Image analysis – Image compression or coding
Reexamination Certificate
1998-11-20
2001-04-17
Grant, II, Jerome (Department: 2624)
Image analysis
Image compression or coding
C382S166000, C382S312000, C382S319000, C382S172000, C382S172000
Reexamination Certificate
active
06219454
ABSTRACT:
BACKGROUND OF THE INVENTION
The present invention relates to an image processing apparatus which composes an image while receiving first compressed data generated by compressing a first piece of image data, second compressed data generated by compressing a second piece of image data that is different from the first piece of image data, and third compressed data generated by compressing selection data to be used for making a selection between the first piece of image data and the second piece of image data. Further, the present invention also relates to color facsimile machines, or to transmission/reception of images of the Internet.
In recent years, the transmission rate of image data has increased rapidly because of development of networks such as the Internet and public lines and advancement of the image compression technology. With the increase in transmission rate, transmittable image data have been improved in resolution and increased in the number of gradation levels, and their colorization has advanced; even a full-color image of 400 dpi (dots per inch) is now a practical transmission object. Incidentally, in transmitting such a full-color image via a network or a public line, image compression is usually performed. In such a case, the JPEG algorithm is generally used as a compression algorithm. Further, image storage such as image filing is now popular, and data that have been compressed, as in the above case, by using the JPEG algorithm or the like are used for storage of color images.
Now, the JPEG algorithm will be described, starting from the compression algorithm. If necessary, input image data is subjected to color space conversion from the RGB color space to the L*a*b* space, for example. Thereafter, the image data is divided into blocks of 8×8 pixels, converted into the frequency domain by DCT (discrete cosine transform) on a block-by-block basis, and then quantized by using a quantization table having the same size as each block. Then, the block data are converted by a zigzagged scan into raster data, from which compressed data is generated by two-dimensional Huffman coding. The expansion algorithm is an algorithm reverse to the compression algorithm. Image data is restored by Huffman decoding, raster-block conversion, dequantization, and IDCT (inverse DCT).
However, since quantization, which is a non-reversible process, is performed in the JPEG algorithm as seen from the above description, an expanded image is deteriorated in image quality from image data before the compression. Usually, to increase the compression ratio, a quantization table in which the quantization step is made larger as the frequency increases is used. Therefore, in particular, the image quality deteriorates remarkably and the compression ratio decreases at portions (characters, line figures, etc.) having edges that include many high-frequency components [of large amplitudes].
SUMMARY OF THE INVENTION
The present invention has been made in view of the above circumstances, and an object of the invention is therefore to provide an image processing apparatus which enables transmission and storage of an image while maintaining high image quality, enables high-speed transmission, and can compose and reproduce an image correctly.
In the invention, compressed data generated by separating image data into a first piece of image data, a second piece of image data, and selection data to be used for making a selection between the first piece of image data and the second piece of image data and coding the respective image data are input and an image is composed and restored. The three kinds of input compressed data can be ones that have been subjected respectively to most suitable image structure conversion processes such as resolution conversion processes, color conversion processes, and compression processes in accordance with their characteristics. By composing image data by receiving compressed data that have been subjected to processes suitable for respective image data in the above manner, the amount of data that are handled at the time of transmission or storage can be reduced and high image quality can be maintained.
Where image data that have been separated in the above manner are transmitted or stored, the characteristics of a compression algorithm do not allow compressed data to be transmitted or stored on a pixel-by-pixel basis and it is necessary to transmit or store compressed data sequentially on a line-by-line basis, a stripe-by-stripe basis (each stripe consists of a plurality of lines), or a page-by-page basis. When compressed data that have been generated through compression on a predetermined unit basis are input, a possibility that pixels of the respective separated image data are obtained at the same time by simply expanding the respective compressed data at the time of composing and restoration is low and hence the original image cannot be restored by this method.
In the invention, input compressed data are temporarily stored in the storing means, or expanded data as generated by subjecting the input compressed data to expansion processes or expanded data generated by subjecting the input compressed data to the expansion processes and image conversion processes are temporarily stored in the storing means, whereby the timing of the respective image data are timed with each other at the time of composing. Alternatively, at the time of composing, previously expanded selection data may be stored in the storing means and then later expanded first and second expanded data may be combined with each other. In this case, the first and second expanded data may be combined with each other in such a manner that one of the first and second expanded data is stored in the storing means first and then the first and second expanded data are combined with each other while being timed with each other by selectively writing the other expanded data the storing means by using the selection data that is the third expanded data.
Since in the above manner the composing process can be executed while three kinds of image data are timed with each other even if they are not timed with each other at the time of input, an image can be composed correctly.
REFERENCES:
patent: 5452102 (1995-09-01), Yokoyama
patent: 5463700 (1995-10-01), Nakazawa
patent: 5999644 (1999-12-01), Sugiura
patent: 6078695 (2000-06-01), Kadono
Anabuki Tetsushi
Hibi Yoshiharu
Kawano Hiroyuki
Matsunoshita Junichi
Fuji 'Xerox Co., Ltd.
Grant II Jerome
Oliff & Berridg,e PLC
Worku Negussie
LandOfFree
Image processing apparatus does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Image processing apparatus, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image processing apparatus will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2483250