Image analysis – Image compression or coding
Reexamination Certificate
2000-02-03
2001-07-03
Couso, Jose L. (Department: 2621)
Image analysis
Image compression or coding
Reexamination Certificate
active
06256417
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to image coding apparatuses, image decoding apparatuses, image coding methods, image decoding methods, image coding program recording media, and image decoding program recording media. More particularly, the invention relates to coding of a color signal representing a color shape of an object and an image signal comprising transparency information and the like, which represents the ratio of each image when plural images are synthesized, for sufficient transmission and storage of these signals.
BACKGROUND OF THE INVENTION
Generally, image data produced by digitizing an image, especially a moving picture, has a great amount of data, so information compression is performed by coding when the data is transmitted or recorded. Such information compression is carried out on the basis of the partial or temporal correlation of the image data.
As an example of compression coding, there is “prediction coding” in which prediction is performed on the basis of an image close in time to an image being the target of coding (target image), a difference between the target image and the prediction image is obtained as differential data, and the differential data is coded. In the prediction, motion compensation is carried out using a motion vector obtained by motion detection of the image. Generally, the higher the correlation is, the higher the coding efficiency of the differential data is.
On the other hand, in order to transmit or store a moving picture with high efficiency, there is proposed a method in which moving picture data is divided into plural layers corresponding to individual objects included in the picture, and each layer is coded. For example, in the case of coding an image composed of a person and a background, initially, an image coding apparatus divides the image into two layers for the person and the background, respectively, encodes the respective layers, and transmits the coded data of the respective layers. On the other hand, in an image decoding apparatus, the transmitted coded data of the respective layers are decoded, and decoded images of the respective layers are synthesized using a prescribed method and displayed.
Further, in the above-mentioned synthesis process, information showing, pixel by pixel, whether the background object is hidden or not by the overlapping of the images, is required. This information is called “a significant signal”, and pixels hiding the background are called “significant pixels”. A large value of the significant signal means that the ratio of synthesis is large and the signal is important visually. To the contrary, a small value of the significant signal means that it is hardly used for synthesis, i.e., it is almost transparent.
As described above, a significant signal shows the shape of an image synthesized with a background, and only significant pixels influence on the quality of the synthesized image. In other words, insignificant pixels have no relation with the image quality, so that the efficiency in coding can be improved by coding only significant pixels.
Meanwhile, a method for orthogonally transforming only significant pixels, called “Shape Adaptive Orthogonal Transform (SADCT)”, is disclosed in IEEE Transactions on Circuits and Systems for Video Technology vol.5, No.1, February 1995. In this method, only significant pixels are subjected to orthogonal transform.
On the other hand, as a method for generating a prediction image, “Overlap Motion Compensation” is disclosed in ITU-T Recommendation H.263, pp. 42-44, October 1995. In this method, an optimum prediction area is calculated by interpolating plural small areas within a prediction image.
However, since compression by coding is based on spatial or temporal correlation in an image, if insignificant pixels are coded as the target of coding together with significant pixels, the coding efficiency is reduced. For example, when all the significant pixels are black, if the insignificant pixels are black as well, the coding efficiency is not reduced. However, if the insignificant pixels are white, since the correlation is reduced, the coding efficiency is reduced. Generally, when insignificant pixels are included in an area surrounded by significant pixels, the inter-pixel correlation is reduced, whereby the coding efficiency is reduced.
Further, also in the above-mentioned SADCT, when, in a target image, insignificant pixels are included in an area surrounded by significant pixels, the correlations in the vertical and horizontal directions are reduced, whereby the coding efficiency is reduced.
As described above, although the pixel values of insignificant pixels do not have much influence on the quality of reproduced image, these pixel values adversely affect the coding efficiency. Therefore, when coding is performed, the pixel values of insignificant pixels must be considered. In the prior art methods, however, this matter has not bee considered at all.
SUMMARY OF THE INVENTION
An object of the present invention is to provide an image coding apparatus that can avoid a reduction in coding efficiency due to pixel values of insignificant pixels when an image signal is coded, thereby to improve the coding efficiency.
Another object of the present invention is to provide an image coding apparatus in which, when coding is performed receiving an image signal and a significant signal that shows whether pixels in the image signal are significant or not, the significant signal is effectively utilized to improve the coding efficiency.
A further object of the present invention is to provide an image decoding apparatus that can perform decoding adaptively to signals which have been effectively coded by the above-mentioned image coding apparatus.
Another object of the present invention is to provide an image coding method that can avoid a reduction in coding efficiency due to influence of pixel values of insignificant pixels, thereby to improve the coding efficiency.
Still another object of the present invention is to provide an image decoding method that can perform decoding adaptively to signals which have been effectively coded by the above-mentioned image coding method.
Yet another object of the present invention is to provide a recording medium in which an image coding program is recorded, which program can realize the above-mentioned image coding apparatus in a general apparatus, such as a personal computer or a work station.
A still further object of the present invention is to provide a recording medium in which an image decoding program is recorded, which program can realize the above-mentioned image decoding apparatus in a general apparatus, such as a personal computer or a work station.
Other objects and advantages of the invention will become apparent from the detailed description that follows. The detailed description and specific embodiments described are provided only for illustration since various additions and modifications within the scope of the invention will be apparent to those of skill in the art from the detailed description.
According to a 1st aspect of the present invention, there is provided an image coding apparatus for coding an input signal comprising an image signal being an array of pixels having pixel values and, a significant signal showing whether the image signal is significant or not, and the apparatus comprises: prediction image generating means for specifying, as a target area, an area of the input image signal to be the target of coding, on the basis of the input significant signal, and generating a prediction image for the target area, using a locally reproduced image which is described later; subtraction means for dividing the prediction image into a plurality of small prediction areas, dividing the target area into a plurality of small target areas, subtracting each small prediction area from corresponding small target area, and outputting a differential signal; area dividing means for dividing a rectangle area including significant pixels from the differential signal output from the subtraction means
Boon Choong Seng
Takahashi Jun
Couso Jose L.
Matsushita Electric - Industrial Co., Ltd.
Parkhurst & Wendel L.L.P.
LandOfFree
Image coding apparatus, image decoding apparatus, image... 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 coding apparatus, image decoding apparatus, image..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image coding apparatus, image decoding apparatus, image... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2559794