JPH02155381A - Moving picture/still picture converter - Google Patents

Abstract

PURPOSE:To prevent deterioration in picture quality due to noise when a hard copy is outputted as a still picture by judging the quantity of a moving picture with a discrimination means and allowing a processing means to apply smooth processing in the direction of time base if the movement of the moving picture is small as the result of discrimination. CONSTITUTION:Each picture element of a picture in 1st and 2nd frames is read from a started memory section 9 to output buses 18, 19 and each picture element of the readout picture is smoothed by a smoothing device 11. Let data of the picture elements in the 1st and 2nd frames be A1, A2 respectively, an output of the smoothing device 11 is expressed as (A1+A2)/2. On the other hand, the inter-frame difference of the picture read from the memory section 9 is calculated for each picture element by a difference device 10 and the calculated inter-frame difference is outputted to a comparator 12. Then the comparator 12 compares a difference data d1 with a threshold level (k), in the case of d1<k as the result of comparison, a processing switching section 13 is thrown to the position of (a). The picture element data B1 outputted to the input bus 20 and smoothed is expressed as B1=(A1+A2)/2.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to a moving image/still image conversion device that outputs moving images input from a television camera, VTI (etc.) as a still image to a CRT, printer, etc. It is.

[Prior Art] Images output by printers and the like were mainly generated by still image input devices such as image scanners, but recently, many media such as television cameras, video cameras, VTRs, etc. are being used as human-powered devices. It has become more complex. In particular, as the definition of high-definition televisions and the like becomes higher, there is a growing tendency to output high-definition images to printers and printing devices.

[Problems to be Solved by the Invention] However, in general, the input image of a television camera is a moving image, whereas the output image of a printer etc. is a still image, so it is necessary to convert the moving image into a still image. be.

The transformation from a moving image to a still image can basically be achieved by utilizing a ^/D converter and semiconductor memory, but in moving images, noise in one frame is reduced due to the integration effect of the visual system in the time axis direction. On the other hand, in a still image extracted from one frame, the visual system has little integration effect and image deterioration due to noise is noticeable. ) is subjected to ^/D conversion and stored in memory, and then when a hard copy is output as a still image from the printer section, there is a problem in that image deterioration due to noise is noticeable.

An object of the present invention is to provide a moving image/still image conversion device that can solve the above problems and prevent image quality deterioration due to noise.

[Means for Solving the Problems] In order to achieve such an object, the present invention provides a discriminating means for discriminating the magnitude of the motion of a moving image, and a method for determining whether the motion of the moving image is small as a result of the discriminating by the discriminating means. , processing means for performing smoothing processing in the time axis direction.

[Function] In the present invention, the magnitude of the motion of the moving image is determined by the determining means, and if the determined result is that the motion of the moving image is small, the processing means performs smoothing processing in the time axis direction.

[Example] Hereinafter, an example of the present invention will be described in detail with reference to the drawings.

FIG. 1 shows an embodiment of the invention.

In the figure, 1 is a moving image/still image conversion device, 2 is a video camera, 3 is a video tape recorder (VTR), 4 is a color printer, and 5 is a display such as a CRT.

FIG. 2 shows the configuration of the moving image/still image conversion device 1 shown in FIG. 1. As shown in FIG.

In the figure, 6 is a manual power switching unit, and 67 is an A/V switch that switches the manual power from the video camera 2 and VTR 3
The D conversion unit converts the video signal from the video camera 2 or VTR 3 switched by the input switching unit 6 into ^/D.
It is something that converts. 8 is a bus switching section. A memory section 9 stores video images obtained via the bus switching section 8. FIG. 3 shows a conceptual diagram of the internal structure of the memory section 9. Reference numeral 10 denotes a subtractor which calculates the difference between pixels read out from the memory section 9 between frames.
Reference numeral 11 denotes a smoother as a processing means, which performs smoothing between frames read out from the memory section 9. A comparator 12 compares the output level of the subtractor 10 with a threshold value. Reference numeral 13 denotes a processing switching unit, which selects either a pixel smoothed by the smoother 11 or a pixel from the memory unit 9 that is not smoothed, based on the output level of the comparator 12. 14 is an output switching unit for outputting the processed image to the color printer 4;
Alternatively, it is outputted to the D/A converter 16 via the monitor switching section 15. Reference numeral 15 denotes a monitor switching unit that outputs an unprocessed image from the video camera 2 or a processed image to the display 5. Reference numeral 17 denotes a control section that performs overall control. Control part 1
Reference numeral 7 is a determining means that determines the magnitude of movement in a moving image.

Next, the processing procedure by the control unit 17 will be explained.

First, a procedure for manually inputting an image into the memory section 9 will be explained.

The input switching unit 6 is switched to side a, and the sampling frequency of the D/D conversion unit 7, the writing speed of the memory unit 9, and the processing speed of the D/A conversion unit 16 are adjusted according to the image transfer rate of the video camera 2. . Further, the bus switching section 8 is switched to the b side, and the monitor switching section 15 is switched to the b side.

In this state, the analog image signal from the video camera 2 is converted into a digital signal by the D/D converter 7, and this digital signal is manually inputted to the D/A converter via the switch 15 to convert the D/A. The signal is again converted into an analog signal by the section 16, and a moving image is displayed on the display 5.

Then, the scene projected on the display 5 or the human custard of the scene input in advance is instructed to the memory unit 9, and the memory unit 9 controls the internal address generator to store n frames (at least 2 frames) in the internal semiconductor memory. (more than one frame) and memorize it.

Next, a noise reduction processing procedure for images stored in the memory section 9 will be explained.

Switch the bus switching section 8 to the a side, and output switching section 1
4 to the b side and the monitor switching unit 15 to the a side,
A threshold value k is set, and the memory section 9 is activated for reading and writing.

Then, the images of the first frame and the second frame are output pixel by pixel from the activated memory unit 9 to the bus 18.19.
The read image is smoothed pixel by pixel by a smoother 11. At this time, if the pixel data of the first frame and the second frame are 81 and 82, the smoother 11
The output of is (AI+A2)/2. On the other hand, memory section 9
The difference between frames of the image read from the image is calculated for each pixel by a difference unit 10, and the calculated difference between frames is output to the comparator 12. The output difference data d1 becomes d+=IA+A2.

Then, the comparator 12 compares the difference data d1 with the threshold value k, and if the result of the comparison is di<k, the processing switching unit 13 is switched to the a side. The smoothed pixel data B output to the human power bus 20 is a+=(^, +A2
)/2.

On the other hand, as a result of the comparison, if d≧, the processing switching section 13 is switched to the b side, the data of the first frame is outputted to the memory section 9 via the human-powered bus 20 and the bus switching section 8, and the data is stored in the memory. Stored in section 9.

From the point of view of saving semiconductor memory, the storage location can be the first frame, or the working area can be the human frame, that is, the first frame, second frame, third frame, etc. It is also possible to provide a special arrangement other than the above.

In this way, the processing and storage of the first frame and the second frame are performed pixel by pixel sequentially for the entire image, and at this time, the processing speed matches the processing speed of the display 5 and the D/A converter 16. If so, you can see the IA embedding process on the display 5.

After completing the processing of the first and second frames, process the new frame 1 (the frame created by processing the first and second frames) and the third frame in the same way as the first and second frames. Then, a new frame 2 (a frame created by processing the new frame 1 and the third frame) is stored in the memory section 9.

However, at this time, the new frame 1 is output to the output bus 18, and the third frame is output to the output bus 19 and processed.

Further, as a method for smoothing the new frame 1 and the N3 frame by the smoother 11, smoothing based on equations (1) and (2) can be considered, but since the smoothing method is not the main focus of the present invention,
Detailed explanation will be omitted.

Either equation (1) or (2) may be adopted.

81= (8,+A3)/2...(1), Iwaffl''Q ttt %, B, -+ 79-・1゛・
・・)/2 )B2= [2Bl+A3)/3 ・
... (2) Iwa 1''・self'' story t“-0”゛)/(+-
)) However, if you consider smoothing as pixel average, formula (2) is more accurate, and if you consider the hardware and configuration, formula (1)
is more advantageous.

Note that when the comparator 12 compares the difference data d2 and the threshold value k, it is also possible to change the threshold value k for each frame to be processed.

Thereafter, a new frame n can be obtained by processing and storing all input frames n.

Display 5 while processing new frame n between frames.
To check this, match the processing speed between frames with the processing speed of the display 5. In addition, since the processing speed of the display system is high, it is necessary to increase the speed of interframe processing, which increases the hardware specifications for interframe processing. Therefore, when the inter-frame processing is made slow, in order to check the new frame n on the display 5, the processing switching section 13 is emphatically switched to side b,
It instructs to read out the image data of the new frame n from the memory section 9 at a high speed that can match the processing speed of the display 5, and executes high-speed reading of the memory section 9. Then, the read image data of the new frame n is sent to the display 5 via the processing switching section 13, the output switching section 14, the monitor switching section 15, and the D/^ conversion section 16.
The image data is output to the display unit 5 and displayed on the display unit 5 based on the image data.

Note that if the display 5 has a built-in frame memory, high-speed reading and high-speed processing are not necessary.

Then, in response to the print request, the output switching unit 14 is switched to side a, and the memory unit 9 is instructed to read out the image data of the new frame n at the timing of data transfer with the color printer 4, and the image data is read from the memory unit 9. Synchronous reading with the printer is performed, the image data is output to the color printer 4 via the processing switching unit 13 and the output switching unit 14, and a hard copy is formed by the color printer 4 based on the image data.

Generally, the image data components are three primary color signals, namely red (
R), green (G), and blue (B), so
Image data flows through the manual switching section 6, A/D conversion section 7, input bus switching section 8, memory section 9, and bus 18.1.
9.20, the inside of each of the difference unit 10, smoother 11, comparator 12, processing switching unit 13, output switching unit 14, monitor switching unit 15, and D/^ conversion unit 16 is R%G.
, B can be processed in parallel.

In this embodiment, a case has been described in which image data from the television camera 2 is processed, but image data from the VTR 3 can also be processed in the same way.

Further, although the unit of one scene of an image has been described in units of one frame of television, it is not limited thereto, and it is possible to perform the unit of one field.

However, in this case, consideration must be given to the scanning method such as interlaced scanning.

In this embodiment, the determination between a still image portion and a moving image portion is mainly performed using a subtractor and a comparator, but the determination may be made based on other motion detection methods.

Furthermore, although the configuration is such that smoothing is performed in the time axis direction using a smoother, it is also possible to perform smoothing using a low-pass filter or the like.

Furthermore, noise reduction processing can be separately performed on the moving image portion.

In addition, processing such as difference, comparison, smoothing, and switching after processing in the memory unit 9 is performed not only by parallel processing hardware but also by software and general-purpose processing units (CPUs, etc.).
It is also possible to configure it with MPU, etc.).

Furthermore, the application system of the video/still image output device of the present invention can be applied not only to human power from television cameras and VTRs, but also to various kinds of video human power devices, and can be used not only for outputting to color printers and displays. (It is also possible to correspond to various types of still image output devices.

The present invention can be applied not only to noise reduction when obtaining a still image from a moving image, but also to creating a moving image with less noise from a moving image.

[Effects of the Invention] As explained above, according to the present invention, since it is configured as described above, there is an effect that image quality deterioration due to noise can be prevented.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of the system, FIG. 2 is a block diagram showing a moving image/still image conversion device according to an embodiment of the present invention, and FIG. 3 is an internal conceptual diagram of the memory section shown in FIG. 2. . 6... Input switching section, 7... A/D conversion section, 8... Bus switching section, 9... Memory section, 10... Differentiator, 11... Smoother, 12... - Comparator, 13... Processing switching section, 14... Output switching section, 15... Monitor switching section, 16... D/A conversion section, 17... Control section. Memory Iku 90 Doumachi Yakusuke Adventure Figure 3

Claims (1)

[Scope of Claims] The present invention includes a determining means for determining whether the motion of a moving image is large or small, and a processing means for performing smoothing processing in the time axis direction if the motion of the moving image is small as determined by the determining means. A video/still image conversion device featuring: