JP2000032467A - Image processor, image processing method, and provision medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To reduce the calculation amount required for motion detection and to reduce circuit scale by sharing a motion detection circuit in an image encoder for performing pixel number conversion. SOLUTION: For source image data inputted from a terminal 11, a pixel number is converted in an image number converter 21, and reduced image data are generated. A motion vector detector 24 detects a motion vector on a generated reduced image and stores it in a vector memory 26. The motion vector detector 25 detects the motion vector on a source image, based on the reduction rate of the motion vector on the reduced image stored in the vector memory 26.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus and method, and a providing medium, and more particularly, to reducing the amount of calculation required for motion detection by sharing a motion detection circuit.
The present invention relates to an image processing apparatus and method capable of reducing a circuit scale, and a providing medium.

[0002]

2. Description of the Related Art When digitally recording image data on an optical disk, a magnetic tape, or the like, or transmitting image data via a predetermined transmission medium, the image data is encoded and compressed to reduce the data amount. That is being done.

For example, MPEG (Moving Picture Coding Exp)
(erts Group) compression method, I (Intra) picture, P
(Predictive) picture, or B (Bidrectionally-predi
ctiv) The picture of each frame is predicted in any of the prediction modes of the pictures, and the prediction error is encoded and transmitted. Basically, since only the prediction error is transmitted, the data amount can be compressed as compared with the case where the image data of each frame is transmitted as it is.

[0004] For example, as shown in FIG.
To fifteen frames to a group of pictures (GOP) as one unit for compression processing. The frames represented by “I”, “P”, and “B” in FIG. 9 represent frames encoded by I pictures, P pictures, and B pictures, respectively.

[0005] As an I-picture frame, its image information is coded and transmitted as it is. This coding is called intra coding.

As shown in FIG. 9A, a P-picture frame is basically a prediction error signal calculated using a past I-picture or P-picture frame as a prediction image. Are encoded and transmitted. This coding is called forward prediction coding.

Further, as a frame of a B picture,
Basically, as shown in FIG. 9 (B), a prediction residual signal calculated using both reference frames in the past and the future as predicted images is encoded. This encoding is called bidirectional encoding.

FIG. 10 is a block diagram showing a configuration example of a conventional image encoding device. The image encoding device in FIG.
It comprises a motion detection circuit 101 and an encoding circuit 102.

A frame memory group 111 in the motion detecting circuit 101 stores the input image data. The motion vector detector 112 calculates the motion of the input image data based on the image data stored in the frame memory group 111 and data indicating a picture type, for example, designation of an I picture, a P picture, or a B picture. Find the vector.

A motion vector is detected for each pixel data called a macro block. Macro block is 16 ×
It is composed of luminance signals corresponding to 16 pixels (dots). This luminance signal is further divided into blocks Y1 to Y4 in units of 8 × 8 dots. And this
A 16 × 16 dot luminance signal is composed of an 8 × 8 dot Cb signal and an 8 × 8 dot Cb signal.
It corresponds to a × 8 dot Cr signal. The pixel data input by the raster scan is converted into a block scan, and output to the prediction encoder 113 in the encoding circuit 102 together with the motion vector.

A prediction encoder 113 determines a prediction mode of a reference block based on a frame memory group 119 for storing predicted image data and a motion vector supplied from the motion vector detector 112, and determines the determined prediction mode. , Forward prediction,
Switching between backward prediction and bidirectional prediction is performed.

The DCT unit 114 converts the prediction error from the prediction encoder 113 into, for example, a DCT (Discrete Cosine Transform
m) (Discrete cosine transform) to generate DCT coefficient data.

The quantizer 115 quantizes the coefficient data from the DCT unit 114 based on information sent from the quantization scale setting unit 122 to generate quantized data.

A VLC (Variable Length Code) unit (variable length encoder) 116 performs variable length coding on the quantized data from the quantizer 115 to generate variable length code data and outputs it to the transmission buffer 121. .

The quantization scale setting unit 122 determines a quantization scale value so as to keep the transmission data amount of the transmission buffer 121 constant, and outputs the value to the quantization unit 115.

The inverse quantizer 117 inversely quantizes the coefficient data quantized by the quantizer 115. IDCT (Inverse D
An iscrete cosine transform unit 118 performs an inverse DCT process on the inversely quantized coefficient data to return an error signal. The adder 123 adds the prediction signal input from the motion compensator 120 to the error signal input from the IDCT unit 118, and locally decodes the error signal. Since the combined image data is used as a reference image for motion compensation prediction, the frame memory group 1
19 is stored.

As a general detection method of the motion detection circuit 101 in FIG. 10, there is a full search method (full research). This full search method is a method for searching all the pixels within the search range in the reference frame. As a detection method of the motion detection circuit 101 in FIG. 10, there is a method called a hierarchical search disclosed in Japanese Patent Application Laid-Open No. Hei 9-82080. By using this method, the amount of calculation required for motion detection can be reduced compared to using the full search method, and the circuit size can be reduced.

In the image coding apparatus of FIG. 10, when the generated code amount increases, the DCT coefficient value is reduced by increasing the quantization step of the quantizer 115, so that the resolution in the DCT space is reduced. In order to solve the problem, the number of pixels of the image data to be encoded is converted, and by reducing the number of macro blocks,
There is a device that encodes image data while maintaining good image quality.

FIG. 11 is a block diagram showing an example of the configuration of an image coding apparatus proposed earlier by the present applicant as Japanese Patent Application No. 10-121531. The image coding apparatus of FIG. 11 converts the number of pixels of image data according to the feature amount of input image data.

The motion detecting circuit 131 has a function similar to that of the motion detecting circuit 101 shown in FIG. 10, and detects a motion vector and a difference signal from the image data input from the terminal 130. The detected motion vector and the difference signal are supplied to the pixel number determination circuit 132.

The pixel number determination circuit 132 determines the number of pixels of the image data to be encoded from the supplied motion vector and the difference signal, and outputs a signal indicating whether or not to convert the number of pixels to the switch 134. , And outputs pixel number information indicating the number of pixels to be converted to the pixel number conversion circuit 135.

The pixel number determination circuit 132 compares the supplied motion vector and the difference signal with, for example, the sum of the absolute value of the motion vector in the frame, the magnitude of the difference signal, etc., with a specific threshold value. It is determined whether to convert the number of pixels of the image data.

FIG. 12 shows frames F0 to F1.
FIG. 14 is a diagram for explaining the transition of the sum of absolute values of the motion vector No. 4; In the example of FIG. 12, the pixel number determination circuit 132 determines the conversion of the number of pixels in the frame F9 because the sum of absolute values of the frames after the frame F9 is larger than the threshold value.

The delay circuit 133 delays the output of the image data input from the terminal 130 until the motion detection circuit 131 and the pixel number determination circuit 132 determine the number of pixels to be encoded. The time of the signal supplied from 132 and the time of the image data to be output are adjusted.

The switching unit 134 is a circuit for determining the number of pixels 13
If the signal from 2 is a signal indicating the conversion of the number of pixels, the output of the image data is switched to the pixel number conversion circuit 135 side. Switch.

The pixel number conversion circuit 135 includes a switch 13
The number of pixels is converted according to the signal input from No. 4. FIG.
FIG. 8 is a diagram for explaining a pixel number conversion process in the pixel number conversion circuit 135. The pixel number conversion circuit 135 calculates the number of pixels determined by the pixel number determination circuit 132, for example, (horizontal n pixels × vertical m pixels) from the input (horizontal N pixels × vertical M pixels) original image data. It is converted into image data of the number of pixels.

Since the macroblock is a rectangular area of 16 pixels × 16 pixels, the number of macroblocks of the original image before pixel number conversion is (N / 16) × (M / 16). On the other hand, the number of macroblocks after pixel number conversion is (n / 16) × (m / 16). That is, the number of macroblocks is converted to (n ×
m) / (N × M) times. Further, since the coding amount of the motion vector is almost proportional to the macroblock, the coding amount of the motion vector can be reduced to (n × m) / (N × M) times.

In the case of the example shown in FIG.
Determines the conversion of the number of pixels of the image data in the frame F9. The motion detection circuit 137 detects the motion of the selected image signal, and outputs a motion vector and image data to the encoding circuit 136.

When the number of pixels is switched in the frame F9 as in the example of FIG. 12, the encoding circuit 136 outputs image data of horizontal N pixels × vertical M pixels up to the frame F8 as shown in FIG. After the frame F9, the pixel data is encoded as pixel data of horizontal n pixels × vertical m pixels.

The encoding circuit 136 transfers the reduced amount of the motion vector code to the increased amount of the DCT coefficient code. For this reason, the quantization step value can be reduced, and a decrease in resolution in the DCT space can be suppressed.

FIG. 15 is a block diagram showing an example of the configuration of an image coding apparatus proposed earlier by the present applicant as Japanese Patent Application No. 10-137676. The image encoding apparatus of FIG. 15 encodes image data of a plurality of pixels in parallel with respect to one input signal, and selects the image data at an output stage.

The original image data input from the terminal 150 is supplied to the motion detection circuit 151 and to the pixel number conversion circuit 154.

The motion detection circuit 151 has the same function as the motion detection circuit 101 in FIG. 10, detects a motion vector from the supplied image data, and outputs the motion vector to the encoding circuit 152. The encoding circuit 152 is also the encoding circuit 10 of FIG.
2 has the same function as that of the first embodiment, and encodes input image data. Then, the encoding circuit 152 outputs the encoded image data to the synchronization delay circuit 153.

The pixel number conversion circuit 154 converts the number of pixels of the input original image data, outputs the converted pixel number to the motion detection circuit 155, and outputs a sequence end code and a sequence header indicating the end of the bit stream. Is output to the encoding circuit 156, which means that.

The motion detection circuit 155 detects a motion vector of the image data whose pixel number has been converted, and
6 is output.

The encoding circuit 156 has a function similar to that of the encoding circuit 102 shown in FIG. 10, and encodes input image data so that a bit stream can be easily switched. For example, in the case of MPEG, the encoding circuit 156
Pixel number conversion circuit 1 before all GOPs as pre-processing
At the end of the GOP, a sequence end code consisting of all "0" is written at the end of the GOP, in addition to writing the sequence header input from 54 and further facilitating insertion of the sequence end code. . In this way, when the number of pixels is converted, a boundary as a sequence is clearly provided there, and a malfunction such as encoding a frame image after the number of pixels conversion using a frame before the number of pixels conversion as a reference image, Bit stream switching is performed smoothly for encoding.
Then, the encoding circuit 156 outputs the encoded bit stream to the synchronization delay circuit 157.

The synchronization delay circuits 153 and 157 detect the synchronization signal in the input bit stream while transmitting the synchronization signal to each other, and delay the bit stream until a new synchronization signal is received. The time of the converted image data is adjusted. In the case of MPEG, for example, the synchronization delay circuits 153 and 157 search for a sequence header and delay the bit stream until a new sequence comes.

The feature quantity extraction circuit 159
A feature amount is extracted based on the information on the encoding input from the second and 156, and the feature amount is output to the pixel number determination circuit 160 in accordance with the synchronization signal input from the synchronization delay circuits 153 and 157. For example, the feature amount extraction circuit 159
Receives the S / N ratio (signal-to-noise ratio) and the like from the encoding circuits 152 and 156, calculates the average S / N ratio in the GOP, and outputs it to the pixel number determination circuit 160 as a feature amount.

The pixel number determining circuit 160 determines a bit stream to be selected based on the feature amount input from the feature amount extracting circuit 159, and outputs a switching signal to the bit stream switching circuit 158. For example, the pixel number determination circuit 160 receives the average S / N ratio as the feature amount from the feature amount extraction circuit 159, and switches the switching signal so as to select the bit stream having the best average S / N ratio. Output to the circuit 158.

The bit stream input from the bit stream switching circuit 158 is temporarily stored in the transmission buffer 159 and then transmitted. Transmission buffer 159
Feeds back information such as the buffer occupancy to the bit stream switching circuit 158.

The bit stream switching circuit 158 includes:
Based on the feedback information from the transmission buffer 159, the transmission buffer 159 is occupied at a constant rate.
Adjust and send out the bitstream. For example, in the case of MPEG, the bit stream switching circuit 158
A sequence code is inserted at the end of GOP1 and adjusted so that the transmission buffer 159 does not break down.
Output GOP2.

As described above, by encoding image data having a plurality of pixels and selecting a bit stream at the output stage, encoded data having good image quality can be output.

[0043]

However, FIG.
Alternatively, an image encoding device that converts the number of pixels of image data as in the image encoding device of FIG. 15 needs to have a plurality of motion detection circuits, and performs motion detection for each image data having a different number of pixels. This has to be performed, the amount of calculation required for motion detection increases, and the circuit scale increases.

The present invention has been made in view of such a situation, and it is possible to reduce the amount of calculation required for motion detection and reduce the circuit scale.

[0045]

An image processing apparatus according to claim 1, wherein a pixel number conversion means for converting the number of pixels of the input first image information, and a pixel converted by the pixel number conversion means. First motion vector detection means for detecting a motion vector of second image information having a number, and the first motion vector detection means based on the motion vector of the second image information detected by the first motion vector detection means. A second motion vector detecting means for detecting a motion vector of the first image information, and a motion vector of the first image information and the first image information, or the second image information and the second image information. It is characterized by comprising encoding means for encoding a motion vector, and output means for outputting encoded information encoded by the encoding means.

According to a third aspect of the present invention, in the image processing method, a pixel number converting step of converting the number of pixels of the input first image information, and a second step having the number of pixels converted in the pixel number converting step are performed. A first motion vector detecting step of detecting a motion vector of the image information; and a motion vector of the first image information based on the motion vector of the second image information detected in the first motion vector detecting step. A second motion vector detecting step of detecting
Image information and a motion vector of the first image information,
Or an encoding step of encoding the second image information and a motion vector of the second image information, and an output step of outputting the encoded information encoded in the encoding step. I do.

The recording medium according to claim 4, wherein a pixel number conversion step for converting the number of pixels of the input first image information, and a second image having the number of pixels converted in the pixel number conversion step A first motion vector detecting step of detecting a motion vector of information; and a motion vector of the first image information based on a motion vector of the second image information detected in the first motion vector detecting step. A second motion vector detecting step of detecting, and a code for coding the motion vector of the first image information and the first image information, or the motion vector of the second image information and the motion vector of the second image information Computer-readable program that causes an image processing apparatus to execute a process including an encoding step and an output step of outputting encoded information encoded in the encoding step. And providing the program.

An image processing apparatus according to claim 1,
In the image processing method described in (1) and the providing medium described in (4), the number of pixels of the input first image information is converted, and the motion vector of the second image information having the converted number of pixels is The detected motion vector of the first image information is detected based on the detected motion vector of the second image information.

[0049]

DESCRIPTION OF THE PREFERRED EMBODIMENTS Embodiments of the present invention will be described below. In order to clarify the correspondence between each means of the invention described in the claims and the following embodiments, each means is described. When the features of the present invention are described by adding the corresponding embodiment (however, an example) in parentheses after the parentheses, the result is as follows. However, of course, this description does not mean that each means is limited to those described.

The image processing apparatus according to the first aspect includes a pixel number conversion means (for example, the pixel number converter 21 in FIG. 2) for converting the number of pixels of the input first image information, and the pixel number conversion means. First motion vector detecting means (for example, the motion vector detector 24 in FIG. 2) for detecting a motion vector of the second image information having the number of pixels converted by the means, and the first motion vector detecting means. A second motion vector detecting means (for example, a motion vector detector 25 in FIG. 2) for detecting a motion vector of the first image information based on the detected motion vector of the second image information; Encoding means for encoding the first image information and the motion vector of the first image information, or the second image information and the motion vector of the second image information (for example, the encoding circuit 16 of FIG. 1) If, and an outputting means for outputting a coded information (e.g., the encoding circuit 16 of FIG. 1) by the encoding means.

FIG. 1 is a block diagram showing a configuration example of an image coding apparatus to which the image processing apparatus of the present invention is applied.

The motion detection circuit 12 converts the number of pixels of the original image data input from the terminal 11, calculates a motion vector, and sets image data having two different numbers of pixels and two corresponding motion vectors. Are output at the same time.

The pixel number determination circuit 13 is
Information, such as the sum of the absolute value of the motion vector in the frame or the magnitude of the difference signal, is compared with a specific threshold value, thereby determining the conversion of the number of pixels and switching the determined information. The signals are output to the switches 14 and 15 as signals.

The switches 14 and 15 are connected to the pixel number determination circuit 1
3 based on the switching signal supplied from
A, 14B, 15A, and 15B are switched.

The encoding circuit 16 encodes the image data input from the switch 14 and the motion vector input from the switch 15 and outputs a bit stream.

FIG. 2 is a block diagram showing a configuration example of the motion detection circuit 12. The motion detection circuit 12 outputs the original image data input from the terminal 11 to the pixel number converter 21 and to the frame memory group 23.

The pixel number converter 21 converts the number of pixels of the original image data input from the terminal 11, generates reduced image data, and stores the generated reduced image data in the frame memory group 2.
2 is output. Frame memory group 2
Reference numeral 2 stores the input reduced image data.

The motion vector detector 24 detects a motion vector from the reduced image data stored in the frame memory group 22 and outputs the detected motion vector to the vector memory 26. Vector memory 2
Numeral 6 stores the input motion vector.

The frame memory group 23 stores the original image data input from the terminal 11.
The motion vector detector 25 detects a motion vector on the original image data from a motion vector on the reduced image data recorded in the vector memory 26.

Next, the operation of the image coding apparatus according to the present embodiment will be described. The original image data input from the terminal 11 is input to the motion detection circuit 12. The original image data input to the motion detection circuit 12 is output to the pixel number converter 21 and also to the frame memory group 23.

The pixel number converter 21 converts the number of pixels of the original image data input from the terminal 11 to generate reduced image data. Then, the pixel number converter 21 outputs the generated reduced image data to the frame memory group 22. The frame memory group 22 stores the input reduced image data.

The motion vector detector 24 detects a motion vector from the reduced image data stored in the frame memory group 22 and outputs the detected motion vector to the vector memory 26. The vector memory 26 stores the input motion vector.

On the other hand, the motion vector detector 25 calculates the motion vector on the original image data from the motion vector on the reduced image data stored in the vector memory 26 by calculation. The processing executed here will be described in detail below.

FIG. 3 is a diagram for explaining the relationship between a reduction ratio and a motion vector when a reduced image is generated by reducing an original image. As shown in FIG. 3, when the number of pixels of the original image is converted and the image is reduced, the amount of motion also decreases in proportion to the reduction ratio.

That is, as shown in FIG.
When the original image having the number of vertical pixels M (FIG. 4A) is converted into a reduced image having the number of horizontal pixels n × the number of vertical pixels m (FIG. 4B), it corresponds to the search range A on the original image. The area on the reduced image that corresponds to the search range B. The search range B is (n × m) / (N
× M). Therefore, when the search is performed using the same motion detection algorithm, when the motion is detected in the search range B, the amount of calculation required for the motion detection can be smaller than when the motion is detected in the search range A.

Therefore, as shown in FIG. 3, the motion vector on the original image and the motion vector on the reduced image are almost proportional to the reduction ratio. In order to reduce the amount of calculation required for the motion vector MV, as shown in FIG.
_B , and the motion vector detector 25 detects the motion vector MV according to the compression ratio as shown in FIG.
_A is calculated, and the search range C around the motion vector MV _A is indicated.
(Narrower than the search range A).

When the motion detection circuit 12 completes the calculation of the motion vector on the image having two different numbers of pixels at the same time, it outputs the motion vector and the image data.
That is, as shown in FIG. 2, a frame memory group 22 to output the reduced image data in which the number of pixels is reduced, vector memory 26 outputs a motion vector MV _B of the reduced image data, a frame memory group 23 is the original outputs the image data, the motion vector 25 to output the motion vector MV _a on the original image data.

The pixel number determination circuit 13 is
Reduced image data supplied from the motion vector MV _B of the reduced image data, the original image data, and information obtained from the motion vector MV _A on the original image data, for example, the absolute value sum or the difference of the motion vectors in the frame By comparing the magnitude of the signal or the like with a specific threshold value, it is determined whether to convert the number of pixels. Then, the pixel number determination circuit 1
3 indicates a case where the number of pixels is converted (for example, a case where the sum of absolute values of the motion vectors is large).
And let switched to the contact point 14A and 15A, respectively, the reduced image data and the motion vector MV _B by selectively output, if you do not convert the number of pixels (e.g., when the absolute value sum of the motion vector is small), the switch 14 and the switch 15 , let switched to the contact point 14B and 15B, respectively, selects the original image data and the motion vector MV _a, to output.

The encoding circuit 16 encodes the image data input from the switch 14 and the motion vector input from the switch 15, and outputs a bit stream.

[0070] As described above, according to the image coding apparatus of the embodiment of FIG. 1, as shown in FIG. 5, on the basis of the motion vector MV _B detected on the reduced image, the motion of the original image when detecting the vector MV _a, since so as to search the search range C of the search range B and the diagram of FIG 4 (B) 5 (a) , than when searching the search range a shown in FIG. 4 (a) The amount of calculation can be reduced.

The motion vector detector 25 detects the motion vector from the motion vector
The motion vector MV _A on the original image is obtained by using the motion vector MV _B on the reduced image stored in, so that the calculation amount can be reduced and the circuit size can be reduced. Can be.

In the above description, a motion vector having two different numbers of pixels is detected.
1, frame memory group 22, motion vector detector 24,
And by increasing the number of vector memories 26,
Further, it is possible to detect a motion vector on image data having a plurality of different numbers of pixels.

FIG. 6 is a block diagram showing another example of the configuration of an image coding apparatus to which the present invention is applied. In the image encoding device of FIG. 1, pixel data having a different number of pixels is selected and then encoded. However, in the image encoding device of FIG. 6,
Image data of two different numbers of pixels are encoded in parallel and selected at the output stage.

The motion detection circuit 61 is configured similarly to the motion detection circuit 12 of FIG. 2, and outputs reduced image data, a motion vector on the reduced image data, original image data, and a motion vector on the original image data. The reduced image data and the motion vector on the reduced image data output from the motion detection circuit 61 are output to the encoding circuit 62, and the original image data and the motion vector on the original image data are encoded by the encoding circuit 63.
Is output to

The encoding circuits 62 and 63 have the same function as the encoding circuit 102 in FIG. 10, and perform encoding so that the bit stream can be easily switched.

The synchronization delay circuits 65 and 66 detect the synchronization code in the bit stream supplied from the encoding circuits 62 and 63 while transmitting the synchronization signal to each other, and delay the bit stream until a new synchronization signal comes. To perform time adjustment of the encoded image data. In the case of MPEG, for example, the synchronization delay circuits 65 and 66 search for a sequence header and delay the bit stream until a new sequence header comes.

The feature quantity extraction circuit 64 includes an encoding circuit 62,
The information about the encoding is received from 63, the feature amount is extracted, and the extracted feature amount is output to the pixel number determination circuit 67.
For example, the feature quantity extraction circuit 64 includes the encoding circuits 62 and 63
May be received, the average S / N ratio in the GOP is calculated, and the average S / N ratio is output to the pixel number determination circuit 67 as a feature amount.

The pixel number determination circuit 67 determines a bit stream to be selected based on the supplied characteristic amount, and outputs a switching signal to the bit stream switching circuit 68.
The bit stream switching circuit 68 selects and outputs a bit stream according to the switching signal.

The transmission buffer 69 stores the supplied bit stream, transmits it, and feeds back information such as the occupancy of the buffer to the bit stream switching circuit 68.

The bit stream switching circuit 68 adjusts the bit stream based on the feedback information from the transmission buffer 69 so that the occupancy of the transmission buffer 69 becomes constant, and outputs the adjusted bit stream.

As described above, the image coding apparatus of the embodiment shown in FIG. 6 has basically the same configuration as the image coding apparatus shown in FIG.
1, 155 and the pixel number conversion circuit 154
1 is different. According to this device,
As in the case of the motion detection circuit 12, the motion detection circuit 61 calculates image data and motion vectors of two different numbers of pixels, encodes them, and selects them as a bit stream at the output stage. Therefore, encoded data of an image having good image quality can be output.

In the image encoding apparatus of the embodiment shown in FIG. 6, two types of image data are encoded and selected at the output stage. The number of pixels is converted in accordance with the image data of, and more circuits corresponding to the encoding circuit 62 and the synchronization delay circuit 65 are provided, and the image data of more different numbers of pixels are encoded. You may make it select.

In the image coding apparatus shown in FIGS. 1 and 6, if the motion detection circuit shown in FIG. 7 is used instead of the motion detection circuits 12 and 61, the motion vectors detected on the images having different numbers of pixels are used. , Motion vectors of images having different numbers of pixels can be obtained.

That is, the pixel number conversion circuit 71 converts the number of pixels of the input original image data. Motion detection circuit 7
2 calculates a motion vector on the image whose number of pixels has been converted. The motion vector conversion circuit 73 obtains the calculated motion vector as a motion vector on an image.

FIG. 8 shows an original image (N horizontal pixels × M vertical pixels).
FIG. 7 is a diagram illustrating a relationship between motion vectors of a reduced image (n horizontal pixels × m vertical pixels). Motion vector on reduced image
When MV _B is multiplied by (N × M) / (n × m), the motion vector on the original picture
MV _A , the motion vector MV _B calculated on the reduced image
In the motion vector conversion circuit 73
By the (N × M) / (n × m) multiplied by 3, it is possible to obtain a motion vector MV _A on the original image.

In the above description, the motion vector on the original image is converted from the motion vector on the reduced image. Alternatively, the motion vector calculated on a different image may be converted.

Note that as a providing medium for providing a user with a computer program for performing the above-described processing,
In addition to recording media such as magnetic disks, CD-ROMs, and solid-state memories, communication media such as networks and satellites can be used.

[0088]

As described above, according to the image processing apparatus of the first aspect, the image processing method of the third aspect, and the recording medium of the fourth aspect, the first image input The number of pixels of the information is converted, a motion vector of the second image information having the converted number of pixels is detected, and a motion vector of the first image information is calculated based on the detected motion vector of the second image information. Since the detection is performed, the detection of the motion vector can be shared, the calculation amount required for detecting the motion vector can be reduced, and the circuit scale can be reduced.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration example of an image encoding device to which the present invention has been applied.

FIG. 2 is a block diagram illustrating a configuration example of a motion detection circuit in FIG. 1;

FIG. 3 is a diagram illustrating a relationship between a reduction ratio of a reduced image and a motion vector.

FIG. 4 is a diagram illustrating a process of detecting a motion vector.

FIG. 5 is another diagram illustrating a motion vector detection process.

FIG. 6 is a block diagram illustrating another configuration example of an image encoding device to which the present invention has been applied.

FIG. 7 is a block diagram illustrating another configuration example of the motion detection circuit.

FIG. 8 is a diagram illustrating a motion vector conversion process.

FIG. 9 is a diagram illustrating predictive coding.

FIG. 10 is a block diagram illustrating a configuration example of a conventional image encoding device.

FIG. 11 is a block diagram illustrating another configuration example of a conventional image encoding device.

FIG. 12 is a diagram illustrating a transition of a sum of absolute values of a motion vector.

FIG. 13 is a diagram illustrating a pixel number conversion process.

FIG. 14 is a diagram illustrating the relationship between pixel number conversion and a motion vector.

FIG. 15 is a block diagram showing still another example of the configuration of a conventional image encoding device.

FIG. 16 is a diagram illustrating a bit stream switching process.

[Explanation of symbols]

11, 11 terminal, 12, 61, 72 motion detection circuit, 13 pixel number determination circuit, 14, 15 switch, 16, 62, 63 encoding circuit, 21 pixel number converter, 22, 23 frame memory group, 24, 2
5 motion vector detector, 26 vector memory,
64 feature extraction circuit, 65, 66 synchronization delay circuit,
67 pixel number judgment circuit, 68 bit stream,
69 transmission buffer, 71 pixel number conversion circuit, 7
3 Motion vector conversion circuit

 ──────────────────────────────────────────────────続 き Continued on the front page (72) Inventor Shigeo Fujishiro 6-35, Kita-Shinagawa, Shinagawa-ku, Tokyo Inside Sony Corporation (72) Inventor Yoshihiro Murakami 6-35, Kita-Shinagawa, Shinagawa-ku, Tokyo Sony Corporation F term (reference) 5C059 KK15 KK19 LB04 MA00 MA05 NN02 NN11 NN28 PP05 PP06 PP07 RC16 RC31 RC37 SS11 SS13 SS20 UA02 UA39

Claims (4)

[Claims] 1. A pixel number conversion means for converting the number of pixels of input first image information, and a second image information detecting means for detecting a motion vector of second image information having the number of pixels converted by the pixel number conversion means. A second motion vector detecting means for detecting a motion vector of the first image information based on a motion vector of the second image information detected by the first motion vector detecting means. Means for coding the first image information and the motion vector of the first image information, or the second image information and the motion vector of the second image information; and the coding means. And an output unit for outputting encoded information encoded according to (1). 2. The method according to claim 1, wherein the second motion vector detecting means uses a conversion rate from the number of pixels of the first image information to the number of pixels of the second image information to calculate the motion of the second image information. The image processing apparatus according to claim 1, wherein a motion vector of the first image information is converted from a vector. 3. A pixel number conversion step of converting the number of pixels of the input first image information, and a second step of detecting a motion vector of the second image information having the number of pixels converted in the pixel number conversion step. A second motion vector detection step of detecting a motion vector of the first image information based on a motion vector of the second image information detected in the first motion vector detection step. Encoding a motion vector of the first image information and the first image information or a motion vector of the second image information and the second image information; and the encoding step. An output step of outputting the encoded information encoded in step (a). 4. A pixel number conversion step of converting the number of pixels of the input first image information, and a second step of detecting a motion vector of the second image information having the number of pixels converted in the pixel number conversion step. A second motion vector detection step of detecting a motion vector of the first image information based on a motion vector of the second image information detected in the first motion vector detection step. Encoding a motion vector of the first image information and the first image information or a motion vector of the second image information and the second image information; and the encoding step. And a computer-readable program for causing an image processing apparatus to execute a process including an output step of outputting encoded information encoded by the computer. Providing medium characterized.