JPS61240785A - Image memory device - Google Patents

Abstract

PURPOSE:To correct skews in case of specific reproduction and to provide better special reproduced images by adding-subtracting corrected values from image memory address counter values according to a specific reproducing mode and also using the above corrected values as image memory writing-reading addresses. CONSTITUTION:The special reproduced images are inputted from an image input terminal 2 and are written to a memory 4 after performing A/D conversion 3. Likewise, special images are read from the memory 4 and are outputted to an image signal output terminal 6 after performing D/A conversion 5. When video signals are written to the memory 4, an AND9 between a signal coming from a timing signal input terminal 7 and a clock coming from a clock input terminal 8 is added to a R/W terminal of the memory 4. And the clock coming from the terminal 8 is added to a data selector 10 as well as an address counter 11. Furthermore, the corrected values setting to the memory 13 according to the specific reproducing mode are added and subtracted to/from a value of the counter 11 by a subtracter 16 and used as the writing and reading addresses of the memory 4 to correct the skews in the case of specific reproduction.

Description

TECHNICAL FIELD OF THE INVENTION The present invention relates to an image memory device, and more particularly to an image memory device for a video tape recorder.

[Technical background of the invention]

In home video tape recorders (VTRs), one field of video signals can be recorded and played back with one scan of the head, so system playback is achieved by stopping the tape and scanning the same track multiple times.

Syslow playback can be performed by scanning at a slower tape speed. However, in order to perform noise-free special playback, it is necessary to constantly scan the appropriate tracks in order to obtain sufficient playback output.1 Since two normal heads cannot cope with this, two or more separate heads are required. Generally, a head is provided to perform special reproduction.

However, this leads to complications in one mechanism and one circuit. Therefore, it has been considered to use a large-capacity semiconductor image memory to store video signals over one field or more, and to perform special reproduction by reading out the signals from the memory at appropriate timing.

Here, the write operation and read operation to the memory will be briefly explained using 7x speed playback (slow playback) as an example. The scanning center locus of the head at this time is ThT in Fig. 6,
Indicated by T,... Here, An indicates a track recorded with a human head, and Bn indicates a track recorded with a B head. Note that (1) is a tape.

FIG. 7(a) shows the reproduction envelope waveform at this time. The reproduced output of the track for which a sufficient envelope output has been obtained is written to the memory (the L level period of the write timing signal shown by Φ in FIG. 7 is the write period to the memory). During the track scanning period when only a small reproduction output is obtained, the signals written in the memory are read out repeatedly. As a result, a sufficient output can be obtained, and a noise-free slow-motion playback image can be obtained. This reproduction output is shown in FIG. 7 (C'').

[Problems with background technology]

By the way, in the above-mentioned special playback, the same field is read out from the memory by turning over the edge eight times. This has the disadvantage that the smoothness of movement is impaired. The reason why the same field is repeated eight times in this way will be described below. In other words, relatively rad output should also be used (if
The same field can be repeated four times, but as shown in Fig. 8, the horizontal synchronization position is shifted by -H between the human head scanning center trajectory T and the B head scanning center trajectory T4. If the scanning output of the head is written in a memory and used, gaps and gaps will occur on the screen.

Therefore, it is necessary to repeat the scanning output of the same field by one head eight times.

[Purpose of the invention]

The present invention has been made in view of the above-mentioned points, and an object of the present invention is to avoid skinny correction that occurs during special playback, and to obtain good special playback images.

[Summary of the invention]

In the present invention, when performing noiseless special playback of a VTR using image memory, the correction value set according to the special playback mode is subtracted or added to the count value of the address counter that specifies the address of the image memory. The above object is achieved by using the obtained value as the write address or read address of the image memory.

[Embodiments of the invention]

An embodiment of the image memory device according to the present invention will be described below with reference to FIGS. 1 to 3.

FIG. 1 is a block diagram showing one embodiment of the invention.

(2) is the video signal input terminal, (3) is the ~Φ converter, (4
) is an image memory having a storage capacity for approximately one field.

SW is a switch, (5) is a D/A converter, and (6) is a video signal output terminal. In addition, (7) is a write timing signal input terminal, and (8) is a clock input terminal.
(9) is an AND circuit. Furthermore, (1@, αυ is a data selector and address counter connected to a clock input terminal, and α is an address controller.

αj is the memory in which the skinny correction value is stored, (1
4) is a code signal input terminal, into which a code with a numerical value of zero is input. Kasumi is a data selector, and (lQ is a subtractor.

Next, the operation of the above embodiment will be explained with reference to FIG. 2 as well. Special playback output input through input terminal (2)
- Output during slow playback) As shown in FIG. 2(a), the playback output is considerably reduced compared to the normal playback output. Therefore, as mentioned above, the output of the field from which a sufficient reproduction output has been obtained is written to the memory (4),
In other fields, a noise-free reproduction screen is obtained by reading out the signals written in the memory (4). In this case, the write timing signal to the memory (4) is input from the terminal (7), and this is a signal synchronized with the head switching pulse (see Fig. 2) (see Fig. 2 (C)). . The write signal of the memory (4λ) is obtained by ANDing the signal from the terminal (7) and the clock from the terminal (8) using an AND circuit (9). Note that this memory (4) is used for writing and It has a structure in which reading can be performed within one cycle of the clock; for example, reading is performed at the H level of the clock, and writing is performed at the L level of the clock.

Based on the write timing signal, the memo (2) is digitized by the ~Φ converter (3) to the memo (January 4).
A video signal Ps during special playback is inputted and written. The write address to the memory (4) at this time is given from the data selector [alpha]. Based on the clock input to the selector terminal S, one is selected and sent as the memory (4) K address.

That is, when reading from the memory (4), input a is selected as the read address, and when writing, input a is selected as the write address. This input value is the value obtained by subtracting the output value of the data selector 9 from the count value of the address counter αD using a subtracter. Data selector a! Reference numeral 9 stores and outputs skid correction values corresponding to the special reproduction mode. That is, when writing the playback output of a single head, input d is selected, this value is subtracted from the count value of address counter αυ, and this subtracted value is given as the write address of memory (4) (actually, Since the input d is zero, the value of the address counter αυ itself becomes the address of the memory (4)). Also, B
When writing the reproduced output of the head, the above-mentioned skiing 3a occurs between the human head signals, so in order to compensate for this, the data selector α9 selects and outputs the input C. With this, the write address is empty, -
The correction value is corrected (the amount Cc in FIG. 2(d) showing the address value to message (4)), and the reproduction output of the B head can also be used.

To summarize the entire operation here, during slow playback, first the output of one head is written, then the memory (4
) reads out the signal written in. During the third read, the read is performed during the H level period within one cycle of the clock, and the L level is
Write the output of the B head that has obtained sufficient output during the level period. Thereafter, the written output is read out repeatedly for four fields. By continuing this operation, the playback output of the B head will also be ski-corrected and can be used for slow playback, making the movement smoother. Note that the switch SW is connected to the terminal b@ when reading from the memory (4) as shown in FIG. 2 (6). The digital video signal obtained via the switch SW is D/A converted by a D/A converter (5) and output to a terminal (6). Here, the address value of the memory (4) at the time of reading is shown in fsz diagram (f), and the content written in the memory (4) is shown in the same diagram (f).

The read contents are shown in the same figure. Figure 3 shows the memory (4).
This is a timing chart shown to make it easier to understand the correction operation of the address given to the terminal (8).
) K input clock, (b) the count value of the address counter I, (C) the input value of the data selector, (d) the write input to the terminal (7) Timing signal, (et) is the selection output of the data selector (to), (f) is the write/read signal,
CI ri) in the same figure is data input to memo January 4), t' in the same figure.
h) is the quantity data output. The figure shows the same address input.

Note that the above description deals with the case where skew correction is performed at the time of writing to the memory (4).

A similar effect can be obtained by performing an address operation when reading from the memory (4).

That is, when writing the output of the B head to the memory (4), the write address is left as is, and when reading, the address obtained by adding the skew correction value to the value of the address counter is used.

Next, skew correction during fast forward (rewind) playback will be explained. For example, VTR Beta 7 Omatsu)
In the βI mode, the tape pattern is not arranged in H order. Therefore, when performing fast-forward playback, when the head crosses the track, a jump of H occurs, and this! ! Until then, a large gap would appear on the playback screen. Here, a method of performing skew correction using the memory (4) and obtaining a noise-free reproduced image even during fast-forward υ reproduction will be explained.

First, the capstan speed must be set to an even multiple of the normal playback speed. If this is set to an odd multiple, the noise part will be fixed and written to a specific address in memory (4) (actually, the noise part will not be written to memory (4)), so the data in this part will not be updated. This rounding avoids the drawback of 4 remaining old data. FIG. 4 shows the head scanning locus at n times the speed, and FIG. 5 shows the timing chart at this time.

First, during special playback, perform envelope detection (see Figure 5 (a)) to detect the field to be written and obtain the writing timing signal.
Detect areas with large playback envelope output. When this envelope detection pulse (see Figure 5(b)) is inverted, it becomes a write timing pulse (see Figure 5(C)).
reference). At the same time, a 0.5H ski jump is detected. This detection pulse (see FIG. 5(d)) is used for switching the output of the data selector Kasumi, which will be explained later.

Next, the address given to memory (4) in this case will be explained. The address counter (11) is reset at the edge of the head switching pulse (see FIG. 5(e)). The output of the data selector QS is switched by the skinny jump detection signal, and when the skew-) Young is detected, the value of the input a (=skip, the output from the memory α floor in which the -correction value is stored) is selected. . The skew correction value at this time is an address value corresponding to 0.5H. Further, the switch SW is always turned to the terminal side. The memory (4) performs reading during the H level period of the operating clock input to the terminal (8), and performs writing during the L level period.

In this case, when a skinny jump is not detected, no address correction is performed, and when a skinny jump is detected, data is written at the previous address position by 0.5H. The write address value in this case is shown in Figure 5(f), the content of the write data to the memory (4) is shown in the same figure (Figure 5), and the read address value is shown in the fifth image. is shown in Figure (i)K.

As mentioned above, in this case, even when performing special playback on a tape with a recording pattern that is not arranged in H order, the outputs of the human and eight heads can be written to the memory, and a special playback screen with smooth movement and no noise can be obtained. be able to.

〔Effect of the invention〕

As described above, according to the present invention, it is possible to obtain a gap between reproduced images showing noise during special reproduction, and to obtain a slow-motion reproduced image with smooth movement. Also, H
By performing special reproduction using a tape with recording patterns that are not arranged in order, it is possible to obtain a reproduced image without gaps or minuses even in case of emergency.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing an embodiment of the picture image memory device according to the present invention, FIGS. 2 and 3 are timing charts of the device shown in FIG. 1, and FIG. 4 is a head scanning center during 12x playback. FIG. 5 is a timing chart of the apparatus shown in FIG. 1 during 12 times speed playback. FIG. 6 is a diagram showing a head scanning locus during 7-slot one reproduction, and FIGS. 7 and 8 are diagrams for explaining problems with conventional image memory devices. 3...A/D converter, 4...memory. 5...D/A converter, 11.15...Data selector. 11...Address counter, 13...Memory. Agent Patent attorney Kensuke Norichika (and 1 other person) Figure 1 b /+＾0/+ha /+ To 3 ze 8 stone 8 death ? Mo: Figure 6 Figure 7 Day n

Claims (1)

[Claims] An image memory device for a video tape recorder, comprising means for writing a digital video signal into an image memory and means for reading the digital video signal from the image memory, and for performing noiseless special playback using the image memory, A means for subtracting or adding a correction value set according to the special playback mode from the count value of an address counter that specifies an address in the image memory, and a means for subtracting or adding the value to the address when writing to the image memory or when reading the image memory. An image memory device characterized by comprising means for use as an address.