edge area based on the transform factor output from said transform unit;
- a quantizing unit quantizing the transform factor for the edge area and the transform factor for the non-edge area by different methods;
- and
- an encoding unit encoding an output of said quantizing unit by an entropy encoding method,
- wherein a total of a number of bits of the high-frequency component and a number of bits of the low-frequency is the same regardless of types of the edge area or the non-edge area, and a number of bits of the high-frequency component for the edge area is the same as a number of bits of the low-frequency component of the non-edge area.
4 more elements
An image processing method comprising the steps of:
- dividing image data into a plurality of n×m pixel matrix block data, where n and m are integers;
- transforming each pixel in the n×m pixel matrix block data by a frequency transform method so as to produce a transform factor including a high-frequency component and a low-frequency component;
- determining whether the block being processed corresponds to an edge area or a non-edge area based on the transform factor output from said transform unit;
- quantizing the transform factor for the edge area and the transform factor for the non-edge area by different methods;
- and
- encoding an output of said quantizing unit by an entropy encoding method,
- wherein a total of a number of bits of the high-frequency component and a number of bits of the low-frequency is the same regardless of types of the edge area or the non-edge area, and a number of bits of the high-frequency component for the edge area is the same as a number of bits of the low-frequency component of the non-edge area.
4 more elements
Dependencies are inferred from the claim text.