Image analysis – Image compression or coding – Adaptive coding
Reexamination Certificate
2000-06-27
2003-12-02
Couso, Jose L. (Department: 2621)
Image analysis
Image compression or coding
Adaptive coding
Reexamination Certificate
active
06658157
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a method and apparatus for converting image information in such a manner that an orthogonal transform is performed on a block-by-block basis, and orthogonal transform coefficients obtained via the orthogonal transform for each block are then quantized thereby converting the bit rate of image information which has been coded into a compressed form, wherein each block is a predefined pixel block called an orthogonal transform block. For example, when image information coded into a compressed form according to the MPEG-2 standard or the like is transmitted via a network for satellite broadcasting, cable television, or the like or when such image information is recorded on a recording medium such as an optical disk or a magnetic disk, the present invention is advantageously employed to reduce the bit rate of the compressed image information.
2. Description of the Related Art
In recent years, it has become popular to treat image information as digital data and compress the digital data by removing redundancy of image information by means of an orthogonal transform and motion compensation. The compressed data is transmitted via a network medium such as satellite broadcasting or cable television, or recorded on a storage medium such as an optical disk or a magnetic disk. In apparatus for such a purpose, a discrete cosine transform according to the MPEG-2 (Moving Picture Experts Group Phase-2) standard is widely used to compress image information.
In recent years, works are being performed to establish standards for digital television broadcasting using the image information compression technique according to MPEG-2 or the like. There are standards for digital television using a normal resolution image (consisting of effective horizontal lines of, for example, 576) and for digital television using a high resolution image (consisting of effective horizontal lines of, for example, 1152).
Because the amount of information of a high-resolution image is huge, a large code size (bit rate) is required to obtain sufficiently high image quality even when the image is coded into a compressed form according to the MPEG-2 standard or the like. For example, when each frame includes 1920 pixels×1080 pixels and is scanned in an interlaced fashion at 30 Hz, a bit rate as high as 18-22 Mbps or higher is required.
Therefore, when such a high-resolution image is transmitted via a network medium for satellite broadcasting, cable television, or the like, it is required to reduce the code size (bit rate) depending on the bandwidth of the transmission line used. When a high-resolution image is recorded on a storage medium such as an optical disk or a magnetic disk, it is required to further reduce the code size (bit rate) depending on the capacity of the storage medium. The reduction in the code size (bit rate) is required not only for high-resolution images but, in some cases, also even for normal-resolution images (for example, an image with a frame consisting of 720 pixels×480 pixels and scanned in an interlaced fashion at 30 Hz).
FIG. 41
illustrates a conventional image information conversion apparatus, including an image decoder
110
and an image coder
120
, for reducing the code size (bit rate) of compressed image information (bit stream).
In this conventional image information conversion apparatus, compressed image information (bit stream) with a high bit rate is input to the image decoder
110
. The image decoder
110
decodes the compressed image information with the high bit rate into baseband video data. The resultant baseband video data is applied to the image coder
120
. The image coder
120
codes the baseband video data received from the image decoder
110
into compressed image information (bit stream) with a low bit rate. A target code size (target bit rate) smaller than the code size of the input compressed image information is set in the image coder
120
, and the image coder
120
performs quantization corresponding to the target code size.
By performing the above-described process, the conventional image information conversion apparatus reduces the code size of compressed image information.
However, in the conventional image information conversion apparatus, because the entire decoding and coding processes are performed, a large scale hardware system is required. As a result, such image information conversion apparatus is expensive and consumes high electric power. Therefore, it is difficult to install such apparatus in consumer devices or portable devices. Furthermore, in the conventional image information conversion apparatus, it is required to completely perform all processes associated with decoding and coding, and thus a very large amount of calculation is required. When processes are performed by software with a general-purpose integrated circuit, the processes cannot be completed in real time if the circuit does not have high enough performance.
Furthermore, in the conventional image information conversion apparatus
120
, in order to convert the input compressed image information with a high bit rate into information having a smaller target code size (target bit rate) than the code size of the input information, the input information is re-quantized using a large quantization step value. As a result, large quantization noise appears as block noise in the decoded image, and thus high image quality cannot be obtained.
Thus, an object of the present invention is to solve the problems described above. More specifically, the object is to provide a method and apparatus for converting image information in such a manner as to reduce the large code size (high bit rate) of the input compressed image information (bit stream) without causing significant degradation in image quality in the decoded image, and outputting the resultant compressed image information with the reduced code size (lowered bit rate).
SUMMARY OF THE INVENTION
According to an aspect of the present invention, there is provided an image information conversion apparatus for converting first compressed image information with a first bit rate to second compressed image information with a second bit rate lower than the first bit rate, the first compressed image information having been produced by coding an image signal into a compressed form by means of performing an orthogonal transform on a block-by-block basis and then quantizing orthogonal coefficients, obtained by the orthogonal transform, within the block, each block being a predetermined pixel block (orthogonal transform block), the image information conversion apparatus comprising: dequantization means for dequantizing the orthogonal transform coefficients, in accordance with the quantization step value of the orthogonal transform coefficients of the input first compressed image information; and quantization means for re-quantizing the orthogonal transform coefficients dequantized by the dequantization means, using a quantization step value which is set-so that the output second compressed image information has the second bit rate.
In this image information conversion apparatus, the orthogonal transform coefficients are dequantized and the resultant dequantized orthogonal transform coefficients are re-quantized using a different quantization step value.
The image information conversion apparatus may further comprise band limiting means for limiting high-frequency component values, in the horizontal direction, of the orthogonal transform coefficients dequantized by the dequantization means, wherein the quantization means re-quantizes the orthogonal transform coefficients including motion compensation error correction coefficients added by the addition means, using a quantization step value which is set so that the output second compressed image information has the second bit rate.
In this image information conversion apparatus, after limiting high-frequency component values of the dequantized orthogonal transform coefficients in the horizontal direction, the re-quant
Kubozono Takeshi
Liu Yu
Okada Shintaro
Satoh Kazushi
Yanagihara Naofumi
Couso Jose L.
Frommer William S.
Frommer & Lawrence & Haug LLP
Sony Corporation
LandOfFree
Method and apparatus for converting image information does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for converting image information, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for converting image information will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3179413