JP4447443B2 - Image compression processor - Google Patents

Description

  The present invention relates to an image compression processing apparatus that compresses an image, and more particularly to an image compression processing apparatus that can reliably maintain high-quality image transmission.

  With digitalization in recent years, moving image data compression and decompression techniques such as MPEG (Moving Picture Experts Group) have begun to be used in surveillance systems, DVDs and BS broadcasts. In addition, not only broadband such as ADSL (Asymmetric Digital Subscriber Line) and CATV (Cable Television) but also narrow band such as mobile phone and ISDN (Integrated Services Digital Network), MPEG-4 which has high compression efficiency and dynamic image quality. Videophone and video content distribution using (ISO / IEC14496-2) is becoming widespread.

In this way, the use of image transmission by digitization has expanded, and the demand for transmitting high-quality images in various infrastructures has increased. A mobile phone TV phone according to an example has a line capacity of 64 kbps, and a bandwidth allocated to image transmission is about 40 kbps. Since voice, control data, and the like are transmitted in other bands, the image data must be controlled to be transmitted at 40 kbps or less.
On the other hand, a moving image compression coding technique such as MPEG-4 is performed by assuming a certain amount of data generated by compression coding, but it is difficult to perform high-precision control such as a unit of several tens of bytes. In the case of compressing a moving image with a constantly changing image complexity, which is a compression algorithm, data several times larger than the assumed band may be output. In the example of the mobile phone described above, if the bit rate of the image data exceeds 40 kbps, the loss of image data, audio data, and other data or an increase in delay will result.

For example, when image data is missing, if the predictive coding method refers to the image data of the previous and subsequent frames as in MPEG, the image data is changed from the frame where the compressed data is lost until the intra frame becomes a prediction reference. Disturbances and collapses occur continuously.
In this specification, an intra frame is referred to as I-VOP, and a frame for performing predictive coding other than I-VOP is referred to as P-VOP.

JP 2003-309841 A

However, since the conventional image compression processing apparatus as described above is a compression algorithm in which it is difficult to accurately control the amount of data that has been compression-encoded, a larger amount of data than the limited bandwidth is stored in the image compression processing apparatus. Output data may be lost, and it may be difficult to reliably maintain high-quality image transmission.
The present invention has been made to solve such a conventional problem, and an object of the present invention is to provide an image compression processing apparatus that can reliably maintain high-quality image transmission.

In order to achieve the above object, the image compression processing apparatus according to the present invention performs the following processing when compressing image data.
That is, the compression processing means compresses the image data for each frame. Bit rate value detection means detects the bit rate value based on the amount of image data output after the compression processing. When the detected bit rate value is equal to or higher than the first predetermined value or exceeds the first predetermined value, the image data discarding unit sets the image data before the output before the compression process or after the compression process to 1 Discard frames or multiple frames. The image quality lowering unit lowers the image quality of the image in the compression processing by the compression processing unit when the discarding of the image data reaches a predetermined number of times within a predetermined period.
Therefore, first, when the amount of image data after the compression processing is large, the image data is discarded to reduce the amount of image data after the compression processing, and according to the number of times such discarding is performed. As a result, the image quality of the image in the compression process is reduced and the amount of image data after the compression process is reduced, so that the amount of image data after the compression process may be increased according to the situation of the amount of image data after the compression process. Therefore, high-quality image transmission can be reliably maintained.

Note that any case may be used as the case where the detected bit rate value is equal to or higher than the first predetermined value or exceeds the first predetermined value.
Also, any image data may be used as the image data before the output before the compression process or after the compression process.
In addition, any aspect may be used as an aspect in which the image data is discarded for one frame or a plurality of frames.

Here, the detection of the bit rate value based on the amount of the image data after the compression processing may be performed with reference to the amount of the image data after the compression processing for one frame, for example. It may be performed on the basis of the average value or the like with reference to the amount of image data after the compression processing for the frame.
Various values may be used as the first predetermined value related to the bit rate value. For example, a target bit rate value is used.
In addition, when the detected bit rate value is equal to or higher than the first predetermined value or exceeds the first predetermined value, the image data before the output before the compression process or after the compression process is discarded for one frame. Then, various frames may be used as the frame to be discarded. For example, the frame when the bit rate value is detected may be used, or the next frame may be used.

In addition, regarding the detection that the discarding of the image data has reached a predetermined number of times within a predetermined period, various periods may be used as the predetermined period. For example, the period is determined using a predetermined time. Alternatively, the period may be determined using a predetermined number of frames. In addition, various times may be used as the predetermined number.
In addition, it is detected that the frequency of discarding image data within a predetermined period exceeds a predetermined value or exceeds a predetermined value as the number of times the image data has been discarded within a predetermined period. Also good.
Various modes may be used as a mode for reducing the image quality of the image in the compression process. For example, the size of the quantization step used when quantizing the image data is increased, and the quantization is performed. A mode that reduces the amount of image data later can be used.

The image compression processing apparatus according to the present invention performs the following processing as one configuration example.
That is, the image quality improvement means improves the image quality of the image in the compression processing by the compression processing means when the detected bit rate value is less than or equal to the second predetermined value or less than the second predetermined value. .
Therefore, when the bit rate value is relatively small and the amount of image data after compression processing can be increased, the image quality of the image in the compression processing can be improved, so that high-quality image transmission can be ensured. Can be maintained.

Note that any case may be used as the case where the detected bit rate value is equal to or less than the second predetermined value or less than the second predetermined value.
Here, various values may be used as the second predetermined value related to the bit rate value. For example, a result value obtained by multiplying the target bit rate value by a value less than 1 is used.
In addition, various modes may be used as a mode for improving the image quality of the image in the compression process. For example, the size of the quantization step used when quantizing the image data is reduced, and the quantization is performed. A mode in which the amount of image data later is increased can be used.

The image compression processing apparatus according to the present invention performs the following processing as one configuration example.
That is, the output is performed on the transmission line. That is, the image data after the compression processing is output to the transmission path directly from the image compression processing device or via another device.
The transmission path bit rate value detecting means detects the transmission path bit rate value. One or both of the first predetermined value and the second predetermined value are determined based on the detected bit rate value of the transmission path.
Therefore, the image data is discarded, the image quality is reduced, and the image quality is improved according to the status of the bit rate value of the transmission path. Therefore, high-quality image transmission is ensured according to the status of the transmission path. Can be maintained.

Here, as the bit rate value of the transmission path, for example, a value representing the vacancy of the transmission path that can be used for communication is used according to the situation where the transmission path is used for communication.
Further, the bit rate value of the transmission path is used as, for example, a target bit rate value, or used as a basis for calculating a target bit rate value.

The present invention can also be provided as a method, a program, a recording medium, and the like.
In the method according to the present invention, each unit executes various processes in the image compression processing apparatus.
The program according to the present invention is executed by a computer constituting the image compression processing apparatus, and various functions are realized by the computer.
In the recording medium according to the present invention, a program to be executed by a computer constituting the image compression processing apparatus is recorded so as to be readable by an input unit of the computer, and the program causes the computer to execute various processes.

More specific description will be given below.
The present invention can provide an image compression processing apparatus according to the following (Configuration Example 1) and (Configuration Example 2).
(Configuration Example 1) In an image compression processing apparatus that processes compression-encoded image data, a bit rate determination processing function for monitoring the number of output bits and outputting an effective bit rate value, and the effective bit rate value and target A frame skip determination processing function for generating a frame skip timing signal from a relationship such as a bit rate ratio, a frame skip processing function for deleting predetermined image data in accordance with the frame skip timing signal, and the frame skip timing signal being predetermined A frame skip number count processing function that outputs a low image quality setting timing signal when a predetermined number of times occur in time, and an image quality parameter value that results in a low image quality in accordance with the low image quality setting timing signal is input to the compression encoding processing unit Parameter control function Image compression processing apparatus.
(Configuration Example 2) In the image compression processing apparatus described in (Configuration Example 1) above, when the effective bit rate is less than a predetermined ratio with respect to the target bit rate, the image quality of compression encoding of the next frame An image compression processing apparatus comprising an image quality parameter control function for changing a parameter value in a high image quality direction.

In the image compression processing apparatus according to the above (Configuration Example 1), image data is compression-encoded as follows.
That is, when the bit rate determination processing function outputs an effective bit rate value based on the compression-encoded compressed data, and the frame skip determination processing function that has input this determines that the effective bit rate value exceeds the target bit rate To supply a frame skip timing signal. The frame skip processing function that has received the frame skip timing signal does not input the image data to the compression encoding processing unit so as not to perform the compression encoding processing of the next frame. As a result, it is possible to control the effective bit rate value below the target bit rate determined by the system for compressed data that has been compression-encoded.
In addition, when the frame skip count signal processing function receiving the frame skip timing signal receives a predetermined number of frame skip timing signals within a predetermined time, the low image quality setting timing signal is supplied to the image quality parameter control function. The image quality parameter control function that receives the low image quality setting timing signal inputs the image quality parameter value for low image quality to the compression coding processing unit, and updates the image quality setting of the compression coding processing to low image quality, so that the subsequent frames The amount of compressed data is suppressed. As a result, frequent occurrence of the frame skip can be suppressed.

  Further, in the image compression processing apparatus according to the above (Configuration Example 2), when the image quality parameter control function determines that the effective bit rate value is too low with respect to the target bit rate, the image quality parameter control function performs the next compression. An image quality parameter value such that the image quality setting of the encoding process is high is input to the compression encoding processing unit. As a result, it is possible to realize an efficient bit rate control function that prevents a decrease in compression efficiency due to the low image quality setting and minimizes a decrease in image quality.

Here, various types of image data to be processed may be used. For example, moving image data composed of a plurality of temporally continuous frames or still image data that is not temporally continuous. Etc. can be used.
As a function for supplying frame skip timing from image data to be processed, various methods may be used. For example, information can be input from an external bit rate measuring device or the like.
In addition, the data compression timing supply conditions include not only when the effective bit rate value exceeds the target bit rate, but also the conditions related to the ratio of the effective bit rate value to the target bit rate. Can be satisfied.
In addition to satisfying the relationship with the target bit rate, the conditions for supplying the data deletion timing satisfy the required specifications of the image compression processing apparatus and the system, such as the condition regarding the ratio of the effective bit rate value to the transmission path bit rate. It can be a condition.

In addition, the supply condition of the image quality parameter value for low image quality is not limited to the timing for supplying the low image quality setting timing signal, but when the effective bit rate value exceeds the target bit rate, or the effective bit rate for the target bit rate. For example, the condition relating to the ratio of the values can be set to satisfy the required specification of the image compression processing apparatus.
In addition, as a condition for supplying image quality parameter values for low image quality, not only the relationship with the target bit rate but also a condition regarding the ratio of the effective bit rate value to the transmission channel bit rate, the requirements of the image compression processing apparatus and system are required. It can be a condition that satisfies the specifications.

Below, the structural example which concerns on this invention is shown further.
As one configuration example, the bit rate determination processing function calculates an effective bit rate value from the amount of compressed data output from the compression encoding processing unit, and supplies the effective bit rate value to the frame skip determination processing unit and the image quality parameter control unit. .
As one configuration example, the frame skip determination function supplies a frame skip timing signal to the frame skip processing unit when it is determined that the effective bit rate value exceeds the target bit rate.
As an example of the configuration, the frame skip processing function supplies image data to the compression encoding processing unit and does not supply only the next frame in which the frame skip timing signal is generated.

As an example of the configuration, when the frame skip timing signal is generated twice while the GOV (Group of VOP) is processed five times, the frame skip number count processing function supplies the low image quality setting timing signal to the image quality parameter control unit.
Here, GOV is one of the processing units in MPEG-4 and refers to a group of VOPs including all P-VOPs from I-VOP to the next I-VOP.
As an example of the configuration, the image quality parameter control function supplies an image quality parameter value for low image quality to the compression coding processing unit when a low image quality setting timing signal is received. Also, when the effective bit rate value becomes 70 percent (%) or less of the target bit rate, an image quality parameter value that provides high image quality is supplied to the compression encoding processing unit.
As an example of the configuration, the compressed data that has been subjected to the compression encoding process to be processed is compressed data that has been compression encoded using predictive encoding.

  According to such an image compression processing device, an effective bit rate value is acquired from the amount of compressed data subjected to compression encoding processing, and when the acquired effective bit rate value exceeds a target bit rate, the next frame is compressed and encoded. By skipping the processing, it becomes possible to maintain the target bit rate or less for the compressed data that has been compression-encoded. In addition, when the effective bit rate value exceeds the target bit rate or depending on the occurrence status of the frame skip, it is possible to reduce the occurrence of frame skip by setting the compression encoding processing to a low image quality setting. It becomes. Furthermore, when it is determined that the effective bit rate value is too low with respect to the target bit rate as a result of the low image quality setting, an image compression processing device with high compression efficiency is realized by setting the compression encoding processing to a high image quality setting. To do.

  As described above, according to the image compression processing apparatus of the present invention, when compressing image data for each frame, the bit rate value is detected and detected based on the amount of image data output after the compression processing. When the bit rate value is equal to or higher than the first predetermined value or exceeds the first predetermined value, the image data before the output before the compression process or after the compression process is discarded for one frame or a plurality of frames, Since the image quality of the image in the compression process is reduced when the number of data discards reaches a predetermined number of times within a predetermined period, the amount of image data after the compression process is excessive depending on the situation of the amount of image data after the compression process. Therefore, high-quality image transmission can be reliably maintained.

  Embodiments according to the present invention will be described with reference to the drawings.

A first embodiment of the present invention will be described.
In this example, for example, a case where the present invention is applied to an image compression processing apparatus using a compression coding processing method based on predictive coding such as MPEG-4 is shown.
FIG. 1 shows a configuration example of the image compression processing apparatus of this example.
The image compression processing apparatus of this example includes a frame buffer 1, a frame skip processing unit 2, a compression encoding processing unit 3, a frame buffer 4, a bit rate determination processing unit 5, a frame skip determination processing unit 6, A frame skip number count processing unit 7 and an image quality parameter control unit 8 are provided.

An example of an operation performed on moving image data by the image compression processing apparatus of this example will be described.
First, the process of the first frame of moving image data will be described.
In a compression encoding method using predictive encoding such as MPEG-4, I-VOP compression encoding that does not perform prediction from previous and subsequent frames is periodically performed, and a frame serving as a reference for prediction is periodically inserted. As a result, the image quality is maintained to a certain extent and the image is restored from image distortion or collapse due to an error. For this reason, since the first frame needs to be a prediction reference, the encoding type is I-VOP. When the first image data is input to the image compression processing apparatus, the image data is stored in the frame buffer 1, and the image data is input to the frame skip processing unit 2 when one frame of image data is prepared.

The frame skip processing unit 2 normally outputs the input image data to the compression encoding processing unit 3, but does not output it only when the frame skip timing signal from the frame skip determination processing unit 6 is generated. Since no frame skip timing signal is generated in the first frame, the frame skip processing unit 2 outputs the input image data to the compression encoding processing unit 3.
The compression encoding processing unit 3 performs a compression encoding process on the input image data. At this time, the image quality parameter used in the compression encoding process uses an arbitrary initial value given when the system is started. When the compression encoding process is completed, the compression encoding processing unit 3 outputs the compressed data obtained by the compression encoding process to the frame buffer 4. At the same time, the amount of compressed data (or the size of the compressed data) related to the compression coding process is input from the compression coding processing unit 3 to the bit rate determination processing unit 5.

The bit rate determination processing unit 5 calculates an effective bit rate value [bps] from the input compressed data amount, and outputs the calculated effective bit rate value to the frame skip determination processing unit 6 and the image quality parameter control unit 8.
Here, various methods may be used as a method of calculating the effective bit rate value. For example, when 1 GOV is composed of 30 frames and an input image of 30 fps (frame per second) is processed, 1 GOV is input in 1 second, and the amount of compressed image data for the past 5 GOV A value obtained by dividing the accumulated value by 5 seconds is set as an effective bit rate value. In this example, when frame skipping is performed, the data amount of the corresponding frame is calculated as zero (0).
In this example, the effective bit rate value is calculated using the value for the past 5 seconds (5 GOV). For example, the effective bit rate value is calculated using a shorter time value such as 1 second. However, the effective bit rate value is likely to fluctuate easily, and frame skipping and image quality parameter changes occur frequently, making it difficult to see the displayed image. It may become. For this reason, an appropriate mode should be used in consideration of these trade-offs.

The frame skip determination processing unit 6 compares the input effective bit rate value with the target bit rate requested by the system, and if the effective bit rate value is lower than the target bit rate (that is, if it is smaller). ) Do not output anything. On the other hand, if the effective bit rate value is higher (that is, larger) than the target bit rate, the frame skip determination processing unit 6 sends the frame skip timing signal to the frame skip processing unit 2 and the frame skip number count processing unit 7. Output.
Here, the target bit rate is set in the frame skip determination processing unit 6 in advance, for example.
Various values may be used as the set value of the target bit rate.
In this example, the frame skipping is performed when the effective bit rate value exceeds the target bit rate, so if there is a margin, a slightly smaller value than the value specified by the system, etc. What is necessary is just to set as the value of the target bit rate in the image compression processing apparatus of this example.

The frame skip number count processing unit 7 outputs a low image quality setting timing signal to the image quality parameter control unit 8 according to the input frequency of the frame skip timing signal from the frame skip determination processing unit 6. Assuming that the effective bit rate value is lower than the target bit rate in the first frame, no frame skip timing signal is generated, so the frame skip number count processing unit 7 does not operate.
In a normal operation, the image quality parameter control unit 8 outputs an image quality parameter value for low image quality to the compression coding processing unit 3 in accordance with the generation state of the low image quality setting timing signal. In the first frame, since the low image quality setting timing signal is not generated, the image quality parameter value is not output.
In this example, for example, a quantization step is used as the image quality parameter in the compression encoding process. In this case, as a change in the image quality parameter, a mode in which the quantization step is changed is used.
In this example, the larger the image quality parameter value, the lower the image quality, and the smaller the image quality parameter value, the higher the image quality.
The above is the processing of the first frame.

Next, processing after the second frame of moving image data will be described.
The second and subsequent frames are P-VOP encoding types for performing predictive encoding, and P-VOP encoding is repeatedly performed until the frame period for the next I-VOP encoding is reached.
When the image data of the second frame is input to the image compression processing device, the image data is stored in the frame buffer 1, and the image data is input to the frame skip processing unit 2 when one frame of image data is prepared. Is done. At this time, since the frame skip timing signal was not generated in the processing of the first frame which is the previous frame, the image data is output from the frame skip processing unit 2 to the compression encoding processing unit 3 in the second frame. .

The compression encoding processing unit 3 performs a compression encoding process on the input image data. Here, since the image quality parameter value is not output from the image quality parameter control unit 8 in the processing of the first frame corresponding to the previous frame, the image quality parameter held by the compression coding processing unit 3 is given at the time of system startup. This is an arbitrary initial value, and this initial value is used in the compression encoding process of the second frame.
When the compression encoding process is completed, the compression encoding processing unit 3 outputs the compressed data to the frame buffer 4. At the same time, the compressed data amount is output from the compression encoding processing unit 3 to the bit rate determination processing unit 5.

The bit rate determination processing unit 5 calculates an effective bit rate value from the compressed data amount input this time and the compressed data amount input up to the previous frame, and the calculated effective bit rate value is used as the frame skip determination processing unit 6 and the image quality. Output to the parameter control unit 8.
Here, if the effective bit rate value in the second frame is equal to or less than the target bit rate requested by the system, the frame skip number count processing unit 6 and the image quality parameter control unit 8 do not output anything.
The above is the processing of the second frame.
In addition, the same processing as described above is repeated for the third and subsequent frames.

Here, the case where the effective bit rate value exceeds the target bit rate for the first time and becomes 110% (%) as a result of the processing in the bit rate determination processing unit 5 will be described as the Xth frame. This is the same processing regardless of I-VOP encoding or P-VOP encoding.
That is, the frame skip determination processing unit 6 generates a frame skip timing signal and outputs it to the frame skip processing unit 2 and the frame skip number count processing unit 7.
The frame skip processing unit 2 to which the frame skip timing signal has been input does not output the image data input to the (X + 1) th frame, which is the next frame, to the compression encoding processing unit 3. For this reason, for the (X + 1) th frame, since the compressed data is not output from the compression encoding processing unit 3, the amount of compressed data is zero (0). Then, the effective bit rate value of the (X + 1) th frame calculated by the bit rate determination processing unit 5 is lower than the effective bit rate value of the X frame.
Further, since the frame skip number count processing unit 7 has input the frame skip timing signal, the frame skip count signal increments the count value of the frame skip count in the past 5 GOV, and the count value becomes 1 at the time of the X frame.

Next, it is assumed that the effective bit rate value exceeds the target bit rate and becomes 110 percent (%) even in the (X + 1) th frame.
In this case, similarly to the processing of the X frame, the frame skip processing unit 2 to which the frame skip timing signal is input outputs the image data to be input to the (X + 2) frame as the next frame to the compression encoding processing unit 3. As a result, the effective bit rate value in the (X + 2) frame is smaller than the effective bit rate value in the (X + 1) frame.
Further, since the frame skip number count processing unit 7 receives the frame skip timing signal, the frame skip number count processing unit 7 increments the count value of the number of frame skips in the past 5 GOV, and the count value becomes 2 at the time of the (X + 1) th frame.

Here, in this example, when the number of frame skips generated in the past 5 GOV is 2, the image quality parameter control unit 8 is set to reduce the image quality of the compression coding process.
Therefore, when the count value of the number of frame skips in the past 5 GOV becomes 2 as described above, the frame skip number count processing unit 7 outputs a low image quality setting timing signal to the image quality parameter control unit 8. In response to this, the image quality parameter control unit 8 outputs an image quality parameter value for low image quality to the compression encoding processing unit 3.
As a result, since the compression encoding processing unit 3 updates the image quality parameter value as low image quality, the compressed data amount of the compressed data output by the compression encoding processing after (X + 3) frames is low. As a result, the effective bit rate value is reduced.
The frame skip count counter 7 resets the frame skip count counter to zero (0) when the count value of the frame skip count reaches a predetermined value (2 in this example). The frame skip count is newly counted.

Further, the case where the effective bit rate value becomes 70 percent (%) of the target bit rate as a result of the processing in the bit rate determination processing unit 5 will be described as the Y frame. This is the same processing regardless of I-VOP encoding or P-VOP encoding.
That is, the image quality parameter control unit 8 compresses and encodes an image quality parameter value that provides high image quality in the Y frame based on the effective bit rate value input from the bit rate determination processing unit 5 and the set target bit rate. Output to the processing unit 3.
As a result, the compression encoding processing unit 3 updates the image quality parameter value as high image quality, so that the amount of compressed data output by the compression encoding processing of the (Y + 1) th frame is Y frame. As a result, the effective bit rate value is improved (that is, increased).
In this example, the image quality parameter control unit 8 compares the effective bit rate value with the target bit rate and determines that the effective bit rate value is less than 70 percent (%) of the target bit rate. In addition, processing for setting high image quality is performed as described above.

In this way, before the effective bit rate value exceeds the target bit rate and affects the system, the frame skip of the compression encoding process and the low image quality setting are used together to simply avoid the buffer overflow. In addition, the quality of the moving image such as frame continuity can be maintained.
Also, if image data with very high complexity is input and low image quality is set, and then image data with low complexity is input, the effective bit rate value will be extremely smaller than the target bit rate. . In this example, in order to avoid the low efficiency of compression encoding due to the change of the input image data, the image quality is improved by performing the high image quality setting only when the effective bit rate value becomes extremely small. As a result, it is possible to realize a highly efficient compression encoding process that also supports changes in moving images.

Various configurations may be used as the configuration for changing the setting from the initial value of the image quality to the low image quality. For example, a configuration in which two levels of the initial value and the value having the low image quality can be set is used. Or a plurality of image quality levels may be provided. As an example, when changing the setting from the initial value to the low image quality, a configuration may be used in which the fluctuation range of the image quality parameter value is fined and the low image quality setting is gradually set. Similarly, various configurations may be used for returning the setting from the low image quality to the high image quality, and a configuration in which the image quality is gradually set may be used.
Also, in this example, when shifting from low image quality to high image quality, instead of returning to the same value as the original image quality, the image quality is returned to a lower image quality than the original image quality. A configuration for returning to the same value may be used. However, if you immediately return to the same value as the original image quality, the same will be repeated, such as occurrence of frame skip, low image quality setting, decrease in effective bit rate value, high image quality setting, occurrence of frame skip again. In some cases, it is also effective to immediately return the image quality to the same value as the original image quality.

With reference to FIG. 2, an example of the flow of compressed data output processing performed by the image compression processing apparatus of this example will be described.
FIG. 2 shows a specific example of the frames A1 to A31 output from the image compression processing apparatus of this example, and the compression output from the compression encoding processing unit 3 to the bit rate determination processing unit 5 An example of the time change of the data amount, an example of the time change of the effective bit rate value output from the bit rate determination processing unit 5, and an example of the time change of the frame skip timing signal output from the frame skip determination processing unit 6 An example of the temporal change of the low image quality setting timing signal output from the frame skip number count processing unit 7 and an example of the temporal change of the image quality parameter value output from the image quality parameter control unit 8 are shown.
The horizontal direction in FIG. 2 indicates the temporal processing flow (flow in frame units) from the first frame (frame A1) which is the first I-VOP to the 31st frame (second frame I-VOP). Frame A31) is shown.

The effective bit rate value does not exceed the target bit rate from the first frame (frame A1) to the fifth frame (frame A5) output from the image compression processing apparatus.
In this example, it is assumed that when the sixth frame (frame A6) is output from the compression encoding processing unit 3, the frame skip determination processing unit 6 determines that the effective bit rate value exceeds the target bit rate. At this time, the frame skip determination processing unit 6 outputs a frame skip timing signal, and the frame skip number count processing unit 7 increments the counter. Since the seventh frame (frame A7) is the next frame from which the frame skip timing signal has been output, the compression encoding process is not performed. For this reason, for the seventh frame, the amount of compressed data is zero (0), and the effective bit rate value can be rapidly lowered.

In the eighth frame (frame A8), it is assumed that compression encoding processing is performed as usual and no event occurs.
In this example, it is assumed that the compressed data amount increases in the ninth frame (frame A9), and the frame skip determination processing unit 6 determines that the effective bit rate value exceeds the target bit rate. At this time, the frame skip determination processing unit 6 outputs a frame skip timing signal, and the frame skip number count processing unit 7 increments the counter. In this example, since the frame skip timing signal is output twice within the 5 GOV period, the low image quality setting timing signal is output from the frame skip number count processing unit 7. The image quality parameter control unit 8 receives this and outputs an image quality parameter value for low image quality, thereby updating the compression parameter setting so as to reduce the amount of compressed data in subsequent frames.

In the 10th frame (frame A10), since it is the next frame from which the frame skip timing signal has been output, compression encoding processing is not performed. For this reason, the amount of compressed data is zero (0), and the effective bit rate value can be rapidly reduced.
In frames subsequent to the eleventh frame (frame A11), since compression coding processing is performed using image quality parameter values for low image quality, a very large amount of compressed data as in the ninth frame is output. Therefore, the effective bit rate value hardly exceeds the target bit rate, and the occurrence of frame skip can be suppressed.

In addition, since the complexity of the input image data is very small and image data that outputs a small amount of compressed data is continuously input, the effective bit rate in the 14th frame (frame A14). Assume that the value falls below 70 percent (%) of the target bit rate.
In this case, the image quality parameter control unit 8 outputs an image quality parameter value that provides high image quality to the compression coding processing unit 3, and updates the compression parameter setting to increase the amount of compressed data for subsequent frames.
In this example, as the image quality parameter value for high image quality at this time, the image quality parameter value for low image quality is set to be larger than the value before setting in the ninth frame (that is, lower image quality is set). The compression encoding process and the bit rate control process are stabilized.
After the fifteenth frame (frame A15), the effective bit rate value changes between 70 and 100 percent (%) of the target bit rate.

  As described above, the image compression processing apparatus of the present example performs compression encoding processing on an input image, and outputs the compressed data generated thereby to a transmission unit having a limited transmission capacity. If the effective bit rate value is determined to exceed the transmission capacity by continuously monitoring the amount of generated data and outputting the compressed data to the transmission unit, the compression encoding process for the next frame is not performed. The frame skip is executed automatically, for example. Also, in the image compression processing apparatus of this example, when a predetermined frame skip occurs within a predetermined time, the image quality parameter is updated so that the amount of compressed data becomes small. Further, when the effective bit rate value falls below a predetermined value within a predetermined time, the image quality parameter is updated so as to achieve high image quality.

As described above, the image compression processing apparatus of the present example implements a bit rate control function for controlling the effective bit rate value so as not to exceed the target bit rate by using both the frame skip and the low image setting of the image quality parameter. . In addition, by determining the lower limit value of the effective bit rate value and setting the image quality parameter below that value, it is possible to increase the efficiency of the compression encoding process by setting the image quality parameter to a high image quality setting.
Here, in this example, the case where moving image data composed of a plurality of continuous frames is processed has been described. However, the image data to be processed is not particularly limited, It is also possible to process data such as frames of still image frames that are continuously arranged.

In this example, when the frame skip occurs a predetermined number of times within a predetermined time, the image quality parameter is changed so that the output image data from the compression encoding processing unit 3 has a low image quality. When setting the image quality, for example, various messages such as “Preventing the occurrence of frame skip” are displayed on the display screen of the client terminal etc. that is playing the image data. You can also In order to display the message, for example, the image parameter control unit 8 may be configured to output a predetermined control signal to the client terminal or the like, or the compression coding processing unit 3 may perform the above-described “frame skipping”. It may be configured to superimpose on the image data itself to output a message such as “Prevented occurrence”.
In addition, as an aspect for determining that a frame skip has occurred a predetermined number of times within a predetermined time, for example, the number of frames corresponding to the predetermined time is used and the frame skip has occurred a predetermined number of times during the predetermined number of frames. It is also possible to use a mode for determining the above.
In addition, the image compression processing apparatus of this example may be configured as a single unit, or may be configured as one of functional units included in the image transmission apparatus, the camera, or the like.

As described above, in the image compression processing apparatus of the present example, the effective bit rate value of the compressed data by the compression encoding process is performed by reducing or expanding the generated bit amount by deleting the image data and changing the image quality setting as necessary. However, control is performed so as not to exceed a given band (in this example, a target bit rate).
Therefore, in the image compression processing apparatus of this example, a high-efficiency bit rate control function can be realized by the combined use of frame skip and image quality setting control, and image quality degradation as a still image or a moving image due to this control can be realized. It can be controlled to be minimal.

Further, in the image compression processing apparatus of this example, when the effective bit rate value exceeds a predetermined value (in this example, the target bit rate), frame skip is performed first, and the effective bit rate value is still sufficiently reduced. If not, low image quality setting is performed. That is, when the effective bit rate value exceeds the target bit rate, it is necessary to skip frames so as not to output compressed data. If compression is performed without performing frame skipping, the effective bit rate value further exceeds the target bit rate, increasing the possibility that the transmission path will fail.
Further, for example, in an application such as a monitoring system in which image quality is important, there is a case where quality (high image quality) is more important than frame continuity. In the bit rate control of this example, the frame skip is preferentially executed over the reduction of the image quality, so that an extreme deterioration in image quality can be prevented.

Here, (Modification 1) to (Modification 4) of the image compression processing apparatus of this example are shown.
(Modification 1)
In the above embodiment, as (Processing 1), in the apparatus that processes image data by a predetermined compression encoding method, the bit rate is monitored based on the amount of image data after compression encoding, and the bit rate value is set as the target. When the value is exceeded, frame skip is performed, and the next frame is discarded without being compressed. As (Process 2), the number of frame skips is counted in (Process 1), and when the frame skip occurs a predetermined number of times within a predetermined time, the image quality parameter is changed to a low image quality setting (for example, compression rate) To increase the amount of image data after the next frame. As (Processing 3), after (Processing 2), when the bit rate value becomes a predetermined value or less, the image quality parameter is changed to a high image quality setting (for example, by reducing the compression rate), and the following The amount of image data after the frame is increased.
In this modified example, first, instead of the above (Processing 1), in a device that processes image data by a predetermined compression encoding method, the bit rate is monitored based on the amount of image data after compression encoding, When the bit rate value exceeds the target value, the image quality parameter is changed to a low image quality setting, and thereafter, the same processing as the above (processing 3) is performed.

(Modification 2)
In this modification, the frame skip number count processing unit 7 outputs the low image quality setting timing signal to the image quality parameter control unit when the frame skip timing signal is generated twice consecutively (in this example, two frames consecutively). Output to 8, do not output in other cases. It should be noted that the frame skip count counting processing unit 7 continues to output the low image quality setting timing signal to the image quality parameter control unit 8 while frame skips once occur continuously and continue to occur thereafter.
For example, if the frame skip does not occur in the (m−1) th frame, the frame skip occurs in the mth frame, and the frame skip occurs in the (m + 1) th frame, the frame skip takes two frames. The frame skip count counting processing unit 7 outputs a low image quality setting timing signal to the image quality parameter control unit 8 because they occur continuously. In response to this, the image quality parameter control unit 8 outputs a parameter value for low image quality to the compression coding processing unit 3. Thereafter, when a frame skip occurs even in the (m + 2) th frame, the frame skip number count processing unit 7 outputs a low image quality setting timing signal to the image quality parameter control unit 8. The image quality parameter control unit 8 outputs a parameter value that further reduces the image quality to the compression encoding processing unit 3, if any.

(Modification 3)
In the above-described embodiment, the frame skip processing unit 2 does not output the image data to the compression encoding processing unit 3, thereby skipping the image data before compression and discarding the image data before compression.
In this modification, the function of the frame skip processing unit 2 is arranged between the compression encoding processing unit 3 and the frame buffer 4, and the image data of the next frame to be skipped is compressed and encoded processing unit 3. The compressed image data output from is discarded. In this case, the bit rate determination processing unit 5 confirms from the compression encoding processing unit 3 or the frame skip determination processing unit 6 or the frame skip processing unit 2 that the amount of compressed image data to be discarded is zero (0). Notice.

(Modification 4)
In this modification, the frame skip processing unit 2 is not provided, and the target compression is not the image data of the next frame but the image data after compression of the current frame, and the corresponding compression in the frame buffer 4 on the output side. The later image data is discarded. In this case, in order to correct the discarded compressed image data so that the data amount becomes zero (0), the bit rate determination processing unit 5 determines the discarded compressed image data from the next calculation of the effective bit rate value. The calculation is performed assuming that the data amount is zero (0). Further, for example, a frame skip timing signal from the frame skip determination processing unit 6 is input to the control unit of the frame buffer 4, and the control unit converts the image data after compression of the frame whose effective bit rate value is excessive into the frame. Delete from buffer 4.

  Here, with regard to the above (Modification 3) and (Modification 4), when a compression encoding method using predictive encoding such as MPEG-4 is used, the compressed image data is simply discarded. If the I-VOP or P-VOP is discarded, the compressed image data of the subsequent frames cannot be restored. For example, the next frame is forcibly set to the I-VOP with respect to the discarded frame. Process. Further, when a compression encoding processing method such as JPEG compression is used, such processing may not be required.

  In the image compression processing apparatus of this example, the compression processing means is configured by the function of the compression encoding processing unit, the bit rate value detection unit is configured by the function of the bit rate determination processing unit 5, and frame skipping is performed. The function of the determination processing unit 6 and the function of the frame skip processing unit 2 constitute an image data discarding unit, and the function of the frame skip number counting processing unit 7 and the function of the image quality parameter control unit 8 constitute an image quality lowering unit. The image quality parameter control unit 8 functions to configure image quality improvement means.

A second embodiment of the present invention will be described.
FIG. 3 shows a configuration example of the image compression processing apparatus of this example.
The image compression processing apparatus of the present example schematically has a configuration in which a transmission path bit rate determination processing unit 11 is further added to the configuration of the image compression processing apparatus shown in FIG.
In FIG. 3, the same components 1 to 8 as shown in FIG. 1 are denoted by the same reference numerals.
Further, in this example, components and operation parts different from the image compression processing apparatus shown in FIG. 1 will be described in detail.

The transmission path bit rate determination processing unit 11 has a function of measuring the transmission capacity remaining in the transmission path in order to perform the main image transmission. In this example, the transmission channel bit rate determination processing unit 11 receives the load information of the transmission channel measured by the network analyzer connected to the transmission channel from the network analyzer, and is left in the transmission channel based on this information. The transmission capacity (in this example, the value of the transmission channel bit rate) is detected and output to the frame skip determination processing unit 6 and the image quality parameter control unit 8.
In this example, the transmission channel bit rate value output from the transmission channel bit rate determination processing unit 11 is used as the target bit rate.
Specifically, a series of processing operations performed on input moving image data is the same as that shown in FIG. 1, and in this example, the frame skip determination processing unit 6 and the image quality parameter control unit 8 are used. The target bit rate used in the above is the transmission channel bit rate value.

The frame skip determination processing unit 6 compares the effective bit rate value from the bit rate determination processing unit 5 with the transmission channel bit rate value from the transmission channel bit rate determination processing unit 11, and determines that the effective bit rate value is the transmission channel bit. When it is higher than the rate value, a frame skip timing signal is generated and output to the frame skip processing unit 2 and the frame skip number count processing unit 7, and the effective bit rate value is lower than the transmission channel bit rate value. Output nothing.
In the image quality parameter control unit 8, the low image quality setting process according to the low image quality setting timing signal is the same as that shown in FIG. 1, and after the image quality is reduced, the transmission channel bit rate value of 70 is input. When the effective bit rate value becomes lower than the percentage (%) value, the image quality parameter is improved.
With such a configuration, the image compression processing apparatus of this example provides a highly efficient bit rate control function by combining frame skipping and image quality setting control even when the transmission capacity of a transmission line such as a network fluctuates. It can be realized, and control can be performed so as to minimize degradation of image quality as a still image or a moving image.

FIG. 4 shows an example of the flow of compressed data output processing performed by the image compression processing apparatus of this example.
That is, FIG. 4 shows a specific example of the frames B1 to B31 output from the image compression processing apparatus of this example, and is output from the compression encoding processing unit 3 to the bit rate determination processing unit 5. An example of the temporal change of the compressed data amount, an example of the temporal change of the effective bit rate value output from the bit rate determination processing unit 5, and the time change of the frame skip timing signal output from the frame skip determination processing unit 6 An example, an example of the temporal change of the low image quality setting timing signal output from the frame skip number count processing unit 7, and an example of the temporal change of the image quality parameter value output from the image quality parameter control unit 8 are shown.
The horizontal direction in FIG. 4 shows the temporal processing flow (flow in frame units) from the first frame (frame B1) as the first I-VOP to the 31st frame (the second I-VOP). This represents the frame B31).
Here, in this example, a transmission channel bit rate value that varies as the target bit rate is used, and the schematic operation is the same as that shown in FIG.
In the image compression processing apparatus of this example, a transmission path bit rate value detection unit is configured by the function of the transmission path bit rate determination processing unit 11.

The background of the technology related to the present invention will be described below. Note that the matters described here are not necessarily limited to the conventional technology.
FIG. 5 shows a configuration example of the image compression processing apparatus.
The image compression processing apparatus of this example includes a frame buffer 21, a compression encoding processing unit 22, and a frame buffer 23, and uses a compression encoding method based on predictive encoding such as MPEG-4.
First, the process of the first frame of moving image data will be described.
Since the first frame needs to be a prediction reference, the encoding type is I-VOP. When the first image data is input to the image compression processing apparatus, the image data is accumulated in the frame buffer 21, and the image data is input to the compression encoding processing unit 22 when one frame of image data is prepared. The compression encoding processing unit 22 performs compression encoding processing on the input image data. At this time, as an image quality parameter value used in the compression encoding process, an arbitrary initial value given at the time of system startup is used. When the compression encoding process is completed, the compression encoding processing unit 22 outputs the compressed data to the frame buffer 23.
Next, processing after the second frame of moving image data will be described.
From the second frame onward, the P-VOP encoding is repeated until the I-VOP encoding cycle is reached. The processes related to the other prediction encoding are the same as those in the first frame.

With reference to FIG. 6, an example of the flow of processing performed by the image compression processing apparatus shown in FIG. 5 will be described.
FIG. 6 shows an example of frames C1 to C31 output as compressed data from the image compression processing apparatus, and shows an example of the temporal change in the compressed data amount and an example of the temporal change in the effective bit rate value. is there. In FIG. 6, the horizontal direction indicates a temporal processing flow (flow in frame units) from the first frame (frame C1) that is the first I-VOP to the 31st frame (second frame I-VOP). This represents a frame C31).

The effective bit rate value does not exceed the target bit rate from the first frame (frame C1) to the fifth frame (frame C5) output from the image compression processing apparatus.
Here, as an example, it is assumed that the effective bit rate value exceeds the target bit rate when the sixth frame (frame C6) is output from the compression encoding processing unit 22. In this case, since the image compression processing apparatus shown in FIG. 5 does not have a bit rate determination processing function, a frame skip determination processing function, or the like, no event that lowers the bit rate occurs. The same applies to the seventh frame (frame C7) and thereafter, and even if the effective bit rate value exceeds the target bit rate, it is output as it is from the image compression processing apparatus.
The present invention improves such a point and can obtain an excellent effect.

Here, the configuration of the image compression processing apparatus according to the present invention is not necessarily limited to the above-described configuration, and various configurations may be used. The present invention can also be provided as, for example, a method or method for executing the processing according to the present invention, a program for realizing such a method or method, or a recording medium for recording the program. It is also possible to provide various devices and systems.
The application field of the present invention is not necessarily limited to the above-described fields, and the present invention can be applied to various fields.
In addition, as various kinds of processing performed in the image compression processing apparatus according to the present invention, for example, the processor executes a control program stored in a ROM (Read Only Memory) in hardware resources including a processor and a memory. May be used, and for example, each functional unit for executing the processing may be configured as an independent hardware circuit.
In addition, the present invention can be grasped as a computer-readable recording medium such as a floppy (registered trademark) disk or a CD (Compact Disk) -ROM storing the control program, or the program (itself). The processing according to the present invention can be performed by inputting the program from the recording medium to the computer and causing the processor to execute the program.

It is a figure which shows the structural example of the image compression processing apparatus which concerns on 1st Example of this invention. It is a figure for demonstrating an example of the process performed by the image compression processing apparatus which concerns on 1st Example of this invention. It is a figure which shows the structural example of the image compression processing apparatus which concerns on 2nd Example of this invention. It is a figure for demonstrating an example of the process performed by the image compression processing apparatus which concerns on 2nd Example of this invention. It is a figure which shows the structural example of an image compression processing apparatus. It is a figure for demonstrating an example of the process performed by an image compression processing apparatus.

Explanation of symbols

  1, 4, 21, 23... Frame buffer, 2.. Frame skip processing unit, 3, 22... Compression encoding processing unit, 5.. Bit rate determination processing unit, 6. .. Frame skip count count processing unit, 8. Image quality parameter control unit, 11. Transmission channel bit rate determination processing unit,

Claims (3)

In an image compression processing apparatus that compresses image data,
Compression processing means for compressing image data for each frame;
Bit rate value detection means for detecting a bit rate value based on the amount of image data output after the compression processing;
When the detected bit rate value is greater than or equal to the first predetermined value or exceeds the first predetermined value, the image data before output after the compression process or after the compression process is discarded for one frame or a plurality of frames. Image data discarding means,
A decrease in image quality means discarded the image data you increase the size of the quantization step in the compression process by the compression processing means when it reaches a predetermined number of times within a predetermined time period,
An image compression processing apparatus comprising: The image compression processing apparatus according to claim 1,
And image quality improving means you reduce the size of the quantization step in the compression process by the compression processing means when the detected bit rate value to be the second predetermined value or less and made or if less than a second predetermined value,
An image compression processing apparatus comprising: The image compression processing apparatus according to claim 1 or 2,
The output is made to a transmission line,
A transmission path bit rate value detecting means for detecting a bit rate value of the transmission path;
One or both of the first predetermined value and the second predetermined value are determined based on the detected bit rate value of the transmission path.
An image compression processing apparatus.

Images