Picture coder and recording medium

Description: TECHNICAL FIELD [Detailed description of the invention]

【0001】

[Industrial application field]

With regard to high-efficiency coding that efficiently encodes signals with a smaller amount of code in recording, transmission, and display devices that involve processing digital signals, images and images, especially when searching from storage media at high speed. The present invention relates to an image coding apparatus and a recording medium that separately encodes priority image information that is preferentially reproduced at the time of reproduction, such as a low-resolution image when the information is layered and encoded.

【0002】

[Conventional technology]

When high-efficiency coding is applied to image information, there is image information (hereinafter, also referred to as priority image information) that is effective to be reproduced preferentially. In this case, as for the image information, only the priority image information is decoded and reproduced according to the purpose. Information other than the priority image information (hereinafter, also referred to as non-priority image information) is information that is meaningless to decode and reproduce only the information. To give a specific example, in the layer coding of the current television signal (hereinafter, television is abbreviated as TV) and HDTV signal such as the NTSC system, the current TV signal and the HDTV signal are used together with the information of the current TV signal. The difference information is also encoded, and the HDTV signal is reproduced by decoding and adding both of the above information, and the current TV signal is reproduced when only the information of the current TV signal is decoded.

【0003】

In this case, the information of the current TV signal is the priority image information, and the difference information is the non-priority image information. Further, in the reproduction of the storage media, a high-speed search while looking at the image is required. Therefore, when the inter-image prediction is performed at the time of coding the image information, one frame is set every few frames. Independently encoded within. When a high-speed search is performed during reproduction, only the independently encoded image information is reproduced. In this case, the independently encoded frame information is the priority image information, and the frame information predicted between images is the non-priority image information.

【0004】

As an example of such hierarchical coding, a case where the current TV signal and the HDTV signal are coded at different resolutions will be described with reference to FIG. FIG. 3 is a block diagram showing an example of a conventional image coding apparatus. In FIG. 3, the image signal input from the image input terminal 1 has a high resolution and is given to the subtracted input of the low-pass filter (hereinafter referred to as LPF) 2 and the subtractor 3. This input image signal is band-limited by LPF2 to about 1/2 in both the vertical and horizontal directions according to the subsample in the subsampler 11, and is given to the subsampler 11. In the subsampler 11, pixels are thinned out by 1/2 in both the vertical and horizontal directions to form a low-resolution image, which is given to the encoder 12 and the interpolator 4.

【0005】

Discrete cosine transform (hereinafter referred to as DCT) is performed in the encoder 12, and the converted coefficient is quantized by the quantization step width given by the code amount controller 14, variable-length coded, and compressed. It becomes image data and is given to the buffer 13. The amount of code of the image data, which varies due to variable length coding, is homogenized in the buffer 13 and given to the multiplexer 8. Information on the sufficiency of the buffer is given to the code amount controller 14 from the buffer 13. In the code amount controller 14, information on the quantization step width is given to the encoder 12 based on the information on the sufficiency of the buffer, and the coder 12 controls so that the generated code amount becomes constant.

【0006】

Specifically, when a large amount of data is accumulated in the buffer, it is judged that the amount of data generated is large and the quantization step is coarsened. Conversely, when the data is close to empty, the quantization step is fine. Will be done. On the other hand, in the interpolator 4, the pixels that have not been thinned out are interpolated to become an interpolated signal having the same number of pixels as the input image signal, which is given to the subtraction input of the subtractor 3. In the subtractor 3, the interpolated signal is subtracted from the input image signal, a signal containing only high frequency components is created, and the signal is given to the encoder 5.

【0007】

The operation of the encoder 5, the buffer 6, and the code amount controller 7 is basically the same as that of the encoder 12, the buffer 13, and the code amount controller 14 that handle the subsampled signal, and the sample to be processed. Only the number is different. The image data encoded by the encoder 5 is given to the multiplexer 8 via the buffer 6. In the multiplexer 8, each image data is multiplexed and output from the data output terminal 9. When the image data obtained from the image coding apparatus as described above is transmitted, it is possible to extract and reproduce only the low-resolution image data from the image data.

【0008】

[Problems to be Solved by the Invention]

However, conventional image coding devices, recording media, and playback devices include low-resolution images in hierarchical coding in which image information is layered and coded, and coding using image-to-image prediction (hereinafter, image-to-image). Since it does not particularly support playback of only priority image information such as images that are independently encoded in (also referred to as predictive coding) (hereinafter, also referred to as independently encoded images), priority is given to high-speed search during playback. Even if only the image information was to be reproduced, it was necessary to read unnecessary information at the same time. In particular, when the code amount of the priority image information is not fixed, the code amount of the priority image information may be equal to the total code amount, for example, when the image is in a stationary state. In this case, even if only the priority image information is read, all the information is read, and high-speed reading of the priority image information becomes impossible. On the other hand, when the ratio of the code amount of the priority image information is fixed, the ratio is not always appropriate in terms of image quality depending on the image, and the image quality may deteriorate. The present invention has been made by paying attention to the above points, and an object of the present invention is to provide an image coding device and a recording medium for enabling high-speed and easy reading of a specific type of image information. It is to be.

【0009】

[Means for solving problems]

The image coding device of the present invention is an image coding device that encodes image information, and is a means for encoding a specific type of image information, that is, priority image information so that an image can be reproduced only from the information at the time of reproduction. And the means for encoding the non-priority image information other than the priority image information, and the generated code amount is controlled so that the total code amount of the priority image information and the non-priority image information falls within a predetermined value. It is an image coding apparatus including means and means for controlling the generated code amount so that the code amount of priority image information falls within a predetermined ratio with respect to the total code amount. The recording medium of the present invention is a recording medium on which encoded image information is recorded, and the priority image information is recorded on a specific track, and the specific track is periodically present every several tracks. It is a recording medium. Further, in the image coding apparatus or recording medium, an image (independently encoded image) or image information that is independently encoded in coding using inter-image prediction (inter-image prediction coding). The priority image information is a low-resolution image in the hierarchical coding in which the image is layered and encoded, or an image having a large quantization value.

【0010】

[Action]

In the present invention, the image coding apparatus encodes the specific type of image information, that is, the priority image information, so that the image can be reproduced only from the information at the time of reproduction so that the code amount is a certain ratio or less with respect to the total code amount. Therefore, in the recording medium, all the priority image information can be recorded on a specific track provided every few tracks. If the code amount of the priority image information is not more than a certain ratio of the code amount of the whole, there is a degree of freedom, so that the optimum image quality can be obtained in that range.

【0011】

[Example]

Hereinafter, a first embodiment relating to the image coding apparatus of the present invention will be described with reference to FIG. FIG. 1 is a block diagram showing a first embodiment of the image coding apparatus of the present invention. In FIG. 1, the same components as those in the conventional example shown in FIG. 3 are designated by the same reference numerals, and the description thereof will be omitted. The main difference between the image coding device shown in FIG. 1 and the image coding device shown in FIG. 3 is that the code amount setting device 15 and the code amount limiter 16 are provided in FIG. The image encoding device shown in FIG. 1 obtains a low-resolution image (for example, a current TV signal such as the NTSC system) from an input high-definition image (for example, an HDTV signal) and encodes it hierarchically. This is to enable playback of images alone.

【0012】

The image coding device shown in FIG. 1 is an image coding device that encodes image information, and includes a encoder 12 that encodes the priority image information and non-priority image information other than the priority image information. The encoder 5 for encoding the above, the first code amount control means 10 for controlling the generated code amount so that the total code amount of the priority image information and the non-priority image information falls within a predetermined value, and the priority A code amount limiter 16 or the like that controls the generated code amount so that the code amount of the image information falls within a predetermined ratio with respect to the total code amount is provided. The first code amount control means 10 is composed of buffers 13 and 6, code amount controllers 14 and 7, code amount setting device 15 and the like, and the code amount limiter 16 and the like provides the second code amount control means. It is configured.

【0013】

In FIG. 1, the image input from the image input terminal 1 has a high resolution and is given to the LPF 2 and the subtractor 3. This input image signal is band-limited by LPF2 to about 1/2 both vertically and horizontally according to the subsample in the subsampler 11, and is given to the subsampler 11. In the subsampler 11, pixels are thinned out to 1/2 both vertically and horizontally to form a low-resolution image, which is given to the encoder 12 and the interpolator 4. DCT is performed in the encoder 12, and the converted coefficient is quantized by the quantization step width given by the code amount controller 14, becomes variable-length coded and compressed image data, and is given to the buffer 13. .. The image data in which the fluctuation of the generated code amount is made uniform in the buffer 13 is given to the multiplexer 8. Further, the code amount generated for each block from the buffer 13 is given to the code amount setter 15 and the code amount controller 14.

【0014】

In the code amount controller 14, the quantization step width is set to the code amount controller 12 and the code amount setter 15 so that the generated code amount generated by the code amount controller 12 becomes the target code amount given by the code amount limiter 16. Is given to. On the other hand, in the interpolator 4, the pixels thinned out by the subsampler 11 are interpolated to become a signal having the same number of pixels as the input image signal, and are given to the subtraction input of the subtractor 3 as an interpolation signal. In the subtractor 3, the interpolated signal is subtracted from the input image signal, a signal containing only high frequency components is created, and the signal is given to the encoder 5.

【0015】

The operation of the encoder 5, the buffer 6, and the code amount controller 7 is basically the same as that of the encoder 12, the buffer 13, and the code amount controller 14 that handle the subsampled signal, and the sample to be processed. Only the number is different. The image data whose fluctuation is equalized in the buffer 6 is given to the multiplexer 8. In the multiplexer 8, the image data of the low resolution image and the image data of only the high frequency components are multiplexed in a format suitable for the recording format and output from the data output terminal 9.

【0016】

The present invention is characterized in the operation of the code amount setting device 15 and the code amount limiter 16 described below. In the code amount setting device 15 and the code amount limiter 16, as long as the code amount of the priority image information does not exceed the limit value, the quality of the reproduced image is prioritized and the code is encoded so as to be encoded with a predetermined quantization balance. The target code amounts TL and TH output from the quantity setter 15 are set, but when the code amount of the priority image information becomes large and limited, the quantization balance is disrupted and the target code amount is controlled. It is a thing. The quantization balance is that an image having a high frequency component has a coarser quantization step width than a low resolution image but is less likely to cause a visual problem. Therefore, the quantization step width for an image having a high frequency component is set to a quantum for a low resolution image. It refers to making it 1.5 times the width of the conversion step.

【0017】

The code amount setter 15 is given information on the quantization step widths for the low-resolution image and the image of the high frequency component from the code amount controllers 14 and 7, and the generated code amount is given from the buffers 13 and 6, respectively. Be done. Since the relationship between the quantization step width and the generated code amount is a general inversely proportional relationship, the quantization step width of the low-resolution image and the high-frequency component image is used by using the above information given to the code amount setter 15. The target code amounts TL and TH for which the values of are in an appropriate balance are set for each frame. The target code amounts TL and TH are given to the code amount limiter 16. In the code amount limiter 16, the TL and TH are limited and output as TL2 and TH2, respectively.

【0018】

Here, if the TL is less than or equal to the specified value (for example, 1/2 of the total code amount), TL2 and TH2 are left as the values of TL and TH, respectively, and if the TL exceeds the specified value, TL2 is specified. The value, TH2, is the value obtained by subtracting the specified value from the total code amount. The specified value depends on the recording form of the recording medium, and if the specific information track on which the priority image information is recorded is one in two, it is 1/2 of the total code amount, or one in three. If it is a ratio, it is 1/3 of the total code amount. In this case, the encoder 12 for a low-resolution image and the encoder 5 for a signal having a high frequency component are usually encoded with a predetermined quantization balance, but when the amount of coding of the low-resolution image is limited, The relationship breaks down.

【0019】

In this way, both usually maintain a good quantization balance in terms of image quality, and the ratio of the code amount changes, but since the upper limit is set for the code amount of the low resolution image, it does not exceed it. .. As a result, both image quality and high-speed reading of priority image information can be achieved. On the other hand, when trying to match the code amount generated by using feedback control such as code amount controllers 14 and 7 with the target code amount, the image quality is improved when the complexity of the image (so-called activity) changes significantly in the screen. It can be compromised. Therefore, it is conceivable to use feedforward processing to estimate the required code amount of each processing block in advance based on the activity of the image, allocate the target code amount accordingly, and then perform feedback control. Such control of the code amount by combining feed forward and feedback is specifically disclosed in Japanese Patent Application Laid-Open No. 2-194734 by the same inventor and the same applicant as the present invention.

【0020】

FIG. 2 is a block diagram showing a second embodiment of the image coding apparatus of the present invention. In FIG. 2, the same components as those in the first embodiment shown in FIG. 1 are designated by the same reference numerals, and the description thereof will be omitted. In the image coding apparatus shown in FIG. 2, predictive coding between images is performed, and independent coding is periodically performed at a ratio of 1 frame to several frames, and the independently coded image information is obtained. High-speed search is possible by recording on the storage medium and reproducing only the independently coded image at the time of reproduction.

【0021】

The image coding device shown in FIG. 2 is an image coding device that encodes image information, and includes a encoder 5 that encodes priority image information and non-priority image information in a time-divided manner, and the priority image information. The first code amount control means 20 that controls the generated code amount so that the total code amount of the non-priority image information falls within a predetermined value, and the code amount of the priority image information is relative to the total code amount. It is equipped with a code amount limiter 16 or the like that controls the generated code amount so as to be within a predetermined ratio. The first code amount control means 20 is composed of a buffer 6, a code amount controller 7, a code amount setter 15, and the like, and the code amount limiter 16 and the like constitutes a second code amount control means. ..

【0022】

In FIG. 2, the input image signal is given to the predictor subtractor 21 and the frame counter 23 through the image input terminal 1. In the prediction subtractor 21, the prediction signal supplied from the switch 22 is subtracted from the input image signal, and the prediction residual signal is given to the encoder 5. In the encoder 5, DCT is performed as in the case of FIG. 1, and the converted coefficient is quantized by the quantization step width given by the code amount controller 7, and the image is variable-length coded and compressed. It becomes data and is given to buffer 6 and decoder 26. In the buffer 6, the fluctuation of the generated code amount of the image data is made uniform, and the data is output from the data output terminal 9.

【0023】

In the decoder 26, the reverse processing of the encoder 5 is performed, and the encoded data is reproduced by the decoder 26 as a predicted residual signal and given to the adder 25. In the adder 25, the predicted residual signal and the predicted signal are added and provided as a reproduced image in the frame memory 24. The signal delayed by one frame in the frame memory 24 is given to the subtraction input of the predictive subtractor 21 via the switch 22. The switch 22 is opened once every 5 to 15 frames, for example, by the independent control signal output from the frame counter 23, in which case the prediction signal is not given to the prediction subtractor 21.

【0024】

Therefore, the input image signal is given to the encoder 5 as it is, and the image of the frame is encoded in the frame as an independent image. The frame counter 23 detects a synchronization signal of the input image signal, and counts it at a predetermined cycle to obtain the independent control signal. Next, control of the code amount will be described. In FIG. 2, the generated code amount for each block to be processed is given from the buffer 6 to the code amount setter 15 and the code amount controller 7, and the information of the quantization step width is encoded from the code amount controller 7. It is given to the device 5 and the code amount setting device 15.

【0025】

In the code amount setting device 15, the target code amount is set from the given information as in the case of FIG. 1, but in the case of this embodiment, the priority image information is an independent image, and the processing is time-divided. Therefore, the above information is obtained for one cycle between the independent image and the next independent image, for example, 5 to 15 frames, and the target code amount TI of the independent image and the target code of the predicted image for the next one cycle. The amount TP and is set. The set target code amounts TI and TP are given to the code amount limiter 16, and the code amount limiter 16 limits the target code amount TI of the independent image as in the case of FIG. The switch 27 is switched according to the output of the frame counter 23 according to the type of the image to be encoded, that is, the independent image or the predicted image, and a limited target code amount, that is, TI2 or TP2, is given to the code amount controller 7. ..

【0026】

Hereinafter, a third embodiment relating to the image coding apparatus of the present invention will be described with reference to FIG. FIG. 6 is a block diagram showing a third embodiment of the image coding apparatus of the present invention. In the image coding apparatus shown in FIG. 6, a coarsely quantized image and a finely quantized high-quality image are hierarchically coded according to the fineness of quantization. As a specific example, the image data of the low transfer rate of coarse quantization and the error component data in which the quantization error of the image is encoded are hierarchically coded, and the image is also displayed only from the image data of the low transfer rate at the time of reproduction. Is obtained, or both are decoded and added to obtain a high-quality image.

【0027】

Such hierarchical processing can be used by obtaining a high-quality image from both data and using it as a material image, and transmitting only a small amount of coded image data when the transfer rate is restricted due to broadcasting or the like. In FIG. 6, the same components as those in the conventional example shown in FIG. 3 are designated by the same reference numerals, and the description thereof will be omitted. The input image signal input from the image input terminal 1 is subjected to the discrete cosine transform (DCT) process by DCT64, and the resulting coefficient is given to the quantizer 65 and the subtractor 3. In the quantization device 65, the coefficient is quantized with a large quantization step width given by the code amount controller 14, and a fixed-length code is given to the variable-length coder 66 and the inverse quantizer 61.

【0028】

In the variable length encoder 66, the fixed length code is variable length coded to obtain low transfer rate image data, which is given to the buffer 13. On the other hand, in the inverse quantizer 61, the fixed-length code is replaced with the corresponding quantization representative value to become the regeneration coefficient, which is given to the subtraction input of the subtractor 3. In the subtractor 3, the reproduction coefficient is subtracted from the output of the DCT 64, an error component signal is generated, and the error component signal is given to the quantizer 62. In the quantizer 62, the error component signal is quantized by the quantization step width given by the code amount controller 7, but the quantization step width is finer than that of the quantizer 65.

【0029】

The output of the quantizer 62 is given to the variable length encoder 63, and the fixed length code is variable length coded to become error component data, which is given to the buffer 6. The error component data output from the buffer 6 and the low transfer rate image data output from the buffer 13 are multiplexed by the multiplexer 8 and output from the data output terminal 9. Since this embodiment is layered according to the fineness of quantization for the same signal, it is not necessary to provide a means for balancing the quantization between layers.

【0030】

Next, a state in which a code string (Data stream) encoded by the image coding apparatus of the present invention is recorded on a recording medium will be described. In VTRs and the like, a large number of information tracks on which information is recorded are formed on a recording medium. In this information track, one track for every few tracks is a specific track on which priority image information such as the low-resolution image and the independent image is recorded. The situation is shown in Fig. 4. FIG. 4 is a diagram showing an example of a recording form on a recording medium. In the example shown in FIG. 4, the specific track and the non-specific track are arranged alternately.

【0031】

FIG. 5 is a diagram showing an example of a data form. As shown in FIG. 5, when the code amount of the priority image information matches the maximum allowable value, the specific track is filled with the priority image information, and the image information other than the priority image information, that is, the non-priority image information It is recorded on a non-specific track other than the specific track. However, when the code amount of the priority image information is less than the maximum value, the non-priority image information is recorded not only on the non-specific track but also on a part of the specific track. Since recording on the recording medium is performed in the above-described form, the multiplexer 8 in FIG. 1 creates code information in a format suitable for the recording form of the information track.

【0032】

[Effect of the invention]

In the present invention, in the image coding apparatus, a specific type of image information is used as priority image information and the code amount is coded so as to be a certain ratio or less with respect to the total code amount. , The priority image information is recorded at a rate of one track for every few tracks, and all of the priority image information is stored in a specific information track formed every few tracks. If the code amount of the priority image information is a certain ratio or less with respect to the total code amount, there is a degree of freedom, and the optimum image quality can be obtained within that range. Further, since the number of information tracks in which priority image information is recorded is small, the information required for high-speed reproduction can be read without waste, and a high-quality high-speed search image can be obtained. Further, a low-resolution image can be easily obtained from a recording medium on which a high-definition image is recorded, and the recording medium and the playback device can be used for both the high-definition image and the low-resolution image. You can.

[Simple explanation of drawings]

[Figure 1]

It is a block diagram which shows 1st Example concerning the image coding apparatus of this invention.

[Figure 2]

It is a block diagram which shows the 2nd Example which concerns on the image coding apparatus of this invention.

[Fig. 3]

It is a block diagram which shows an example of the conventional image coding apparatus.

[Fig. 4]

It is a figure which shows an example of the recording form in a recording medium.

[Fig. 5]

It is a figure which shows the example of a data form.

[Fig. 6]

It is a block diagram which shows the 3rd Example which concerns on the image coding apparatus of this invention.

[Explanation of symbols]

1 ... Image input 2 ... LPF 3 ... subtractor 4 ... Interpolator 5, 12 ... Encoder 6, 13 ... buffer 7, 14 ... Quantization controller 8 ... Multiplexer 9 ... data output 10, 20 ... 1st code amount control means 11 ... Sub sampler 15 ... Code amount setter 16 ... 2nd code amount control means (code amount limiter) 21 ... Predictive subtractor 22, 27 ... Changeover switch 23 ... frame counter 24 ... frame memory 25 ... adder 26 ... Decoder