KR20060099026A - Method for processing zoom-in function of image - Google Patents

Abstract

The present invention relates to an image magnification processing method. When the screen mode of various image processing apparatuses is a k times magnification mode, only the image data corresponding to the k times the size of the display screen is subsampled and stored in the decoding buffer. In addition, when the region of interest of the user changes, only the changed region is converted into image data before subsampling and output to the screen.

According to the present invention, it is possible to enlarge and display an image in real time while minimizing memory use, and to maintain a good screen quality when the screen is enlarged. Accordingly, an efficient zoom-in function may be implemented in various image processing apparatuses that do not have enough memory.

Image magnification, zoom-in, JPEG, memory, subsampling

Description

Method for processing zoom-in function of image}

1 is an overview of an image data processing process in an image processing apparatus;

2 is a flow chart of one embodiment according to the present invention;

3 is an overview of enlargement of a screen;

4 is a flow chart of another embodiment according to the present invention;

5 and 6 are schematic diagrams for explaining when changing the ROI.

* Explanation of symbols for main parts of the drawings

11: JPEG file 12: Decoder

13: decoding buffer 14: display screen

15: original video

16: Enlarged image data stored in decoding buffer

52,53: Image data displayed on the actual screen

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image magnification processing method. In particular, in various image processing apparatuses having insufficient memory, zoom-in processing of image data recorded in an image file such as JPEG in real time using a minimum memory It's about how you can do it.

Currently, various kinds of devices that process an image and output it on a display screen are used. The image data is recorded through various compression methods for efficiently compressing image data such as JPEG, and the image processing apparatus decodes the compressed image data and outputs it to the display screen.

Such various image processing apparatuses may be configured not only to output image data to a display screen but also to be processed in various forms such as a slide show function or a zoom-in function.

On the other hand, in the case of an image processing apparatus that has sufficient memory capacity to process an image, such as a digital camera or a personal computer (PC), the entire image can be decoded and resided in a memory and the screen can be enlarged in real time. In an image processing apparatus having insufficient memory capacity, it is not easy to enlarge an entire image in real time due to memory limitations.

An example of an image processing apparatus having insufficient memory is a personal video recorder (PVR). The PVR is a new concept of digital recorder that records video signals to a hard disk drive (HDD), unlike a VCR (Video Cassette Recorder) that records video signals to magnetic tape.

These PVRs basically provide a built-in hard disk drive (HDD) in the set-top box or television set main unit to record broadcast programs, but MP3 in the TV set using external memory can be used by users who want various functions. Various functions are added, such as an electronic album function that allows you to listen to music or view stored pictures through a digital camera. Accordingly, a zoom-in function such as a photo is also required.

However, when the memory is limited as described above, real-time magnification of the entire image is almost impossible.

Of course, in order to suppress the use of additional memory, it is possible to enlarge the area selected by the user more than twice on the currently displayed screen, but in this case, deterioration of image quality occurs, so that when the size of the screen is large, It has a big impact. In addition, before the screen is enlarged, the user may first select an area to be enlarged, and the decoding process may be performed only on the selected area, but the disadvantage of having to select the area again each time is to view another area.

Accordingly, the present invention has been made to solve the above problems, and a method for zooming-in image data in real time without deterioration of image quality even in various image processing apparatuses such as a PVR that does not have sufficient memory capacity. The purpose is to provide.

In order to achieve the above object, a method for processing an enlargement of an image according to the present invention includes: decoding image data recorded in an image file in units of blocks; Determining whether the currently selected screen mode is a kx magnification mode; As a result of the determination, sub-sampling the decoded data to fit the size (M * N) of the display screen when the screen is not in the magnification mode, and kM * kN when the magnification is k times. Subsampling to fit; Storing the subsampled image data in a decoding buffer; And returning to the decoding or outputting the image data stored in the decoding buffer on the display screen according to whether or not decoding of the last block of the image data is completed.

The subsampling may be configured to consist of four Y signals per four pixels and one U signal and one V signal, respectively.

Another embodiment of the image magnification processing method according to the present invention is to decode the image data recorded in the image file and sub-sample it to fit kM * kN, which is k times the display screen size (M * N). Storing in a decoding buffer; A conversion / output step of selecting M * N size image data based on a predetermined reference point among the image data stored in the decoding buffer, converting the image data into the image data before the subsampling is performed, and outputting the image data to the display screen; And when the region of interest of the user is changed, performing the conversion / output step again based on a reference point indicating the change.

The conversion / output step may be configured to process based on the center of the image at the first execution.

In the converting / outputting step, when the ROI is changed, only the data of the changed region may be converted into image data before the subsampling is performed.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

Referring to FIG. 1, an example of a process of processing an image in the image processing apparatus will be described briefly. Image data read from an image file storing compressed image data, such as a JPEG file 11, is decoded by the decoder 12. Once stored in the decoding buffer 13, the image data stored in the decoding buffer 13 is output through the display screen (14).

FIG. 1 shows an embodiment in which the size of the display screen 14 is M * N, the size of the image recorded in the JPEG file is 5M * 5N, and the size of the decoding buffer 13 is 2M * 2N. Here, M and N are integers, and the number of pixels constituting the display screen 14, and '2' means a multiple to enlarge the image. Further, it is assumed that the image processing apparatus decodes the compressed image data 15 recorded in the JPEG file in units of scans of blocks consisting of 8 * 8 pixels.

These assumptions are set for better understanding of the description.The size of the original image (5M * 5N), the size of the decoding buffer (2M * 2N), the multiple to be enlarged (k = 2), and the size of the block in scan units ( 8 * 8) etc. can be arbitrarily comprised as needed.

Referring to FIG. 2, a decoding process for realizing an image magnification function according to the present invention will be described.

First, in order to be able to decode regardless of the size of the image data, decoding is performed in units of scans of 8x8 blocks (S21). Here, the data size when decoding one block is 192 (8 * 8 * 3) bytes in total since 3 bytes (Y signal, U signal, V signal) are required for each pixel.

Then, it is checked whether the currently set screen mode is a normal screen mode or a 2x zoom-in mode (S22).

As a result of the check in step S22, in the normal screen mode, the data decoded in units of scans of 8 * 8 blocks in step S21 is sub-sampled to match the size of the display screen M * N. That is, in the normal screen mode, the size of the image finally seen on the screen is related to the size of the display screen. Therefore, in step S21, the data decoded in units of scans of 8 * 8 blocks is used to match the display screen size M * N Will be sampled.

However, as a result of the confirmation in step S22, in the 2x magnification mode, the maximum area that can be output on the display screen is 4 times the screen size (2M * 2N), so sub-sampled decoded image data to fit this size. (S24).

At this time, assuming that 4: 2: 0 subsampling is performed in steps S23 and S24, since four Y signals, one U signal, and one V signal are output per four pixels, it occurs when decoding one block. The size of the data is 96 (8 * 8 + 4 * 4 * 2) bytes.

The video data sub-sampled in the step S23 or the step S24 is stored in the decoding buffer for output to the display screen (S25).

Steps S21 to S26 described above are executed for every one block scan unit of the corresponding image data (S26), and when the decoding of the corresponding image data is finished, the image data stored in the decoding buffer is output to the display screen (S27). .

That is, when the user selects the 2x magnification mode as shown in the example shown in FIG. 3, the M * N partial region 15-1 of the original image data 15 is enlarged by 2 times to sub-sample the image data 16. As a decode buffer. Therefore, the 2X magnification image data 16 stored in the decoding buffer may be displayed on the display screen in real time as the user changes the ROI with various direction keys.

Another embodiment according to the present invention will now be described with reference to FIG. 4.

First, the video data recorded in the video file is decoded and subsampled to fit kM * kN, which is k times the display screen size M * N, and stored in the decoding buffer (S41). In this case, assuming 4: 2: 0 subsampling is performed on an 8 * 8 block, it occurs when decoding one block because four Y signals, one U signal, and one V signal are generated per four pixels. The data size is 96 (8 * 8 + 4 * 4 * 2) bytes.

Now, the enlarged image data stored in the decoding buffer is output to the display screen. The decoding buffer has k times enlarged image data (kM * kN) and the display screen size is M * N. Which part of the stored image data should be determined.

This part can be selected in various ways, but it is preferable to set the center of the image as a reference point at the time of initial output (S42).

Now, data of size M * N is selected with respect to the image data stored in the decoding buffer based on the reference point, and the selected image data is converted into image data before subsampling is performed (S43). As an example, when the subsampling in step S41 is 4: 2: 0, the pixel converts the data into 4: 4: 4 data having both a luminance signal and a chrominance signal, and outputs the data to the screen.

When the ROI is changed from the user, the process of executing step S43 is repeated based on the changed reference point (S44).

For better understanding of the description, referring to FIG. 5, since the M * N size image data is enlarged twice by subsampling and stored in the decoding buffer after step S41, the image data having 2M * 2N size is included in the decoding buffer. 51 is stored.

When step S43 is executed for the first time, image data 52 of M * N size among the image data stored in the decoding buffer is output on the display screen. At this time, if the first reference point is set to the center of the image in step S42, the user sees a screen enlarged twice with respect to the center (A) of the image.

Now, when the user moves the region of interest to the upper left as shown in the example shown in FIG. 6 by using the arrow keys, the image data 53 of twice the M * N size stored in the decoding buffer with respect to the new reference point B is shown. Is displayed on the display screen. In other words, the screen can be enlarged in real time.

On the other hand, in step S43, the video data in the subsampled state stored in the decoding buffer is converted into video data before subsampling is output and output. In this case, when the user changes the region of interest, the new M * N region 53 may be configured to convert only the changed region such as the hatched portion without data conversion. Other parts can be output using image data which has already been converted.

The present invention is not limited to the above embodiments and can be variously modified and implemented by those skilled in the art without departing from the technical spirit of the present invention.

According to the present invention, since the image can be enlarged and displayed in real time while minimizing the memory usage, an efficient zoom-in function can be implemented in various image processing apparatuses in which there is not enough memory. In addition, it is possible to maintain excellent image quality even when the image is enlarged because the image is decoded and displayed in the original image instead of being enlarged by simply applying interpolation to the area currently displayed on the screen.

In addition, 4 pixels without storing the decoded data of the entire image and sub-sampling only the area (2M * 2N) according to the magnification factor, or without storing all the data necessary to represent one pixel Storing four luminance signals and one color difference signal per unit can greatly reduce the total memory usage.

Claims (6)

Decoding the image data recorded in the image file in units of blocks; Determining whether the currently selected screen mode is a kx magnification mode; As a result of the determination, the sub-sampling of the decoded data to fit the size (M * N) of the display screen when the screen is not in the magnification mode, and kM * kN in the case of the k-fold magnification mode. Subsampling to fit; Storing the subsampled image data in a decoding buffer; And And returning to the decoding step or outputting the image data stored in the decoding buffer to a display screen according to whether or not decoding of the last block of the image data is completed. The method of claim 1, And the sub-sampling comprises four Y signals per four pixels and one U signal and one V signal, respectively. Decoding the image data recorded in the image file, subsampling to fit kM * kN, which is k times the display screen size (M * N), and storing the result in a decoding buffer; A conversion / output step of selecting M * N size image data based on a predetermined reference point among the image data stored in the decoding buffer, converting the image data into the image data before the subsampling is performed, and outputting the image data to the display screen; And If the user's region of interest changes, the step of performing the conversion / output step based on the reference point indicating the change again. The method of claim 3, wherein And the converting / outputting step is configured to process based on the center of the image at the first execution. The method of claim 3, wherein And the converting / outputting step is configured to convert only the data of the changed region into the image data before the subsampling when the region of interest is changed. The method according to any one of claims 1 to 5, And the image file comprises a JPEG file.

Images