JPH0877661A - Autotracking method for vtr - Google Patents

Abstract

PURPOSE: To highly accurately control tracking in a short time by changing a target tracking value continuously at high speeds and obtaining the target value in a step wherein a reproduced output of a rotary head is changed from an increase to a decrease and from a decrease to an increase. CONSTITUTION: A tracking value generator 5 increments and generates a tracking value per one rotation of a head drum 1 by a pulse of a drum servo circuit 8 which is synchronized with the rotation of the head drum, and supplies the value as a target value to a servo circuit 7 thereby to control a capstan motor 12. At this time, a neural network 4 detects maximum and minimum points of an output of a video head 9 from an envelope detector 2, reversing a changing direction of the tracking value generator 5 at the maximum point. Then, the neural network 4 changes at the minimum point the output by a count of steps so that an operated output becomes maximum, thereby making the output agree with the target value. Accordingly, the maximum point can be detected quickly and a correct correction is executed to a change with time.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a system for automatically tracking a video tape recorder.

[0002]

2. Description of the Related Art The conventional automatic tracking method is a so-called hill climbing control method. By changing the tracking value, it is checked whether or not the video head output is larger than the previous value, and the top of the mountain is searched. It is a method.

[0003]

The conventional method is to check whether the video head output increases when the tracking value is changed by a constant value with reference to the video head output for the current tracking value. The output to is required to be accurate. This is related to the movement range, and if the tracking value is changed significantly, the output change will basically be large and it will be easy to judge whether it has increased or decreased, but the accuracy of the point where the output becomes the maximum is poor. Become.

Therefore, in order to improve the accuracy, it is indispensable to reduce the movement width, wait for the system to settle and check the output change, and it takes a long time to complete the tracking control. An object of the present invention is to solve the problems of the conventional method and to provide a method for automatically performing highly accurate tracking control in a short time.

[0005]

In order to achieve the above object, in the automatic tracking system of the present invention, the target tracking value of the tracking servo is synchronized with the rotation of the head drum of the VTR and the response speed of the servo system is changed. Continuously change at a constant speed faster than that, find the number of steps (A) from the point where the playback output of the rotary head changes from rising to falling, change the target tracking value in the opposite direction from that point, and rotate again. The number of steps (B) from the point where the reproduction output of the head goes from the fall to the rise is obtained, and the tracking value that maximizes the output of the rotary head is obtained using A and B.

Further, the number of steps (B) is calculated from the point where the reproduction output of the rotary head goes from the downward direction to the upward direction. The method is also an effective means.

[0007]

The principle of the operation will be described in detail in order to deepen the understanding of the operation of the above configuration. Tracking of VTR When the tracking value is continuously changed at a constant speed faster than the response speed of the servo system, the relationship between the target value and the actual value may vary with time during a short time as shown in Fig. 1. It becomes a car to go.

That is, when the tracking value is moved by one step, if the tracking value continuously changes before the bed has moved to the corresponding position, the tracking value changes continuously to the next value. The relationship between the two becomes a car that opens up over time. The extent of this opening depends on the speed, and if it is slow enough, there will be no open car and it will follow.

FIG. 2 is a schematic diagram for explaining the principle of the present invention. The horizontal axis represents time, and the vertical axis represents tracking value and reproduction output level.

Assuming that the reproduction output changes from rising to falling at time t _{1 when the} target tracking value is changed as shown in the figure, the actual head has a response delay at this point, so that A- (B + C) In place. If the direction in which the target tracking value is changed is reversed at this point, it coincides with the actual tracking value at t ₂ . From this point, the actual tracking value also changes in the opposite direction, and the reproduction output changes from falling to rising.

At t _{3, the} target value becomes A- (B + C) and the change is stopped. The actual value is delayed and stops changing at t ₄ . Here, t ₁ and t ₂ can be understood by observing the change of the reproduction output, but t ₃ cannot be understood by it. However, when this figure is solved as a figure, it becomes as follows. That is, if only A and B are known, it is possible to obtain a tracking value that maximizes the output of the rotary head. As is apparent from the figure, this value C is determined only by t ₁ , the changing speed of the target value and the changing speed of the actual value. That is, it can be said that it also plays the role of correcting the delay in the response of the system.

A normal VTR does not always have the configuration shown in FIG. 2, and a time delay occurs in determining the maximum point of the reproduction output. In particular, when a neural network is used for determination, a delay of about 3 steps is inevitable because data up to several steps past the maximum point is used. Further, since the erroneous determination due to noise is reduced, the use of a moving average also causes a delay. This delay moves from the point in which the direction of change of the tracking value is reversed to the direction in which the tracking value is corrected. Therefore, the number of steps smaller than the number of steps (C) in the above equation is t
If you proceed from ₂ , you will reach A- (B + C).

In the automatic tracking system constructed as described above according to the above-mentioned principle, the target tracking value is changed at a constant speed one after another without waiting for the tracking servo to settle, so that the rotary head is shortened in a short time. You can find the maximum point of the playback output of. The true tracking value corresponding to the maximum point found can be easily calculated by calculating the response delay.

[0014]

EXAMPLE FIG. 3 shows an example of the present invention. This figure is VT
Of the block diagram of R, only the part related to the present invention is extracted and written. In the figure, the video head 9. The output of is the preamplifier 1. FM demodulator 10. Is demodulated to a video signal and output. Preamplifier 1. The output of the envelope detector 2. Is also introduced into the A / D converter 3. Is input to The envelope which became a digital signal is a filter-6. 3. A neural network that has been trained in advance via Input which is input to and shows an upward trend 1. And output showing a downward trend 2. And a tracking value generator 5. Be led to. Here filter 6. Can produce the same effect by using the moving average of data.

On the other hand, drum motor servo circuit 8. The switching pulse synchronized with the rotation of the head drum is the tracking value generator. Is used to synchronize the tracking value generation.

Tracking value generator 5. Is generated by incrementing the tracking value once for each rotation of the head drum, and the tracking servo circuit 7. To the target value. The servo is a capstan motor 12. Control the drum motor 11. And the phase of the CTL signal (not shown) reproduced from the tape are brought close to the target value. Actually, the tracking servo controls the time interval between the switching pulse of the head drum and the CTL signal to be constant.

Next, the neural network which plays an important role in this embodiment is shown in FIG. Figure 6. Figure 7.
Add a general explanation using. This network has a plurality of signal inputs (in) as shown in FIG. 5, and the output takes a value of 0 to 1 with a value obtained by adding weights (W _N ) to each signal and adding them as input variables. Is. A specific waveform is input to the input of this network (an example is shown in FIG. 6), and W _N is set so that the output becomes 1 only when the waveform is present and 0 otherwise. By doing so, it is possible to know that a specific waveform, that is, a desired waveform has been input. Here, setting W _N so that the desired output is output is called learning.

In order to actually perform a desired operation in such a network, as shown in FIG. 7, a network having a three-layer structure including an input layer, an intermediate layer, and an output layer, each of which is connected by a settable coupling coefficient. Use the one that is. Usually, this network is often realized by software using a micro computer, and the learning result is stored in ROM.

In this embodiment, this neural network is used to detect the maximum and minimum points of the envelope, that is, the reproduction output. Figure 4 shows the case of the maximum point in this operation. Will be explained. The horizontal axis indicates the tracking value, which goes to the right one step at a time in synchronization with the rotation of the head drum when the automatic tracking is started. The vertical pongee indicates the output, and the envelope in the figure is the envelope detector 2. Is a schematic representation of the output of. Output 1 and output 2 in FIG.
2 shows the output waveforms of the output 1 and the output 2 of FIG.

While the envelope is rising, output 1 is
It outputs "1" and outputs "0" when the rise is stopped. The tracking value at this time is T ₁ . After T _1, it does not show an upward trend and thus remains “0”. On the other hand output 2 is, at first, but the envelope is outputting a "0" for not shown a downward trend and outputs a "1", so continue to fall after showing a downward trend from T _2.

Since the envelope exhibits left-right symmetry when the tracking value is changed, (T ₁ + T ₂ ) / 2 can be used as the tracking value of the maximum point.

When the automatic tracking is started, the tracking value generator 5. Tracking servo target value is output by incrementing or decrementing step by step in synchronization with the rotation of the head drum. Sent to 3. The tracking servo starts the tracking operation and the envelope shows the maximum value, that is, the neural network 4. Output of 1. Does not show an upward trend, and further output 2. Tracking value generator 5. Reverse the direction of change.

Next, when the envelope shows the minimum point, that is, the neural network 4. Output 2. Output does not show a downward trend. When the value starts to show an increasing tendency, the automatic tracking is ended by further changing the step number (C) from that point.

[0024]

Since the present invention is constructed as described above, it has the following effects.

(1) Since the tracking value can be changed at a speed faster than the response speed of the servo system, the maximum point can be found quickly. (2) Since the response delay is corrected using both the speed at which the target value of the servo system changes and the speed at which the actual value changes, there is no need to measure and prepare the response delay of the system in advance. Even if occurs, it is possible to make accurate corrections.

[Brief description of drawings]

FIG. 1 is an operation explanatory diagram when a tracking value is changed faster than a response speed of a servo system.

FIG. 2 is an explanatory diagram showing the principle of the present invention.

FIG. 3 is a block diagram of an embodiment of the present invention.

FIG. 4 is an explanatory diagram of maximum point calculation according to the embodiment of this invention.

FIG. 5 is an explanatory diagram of the operation of the neural network.

FIG. 6 is an explanatory diagram when a waveform is input to a neural network.

FIG. 7 is an explanatory diagram of a three-layered neural network including an input layer, an intermediate layer, and an output layer.

[Explanation of symbols]

 1 Preamplifier 2 Envelope detector 3 A / D converter 4 Neural network 5 Tracking value generator 6 Filter 7 Tracking servo circuit 8 Drum motor Servo circuit 9 Video head 10 FM demodulator 11 Drum motor 12 Capstan motor

Claims (2)

[Claims] 1. A reproducing output of a rotary head is increased by continuously changing a target tracking value of a tracking servo in a VTR at a constant speed faster than a response speed of a servo system in synchronization with rotation of a head drum. The number of steps (A) from the point to the point where the reproduction output of the rotary head changes from the downward to the upward is obtained by changing the target tracking value in the opposite direction from that point. , A and B are used to obtain the tracking value that maximizes the rotary head output. 2. From the point where the reproduction output of the rotary head goes from downward to upward, The automatic tracking method described in claim 1.