JP3154741B2 - Image processing apparatus and system - Google Patents

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 for enlarging and reducing moving images and still images, and a method thereof.

[0002]

2. Description of the Related Art In recent years, many presentation images using natural images have been seen at various events and presentations, but in these applications, real-time enlargement of moving images, reduction of rotation and still images, and the like have been performed. Processing functions are required, and many image processing apparatuses having those functions have been commercialized. In this image processing apparatus, a coordinate conversion process of an image is performed. The coordinate conversion process includes the following two processes.

(1) Coordinate conversion processing for calculating the corresponding coordinates of the screen before processing from the coordinates on the screen after processing. (2) Interpolation processing for calculating data of coordinates before processing obtained by the above-described coordinate conversion (data generally does not exist because it is a non-grid point) from data around it. In particular, the interpolation process (2) is important because it affects the quality of a completed image, and a high-precision interpolation method is required to generate a high-quality image. As the interpolation method of the image data, from the coordinates of the original image obtained by the coordinate conversion, the “nearest neighbor method” in which the data of the nearest pixel is replaced with the processed data, or based on the four nearest pixels. There are known a "bilinear interpolation method" for performing interpolation by interpolation and a "cubic interpolation method" for complementing based on 16 pixels.

FIG. 3 is a block diagram showing a conventional image processing apparatus. 11 is a frame memory for temporarily pooling an input moving image or a still image, 12 is a buffer memory for a display image, and 13 is a coordinate conversion for performing coordinate conversion. The unit 14 is a data processing unit, and the coordinate conversion unit 13 generates coordinates “address” for performing “bilinear interpolation” or “cubic interpolation”.

[0005]

By the way, in the conventional image processing apparatus, the same complementary system is used in both the moving image processing and the still image processing.
In order to improve the quality of the processed image, it is necessary to use the "bilinear interpolation method" or the "cubic interpolation method". Need data. Therefore, when these interpolation methods are applied to moving image processing, the frame memory requires 4 or 16 accesses to one dot clock (10 MHz), so that 4 or 16 memories are actually used. And access in parallel. Therefore, not only is a memory for 4 or 16 screens necessary, but also data needs to be generated for each dot clock, so that the scale of hardware for processing them increases, resulting in a very large device. There was a problem that had to be.

The present invention has been made in view of such circumstances, and the circuit and apparatus can be reduced in scale by switching the data interpolation method associated with image processing for moving images and still images. It is an object of the present invention to provide a measurable image processing apparatus and its method.

[0007]

According to the present invention, a frame memory 101 for storing input image data as shown in FIG.
And a buffer memory 102 for storing display output data.
A coordinate conversion unit 103 for calculating original image coordinates from an address in the buffer memory 102 and outputting corresponding reference data to the frame memory 101;
Interpolation processing based on the reference data output from
A data processing unit 104 for generating data for each pixel and inputting the data to the buffer memory 102. When the input image is a still image, the data processing unit 104 performs an interpolation process based on a bilinear interpolation method or a cubic interpolation method. In this case, an image processing apparatus is characterized in that interpolation processing is performed based on the nearest neighbor method.

Further, the present invention relates to an image processing method for enlarging and reducing a moving image and a still image, wherein a data interpolation method accompanying the image processing is switched corresponding to each of the moving image and the still image. An image processing method is provided in which an image after processing is processed by an interpolation method that has lower quality than a still image. Preferably, the data interpolation processing of the moving image is performed based on the nearest neighbor method, and the data interpolation processing of the still image is performed based on the bilinear interpolation method or the cubic interpolation method.

[0009]

In FIG. 1, at the time of moving image processing, the coordinate conversion unit 10
3 generates coordinates of the “nearest neighbor method” and stores the coordinates in the frame memory 1.
Data is read from 01. The read data becomes a low quality display image by interpolation by the “nearest neighbor method” input to the buffer memory 102. However, there is no problem because a moving image as a characteristic of human vision cannot perceive the details, so that even a low quality image can withstand viewing.

At the time of still image processing, since the human sensation can be perceived to a considerable degree in detail, the coordinate conversion unit 103 generates coordinates of "bilinear interpolation" or "cubic interpolation".
The output data of the frame memory 101 is
At 04, interpolation processing is performed and input to the buffer memory 102. Since this still image processing operation is allowed to be performed at a somewhat low speed, the frame memory 101 is stored four times or 16 times.
No inconvenience occurs even if data is generated after multiple accesses.

[0011]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below in detail with reference to the embodiments shown in the drawings. This does not limit the present invention. FIG. 2 is a block diagram showing one embodiment of the present invention. In FIG. 2, reference numeral 1 denotes a frame memory which is composed of a dual port memory and stores input image data as it is, and 2 is a buffer memory which is composed of a dual port memory and stores display screen data. Reference numeral 5 denotes an image input unit for generating an input-side write command and an address for writing input image data to the frame memory 1. Reference numeral 3 denotes a coordinate conversion unit which calculates original image coordinates from display screen coordinates (addresses in the buffer memory 2) by a well-known affine transformation method by a coordinate conversion method at the time of moving image processing, and outputs reference data, that is, the nearest neighbor data. An address and a read command for extracting one piece of data required for the method from the frame memory 1 are generated. Further, the coordinate conversion unit 3
At the time of still image processing, the original image coordinates are calculated from the display screen coordinates (the address of the buffer memory 2) by the affine transformation method, and the reference data, that is, the data of four points required for the bilinear interpolation method or the tertiary interpolation method is required. An address and a read command for extracting 16 points of data from the frame memory 1 are generated.

During the still image processing, the degree of the shift of the non-grid points from the grid points generated during the affine transformation is output from the coordinate conversion unit 3 to the data processing unit 4. Reference numeral 4 denotes a data processing unit that operates only during still image processing. When the reference data is output from the frame memory 1 to the data bus B by the address and the read command output from the coordinate conversion unit 3, the data processing unit 4 calculates the data. Interpolation is performed using the reference data thus obtained and the deviation from the lattice point, data for each pixel is generated, and the data is output to the data bus B.

Reference numeral 6 denotes a light control unit which receives from the coordinate conversion unit 3 the data output from the frame memory 1 to the data bus B by the coordinate conversion unit 3 at the time of moving image processing.
An address and a write command for writing the data to the buffer memory 2 are output. Further, at the time of still image processing, the write control unit 6 receives the fact that the data processing unit 4 has output data on the data bus B,
An address and a write command for writing the data to the buffer memory 2 are output.

The operation in such a configuration will be described below. First, when moving image processing or still image processing is instructed to the coordinate conversion unit 3 and the data processing unit by the switching signal, image processing is performed correspondingly. At the time of moving image processing, the image input unit 5 continuously writes input image data in the frame memory 1 and constantly updates the content. The coordinate conversion unit 3 generates coordinates of the “nearest neighbor method” and reads data from the frame memory 1. The read data is the data bus B
And written to the buffer memory 2 by the write control unit 6.

During still image processing, the image input unit 5 stops writing input image data to the frame memory 1. The coordinate conversion unit 3 generates coordinates of “bilinear interpolation method” or “cubic interpolation method”, and the output data of the frame memory 1 is input to the data processing unit 4 one byte at a time, then complemented and subjected to the data bus B Via the write control unit 6
Is written to the buffer memory 2.

[0016]

According to the present invention, the nearest neighbor method is used in moving image processing, and the interpolation method with higher accuracy than the nearest neighbor method is used in still image processing. Therefore, the number of frame memories can be reduced. The data processing unit can be simplified, and the processing circuits and devices can be reduced in size.

[Brief description of the drawings]

FIG. 1 is a block diagram showing the principle of the present invention.

FIG. 2 is a block diagram showing an embodiment of the present invention.

FIG. 3 is a block diagram showing a conventional example.

[Explanation of symbols]

 Reference Signs List 1 frame memory 2 buffer memory 3 coordinate conversion unit 4 data processing unit 5 image input unit 6 write control unit B data bus

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁷ identification code FI H04N 5/262 G06F 15/66 355C (58) Investigated field (Int.Cl. ⁷ , DB name) G09G 5/36 G06F 3 / 153 320 G06T 3/40 G06T 13/00 H04N 5/262

Claims (1)

(57) [Claims] 1. A frame memory (101) for storing input image data, a buffer memory (102) for storing display output data, and reference image data corresponding to original image coordinates calculated from addresses of the buffer memory (102). Is output to the frame memory (101).
3) and a data processing unit (104) that performs interpolation processing based on reference data output from the frame memory (101), generates data for each pixel, and inputs the data to the buffer memory (102). An image processing apparatus, wherein an interpolation process is performed based on a bilinear interpolation method or a cubic interpolation method when an image is a still image, and an interpolation process is performed based on a nearest neighbor method when an input image is a moving image.