JP2002369194A - Image encoding equipment and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To maintain high encoding efficiency in a scene in which prediction of movement is difficult. SOLUTION: In an alternative frame producing part 10, one or a plurality of frames of image data stored in a first frame memory 1 are inputted, the inputted frames are processed, and an alternative frame is produced. In a frame substituting part 11, the alternative frame produced by the alternative frame producing part 10 is compared with an original frame stored in the first frame memory 1 in the following manner, or one or both of the frames are estimated, or estimation and comparison are performed. When it is judged that the alternative frame is more suitable for encoding, the alternative frame substitutes for the frame stored in the first frame memory 1. As a result, the alternative frame whose encoding efficiency is higher than that of the inputted image frame is encoded from the first frame memory 1.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image encoding apparatus and an image encoding method for compressing and encoding image data,
The present invention relates to an image transmission apparatus that transmits compressed image data having the functions or procedures of these apparatuses or an image recording apparatus that records compression-encoded image data on a medium such as a magnetic or optical medium.

[0002]

2. Description of the Related Art FIG.
49, No. 4 shows a conventional image coding apparatus. 1 is a first frame memory for temporarily storing input image data, 2 is an orthogonal transform unit for orthogonally transforming the image data stored in the first frame memory 1, and 3 is an orthogonal transform coefficient which is an output of the orthogonal transform unit 2. Quantizing unit 4 quantizes the output data from quantizing unit 3 and other encoding-related data in a predetermined format using variable length coding (hereinafter referred to as VLC: Variable Length Cod).
ing) a VLC unit, 5 is a transmission buffer for temporarily storing the output of the VLC unit 4 for transmission to the outside of the device, 6 is an inverse quantization unit that inversely quantizes the output data of the quantization unit 3, 7
Is an inverse orthogonal transform unit that performs inverse orthogonal transform on the orthogonal transform coefficient output from the inverse quantization unit 6; 8 is a second frame memory that temporarily stores image data output from the inverse orthogonal transform unit 7; Reference numeral 9 denotes a motion prediction unit that uses the image data stored in the second frame memory to generate predicted reference image data of image data to be encoded next.

Next, the operation will be described. First, image data input from outside the device is stored in the first frame memory 1. When the image coding apparatus is compliant with, for example, ISO / IEC 13818-2, the frame order of the image data may be different between the input order and the coding order. In that case, the image data is rearranged in the first frame memory 1. Is performed. Next, one frame according to the encoding order is selected from the image data stored in the first frame memory 1 and output to the orthogonal transform unit 2. The image data output from the first frame memory is directly input to the orthogonal transformation unit 2 as it is, or is compared with the prediction reference image data generated by the motion prediction unit 9, and only the difference data is sent to the orthogonal transformation unit 2. Is entered. The orthogonal transformation unit 2 performs orthogonal transformation on the input image data. The quantization unit 3 quantizes the orthogonal transform coefficients output from the orthogonal transform unit 2 with an appropriate step size. The step size can be changed in units of slices or macroblocks, for example,
Normally, a smaller step size results in a high-definition image but a larger code amount. Conversely, a larger step size reduces the image quality but reduces the code amount. It is necessary to set an appropriate value according to the conversion efficiency.

The orthogonal transform coefficients quantized by the quantization unit 3 are variable-length coded together with other coding-related data in the VLC unit 4. The data variable-length coded by the VLC unit 4 is temporarily stored in the transmission buffer 5 as a coded bit string. The data stored in the transmission buffer 5 is then output outside the device.

[0005] The orthogonal transform coefficients quantized by the quantization unit 3 are also input to an inverse quantization unit 6. The inverse quantization unit 6 inversely quantizes the input orthogonal transform coefficients with a step size equal to that of the quantization unit 2. In the inverse orthogonal transform unit 7, the inverse quantization unit 6
The inverse orthogonal transform is performed on the orthogonal transform coefficient inversely quantized by, and is converted into image data. The image obtained by the inverse orthogonal transform unit 7 is stored in the second frame memory 8. The image data stored in the second frame memory 8 is transmitted to the orthogonal transformation unit 2,
It is equivalent to a decoded coded bit string generated through a series of processing in the quantization unit 3 and the VLC unit 4. Motion prediction unit 9
Then, using the image data stored in the second frame memory, predicted reference image data of image data to be encoded thereafter is generated. Using this predicted reference image data, only the difference data from the actual image data is encoded, thereby realizing a high compression rate.

[0006]

However, in the conventional image coding apparatus, it is high in that only the difference data from the actual image data is coded using the predicted reference image data generated by the motion prediction section 9. Although it achieves a compression ratio, it is difficult to generate accurate predicted reference image data in scenes with irregular motion or scenes in which frames with low correlation with neighboring frames suddenly appear, such as flashes. There is a problem that efficiency is reduced.

For this reason, in order to maintain a high compression rate in a state where the coding efficiency is reduced, it is possible to perform processing such as increasing the step size in the quantization unit 3, for example. There is a new problem that image quality is deteriorated.

Accordingly, the present invention has been made to solve such a problem, and provides an image coding apparatus capable of maintaining high coding efficiency even in a scene where motion estimation is difficult. The purpose is to:

[0009]

According to the present invention, there is provided an image coding apparatus for compressing and coding image data, comprising: a substitute frame generating means for processing a frame of image data to generate a substitute frame; Frame replacement means for replacing the image frame with the substitute frame,
It is characterized by having.

[0010] Further, the alternative frame generating means converts the luminance level or the color difference level of each frame of the image data and its adjacent frame when generating the alternative frame.

[0011] Further, the substitute frame generating means is characterized in that when generating the substitute frame, the input image frame is combined with its neighboring frame or a neighboring frame of the input image frame.

Further, when generating the substitute frame, the substitute frame generating means detects a motion between each frame of the image data and its neighboring frame, and processes the frame based on the detection result. And

[0013] Further, the frame replacement means evaluates the substitute frame, and replaces the image frame with the substitute frame based on the evaluation result.

Further, the frame replacing means calculates the coding efficiency of the substitute frame as the evaluation of the substitute frame, and replaces the image frame with the substitute frame based on the calculation result. .

Further, the frame replacing means evaluates the image frame and replaces the image frame with the substitute frame based on the evaluation result.

Further, the frame replacement means calculates the coding efficiency of the image frame as the evaluation of the image frame, and replaces the image frame with the substitute frame based on the calculation result. I do.

The frame replacing means evaluates the image frame and the substitute frame, compares the evaluation results, and replaces the image frame with the substitute frame based on the comparison result. Features.

Further, the frame replacing means evaluates the image frame and the substitute frame based on the coding efficiency and the degree of approximation, and compares the evaluation results to determine whether the coding efficiency of the substitute frame is the code of the image frame. When the replacement efficiency is higher and the degree of approximation between the substitute frame and the image frame is high, the image frame is replaced with the substitute frame.

Still further, the present invention further comprises a holding means for holding an input image frame required for generating the substitute frame when the substitute frame generating means processes the frame of the image data and generates the substitute frame. Features.

Further, the substitute frame generating means outputs information on a method of generating a substitute frame to another processing unit in the encoding device.

Further, the substitute frame generation means is characterized in that information on a method of generating a substitute frame is multiplexed on encoded data and transmitted to a decoding side.

[0022]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiment 1 FIG. 1 is a block diagram showing Embodiment 1 of an image encoding device according to the present invention. In the figure, reference numeral 1 denotes a first frame memory for temporarily storing input image data and rearranging it in the encoding order, and 2 denotes a first frame memory.
An orthogonal transform unit for orthogonally transforming the image data stored in the frame memory 1, a quantizing unit 3 for quantizing the orthogonal transform coefficient output from the orthogonal transforming unit 2, and output data from the quantizing unit 3 and other data A VLC unit for performing variable-length encoding of the encoding-related data in a predetermined format, 5 a transmission buffer for temporarily storing the output of the VLC unit 4 for transmission to the outside of the apparatus, and 6 an output of the quantization unit 3 Inverse quantization unit for inversely quantizing data, 7
Is an inverse orthogonal transform unit that performs inverse orthogonal transform on the orthogonal transform coefficient output from the inverse quantization unit 6; 8 is a second frame memory that temporarily stores image data output from the inverse orthogonal transform unit 7; Reference numeral 9 denotes a motion prediction unit that uses the image data stored in the second frame memory to generate predicted reference image data of image data to be encoded next, and 10 denotes a first frame memory 1
An alternative frame generation unit 11 generates an alternative frame of the image data stored in the first frame memory 1, and a frame replacement unit 11 that replaces the image data stored in the first frame memory 1 with the alternative frame generated by the alternative frame generation unit 10.

Next, the operation will be described. First, image data input from outside the device is stored in the first frame memory 1. The image encoding device is, for example, ISO / IEC 13818-2
, The frame order of the image data may be different between the input order and the encoding order. In that case, the image data is rearranged in the first frame memory 1.

Next, one frame according to the encoding order is selected from the image data stored in the first frame memory 1 and output to the orthogonal transform unit 2. The image data output from the first frame memory 1 is directly input to the orthogonal transformation unit 2 as it is, or is compared with the prediction reference image data generated by the motion prediction unit 9, and only the difference data is compared with the orthogonal transformation unit 2. Is input to

The orthogonal transformation unit 2 performs an orthogonal transformation on the input image data. The quantization unit 3 quantizes the orthogonal transform coefficients output from the orthogonal transform unit 2 with an appropriate step size. The orthogonal transform coefficients quantized by the quantization unit 3 are variable-length coded together with other coding-related data in the VLC unit 4. The data variable-length coded by the VLC unit 4 is temporarily stored in the transmission buffer 5 as a coded bit string. The data stored in the transmission buffer 5 is output outside the device.

The orthogonal transform coefficients quantized by the quantization unit 3 are also input to the inverse quantization unit 6. In the inverse quantization unit 6,
The input orthogonal transform coefficients are inversely quantized with a step size equal to that of the quantization unit 2. The inverse orthogonal transform unit 7 performs an inverse orthogonal transform on the orthogonal transform coefficient inversely quantized by the inverse quantization unit 6, and converts the orthogonal transform coefficient into image data. The image obtained by the inverse orthogonal transform unit 7 is stored in the second frame memory 8. The motion prediction unit 9 uses the image data stored in the second frame memory 8 to generate predicted reference image data of image data to be encoded thereafter.

Here, in the first embodiment, the substitute frame generation unit 10 inputs one or a plurality of frames of the image data stored in the first frame memory 1 and processes the input frame to generate a substitute frame. Generate The frame replacement unit 11 compares the replacement frame generated by the replacement frame generation unit 10 with the original frame stored in the first frame memory 1 as described below, or evaluates one or both of them. In the evaluation and comparison, if it is determined that the substitute frame is more suitable for encoding, the frame stored in the first frame memory 1 is replaced with the substitute frame.

FIG. 2 shows an example 1 of processing an image in the substitute frame generating section 10. The frame # 2 of the image data before processing is, for example, a frame at the moment when the screen background flashes, as shown in FIG. Since such a frame has a low correlation with neighboring frames # 1 and # 3, high coding efficiency cannot be expected.

On the other hand, the processed frame # 21 is a substitute frame of the frame # 2 generated by performing the conversion of the luminance level or the color difference level on a part of the frame # 1. Since the frame # 21 is generated by processing the frame # 1, a high correlation with the frame # 1 is maintained. Therefore, if the frame # 2 is replaced with the frame # 21 for encoding, high encoding efficiency can be expected in the frame # 21 and the subsequent frames.

In this example, the immediately preceding frame #
1 is processed to generate an alternative frame, but the frame to be processed need not be limited to the immediately preceding frame, and may be, for example, the immediately succeeding frame (in the case of FIG. 2, frame # 3). In addition, a frame at a position several frames away may be processed.

FIG. 3 shows a second example of image processing in the substitute frame generator 10. The frame # 2 of the image before the processing is a frame having a low correlation with the neighboring frames # 1 and # 3 generated by, for example, noise or the like generated at the time of recording. The processed frame # 21 is the frame #
# 2 generated by synthesizing frame # 1 and frame # 2
Is an alternative frame. Frame # 21 is frame #
Since the correlation with frame # 1 is higher than that of frame # 2, if frame # 2 is replaced with frame # 21 and encoded, high encoding efficiency can be expected in frame # 21 and subsequent frames.

In this example, two consecutive frames including the original frame are synthesized, but the present invention is not necessarily limited to this. For example, frame # 1 and frame # 3
May be used as an alternative frame of the original frame # 2. In this case, there is a possibility that the picture of the substitute frame # 21 is far away from the original frame # 2, but it is of course possible to substitute such a substitute frame and encode it. This is because, in a scene where the pattern is complicated and the movement is intense, it is often impossible to perceive even if a frame far away from the original frame # 2 is inserted. Because there is. Further, even if the pattern of the substitute frame # 21 is far from the original frame # 2, it is determined that the substitute frame # 21 does not conform to the motion prediction by the evaluation of the substitute frame in the frame replacement unit 11 described later and the comparison with the original frame. In this case, the replacement of the original frame by the substitute frame is not performed. This is the same for other image processing examples.

For example, frame # 1 and frame # 2
And frame # 3 may be used to combine three or more frames. Further, it is possible to perform the above-described conversion of the luminance level or the conversion of the color difference level on the synthesized frame.

FIG. 4 shows a third example of image processing in the substitute frame generator 10. The motion of the object in frames # 1 and # 2 of the image before processing is detected, and frame # 1 is processed using the detected motion information to generate frame # 21. Since the frame # 21 is generated by processing the frame # 1, a high correlation is maintained between the frame # 1 and the frame # 21. Coding efficiency can be expected. Further, since the frame # 2 is a frame processed from the frame # 1 using the motion information, it is easy to perform the motion prediction between the frame # 1 and the frame # 21. The probability that an incorrect prediction is made can be reduced.

In this example, motion prediction is performed between two consecutive frames. However, the present invention is not necessarily limited to this policy. For example, prediction may be performed between frame # 2 and a frame earlier than frame # 1. This point is the same in the above-described processing examples 1 and 2.

Further, the prediction is not necessarily performed only with the previous frame. For example, the prediction of the frame # 2 may be performed with the frames after the frame # 3, and the prediction of the frame # 1 may be performed. The prediction may be performed by using both the frame # 3 and the frame # 3. This point is also the same in the processing examples 1 and 2 described above.

Next, a method of evaluating the substitute frame generated by the substitute frame generation unit 10 in the frame replacement unit 11 will be described. Items to be evaluated include, for example, the coding efficiency of the original frame, the coding efficiency of the substitute frame, and the degree of approximation between the substitute frame and the original frame. The coding efficiency is the autocorrelation between horizontal pixels,
Alternatively, the evaluation can be made by calculating the autocorrelation between vertical pixels or the autocorrelation between frames. The degree of approximation can be evaluated by calculating the sum of absolute differences between the substitute frame and the original frame.

Here, when using these alternative frame evaluation methods, the frame replacing unit 11 obtains and evaluates only the coding efficiency of the original frame without obtaining the coding efficiency of the alternative frame, and evaluates the code of the original frame. Only when the coding efficiency exceeds a predetermined threshold, the coding efficiency of the replacement frame is replaced with the replacement frame, or the coding efficiency of the replacement frame is obtained by evaluating only the coding efficiency of the replacement frame without obtaining the coding efficiency of the original frame. Is replaced with an alternative frame only when the value falls below
Only one of the frames, such as the coding efficiency of the original frame or the substitute frame, may be evaluated, and the replacement of the substitute frame may be performed based on the evaluation result. It is also possible to evaluate the conversion efficiency and the like, compare the evaluation results, and replace the replacement frame with the replacement frame if the replacement frame has better coding efficiency. In this way, when an alternative frame is evaluated and the input image frame is replaced with the alternative frame based on the evaluation result, it is possible to prevent an inappropriate alternative frame from being selected, while evaluating the input image frame and When the input image frame is replaced with the substitute frame based on the result, it is possible to prevent the replacement with the substitute frame without meaning.

Further, the frame replacement unit 11 uses an approximation degree other than the coding efficiency as an evaluation method, and evaluates only one approximation degree of the approximation degree of both frames without evaluating the coding efficiency. Alternatively, the evaluation may be performed by combining two evaluation parameters of the coding efficiency and the degree of approximation. For example, when evaluating based on both the coding efficiency and the degree of approximation, if it is determined that the coding efficiency of the substitute frame is better than the coding efficiency of the original frame and that the degree of approximation of both is high. The image frames stored in the first frame memory are replaced with substitute frames.

As described above, the frame replacement unit 11
Is evaluated by comparing the substitute frame generated by the substitute frame generation unit 10 with the original frame stored in the first frame memory 1, and when it is determined that the substitute frame is more suitable for encoding, , The original frame stored in the first frame memory 1 is replaced with an alternative frame.

Therefore, according to the first embodiment, even when a scene in which the motion is irregular or a scene in which a frame having a low correlation with the neighboring frame appears due to flash or the like occurs, for example, the alternative frame generation unit 10 Creates a substitute frame, and the frame replacing unit 11 compares the substitute frame with the original frame stored in the first frame memory 1, and replaces the substitute frame with the substitute frame when it determines that the substitute frame is more suitable. As a result, accurate prediction reference image data can be generated by motion prediction using the substitute frame, and a decrease in coding efficiency can be prevented.

In the above description, the alternative frame generation method in the alternative frame generation unit 10 has been described not to be used in a processing unit other than the alternative frame generation unit 10, for example, the motion prediction unit 9 or the like. Of course, a processing unit other than 10 may input a method of generating a substitute frame in the substitute frame generating unit 10 and use it for its own processing. That is, for example, when generating the predicted reference image data to be compared with the substitute frame, the motion prediction unit 9 may input the method of generating the substitute frame from the substitute frame creation unit 10 to improve the prediction accuracy. good. In this way, even when the alternative frame generation unit 10 performs the conversion of the luminance level or the chrominance level, the motion prediction unit 9 performs the motion prediction while considering the change amount of the luminance level or the chrominance level caused by the conversion. Is performed, the prediction accuracy can be improved.

For example, when the substitute frame generation unit 10 performs processing using motion prediction, the motion prediction unit 9
By inputting a method of generating a substitute frame to be processed using motion prediction and matching the substitute frame generation unit 10 with the prediction parameters, the prediction accuracy can be improved.

The substitute frame generation unit 10 not only outputs the method of generating the substitute frame to the motion prediction unit 9 and the like in the substitute frame generation unit 10, but also outputs the method to the decoding side via the VLC unit 4 and the transmission buffer 5. May be transmitted. With this configuration, when the luminance level and the chrominance level are converted by the substitute frame generation unit 10 of the encoding apparatus, the decoding apparatus considers the amount of change in the luminance level and the chrominance level caused by the conversion. In addition, processing such as image quality correction can be performed. This is the same in Embodiment 2 described later.

Embodiment 2 FIG. 5 is a block diagram showing Embodiment 2 of an image coding apparatus according to the present invention. In the figure, reference numeral 2 denotes an orthogonal transform unit for orthogonally transforming input image data, 3 denotes a quantizer for quantizing orthogonal transform coefficients output from the orthogonal transform unit 2, and 4 denotes output data from the quantizer 3; In addition, a VLC unit for performing variable-length encoding of encoding-related data in a predetermined format, 5 is a transmission buffer for temporarily storing the output of the VLC unit 4 for transmission to the outside of the apparatus, and 6 is a transmission buffer for the quantization unit 3. An inverse quantization unit that inversely quantizes the output data, 7 is an inverse orthogonal transform unit that performs inverse orthogonal transform on the orthogonal transform coefficient output from the inverse quantizer 6, and 8 is an output of the inverse orthogonal transform unit 7. A second frame memory for temporarily storing image data, 9
Is a motion prediction unit that uses the image data stored in the second frame memory 8 to generate predicted reference image data of image data to be encoded next, and 10 is a substitute frame that generates a substitute frame of the input image data. The generation unit 11 selects one of the input image data and the alternative frame generated by the alternative frame generation unit 10, and outputs a frame replacement unit 12 that records the image data input to the alternative frame generation unit 10. This is a memory for storing. The difference from the first embodiment is that the image encoding apparatus according to the second embodiment does not have the first frame memory 1 and cannot use the first frame memory 1, so that the memory 12 for holding the input image frame necessary for creating the substitute frame is used. Is provided.

Therefore, in the case of the image coding apparatus according to the second embodiment, image data input from outside the apparatus is directly input to each of the substitute frame generation unit 10 and the frame replacement unit 11.

Here, the substitute frame generation unit 10 has an interface with the memory 12 and stores the image data input through the interface in the memory 12. For this reason, the memory 12 stores the image data input to the alternative frame generation unit 10 in the past.
The input image frames to be stored may be only one or more frames required for generating the substitute frame. The substitute frame generation unit 10 processes past input image data and the like recorded in the memory 12 in the same manner as in the first embodiment and the like, generates a substitute frame, and outputs the substitute frame to the frame replacement unit 11.

The frame replacement unit 11 evaluates and compares the image data input from the outside of the apparatus and the substitute frame generated by the substitute frame generation unit 10 in the same manner as in the first embodiment. One is selected and output to the orthogonal transform unit 2.

Therefore, according to the second embodiment, even when a scene in which the motion is irregular or a scene in which a frame having a low correlation with a neighboring frame appears due to a flash or the like occurs, for example, the first embodiment can be used. As in the case, the substitute frame generating unit 10 creates a substitute frame, and the frame replacing unit 11 compares the substitute frame with the original frame stored in the first frame memory 1, and the substitute frame is more suitable. When it is determined that the frame is replaced with the substitute frame, accurate prediction reference image data can be generated by motion prediction using the substitute frame, and a decrease in coding efficiency can be prevented.

Further, in the second embodiment, since the memory 12 for temporarily storing the input image frame used for generating the substitute frame is provided, unlike the image encoding apparatus of the first embodiment shown in FIG. Even an image encoding device that does not have the first frame memory 1 that rearranges image data in the encoding order can generate a substitute frame.

In the second embodiment, since the first frame memory 1 is not provided and cannot be used, the memory 12 is provided as holding means for holding an input image frame necessary for creating a substitute frame. However, the present invention is not limited to this, and it is sufficient if an input image frame necessary for creating an alternative frame can be held. For example, when only the previous input image frame is needed for creating an alternative frame, May provide an input image frame delay circuit in the substitute frame creating unit 10 instead of the memory 12 or, if a plurality of previous input image frames are necessary for creating the substitute frame, substitute the memory 12 The frame creating unit 10 may be provided with a plurality of stages of delay circuits for the input image frame.

In the first and second embodiments, the image encoding apparatus and the image encoding method have been described. However, according to the present invention, compressed image data having the function or procedure of these apparatuses is transmitted. The present invention is also applicable to an image transmission apparatus that performs compression and encoding, and an image recording apparatus that records compression-encoded image data on a medium such as magnetism or light.

[0053]

According to the present invention, a frame of image data is processed, a substitute frame is generated, and the image frame is replaced with the substitute frame. It is possible to perform coding by substituting with a substitute frame having good coding efficiency, and maintain high coding efficiency.

In particular, when the substitute frame is generated, the luminance level or the color difference level of each frame of the image data and its neighboring frames is converted, so that the substitute frame which is close to the original frame and has high coding efficiency can be obtained. Can be generated.

Further, when generating the substitute frame, the input image frame and its neighboring frame or the neighboring frame of the input image frame are synthesized, so that the substitute frame which is close to the original frame and has high coding efficiency can be obtained. Can be generated.

Further, when generating the substitute frame, the motion between each frame of the image data and its neighboring frame is detected, and the frame is processed based on the detection result. In addition, it is possible to generate a substitute frame having high coding efficiency.

Further, since the substitute frame is evaluated based on the coding efficiency and the like, and the image frame is replaced with the substitute frame based on the evaluation result, it is possible to prevent an inappropriate substitute frame from being selected.

Also, the input image frame is evaluated based on the coding efficiency and the like, and the image frame is replaced with the substitute frame based on the evaluation result. Therefore, it is possible to prevent the substitute frame from being meaninglessly replaced.

When the substitute frame generation means processes a frame of image data to generate a substitute frame,
Since a holding means for holding an input image frame necessary for generating the substitute frame is provided, an appropriate substitute frame can be efficiently generated without a frame memory for rearranging the encoded data in the encoding order. be able to.

Further, since information relating to the method of generating a substitute frame is transmitted to each section of the coding apparatus, the substitute frame can be coded efficiently.

Further, since the information on the method of generating the substitute frame is multiplexed with the encoded data and transmitted to the decoding side, the decoding side can also process the substitution frame efficiently.

[Brief description of the drawings]

FIG. 1 is a block diagram showing Embodiment 1 of an image encoding device according to the present invention.

FIG. 2 is a diagram showing an image processing example 1 in an alternative frame generation unit.

FIG. 3 is a diagram showing an image processing example 2 in the alternative frame generation unit.

FIG. 4 is a diagram illustrating a third example of image processing in the alternative frame generation unit.

FIG. 5 is a block diagram showing Embodiment 2 of the image encoding device of the present invention.

FIG. 6 is a block diagram showing a conventional image encoding device.

[Explanation of symbols]

 REFERENCE SIGNS LIST 1 first frame memory 2 orthogonal transformation unit 3 quantization unit 4 VLC unit 5 transmission buffer 6 inverse quantization unit 7 inverse orthogonal transformation unit 8 second frame memory 9 motion compensation unit 10 alternative frame generation unit (alternate frame generation means) 11 Frame replacement unit (frame replacement means) 12 Memory (recording means)

Continued on the front page F term (reference) 5C059 KK22 LA00 LA09 MA23 MC11 MC38 ME01 NN21 PP16 SS06 SS11 TA03 TB04 TC12 TC42 TD04 UA02 UA05 UA32 5J064 AA01 BA16 BB03 BC01 BC14 BC16 BD03

Claims (14)

[Claims] 1. An image encoding apparatus for compressing and encoding image data, an alternative frame generating means for processing a frame of the image data and generating an alternative frame, and a frame replacing means for replacing the image frame with the alternative frame. An image encoding device, comprising: 2. The image according to claim 1, wherein the substitute frame generating means converts a luminance level or a color difference level of each frame of the image data and its neighboring frames when generating the substitute frame. Encoding device. 3. The substitute frame generating means, when generating a substitute frame, combines an input image frame and its neighboring frame or a neighboring frame of the input image frame. Image encoding device. 4. The method according to claim 1, wherein, when generating the substitute frame, the substitute frame generation means detects a motion between each frame of the image data and its neighboring frame, and processes the frame based on the detection result. The image encoding device according to any one of claims 1 to 3, wherein 5. The frame replacement unit according to claim 1, wherein the frame replacement unit evaluates the substitute frame, and replaces the image frame with the substitute frame based on an evaluation result. Image encoding device. 6. The frame replacement means calculates the coding efficiency of the replacement frame as the evaluation of the replacement frame, and replaces the image frame with the replacement frame based on the calculation result. The image encoding device according to claim 5. 7. The image processing apparatus according to claim 1, wherein the frame replacement unit evaluates the image frame, and replaces the image frame with the substitute frame based on an evaluation result. Image encoding device. 8. The frame replacement means calculates, as the evaluation of the image frame, a coding efficiency of the image frame, and replaces the image frame with the substitute frame based on the calculation result. The image encoding apparatus according to claim 7, wherein: 9. The frame replacement means evaluates the image frame and the substitute frame, compares the evaluation results, and replaces the image frame with the substitute frame based on the comparison result. The image encoding device according to any one of claims 1 to 4, wherein 10. The frame replacement means evaluates the image frame and the substitute frame based on the coding efficiency and the degree of approximation, and compares the evaluation results to determine whether the coding efficiency of the substitute frame is the code of the image frame. 10. The image encoding device according to claim 9, wherein the image frame is replaced with the alternative frame when the replacement efficiency is higher and the degree of approximation between the alternative frame and the image frame is high. 11. When the substitute frame generating means processes a frame of image data and generates a substitute frame, the substitute frame generating means includes a holding means for holding an input image frame required for generating the substitute frame. The image encoding device according to any one of claims 1 to 10, wherein 12. The image encoding apparatus according to claim 1, wherein said alternative frame generating means outputs information relating to a method of generating an alternative frame to another processing unit in the encoding apparatus. apparatus. 13. The image encoding apparatus according to claim 1, wherein said alternative frame generating means multiplexes information on a method of generating an alternative frame into encoded data and transmits the multiplexed data to a decoding side. apparatus. 14. An image encoding method for compressing and encoding image data, wherein a frame of the image data is processed, an alternative frame is generated, and the image frame is replaced with the alternative frame. Method.