Optical information processing apparatus having a neural network for inducing an error signal

• US 5,333,125 A
• Filed: 11/05/1993
• Issued: 07/26/1994
• Est. Priority Date: 05/25/1990
• Status: Expired due to Fees

• Current Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Original Assignee: Canon Kabushiki Kaisha (Canon Inc.)
• Inventors: Yamamoto, Masakuni
• Primary Examiner(s): Psitos, Aristotelis
• Assistant Examiner(s): Edun, Muhammad N.
• Application Number:
  US08/147,447
• Time in Patent Office: 263 Days
• Field of Search: 369/44.41, 369/44.42, 369/44.12, 369/44.25, 369/44.26, 369/44.29, 369/44.35, 369/44.11, 369/54, 369/58, 369/120, 369/44.34, 369/44.36, 235/454, 235/455, 235/460, 235/459, 235/470, 395/22
• US Class Current: 369/44.34
• CPC Class Codes: G11B 20/10518 using neural networks
• CPC Class Codes: G11B 7/09 with provision for moving the light beam or focus plane for the purpose of maintaining alignment of the light beam relative to the record carrier during transducing operation, e.g. to compensate for surface irregularities of the latter or for track following

Abstract

An optical information processing apparatus is provided for at least either recording or reproducing information by projecting a converging light beam on a recording medium. The optical information processing apparatus includes a light detector for receiving the light beam, a neural network for inducing a tracking or focusing error signal from an output signal of the light detector, and a servo device for performing a tracking or focusing servo operation in accordance with the tracking or focusing error signal induced by the neural network.

Claims

1. An optical information processing apparatus for at least one of recording and reproducing information by irradiating a converged light beam on a recording medium, comprising:

a light detector for receiving a light beam which has been reflected or transmitted by the medium;

a neural network for receiving an output signal of said light detector and for outputting at least one of a tracking error signal and a focusing error signal, said neural network comparing said at least one of the tracking error signal and the focusing error signal with an instructor signal, and performing a learning operation so that the compared signals coincide; and

servo means for performing at least one of a tracking servo operation in accordance with the tracking error signal induced by said neural network and a focusing servo operation in accordance with the focusing error signal induced by said neural network.

2. The optical information processing apparatus according to claim 1, wherein:

said light detector comprises a plurality of photoelectric conversion elements; and

said neural network comprises a circuit for weighting the output for each of said plurality of photoelectric conversion elements to produce a plurality of weighted outputs, and a circuit for calculating at least one of the focusing error signal and the tracking error signal by operating on the plurality of weighted outputs.

3. The optical information processing apparatus according to claim 2, wherein

said light detector comprises l pieces of photoelectric conversion elements;

and

said neural network comprises;

memory means for storing data on weights for weighting the output for each of said plurality of photoelectric conversion elements;

l pieces of input cells for receiving the output for each of said plurality of photoelectric conversion elements;

a plurality of pieces of first weight cells for weighting an output signal for each input cell in accordance with the data stored in the memory means;

m pieces of intermediate cells for operating on the output signal for the first weight cells, wherein the number of first weight cells is the product of the number l of input cells and the number m of intermediate cells;

a plurality of pieces of second weight cells for weighting an output signal for each intermediate cell in accordance with the data stored in the memory means;

n pieces of output cells for operating on an output signal for the second weight cells, wherein the number of second weight cells is the product of the number m pieces of intermediate cells and the number n pieces of output cells; and

an output conversion cell for converting the output signal for the output cells into at least one of the focusing error signal and the tracking error signal and transmitting the at least one of the focusing error signal and the tracking error signal to said servo means.

4. The optical information processing apparatus according to claim 3, wherein each of said first and second weight cells comprises an amplifier element which amplifies each at least one focusing error signal and tracking error signal by an amplification factor complying with the data stored in said memory means.

5. The optical information processing apparatus according to claim 3, wherein said intermediate cells and said output cells comprise a non-linear element which adds to a positive output signal from said first and second weight cells, subtracts from a negative output signal from said first and second weight cells, and outputs a signal when the result of the adding or subtracting exceeds a specified threshold value.

6. The optical information processing apparatus according to claim 1, wherein:

said light detector comprises a plurality of photoelectric conversion elements; and

said neural network comprises a microcomputer for weighting the output of each of said plurality of photoelectric conversion elements to produce a plurality of weighted outputs, and for operating on the plurality of weighted outputs, and an output conversion cell for converting an arithmetic output of said microcomputer into at least one of the tracking error signal and the focusing error signal and transmitting the at least one of the tracking error signal and focusing error signal to said servo means.

7. The optical information processing apparatus according to claim 6, wherein said neural network further comprises memory means in which data concerning the weight for weighting the output of each of said plurality of photoelectric conversion elements has been stored.

8. The optical information processing apparatus according to claim 6, wherein said neural network further comprises an input conversion cell which converts output signals from a plurality of photoelectric conversion elements into a digital signal and inputs the digital signal into said microcomputer.

9. The optical information processing apparatus according to claim 1, wherein said servo means comprises an actuator for changing the tracking and focusing states of the light beam, and a drive circuit for driving said actuator in accordance with at least one of the tracking error signal and the focusing error signal.

10. The optical information processing apparatus, comprising:

a light source;

an objective lens for converging a light beam emitted from said light source on a recording medium;

an actuator which moves said objective lens at least in one of a direction of the optical axis and a direction perpendicular to the optical axis;

a light detector for receiving the light beam which has been reflected or transmitted by the medium;

a neural network for receiving an output signal of said light detector and for outputting at least one of a tracking error signal and a focusing error signal, said neural network comparing said at least one of the tracking error signal and the focusing error signal with an instructor signal, and performing a learning operation so that the compared signals coincide; and

a servo circuit which feeds back at least one of the tracking error signal and the focusing error signal induced by said neural network into said actuator, and performs at least one of a tracking servo operation and a focusing servo operation.

11. The optical information processing apparatus according to claim 10, wherein:

said light detector comprises a plurality of photoelectric conversion elements; and

said neural network comprises a circuit for weighting the output for each of said plurality of photoelectric conversion elements to produce a plurality of weighted outputs, and a circuit for calculating at least one of the tracking error signal and the focusing error signal by operating on the plurality of weighted outputs.

12. The optical information processing apparatus according to claim 11, wherein:

said light detector consists of l pieces of photoelectric conversion elements; and

said neural network comprises;

memory means for storing data on weights for weighting the output of each of said plurality of photoelectric conversion elements;

l pieces of input cells for receiving the output for each of said plurality of photoelectric conversion elements;

pieces of first weight cells for weighting an output signal for each input cell in accordance with the data stored in the memory means;

m pieces of intermediate cells for operating on the output signal for the first weight cells, wherein the number of first weight cells is the product of the number l of input cells and the number m of intermediate cells;

a plurality of pieces of second weight cells for weighting an output signal for each intermediate cell in accordance with the data stored in the memory means;

n pieces of output cells for operating on an output signal for the second weight cells, wherein the number of second weight cells is the product of the number m pieces of intermediate cells and the number n pieces of output cells; and

an output conversion cell for converting the output signal for the output cells into at least one of the focusing error signal and the tracking error signal and transmitting the at least one of the focusing error signal and the tracking error signal to said servo means.

13. The optical information processing apparatus according to claim 12, wherein each of said first and second weight cells comprises an amplifier element which amplifies each at least one focusing error signal and tracking error signal by an amplification factor complying with the data stored in said memory means.

14. The optical information processing apparatus according to claim 12, wherein said intermediate cells and said output cells comprise a non-linear element which adds to a positive output signal from said first and second weight cells, subtracts from a negative output signal from said first and second weight cells and outputs a signal when the result of the adding or subtracting exceeds a specified threshold value.

15. The optical information processing apparatus according to claim 10, wherein:

said light detector comprises a plurality of photoelectric conversion elements; and

said neural network comprises a microcomputer for weighting the output of each of said plurality of photoelectric conversion elements to produce a plurality of weighted outputs, and for operating on the plurality of weighted outputs, and an output conversion cell for converting an arithmetic output of said microcomputer into at least one of the tracking error signal and the focusing error signal and transmitting the at least one of the tracking error signal and focusing error signal to said servo means.

16. The optical information processing apparatus according to claim 15, wherein said neural network further comprises memory means in which data concerning the weight for weighting the output of each of said plurality of photoelectric conversion elements has been stored.

17. The optical information processing apparatus according to claim 15, wherein said neural network further comprises an input conversion cell which converts output signals from a plurality of photoelectric conversion elements into a digital signal and inputs the digital signal into said microcomputer.

18. An optical information processing apparatus for at least one of recording and reproducing information by irradiating a converged light beam on a recording medium, comprising:

a light detector for receiving a light beam which has been reflected or transmitted by the medium, which comprises a plurality of photoelectric conversion elements;

a neural network for inducing at least one of a tracking error signal and a focusing error signal from an output signal from said light detector, said neural network comprising a circuit for weighting the output of each of said photoelectric conversion elements to produce a plurality of weighted outputs and for calculating at least one of the tracking error signal and the focusing error signal by operating on the plurality of weighted outputs after weighting the output of each of said plurality of photoelectric conversion elements, said neural network further comprising a circuit for comparing said at least one of the tracking error signal and the focusing error signal with an instructor signal, and changing the weight for weighting said output so that the compared signals coincide; and

servo means for performing at least one of a tracking servo operation in accordance with the tracking error signal induced by said neural network and a focusing servo operation in accordance with the focusing error signal induced by said neural network.

19. The optical information processing apparatus according to claim 18, wherein:

said light detector consists of l pieces of photoelectric conversion elements; and

said neural network comprises;

memory means for storing data on weights for weighting the output for each of said plurality of photoelectric conversion elements;

l pieces of input cells for receiving the output for each of said plurality of photoelectric conversion elements;

pieces of first weight cell for weighting an output signal for each input cell in accordance with the data stored in the memory means;

m pieces of intermediate cells for operating on the output signal for the first weight cells, wherein the number of first weight cells is the product of the number l of input cells and the number m of intermediate cells;

a plurality of pieces of second weight cells for weighting an output signal for each intermediate cell in accordance with the data stored in said memory means;

n pieces of output cells for operating on an output signal for the second weight cells, wherein the number of second weight cells is the product of the number m pieces of intermediate cells and the number n pieces of output cells; and

an output conversion cell for converting the output signal for the output cells into and transmitting the at least one of the focusing error signal and the tracking error signal to said servo means.

20. The optical information processing apparatus according to claim 19, wherein each of said first and second weight cells comprises an amplifier elements which amplifies each at least one focusing error signal and tracking error signal by an amplification factor complying with the data stored in said memory means.

21. The optical information processing apparatus according to claim 19, wherein said intermediate cells and said output cells comprise a non-linear element which adds to a positive output signal from said first and second weight cells, subtracts from a negative output signal from said first and second weight cells and outputs a signal when the result of the adding or subtracting exceeds a specified threshold value.

22. The optical information processing apparatus according to claim 18, wherein:

said light detector comprises a plurality of photoelectric conversion elements; and

said neural network comprises a microcomputer for weighting the output of each of said plurality of photoelectric conversion elements to produce a plurality of weighted outputs, and for operating on the plurality of weighted outputs, and an output conversion cell for converting an arithmetic output of said microcomputer into at least one of the tracking error signal and the focusing error signal and transmitting the at least one of the tracking error signal and focusing error signal to said servo means.

23. The optical information processing apparatus according to claim 22, wherein said neural network further comprises memory means in which data concerning the weight for weighting the output of each of said plurality of photoelectric conversion elements has been stored.

24. The optical information processing apparatus according to claim 22, wherein said neural network further comprises an input conversion cell which converts output signal from a plurality of photoelectric conversion elements into a digital signal and inputs the digital signal into said microcomputer.

25. The optical information processing apparatus according to claim 18, wherein said servo means comprises an actuator for changing the tracking and focusing states of the light beam, and a drive circuit for driving said actuator in accordance with at least one of the tracking errors signal and the focusing error signal.

26. A method of adjusting an optical information processing apparatus, comprising means for irradiating a converged light beam on a recording medium, a light detector for receiving a light beam which has been reflected or transmitted by the medium, which comprises a plurality of photoelectric conversion elements, a neural network for inducing at least one of a tracking error signal and a focusing error signal from an output signal from the light detector, wherein the neural network comprises a circuit for weighting the output of each of the photoelectric conversion elements to produce a plurality of weighted outputs and for calculating at least one of the tracking error signal and the focusing error signal by operating on the plurality of weighted outputs after weighting the output of each of the plurality of photoelectric conversion elements, and servo means for performing at least one of a tracking servo operation in accordance with the tracking error signal induced by the neural network and a focusing servo operation in accordance with the focusing error signal induced by the neural network, comprising the steps of:

locating the medium and the light beam at specified relative positions;

comparing at least one of the tracking error signal and the focusing error signal output from the neural network with a reference value; and

changing the weight given to the output of each of the plurality of photoelectric conversion elements in accordance with the comparison result in said comparing step.

27. The method of adjusting an optical information processing apparatus according to claim 26, further comprising the steps of repeating said comparing and changing steps until at least one of the tracking error signal and the focusing error signal coincides with the reference value.

28. The method of adjusting an optical information processing apparatus according to claim 26, wherein said locating, comparing, and changing steps are performed with reference to each of a plurality of predetermined locations different from each other for the medium and the light beam.

29. The method of adjusting an optical information processing apparatus according to claim 26, wherein the neural network further comprises memory means, in which data on weights for weighting the output of each of the photoelectric conversion elements has been stored, and wherein said method further comprises the step of changing the weight given to the output of each of the plurality of photoelectric conversion elements by rewriting the data stored in the memory means.

30. An apparatus for adjusting an optical information processing apparatus, comprising means for irradiating a converged light beam on a recording medium, a light detector for receiving a light beam which has been reflected or transmitted by the medium, which comprises a plurality of photoelectric conversion elements, a neural network for inducing at least one of a tracking error signal and a focusing error signal from an output signal from the light detector, wherein the neural network comprises a circuit for weighting the output of each of the plurality of photoelectric conversion elements to produce a plurality of weighted outputs, and a circuit for calculating at least one of the tracking error signal and the focusing error signal by operating on the plurality of weighted outputs, and servo means for performing at least one of a tracking servo operation in accordance with the tracking error signal induced by the neural network and a focusing servo operation in accordance with the focusing error signal induced by the neural network, comprising:

means for changing the relative positions of the medium and the light beam;

means for comparing at least one of the tracking error signal and the focusing error signal output from the neural network with a reference value; and

means for changing the weight given to the output of each of the plurality of photoelectric conversion elements in the weighting circuit in accordance with the comparison result.

31. The apparatus for adjusting an optical information processing apparatus according to claim 30, wherein the neural network further comprises memory means, in which data on weights for weighting the output of each of the photoelectric conversion elements has been stored, and wherein said changing means changes the weight given to the output of each of the plurality of photoelectric conversion elements by rewriting the data stored in said memory means.