JP2014082603A - Image processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce capacity of a memory of a captured destination and access band width while occurrence of overwriting with overflow of data is avoided when video data being an object of high image quality video processing is captured.SOLUTION: An image processor includes: an encoding section for encoding video data in a raster scanning direction and writing it in a capture memory; and a decoding section for reading data for the number of fields required for high image quality video processing from the memory to decode data. The encoding section is allowed to estimate a size of encoded data of the video data for one screen on the basis of a size of the encoded data to front division data by one when division data obtained by dividing the video data for one screen into a plurality of pieces of data in the raster scanning direction is encoded and is allowed to perform processing for increasing a quantization parameter when the estimated encoded data size exceeds idle capacity of the memory when encoding for one screen is started.

Description

  The present invention relates to an image processing technique, and more particularly to a technique for improving the image quality of a video.

  Liquid crystal displays are rapidly becoming popular in place of conventional CRTs. The liquid crystal display employs a progressive method as a video display method. Since TV broadcast video signals and DVD-Video video signals are compliant with the interlace method, a display device using a liquid crystal display as a display device displays an interlaced image with high image quality. Some have a high-quality video processing circuit that makes this possible. Examples of processing in this type of high-quality video processing circuit include motion adaptive IP conversion and motion adaptive noise reduction.

  Motion adaptive IP conversion is one type of IP conversion that converts interlaced video data into progressive video data. In motion-adaptive IP conversion, the size of a subject's movement over the field to be converted and the fields before and after the field to be converted is detected. If the movement is large, IP conversion is performed using information in the preceding and following fields, but the movement is small. For example, IP conversion using only the information of the conversion target field is performed. Refer to Patent Document 1 for details of IP conversion. Similarly, in motion adaptive noise reduction, the process is switched according to the magnitude of the movement of the subject. In this way, in motion adaptive IP conversion and motion adaptive noise reduction, it is necessary to refer to the fields before and after the field to be processed in order to obtain the magnitude of the subject's motion. It is necessary to execute the processing while holding the video data representing the video of each field by the number of fields to be referred to at a specific time (hereinafter referred to as “video capture”).

Japanese Patent Laid-Open No. 10-145817

When the resolution of a high-quality video processing target video is increased, it is necessary to use a large-capacity video capture destination memory, and the bandwidth for accessing the video capture destination memory (that is, It is necessary to sufficiently increase the bandwidth when reading data from the memory and when writing data to the memory. In recent years, the PiP (Picture in picture) method or PoP (Picture on
It is also common to combine images into a single image and display them on a display device. When the number of images to be combined increases, the storage capacity and bandwidth of the video capture destination memory can be further increased. Necessary. However, in order to cope with such an increase in memory capacity and bandwidth, there is a problem that advanced and expensive technology is required for designing the substrate of the image processing apparatus.

  The present invention has been made in view of the above problems, and when storing video data subject to image processing such as high-quality video processing in a buffer, the buffer capacity is overwritten and overwritten. It is an object of the present invention to provide a technique capable of reducing the storage capacity of the buffer and the bandwidth when accessing the buffer while avoiding the occurrence of a failure.

  In order to solve the above-described problems, the present invention provides an encoding unit that encodes video data to be processed in a raster scan direction and writes it in a video capture memory, and data for a screen required for image processing from the video capture memory. A decoding unit that reads out and decodes the video data of each screen, and an image processing unit that performs image processing on the video data output from the decoding unit and outputs the processed video data. When the data obtained by dividing the video data into a plurality of parts in the raster scan direction is divided data, the encoding unit encodes each divided data into one screen from the encoding result up to the previous divided data. Estimate the data size when video data for a minute is encoded, and the data size and the video capture memo It provides an image processing apparatus according to claim containing a compression ratio adjusting means for adjusting a parameter for encoding the divided data based on the free space of. Specific examples of the image processing by the image processing unit include high-quality video processing such as motion adaptive IP conversion and motion adaptive noise reduction, and one screen of video data represents one field of video. Video data or video data representing one frame of video.

  According to the present invention, since the video data to be processed in the image processing is stored in the video capture memory after being encoded by the encoding unit, the capacity of the memory and the bandwidth when accessing the memory are encoded. The size can be reduced as compared with the conventional technique which is not performed. In addition, when a video capture memory having the same capacity as the conventional one is used, the number of fields or frames that can be stored in the memory is larger than the conventional one. It is possible to increase the number of reference frames as compared with the conventional case and perform more precise high-quality video processing.

  In the present invention, each time encoding of divided data is performed, the data size when video data for one screen is encoded is estimated from the encoding result up to the previous divided data, and the estimated data size is When the free space of the video capture memory is exceeded, the parameters for encoding the divided data are adjusted so that the compression efficiency becomes high (that is, the data size after encoding becomes smaller). . For this reason, when writing the encoded data to the video capture memory, it is possible to reliably avoid the occurrence of a failure such as overwriting due to the overflow.

  In a more preferred aspect, the encoding unit sets the parameter so as to increase the compression efficiency when the estimated data size exceeds a predetermined maximum encoded data size when encoding each divided data. It includes a second compression rate adjusting means for adjusting. Although details will be clarified in the description of the embodiment, according to such an aspect, the target maximum compression rate is set within a range in which the required image quality can be obtained for the video represented by the video data processed by the image processing unit. If it is determined, it is possible to achieve both avoidance of failure such as overwriting due to overflow and maintenance of image quality.

It is a figure which shows the structural example of image processing LSI1 of one Embodiment of the image processing apparatus of this invention. It is a figure for demonstrating the synthetic | combination process which the display control part 80 of the image processing LSI1 performs. 4 is a diagram for explaining an encoded data buffer secured in an external memory connected to the image processing LSI 1. FIG. It is a figure which shows an example of the time transition of the compression rate in this embodiment, a quantization parameter, and the evaluation value of an image quality. It is a figure for demonstrating the 1st compression rate adjustment process which the encoder 30 for external video data of the image processing LSI1 performs. It is a figure for demonstrating the 2nd compression rate adjustment process which the encoder 30 for external video data of the image processing LSI1 performs.

Hereinafter, embodiments of the present invention will be described with reference to the drawings.
FIG. 1 is a diagram illustrating a configuration example of an image processing LSI 1 according to an embodiment of the image processing apparatus of the present invention. The image processing LSI 1 performs high-quality video processing on interlaced analog video signals and digital video data, synthesizes the video after high-quality video processing and the OSD menu, and displays them on a display device such as a liquid crystal display in real time. It is. As shown in FIG. 1, the image processing LSI 1 includes an analog front end (in FIG. 1, expressed as “AFE”, hereinafter the same in this specification) 10, a video decoder 20, an external video data encoder 30, external video data Decoder 40, high-quality video processing unit 50, correction processing unit 60, OSD drawing unit 70, display control unit 80, host CPU-I / F 90, SDRAM-I / F 100, ROM-I / F 110, and shared work memory 120 Is included. A host CPU that controls the operation of the image processing LSI is connected to the host CPU-I / F 90. An external memory such as an SDRAM is connected to the SDRAM-I / F 100. The ROM-I / F 110 is connected to a pattern ROM that stores pattern data such as sprites constituting the OSD menu. The shared work memory 120 stores external video data that has undergone high-quality video processing by the high-quality video processing unit 50 and correction processing by the correction processing unit 60 and OSD menu video data generated by the OSD drawing unit. Function as.

  For example, an interlaced analog video signal such as a video signal of a television broadcast, a video signal of a DVD-Video, or an output video signal of a video camera is given to the AFE 10. The AFE 10 performs A / D conversion on the given analog video signal, and provides the video decoder 20 with digital format video data (that is, video data of each field in the interlace method) as a conversion result. The video decoder 20 separates the synchronization signal, the luminance component, and the color component from the video data output from the AFE 10 and supplies digital data representing each to the external video data encoder 30.

  In addition to the video data supplied from the video decoder 20, the external video data encoder 30 is supplied with digital video data (for example, video data in ITU601, ITU656 format) that represents a video to be combined with the video represented by this video data. It is done. The external video data encoder 30 executes an encoding process for irreversibly compressing each of the video data of a plurality of channels in a time-sharing manner, and connects an encoding result (hereinafter, encoded data) of each video data to the SDRAM-I / F 100. To the external memory In the input stage of the external video data encoder 30, a buffer is provided for each channel in order to perform time division processing smoothly.

  The external video data decoder 40 reads out and decodes the encoded data from the external memory in accordance with the video display timing (for example, the timing instructed by the display control unit 80) in the image processing LSI 1, and performs luminance (Y), color (Cb , Cr) is provided to the high quality video processing unit 50 as data. That is, in the present embodiment, an external memory connected to the SDRAM-I / F 100 is used as a “video capture memory”. In this embodiment, the external memory is used as a “video capture memory”. However, the shared work memory 120 may of course be used as a video capture memory. Which of the two memories is used as a video capture memory? It may be determined according to the number of video input channels, the resolution of each video, and the like.

  The encoding process by the external video data encoder 30 and the decoding process by the external video data decoder 40 (hereinafter sometimes collectively referred to as “codec processing”) are divided into a plurality of lines in the raster scan direction. Done. For this reason, a small amount of memory (that is, a plurality of lines) is also used for the processing in each of the subsequent stages of the external video data decoder 40 (high-quality video processing unit 50, correction processing unit 60, OSD drawing unit 70, and display control unit 80). Can be executed in raster units.

The high quality video processing unit 50 subjects the video data supplied from the external video data decoder 40 to high quality video processing such as motion adaptive IP conversion, motion adaptive noise reduction, and the like, and supplies the video data to the correction processing unit 60. The correction processing unit 60 performs various correction processing such as enlargement / reduction processing, rotation processing, or camera video distortion correction on the video data that has undergone the processing by the high-quality video processing unit 50 to obtain video data of the external video layer. Write to shared work memory 120. The OSD drawing unit 70 generates, in raster units, video data representing a menu or video of attached information that is displayed on the display device over the external video, and writes it in the shared work memory 120 as video data of the OSD layer. The display control unit 80 reads the video data of each layer from the shared work memory 120, and superimposes them in raster (multiple line buffer) units as shown in FIG. 2A, thereby displaying one screen of the layout shown in FIG. Video data for the remaining minutes is generated and output to a display device (not shown). In addition, what is necessary is just to let a user specify freely about the priority at the time of overlapping each layer.
The above is the configuration of the image processing LSI 1.

  A feature of the present embodiment is that an external video data encoder 30 that encodes video data to be processed in high-quality video processing and writes the encoded data of the encoding result to a video capture memory (in this embodiment, an external memory). And an external video data decoder 40 that reads and decodes the encoded data stored in the video capture memory and supplies the decoded video data to the high-quality video processing unit 50. As described above, since the video data to be processed in high-quality video processing is encoded and stored in the video capture memory, according to the present embodiment, the conventional technology for capturing video without performing the encoding. Compared to the above, the capacity of the video capture memory is small, and the bandwidth for accessing the video capture memory (in this embodiment, the bandwidth of data transmission via the SDRAM-I / F 100) However, it is narrower than before. In addition, in the present embodiment, by causing the external video data encoder 30 to execute processing that clearly shows the features of the present embodiment, overwriting due to overflow when capturing the video to be processed in high-quality video processing is performed. It is possible to avoid the occurrence of such failures. The following description focuses on the external video data encoder 30 that clearly shows the features of the present embodiment.

  In this embodiment, since it is necessary to perform codec processing of video data in real time without delay, a one-pass encoding algorithm (that is, analysis processing of video data to be encoded) is used as an encoding algorithm in the external video data encoder 30. An encoding algorithm that advances the compression process simultaneously and in parallel is employed. The compression rate r by this encoding algorithm (value obtained by dividing the data size of the encoding result by the data size of the video data before encoding × 100) is determined according to the quantization parameter qP as in the case of MPEG or the like. Specifically, the larger the quantization parameter qP, the lower the compression ratio r (the data size of the encoding result is smaller, that is, the higher the compression efficiency), and the image quality of the video represented by the video data obtained by decoding the encoded data Will decline. Note that it is conceivable to use PSNR (peak signal to noise ratio) as an index value for evaluating the image quality.

  In the present embodiment, the maximum value of the target compression rate r (hereinafter referred to as the target maximum compression rate) is set by the user according to the image quality allowed for the video displayed on the display device through the processing by the image processing LSI 1. Is set. On the other hand, in the external video data encoder 30, the initial value of the quantization parameter qP is set according to the target maximum compression rate. In other words, the external video data encoder 30 is configured so that the initial value of the quantization parameter can be variably set through the setting of the target maximum compression rate. As a configuration for variably setting the initial value of the quantization parameter via the setting of the target maximum compression rate, the initial value of the quantization parameter qP is associated with each of a plurality of types of values that can be assumed as the target maximum compression rate. Is stored in the external video data encoder 30, and the external video data encoder 30 executes a process of reading the quantization parameter corresponding to the target maximum compression rate set by the user from the table. Conceivable.

Further, the external video data encoder 30 secures a storage area in the external memory having a capacity necessary for storing video data for one channel based on the target maximum compression rate. In this embodiment, the video data to be processed by the high-quality video processing is obtained by subjecting an interlaced analog video signal to A / D conversion, and the external video data encoder 30 has one field. A buffer having a capacity corresponding to the target maximum encoded data size is secured for 8 fields (that is, 4 frames) for each channel (see FIG. 3). This is because the video input and the synthesized video output are asynchronous, and the encoding result (compression rate) of each video data varies. Note that the target maximum encoded data size (bytes) of one field is expressed as a pixel value of 24 bits, where the target maximum compression rate is r%, the horizontal resolution of the video is X, and the vertical resolution is Y. Then, it calculates with the following formula | equation (1).
Target maximum encoded data size for 1 field (bytes)
= X × Y / 2 × 24 × r / 100/8 (1)
As shown in FIG. 3, the writing of encoded data to each buffer secured in the external memory and the reading of encoded data from the same buffer are performed asynchronously, and these buffers function as a FIFO.

  In the encoding of video data, a quantization parameter qP for encoding the next field is usually determined from the encoded data size of the previous field (initial value is used for the first field). It is common to encode video data from top to bottom in the raster scan direction using the parameter qP. In the external video data encoder 30 of this embodiment, the quantization parameter qP is similarly controlled. This is to prevent the compression rate of each field from varying from the target maximum compression rate. FIG. 4 is a diagram illustrating the relationship between the quantization parameter qP, the compression ratio r, and the index value PSNR indicating the image quality in the time axis direction when encoding a certain video. In the example shown in FIG. 4, since the compression rate r instantaneously exceeds the target maximum compression rate at the time of encoding the field F1, the quantization parameter qP is set so that the compression rate r decreases in the encoding of the next field F2. Has been raised. On the contrary, since the compression rate r is lower than the target maximum compression rate in the encoding of the field F3, the quantization parameter qP is lowered so that the compression rate r approaches the target maximum compression rate in the encoding of the next field F4. .

  However, in the aspect in which the quantization parameter qP is fixed in units of fields, when the video to be processed includes a field of video that is difficult to compress, such as snow noise, the compression rate in the field is the target maximum compression rate. There is a possibility that failure such as overwriting accompanying overflow will occur depending on the remaining FIFO. In order to avoid the occurrence of such a failure, the external video data encoder 30 of the present embodiment adjusts the quantization parameter qP more finely (in other words, the compression rate is finely adjusted. A first compression rate adjusting means 30A and a second compression rate adjusting means 30B are provided.

  As described above, the encoding process in the external video data encoder 30 is performed for each divided data obtained by dividing video data for one field into a plurality of parts in the raster scan direction. When encoding the divided data, the first compression rate adjusting unit 30A estimates the data size of the encoded data when the video data for one field is encoded from the data size of the encoding result up to the previous divided data. If the estimated data size exceeds the FIFO free space, the quantization parameter qP is adjusted so that the divided data is encoded at a lower compression rate. That is, when the first compression rate adjustment unit 30A determines that the FIFO remaining amount is insufficient, the first compression rate adjustment unit 30A gradually increases the quantization parameter even within one field. If the quantization parameter qP is increased, the compression ratio r decreases (see FIG. 4), and FIFO overflow is avoided.

  FIG. 5 is a diagram for explaining the operation of the first compression ratio adjusting unit 30A when encoding video data of one field by dividing it into four equal parts in the vertical direction. Whether or not the quantization parameter qP is to be adjusted is determined every time the division data encoding result is written into the FIFO, as shown in FIG. More specifically, the first compression ratio adjusting unit 30A determines the data size of the encoded data when the video data for one field is encoded each time writing the encoded result of the divided data into the FIFO. If the estimated data size is larger than the FIFO free capacity at the start of encoding of that one field (FIFO capacity), quantization is performed. Increase the parameter by a predetermined amount (or at a predetermined rate).

  For example, at the time when the encoding result of the divided data corresponding to the area A in FIG. 5 is written into the FIFO, the first compression ratio adjustment unit 30A obtains the data size E1 of the encoding result by multiplying by the coefficient K1. This value is compared with the FIFO free space as an estimated value of the data size of the encoded data when video data for one field is encoded. Similarly, at the time when the encoding result of the divided data corresponding to the area B in FIG. 5 is written to the FIFO, the first compression ratio adjusting unit 30A encodes the data size of the encoding result and the encoded data corresponding to the area A. A value obtained by multiplying the sum E12 of the data size by the coefficient K2 is compared with the FIFO free space as an estimated value of the data size of the encoded data when one field of video data is encoded. Then, at the time when the encoding result of the divided data corresponding to the area C in FIG. 5 is written to the FIFO, the first compression ratio adjusting unit 30A sets the data size of the encoding result and each of the area A and the area B. The value obtained by multiplying the corresponding encoded data size sum E123 by the coefficient K3 is compared with the FIFO free space as the estimated data size of the encoded data when one field of video data is encoded. . Here, what value should be used as each of the coefficients K1, K2, and K3 may be determined through appropriate experiments. Specifically, in this embodiment, since the video data to be encoded is divided into four equal parts in the raster scan direction, K1 = 4 + safety factor (or 4 × safety factor), K2 = 4/2 + safety factor. (Or 4/2 × safety factor), K3 = 4/3 + safety factor (or 4/3 × safety factor), and each safety factor may be determined by experiment.

Next, the second compression rate adjusting unit 30B will be described.
As described above, in the present embodiment, it is possible to avoid the occurrence of failure such as overwriting due to overflow when writing the video data to the FIFO by the function of the first compression ratio adjusting unit 30A. However, it should be noted here that there is sufficient space in the video capture memory, and it is determined that adjustment of the quantization parameter qP (in other words, the compression rate) is not necessary in comparison with the free space. Even in such a case, it is not preferable that encoded data encoded at a compression ratio exceeding the target maximum compression ratio is written indefinitely in the video capture memory. The reason is as follows. If writing of encoded data encoded at a compression rate exceeding the target maximum compression rate occurs frequently, the free space in the video capture memory rapidly decreases with time. Even if the free capacity of the video capture memory rapidly decreases with time, in the present embodiment, the occurrence of failure such as overwriting due to overflow is avoided by the action of the first compression rate adjusting means 30A. However, when the compression rate is rapidly lowered, the image quality of the video displayed on the display device is rapidly lowered. In other words, even if there is enough space in the video capture memory, if you allow unlimited writing of encoded data encoded at a compression rate that exceeds the target maximum compression rate, the image quality of the subsequent video will be greatly reduced. There is a risk of letting you. The second compression rate adjusting means 30B is provided to avoid the occurrence of such a situation.

  More specifically, each time the second compression ratio adjustment unit 30B encodes the divided data and writes it to the FIFO, the data size of the encoded data of the divided data encoded so far is the maximum encoding for the divided data. It is determined whether or not the data size is exceeded. If so, the quantization parameter is increased by a predetermined amount (or at a predetermined rate). Here, the maximum encoded data size is a data size for one field that is set in advance so that a large deterioration in image quality can be prevented by performing the above-described adjustment by the second compression ratio adjusting unit 30B.

  For example, the second compression rate adjusting unit 30B, when writing the encoded result of the divided data corresponding to the area A in FIG. 6 to the FIFO, the data size E1 of the encoded result and the target maximum encoding for one field. The maximum encoded data size corresponding to the area A calculated by multiplying the data size (formula (1)) by a predetermined coefficient M1 is compared, and if the former exceeds the latter, the quantization parameter is raised. Note that the encoding data size of one field is estimated from the data size E1, the size of the estimated encoding data size is compared with the maximum encoding data size for one field, and the quantization parameter qP is increased when the former is larger. You may do it. Further, at the time when the encoding result of the divided data corresponding to the area B in FIG. 6 is written to the FIFO, the second compression ratio adjusting unit 30B determines the data size of the encoding result and the encoded data corresponding to the area A. The data size sum E12 is compared with the maximum encoded data size corresponding to the region A + region B calculated by multiplying the target maximum encoded data size for one field by a predetermined coefficient M2, and the former exceeds the latter. Increase the quantization parameter. Then, at the time when the encoding result of the divided data corresponding to the area C in FIG. 6 is written into the FIFO, the second compression ratio adjusting unit 30B determines the data size, area A, and area B of the encoded result of the divided data. The maximum encoded data size corresponding to the region A + region B + region C calculated by multiplying the sum E123 of the data size of each encoded data corresponding to each of the above and the target maximum encoded data size for one field by a predetermined coefficient M3 And the quantization parameter is increased when the former exceeds the latter.

  As described above, in the image processing LSI of this embodiment, the video data to be processed in the high-quality video processing is stored in the video capture memory after being encoded by the external video data encoder 30. For this reason, it is possible to reduce the capacity of the external memory that plays the role of the video capture memory and the bandwidth when accessing the external memory as compared with the conventional technique that does not perform encoding. In addition, when the external memory having the same storage capacity as that of the conventional memory is used, the number of encoded data that can be stored in the external memory is larger than that of the conventional memory. It becomes possible to perform more precise high-quality video processing than before.

  In addition, in this embodiment, since the quantization parameter (in other words, the compression rate of the video data) is adjusted according to the free space of the video capture memory, overflow occurs when writing encoded data to the external memory. There will be no failure such as overwriting.

Although one embodiment of the present invention has been described above, it goes without saying that the following modifications may be added to this embodiment.
(1) In the above embodiment, the case where the quantization parameter adjustment by the first compression rate adjustment unit 30A and the quantization parameter adjustment by the second compression rate adjustment unit 30B are performed has been described. However, since only the adjustment by the first compression rate adjustment unit 30A can avoid the memory destruction at the time of writing the encoded data to the video capture memory, the second compression rate adjustment unit 30B may be omitted. . In the above embodiment, the compression rate is adjusted by adjusting the quantization parameter qP. However, the parameters for adjusting the compression rate may be determined in relation to the encoding algorithm.

(2) In the above-described embodiment, the image processing unit including the high-quality video processing unit 50 and the correction processing unit 60 is provided in the subsequent stage of the external video data decoder 40. However, image processing for executing other types of image editing processing May be provided after the external video data decoder 40. In the above embodiment, the case where the compression rate is adjusted for each divided data obtained by dividing video data for one field in the interlace method into a plurality of raster scan directions has been described. However, one frame in the progressive method is used. The compression ratio may be adjusted for each piece of divided data obtained by dividing the video data for a plurality of pieces in the raster scan direction. In short, any mode may be used as long as the compression ratio is adjusted for each divided data obtained by dividing video data for one screen into a plurality of raster scan directions.

(3) In the above embodiment, an external memory connected to the SDRAM-I / F 100 is used as a memory for capturing a video that is a target of high-quality video processing. However, as described above, the shared work memory 120 (that is, the memory built in the image processing LSI 1) may serve as the video capture memory. According to such an aspect, the external memory is omitted. Thus, the manufacturing cost of the entire image display system can be reduced.

(4) DSP (Digital
A program that causes a computer such as a signal processor to function as an image processing unit for performing various image processing on the video data output from the external video data encoder 30, the external video data decoder 40, and the external video data decoder 40. May be provided. Here, as a specific provision mode of such a program, a mode in which the program is recorded and distributed on a computer-readable recording medium such as a CD-ROM (Compact Disk-Read Only Memory) or an electric such as the Internet A mode of distribution by downloading via a communication line is conceivable.

  DESCRIPTION OF SYMBOLS 10 ... Analog front end, 20 ... Video decoder, 30 ... External video data encoder, 40 ... External video data decoder, 50 ... High quality video processing part, 60 ... Correction processing part, 70 ... OSD drawing part, 80 ... Display Control unit 90 ... Host CPU-I / F, 100 ... SDRAM-I / F, 110 ... ROM-I / F, 120 ... Shared work memory.

Claims (3)

An encoding unit that encodes video data to be processed in a raster scan direction and writes it to a video capture memory;
Read and decode data for the screen necessary for image processing from the video capture memory, and output the video data of each screen;
An image processing unit that performs image processing on the video data output from the decoding unit and outputs the video data;
When the data obtained by dividing the video data for one screen to be processed into a plurality of parts in the raster scan direction is divided data, the encoding unit includes:
When each divided data is encoded, the data size when the video data for one screen is encoded is estimated from the encoding result up to the previous divided data, and the data size and the free space of the video capture memory are estimated. An image processing apparatus comprising: a compression ratio adjusting unit that adjusts a parameter for encoding the divided data based on the above. The encoding unit is
A second compression ratio adjusting unit configured to adjust the parameter so that the compression efficiency is increased when the estimated data size exceeds a predetermined maximum encoded data size when encoding each divided data; The image processing apparatus according to claim 1, further comprising: The video data input to the encoding unit is data obtained by subjecting an analog video signal to A / D conversion,
The video data output from the image processing unit and other video data are combined to generate and output video data for one screen to be displayed on a display device. The image processing apparatus according to claim 1 or 2.

Images