JP3078151B2 - Image processing device and recording / reproducing device - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus and a recording / reproducing apparatus and, more particularly, to a physical device generated between a recording medium and a probe electrode while causing the probe electrode to perform two-dimensional scanning relative to the recording medium. The present invention relates to a recording / reproducing apparatus that records information on a recording medium using a signal due to interaction and reproduces the recorded information as image information, and an image processing apparatus suitable for the recording / reproducing apparatus.

[0002]

2. Description of the Related Art In recent years, the recording capacity of data required for a recording / reproducing apparatus has tended to increase. For that purpose, it is essential to reduce the size of a recording unit and improve the recording unit density. . For example, in a digital audio disk using optical recording, the size of a recording unit reaches about 1 μm ² .

On the other hand, recently, a scanning tunneling microscope (ST) capable of directly observing the electronic structure on and near the surface of a substance has been developed.
M) has been developed and has the following advantages, and is expected to be widely applied. (1) A high-resolution measurement of a real space image can be performed regardless of whether it is a single crystal or amorphous. (2) Observation can be performed with low power without damaging the recording medium due to current. (3) Operates not only in ultra-high vacuum but also in air and in solution. (4) It can be used for various materials.

In a scanning tunneling microscope, when a voltage is applied between a probe electrode (metal probe) and a conductive material, the probe electrode is brought closer to the conductive material by a distance of about 1 nm. It utilizes the fact that a tunnel current (a signal due to a physical interaction generated between the probe electrode and the conductive substance) flows between the probe electrode and the conductive substance. Tunnel current is very sensitive to changes in distance between the probe electrode and conductive material,
By scanning the probe electrode relative to the surface of the conductive material two-dimensionally so as to keep the average distance or tunnel current between the probe electrode and the conductive material constant, Surface information of the space can be obtained. At this time, the resolution of the conductive material in the in-plane direction is 10
nm or less.

The principle of such a scanning tunneling microscope is described below.
If applied, as a recording medium, switching of voltage and current
Materials that have a memory effect on properties (for example, π electrons
Organic compounds and thin layers of chalcogenides)
By doing so, 0.01 μm ^Two The information is recorded in the following recording units.
Recording and reproduction become possible. In addition, power such as electron beams and light
A method to change the surface shape of a recording medium using magnetic waves
If used, record from the limit of convergence of electron beam etc.
The unit is larger, but at the same recording density as the current optical recording
Can be recorded and reproduced.

[0006]

However, in a conventional recording / reproducing apparatus to which the above-mentioned principle of the scanning tunneling microscope is applied, high-density information is provided over the entire surface (recording surface) of a recording medium having a predetermined area. In the case of performing recording / reproducing, the unevenness existing on the recording surface deteriorates the S / N ratio and the error rate of the signal at the time of reproducing, which is a major obstacle to achieving a higher recording capacity.

As a method for reducing physical irregularities on the surface of the recording medium, a method of growing a thin Au film on an open surface of mica or the like to form a single crystal to obtain a smooth surface can be applied. However, even when a recording medium according to such a method is used, irregularities almost equal to the recording bit size having a diameter of 10 nm or less rarely occur, but the shape of the recording bit changes due to atomic-order steps or ridges. As a result, there remains a problem that the SN ratio and the error rate of the signal at the time of reproduction are reduced.

As a method of removing noise from a reproduced signal, there is a method of filtering in a frequency domain belonging to a linear filter. In this method, the frequency band of the signal and noise at the time of reproduction is known and can be linearly separated. Only valid if. However, regarding the recording bits, even though the frequency band for each shape is known, it is not known where they are located. In addition, steps and ridges have portions (that is, edges) having large differential coefficients with respect to the spatial coordinates, so that these frequency components are spread over the entire frequency domain and the spatially compact frequency components of the recording bits are used. Are also spread over the entire frequency band, so that these cannot be band-separated in the frequency space. Furthermore, since the positions of recording bits, steps, and ridges cannot be specified, it is difficult to extract the power spectrum by template matching.

On the other hand, as a conventional image processing technique emphasizing spatial correlation, an extracted image is created by extracting only steps and ridges from an image, and the extracted images are subtracted from the original image to remove steps and ridges. And those belonging to a non-linear filter. However, the method of extracting only steps and ridges from an image is a method in which the processing mode is iterative or serial, and strongly depends on the number of pixels, such as a thinning algorithm or an algorithm for calculating the number of connected black pixels. Therefore, it is very disadvantageous in the recording / reproducing apparatus in which the reading speed is the most important function.

An object of the present invention is to provide a recording / reproducing apparatus which can reduce a reduction in an error rate at the time of reproduction due to steps or ridges of atomic order.

Another object of the present invention is to provide an image processing apparatus having high parallelism independent of the number of pixels and capable of improving the operation speed.

[0012]

According to the present invention, there is provided an image processing apparatus comprising: a total curvature numerator calculating means for calculating a numerator value of a total curvature at each pixel of image information; and an average curvature numerator at each pixel of the image information. An average curvature numerator calculating means for calculating a value of a numerator; an AND operation means for calculating an AND of an output signal of the numerator calculating means of the total curvature and an output signal of the numerator calculating means of the average curvature; A median filter to which an output signal of the product calculating means is input.

The recording / reproducing apparatus of the present invention utilizes a signal caused by physical interaction between the probe electrode and the recording medium while scanning the probe electrode relative to the recording surface of the recording medium in a two-dimensional manner. Then, while recording information on the recording medium, in a recording and reproducing apparatus for reproducing the recorded information as image information, the numerator of the total curvature in each pixel of the image information reproduced from the signal due to the physical interaction Image information creating means for calculating the value and the value of the numerator of the average curvature to create new image information.

[0014] The image information creating means calculates a value of a numerator of a total curvature in each pixel of the image information reproduced from the signal due to the physical interaction, and a total curvature calculating means for each pixel of the image information. Average curvature calculation means for calculating the value of the numerator of the average curvature in, and AND operation means for calculating the AND of the output signal of the numerator calculation means of the total curvature and the output signal of the numerator calculation means of the average curvature, A median filter to which an output signal of the AND operation means is input.

[0015]

The image processing apparatus according to the present invention regards image information having a two-dimensional spread as a function defined on a two-dimensional plane, and calculates all the functions of each function (each pixel of the image information) at each point on the two-dimensional plane. By finding the curvature, it can be seen that points where the value of the total curvature is positive are elliptical points and points where the value of the total curvature is not positive are parabolic and hyperbolic points, and
By calculating the average curvature of the function at each point on the two-dimensional plane, an elliptical point having a negative average curvature value is convex to the recording surface, and an elliptical point having a positive average curvature value is recorded. Although the fact that it is known that the surface is concave is used, the following processing is performed to improve the arithmetic processing.

In general, the total curvature and the average curvature of a function at each point on a two-dimensional plane are respectively described as follows.
It is expressed as a fraction, and the denominator always has a positive value. Therefore, whether or not the value of the total curvature is positive can be determined only by calculating the value of the numerator of the total curvature, and whether or not the value of the average curvature is negative is obtained by determining the value of the numerator of the average curvature. , So instead of the value of the total curvature and the value of the average curvature,
The value of the numerator of the total curvature and the value of the numerator of the average curvature are respectively obtained. That is, if it is known in advance that an elliptical point on a convex curved surface with respect to the recording surface indicates necessary information, the value of the numerator of all curvatures is obtained by the numerator of full curvature calculation means,
The value of the numerator of the average curvature is obtained by the average curvature numerator calculation means,
By calculating the logical product of the output signal of the numerator of the mean curvature and the output signal of the numerator of the mean curvature, erroneous information included in the image information is eliminated, and only necessary information is extracted. be able to. Further, by inputting the output signal of the AND operation means to the median filter, it is possible to eliminate erroneous information due to an error caused by discretization of spatial coordinates.

The recording / reproducing apparatus according to the present invention calculates the value of the numerator of the total curvature and the value of the numerator of the average curvature at each pixel of the reproduced image information from the signal due to the physical interaction, and generates new image information. Since the recording bit becomes an elliptical point on a curved surface that is convex with respect to the recording surface by including the image information generating means to be generated, the recording bit becomes a parabolic point included in the image information.
By removing hyperbolic points and elliptical points on a curved surface that is concave with respect to the recording surface, new image information in which only recording bits are extracted can be obtained.

The value of the numerator of the total curvature and the value of the numerator of the average curvature of each pixel can be obtained by calculating the first basic amount and the second basic amount near each pixel, respectively.
Here, since the first basic amount and the second basic amount are local amounts, the processing mode has high parallelism independent of the number of pixels.

[0019]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing an embodiment of the image processing apparatus of the present invention.

The image processing apparatus 100 includes an image information input unit 101
And the numerator of the total curvature 102 and the numerator of the average curvature
103, AND operation unit 104, Median filter 105
And Here, the full curvature numerator calculator 102 calculates the value of the numerator of the total curvature of each pixel using the image information signal sent from the image information input unit 101, and the value of the numerator of the total curvature is positive. In some cases, the pixel value of the pixel is set to “1”, and when the value of the numerator of all curvatures is not positive, the pixel value of the pixel is set to “0”.
Thus, non-linear binarization processing is performed on the image information signal for each pixel. The average curvature numerator calculation unit 103 calculates the value of the numerator of the average curvature of each pixel using the image information signal sent from the image information input unit 101, and when the value of the numerator of the average curvature is negative, the pixel of that pixel is calculated. Is set to "1", and when the value of the numerator of the average curvature is not negative, the pixel value of the pixel is set to "0", so that the image information signal is subjected to nonlinear binarization processing for each pixel. It is for. The logical product operation unit 104 is for calculating the logical product of the output signal of the numerator computer 102 for the total curvature and the output signal of the numerator computer 103 for the average curvature for each pixel. The median filter 105 includes an AND operation unit 104
In order to remove an isolated point included in the output signal of FIG.

Next, the principle of the image processing method in the image processing apparatus 100 will be described.

(1) Total curvature and average curvature As shown in FIGS. 2A and 2B, the ridges and steps of the atomic order existing on the recording surface of the recording medium have steps in only one direction. Is considered to be composed of parabolic points. On the other hand, whether the method is to change the surface shape of the recording medium by applying the principle of the scanning tunneling microscope or to change the surface shape of the recording medium using an electromagnetic wave such as an electron beam or light, the recording bit is As shown in FIG. 2 (C), the shape is considered to have a unimodal shape, and therefore is considered to be constituted by elliptical points at least around the maximum value.
Note that the “monomodal shape” specifically refers to a shape in which the height monotonically decreases as the position moves away from the position of the unique maximum value.

Therefore, the total curvature K and the elliptic point (eliptic
point), parabolic point and hyperbolic point (h
(b) a parabolic point when K = 0, and (c) a hyperbolic point when K <0. By calculating the total curvature K at each point on the recording surface (that is, each pixel of the image information), when the sign of the total curvature K is positive (K> 0), the point belongs to the recording bit area, and the total curvature K But"
When the sign of “0” or the total curvature K is negative (K> 0), it is understood that the point does not belong to the recording bit area. Is negative (H <0), whereas the sign of the average curvature H at the elliptical point on the curved surface that is concave with respect to the recording surface is positive (H> 0). The direction of the recording bit can be detected by calculating the average curvature H of the points belonging to.

(2) Calculation of Total Curvature K and Average Curvature H The total curvature K and average curvature H are expressed by equations (1) and (2), respectively.

[0026]

(Equation 1) Here, E, F, and G indicate a first basic amount, and P, Q, and R indicate a second basic amount, which are represented by equations (3) to (8), respectively.

[0027]

(Equation 2) Here, for example, x _vv represents 2 order differential coefficient of x by v, for example x _u represents 1 order differential coefficient of x by u.

Now, when treating an image as a function defined on the recording surface (two-dimensional plane) of the recording medium, x =
u, y = v and z = f (x, y), so that each first basic amount E, F, G and each second basic amount P, Q, R
Are represented by equations (9) to (14), respectively.

[0029]

(Equation 3) Therefore, the total curvature K and the average curvature H can be obtained from equations (15) and (16), respectively.

[0030]

(Equation 4) (3) Method for Simplifying the Operation If the above operation is to be realized by hardware, it is difficult to perform the exponentiation operation. Incidentally, the more (15) and above (16), the denominator of the total curvature ^{_{K (1 + f X 2 +}} f y 2)
Since ² and average denominator of curvature ^{_{H (1 + f X 2 +}} f y 2) 3/2 both always takes a positive value, total curvature K and mean curvature H
Paying attention only to the sign of, only the value of the numerator of the total curvature K and the value of the numerator of the average curvature H need to be calculated.
Then, the value of the numerator of the total curvature K and the value of the numerator of the average curvature H can be obtained only by the four arithmetic operations, so that the circuit can be simplified.

In actual calculations, image data on a two-dimensional plane
Is obtained by differentiating the two-dimensional space and f = ｛f_ijGiven by｝
Often. In this case, three vertical pixels and one horizontal pixel
When the approximate calculation is performed in a range of three pixels, for example, the pixel f
_ij(15) necessary to calculate the curvature of
Each derivative f of the equation (16)_u, F_v, F_uu, F_vvAnd f _uv
Can be obtained from equations (17) to (21), respectively.
it can.

[0032] _{f u = f x = (f} i + 1, j -f i-1, j) / 2 (17) f v = f y = (f i, j + 1 -f i, j-1) / 2 (18) f _uu = f _xx = (f _{i + 1, j} + f _{i-1, j} -2f _{i, j} ) / 2 (19) f _vv = f _yy = (f _{i, j + 1} + f _{i , j-1} -2f _{i, j} ) / 2 (20) f _uv = f _xy = (f _{i + 1, j + 1} + f _{i-1, j-1} -f _{i + 1, j-1} -f _{i -1, j + 1} ) / 4 (21) Accordingly, the value K ₁ of the numerator of the total curvature can be obtained from the equation (21).

K ₁ = f _xx f _yy −f _xy ² = (f _{i + 1, j} + f _{i-1, j} −2f _{i, j} ) (f _{i, j + 1} + f _{i, j−1} −2f _{i , j) / 4 - (f} i + 1, j + 1 + f i-1, j-1 -f i + 1, j-1 -f i-1, j + 1) 2/16 = K 11 + K 12 _{+ K 13 + K 14 + K} 15 (21) _{where, K 11 = (f i +} 1, j f i, j + 1 + f i + 1, j f i, j-1 + f i-1, j f i, j _{+1 + f i-1, j} f i, j-1) / 4 (22) K 12 = - (f i + 1, j f i, j + f i-1, j f i, j + f i, j + _{1 f i, j + f i} , j-1 f i, j) / 2 (23) K 13 = (f i, j) 2 (24) K 14 = - ((f i + 1, j + 1) 2 _{+ (f i + 1, j} -1) 2 + (f i-1, j + 1) 2 + (f i-1, j-1) 2) / 16 (25) K 15 = - (f i + _{1, j + 1} fi _{-1, j-1} + fi _{+ 1, j-1} fi _{-1, j + 1} -fi _{+ 1, j + 1} fi _{+ 1, j- 1} -fi _{+ 1, j + 1} fi _{-1, j + 1} -fi _{+ 1, j-1} fi _{-1, j-1} -fi _{-1, j + 1} fi _{-1, j-1} ) / 8 (26) the value H ₁ molecule of mean curvature can be obtained from (27) .

[0034] _{H 1 = (1+ (f y} ) 2) f xx + (1+ (f x) 2) f yy -2f x f y f xy = H 11 + H 12 + H 13 (27) where, H ₁₁ = (1+ ( _fy ) ² ) _fxx = [1 + {(fi _{, j + 1-} fi _{, j-1} ) / 2} ² ] (fi _{+ 1, j} + fi _{-1, j} -2f _{i, j) / 2 (28} ) H 12 = (1+ (f x) 2) f yy = _{[1 + {(f i + 1} , j -f i-1, j) / 2} 2 ] (f _{i, j + 1 + f i, j} -1 -2 f i, j) / 2 (29) H 13 = -2f x f y f xy = - (f i + 1, j -f i-1, j) (f _{i, j + 1} -f _{i, j-1} ) (f _{i + 1, j + 1} + f _{i-1 , j-1} -f _{i + 1, j-1} -f _{i-1, j + 1} ) / 8 (30) Next, the value K of the numerator of the total curvature represented by the above equation (21)
₁ for calculating the total curvature numerator 102 shown in FIG.
Will be described with reference to FIG.

The numerator calculator 102 for the total curvature includes the first to ninth
Second computing block for computing a shift register 111 ₁ to 111 ₉ of a first calculation block for calculating the variable K ₁₁ represented by the equation (22), a variable K ₁₂ represented by the equation (23) When a third computing block for computing a variable K ₁₃ represented by the formula (24), and a fourth computing block for computing a variable K ₁₄ represented by the equation (25), above (26) a fifth computing block for computing a variable K ₁₅ represented, the calculation of the sixth for computing the value K ₁ of the total curvature molecule represented by the above formula (21) from the output signal of the first to fifth calculation block Block and numerator value K ₁ of total curvature
And a first coded block for performing the above coding.

Here, the first to ninth shift registers 11
1 ₁ to 111 _9, pixel f _{i-1, j-1,} pixel f _{i, j-1,} pixel f
_{i + 1, j-1} , pixel fi _{-1, j} , pixel fi _{, j} , pixel fi _{+ 1, j} ,
The pixel values of the pixel _{fi-1, j + 1} , the pixel fi _{, j + 1,} and the pixel fi _{+ 1, j + 1} are stored, respectively.

[0037] The first computing block for computing a variable K _11, pixel f _{i + 1,} pixel values of _j and pixel f _i, the first multiplying circuit 112 ₁ for multiplying the pixel value of _{j + 1} , pixel f _{i + 1,} pixel values of _j and pixel f _i, and ₂ second multiplier circuit 112 for multiplying the pixel values of _j-1, pixel f _i-1, the pixel value of _j and the pixel f _{i ,} a third multiplier circuit 112 ₃ for multiplying the pixel value of _{j + 1,} pixel f _{i-1, j}
Pixel value and the pixel f _i, and a fourth multiplier circuit 112 ₄ for multiplying the pixel values of _j-1, the first multiplying circuit 112 ₁ of the output signal, the second
Multiplier circuit 112 and _second output signals, a first adder circuit 113 ₁ for adding the third multiplier circuit 112 _third output signal and the fourth multiplication circuit 112 ₄ of the output signal, the first adder circuit 113 ₁ It becomes the value of the output signal from the first divider circuit 114 ₁ Tokyo to 1/4.

[0038] The second computing block for computing a variable K _12, pixel f _{i + 1,} pixel values of _j and pixel f _i, a multiplier circuit 112 ₅ 5 for multiplying the pixel values of _j, the pixel Sixth multiplication circuit 11 for multiplying the pixel value of f _{i-1, j by} the pixel value of pixel f _{i, j}
2 _6, pixel f _{i, j + 1} pixel value and the pixel f _i, and the seventh multiplier circuit 112 ₇ for multiplying the pixel values of _j, pixel f _{i, j-1} of the pixel value and the pixel f Eighth multiplication circuit that multiplies the pixel values of _{i and j}
And 112 _8, the output signal of the fifth multiplier circuit 112 _5, the output signal of the multiplier circuit 112 ₆ sixth, the addition of the seventh multiplier circuit 112 ₇ output signal and the output signal of the eighth multiplier circuit 112 ₈ of the a second adder circuit 113 ₂ for performing the value of the second adder circuit 113 _{and second} output signal 1
/ 2, a second division circuit 114 _2, and a second division circuit 114 ₂
Consisting of a first sign inversion circuit 115 _{Tokyo and} to a negative value inverts the output signal of the code.

The third computing block for computing a variable K ₁₃ consists ninth multiplier circuit 112 ₉ for squaring the pixel values of the pixel f _{i, j.}

[0040] The fourth computing block for computing a variable K _14, pixel f _{i + 1,} the first 10 multiplication circuit 112 ₁₀ which performs the square operation _{j + 1} pixel values, pixel f _{i + 1, j} a multiplier circuit 112 ₁₁ of the 11 for squaring the pixel values of _-1, the twelfth multiplier circuit 112 ₁₂ to perform the square operation on the pixel value of the pixel f _{i-1, j + 1,} pixel f _i Thirteenth multiplication circuit that performs a square operation on pixel values of _{-1, j-1}
112 and _13, the output signal of the ninth multiplier circuit 112 _9, the output signal of the tenth multiplier circuit 112 _10, the addition of a 11 multiplying circuit 112 ₁₁ output signal and a 12 multiplier circuit 112 ₁₂ output signal of the the a third adding circuit 113 ₃ to perform, and the third divider circuit 114 ₃ to the value of the third addition circuit 113 _third output signal to 1/16, the sign of the third divider circuit 114 _third output signal reversed and made of a second sign inversion circuit 115 ₂ which is a negative value.

Variable K₁₅Fifth calculation block for calculating
Is the pixel f_{i + 1, j + 1} Pixel value and pixel f _{i-1, j-1} Pixel value of
13th multiplication circuit 112 for multiplying by₁₃ And the pixel f
_{i + 1, j-1} Pixel value and pixel f_{i-1, j + 1} Multiplication with the pixel value of
Fourteenth Multiplying Circuit 112₁₄ And the pixel f_{i + 1, j + 1} Pixel
Value and pixel f_{i + 1, j-1} 15th power of multiplying by the pixel value of
Arithmetic circuit 112₁₅ And the pixel f_{i + 1, j + 1} Pixel value and pixel f
_{i-1, j + 1} 16th multiplication circuit 112 that performs multiplication with the pixel value of
₁₆ And the pixel f_{i + 1, j-1} Pixel value and pixel f_{i-1, j-1} Painting
Seventeenth multiplying circuit 112 for multiplying by a prime value₁₇ And the pixel f
_{i-1, j + 1} Pixel value and pixel f_{i- 1, j-1} Multiplication with the pixel value of
Eighteenth Multiplying Circuit 112₁₈ And the fifteenth multiplication circuit 112
₁₅ Output signal of the sixteenth multiplication circuit 112₁₆ Output signal,
Seventeenth multiplication circuit 112₁₇Output signal and eighteenth multiplication
Circuit 112₁₈ Adding circuit 113 for adding the output signals of_Four
And the fourth adder circuit 113_FourInvert the sign of the output signal of
Of the third sign inversion circuit 115_ThreeAnd the thirteenth multiplication
Road 112₁₃ Output signal of the fourteenth multiplication circuit 112₁₄ Output signal
Signal and third sign inverting circuit 115_Three Add the output signal of
Fifth adding circuit 113_FiveAnd the fifth adder circuit 113_FiveOutput
Fourth division circuit 114 for reducing signal value to 1/8_FourAnd the fourth
Division circuit 114_FourInverts the sign of the output signal to a negative value
Fourth sign inversion circuit 115_FourConsists of

The sixth computing block for computing a molecular K ₁ of total curvature, the first sign inversion circuit 115 ₁ of the output signal, the first sign inversion circuit 115 ₁ of the output signal, the second sign inversion circuit 115
Performs addition of ₂ the output signal and the fourth sign inversion circuit 115 ₄ of the output signal consists of the addition circuit 113 ₆ sixth.

_First encoding of the numerator value K ₁ of the total curvature
Coding block, a sixth pixel f when the value of the adding circuit 113 ₆ output signal (the value K ₁ of the total curvature molecule) is a positive
The pixel value of _{i, j} is set to “1” and the sixth adder circuit
When 113 of the ₆ output signal value (a value K ₁ of the total curvature molecule) is not positive it consists of a first comparator 116 _1, pixel f _i, the pixel value of the _j "0".

Next, the specific configuration of the average curvature numerator calculator 103 shown in FIG. 1 for obtaining the average curvature numerator H ₁ represented by the above equation (27) will be described with reference to FIG.

The numerator calculator 103 for the average curvature calculates the above (2)
8) a seventh computing block for computing a variable H ₁₁ represented by formula, and an eighth computing block for computing a variable H ₁₂ represented by the formula (29), the variable represented by the formula (30) H ₁₃ , a ninth operation block for calculating the numerator value H ₁ of the average curvature represented by the above equation (27) from the output signals of the seventh to ninth operation blocks, and a second encoding block for coding the value H ₁ of curvature molecule. Incidentally, in FIG. 4, for convenience of description, the first to ninth shift register 111 ₁ to 111 ₉ also described.

Variable H₁₁The seventh operation block for calculating
Is the pixel f_{i, j-1} Invert the sign of the pixel value of
Fifth sign inversion circuit 115_FiveAnd the pixel f_{i, j + 1} Pixel value
And fifth sign inversion circuit 115_FiveOutput signal
7 adder circuit 113₇And the seventh adder circuit 113₇Output signal
Fifth division circuit 114 for halving the value_FiveAnd the fifth division
Road 114_Five19th multiplication that performs the square operation of the value of the output signal of
Road 112₁₉ And a nineteenth multiplication circuit 112₁₉ Output signal value of
Adder 113 for adding "1" to the₈And the pixel f _{i, j} 
Sixth sign inversion that inverts the sign of the pixel value of
Circuit 115₆And the sixth sign inverting circuit 115₆Output signal value
Twentieth multiplier circuit 112 for doubling₂₀ And the pixel f_{i + 1, j} of
Pixel value, pixel f_{i-1, j} Pixel value and twentieth multiplication circuit
112₂₀ Ninth adding circuit 113 for adding the values of the output signals of₉
And the ninth adder circuit 113₉Halves the output signal value of
Sixth division circuit 114₆And an eighth adder circuit 113₈Output signal
And the sixth division circuit 114₆Multiply with the output signal value of
20th multiplication circuit 112₂₀ Consists of

The eighth computing block for computing a variable H _12, a seventh sign inversion circuit 115 _7, negative values by inverting the sign of the pixel value of the pixel f _{i-1, j,} pixel f _{i + 1,} and the 10 adder circuit 113 ₁₀ which performs addition of _j pixel value and the sixth sign inversion circuit 115 ₆ output signals of the, the value of the output signal of the tenth adder circuit 113 ₁₀ 1/2 The seventh division circuit 114 ₇ and the seventh division circuit
21 to perform the square operation value of the division circuit 114 ₇ output signal
A multiplication circuit 112 _21, and the 11 adder circuit 113 _{1 1} of which adds "1" to the value of the output signal of the 21 multiplication circuit 112 _21,
The pixel value of the pixel f _{i, j + 1,} the pixel value of the pixel f _{i, j-1}
12 to perform the addition of the value of the output signal of the multiplier circuit 112 ₂₀
The adder circuit 113 _12, a divider circuit 114 ₈ of the eighth to the value of the output signal of the first 12 adder circuit 13 ₁₂ of the half, the output signal of the eleventh adder circuit 113 ₁₁ values and the eighth consisting 22nd multiplication circuit 112 ₂₂ for performing multiplication of the value of the output signal of the divider circuit 114 _8.

Variable H₁₃Ninth operation block for calculating
Is the pixel f_{i + 1, j-1} Pixel value and pixel f _{i-1, j + 1} Pixel value of
13th adder circuit 113 for adding₁₃ And the thirteenth
Arithmetic circuit 113₁₃ Invert the sign of the output signal value of
Eighth sign inversion circuit 115₈And the pixel f_{i + 1, j + 1} Pixel
Value, pixel f_{i-1, j-1} Pixel value and the thirteenth addition circuit 11
Three₁₃ Fourteenth adder circuit 113 for adding the values of the output signals of
₁₄ And the fourteenth adder circuit 113₁₄ Output signal value is 1 /
Ninth division circuit 114 for making 2₉And the seventh division circuit 114₇of
Output signal value and eighth division circuit 114₈Output signal value and
9 division circuit 114₉23rd multiplication with the value of the output signal of
Multiplication circuit 112_{twenty three} And the twenty-third multiplication circuit 112_{Two Three} Output signal
Ninth sign inversion cycle in which the sign of the sign value is inverted to a negative value
Road 115₉Consists of

The tenth calculation of the numerator value H ₁ of the average curvature is performed.
Calculation block of the performing addition of the first 20 multiplication circuit 112 ₂₀ value of the output signal of the value of the 22 values of the multiplying circuit 112 ₂₂ output signal and the ninth output signal of the sign inversion circuit 115 ₉ of 1
Consisting of five of the adding circuit 113 _15.

The numerator value H of the average curvature₁ The encoding of
The second coding block is a fifteenth adder 113₁₅ Output
Signal value (average curvature numerator value H₁ ) Is negative when
Element f _{i, j} Is set to “1” and the fifteenth processing
Arithmetic circuit 113₁₅ Output signal value (average curvature numerator value H
₁ ) Is not negative, the pixel f_{i, j} Pixel value of “0”
Second comparator 116_TwoConsists of

Next, the results of an experiment using the image processing apparatus 100 will be described with reference to FIG.

The original image 200 displayed by the contour lines input by the image information input unit 101 is composed of a combination of an elliptical point area 201 indicating recording bits and a parabolic point area 202 indicating steps or ridges. Is, by the combination of the elliptical point area 201 and the parabolic point area 202, and by discretizing the spatial coordinates as the domain of the image,
There is also a hyperbolic point region 203. By inputting the image signal indicating the original image 200 to the numerator of full curvature 102, the output image 210 of the numerator of full curvature where the parabolic point region 202 and the hyperbolic point region 203 are isolated points is obtained. Is obtained. Also, by inputting the image signal indicating the original image 200 to the average curvature numerator calculator 103, an output image 220 of the numerator of the average curvature in which the concave area and the flat area are isolated points is obtained. Therefore, the output signal of the numerator of total curvature 102 and the output signal of the numerator of average curvature 103 are ANDed.
104, the elliptical point area 201 indicating the recording bit is extracted and the logical product operation unit output image 23
0 is obtained. However, in the AND operation unit output image 230, in addition to the elliptical point area 201 indicating recording bits, an elliptical point area due to an error due to discretization of spatial coordinates is also extracted. Therefore, by inputting the output signal of the AND operation unit 104 to the median filter 105, an elliptical point extraction image 240 in which only the elliptical point area 201 indicating the recording bit is extracted is obtained.

FIG. 6 is a graph showing an experimental example of a recording bit recognition map by the image processing apparatus 100. In the figure, the horizontal axis represents the logarithmic value of the SNR _h related to the height (= l
og (record bit height / ridge height)), and the vertical axis indicates the logarithmic value of SNR _W related to the diameter (= log (record bit diameter / ridge width)). The boundary line L ₁ is a 255-level height of the recording bit, the diameter of the recording bits indicates the recognition boundary in the case of 5 pixels,
A boundary line L ₂ indicates a recognition boundary line when the recording bit height is 255 levels and the recording bit diameter is 7 pixels, and a boundary line L ₃ is when the recording bit height is 255 levels. The recognition boundary line when the recording bit diameter is 9 pixels is shown. Note that 100% of the recording bits are recognized under the conditions on the right side of each of the boundary lines L ₁ , L ₂ , and L ₃ .

From the results shown in FIG. 6, for a ridge having a certain height and a certain width (same for steps), the S
By logarithm of SNR _W about logarithm and diameter of NR _h choose the height and diameter of the recording bit to come to the right of each boundary line _{L 1, L 2, L 3} , the recording bit 100%
Can be recognized.

FIG. 7 is a schematic diagram showing one embodiment of the recording / reproducing apparatus of the present invention.

The recording / reproducing apparatus 10 includes a recording medium 11, a coarse movement control mechanism 12, a structure 13, a large coarse movement mechanism 14,
Isolation table 15, probe electrode 16, fine movement control mechanism 17
, An XY controller 18, a voltage application device 19, a probe current amplifier 20, and an arithmetic device 21. Each component of the recording / reproducing device 10 will be described in detail below.

(1) Recording Medium 11 The recording medium 11 is formed by accumulating eight layers of squarylium-bis-6-octylazulene on a graphite substrate by using the LB method. Here, squarylium-bis-
6-octylazulene has a memory effect on the switching characteristics of voltage and current. Note that the recording medium 11 is
In an initial state, it is in a non-recording state (off state).
As shown in FIG. 8, the recording surface of the recording medium 11
One general recording area (first to fourth general recording areas 30)
₁ to 30 ₄ ), and _first to _fourth position reference pattern areas 31 _{1 to} 31 ₄ are secured on the left side of the _first to _fourth general recording areas 301 to 304 in the drawing. ing.

(2) Coarse Motion Control Mechanism 12, Structure 13, Large Coarse Motion Mechanism 14, and Vibration Isolation Table 15 The coarse motion control mechanism 12 on which the recording medium 11 is placed has a parallel spring using an elastic hinge. Recording medium 1
1 for performing coarse movement control in the illustrated X-axis direction and Y-axis direction. The structure 13 is composed of Inbar for supporting each structural part of the recording / reproducing apparatus 10. The large coarse movement mechanism 14 is for performing coarse movement control of the recording / reproducing apparatus 10 in the illustrated X-axis direction and the Y-axis direction outside the control range of the coarse movement control mechanism 12. The seismic isolation table 15 on which the large coarse movement mechanism 14 is mounted is for preventing external vibrations from being transmitted to the recording / reproducing apparatus 10 and for preventing malfunction of the recording / reproducing apparatus 10 due to external vibrations. .

(3) Probe electrode 16 and fine movement control mechanism 1
7 and XY controller 18 The probe electrode 16 is used for recording and reproducing information. The tip of the probe electrode 16 is obtained by mechanically polishing or electrolytically polishing the tip of a tungsten needle in order to increase the resolution of recording and reproduction. However,
The material of the probe electrode 16 may be Pt-Ir, Pt, or the like, and the processing method is not limited thereto. The fine movement control mechanism 17 to which the probe electrode 16 is attached is configured by using a cylindrical piezoelectric element, and moves the probe electrode 16 in the Z-axis direction in the figure so that the distance between the probe electrode 16 and the recording medium 11 is increased. Is finely adjusted, and the probe electrode 16 is moved in the X-axis direction and the Y-axis direction in the drawing, so that the probe electrode 16 is two-dimensionally scanned on the surface (recording surface) of the recording medium 11. The XY controller 18 drives and controls the coarse movement control mechanism 12 and the fine movement control mechanism 17 in accordance with a command from the arithmetic unit 21. The XY controller 18 adjusts each operation voltage indicating the drive amount of the coarse movement control mechanism 12 and the fine movement control mechanism 17 by coarse movement. Output to the control mechanism 12 and the fine movement control mechanism 17, respectively.

(4) Voltage Applying Apparatus 19 and Probe Current Amplifier 20 The voltage applying apparatus 19 applies a recording voltage, between the probe electrode 16 and the recording medium 11 in accordance with a command from the arithmetic unit 21.
A reproducing voltage and an erasing voltage are applied, respectively. Note that the voltage applying device 19 can also apply an arbitrary bias voltage between −10 V and +10 V between the probe electrode 16 and the recording medium 11. The probe current amplifier 20 detects a current flowing between the probe electrode 16 and the recording medium 11 and amplifies the detected current. The amplified current is supplied to the probe current amplifier 20.
After that, it is output from the probe current amplifier 20 to the arithmetic unit 21.

(5) Arithmetic Unit 21 The arithmetic unit 21 includes an A / D converter for converting the voltage sent from the probe current amplifier 20 into a digital signal, and an XY converter.
A microprocessor for controlling the controller 18 and the voltage application device 19; a position reference memory for storing position references on the recording medium 11; a processor for pattern-recognizing and coding information to be recorded; and A / D conversion An image processing circuit that creates image information from a digital signal output from the circuit and extracts only a recording bit area from the created image information. Note that image processing in this image processing circuit is performed in the same manner as in the image processing apparatus 100 shown in FIG.

Next, the position reference pattern and the recording signal area on the recording medium 11 in the recording / reproducing apparatus 10 will be described.
This will be described with reference to FIG.

In the recording / reproducing apparatus 10, before information is recorded in the _first to _fourth general recording areas 30 ₁ to 304 of the recording medium 11, the position reference pattern is changed according to the following procedure. Fourth position reference pattern areas 31 _{1 to} 31 ₄
Respectively.

According to the mechanical accuracy of the structure 13, the recording medium 1
Tip of the probe electrode 16 on one is a home position approximately 200μm from H _P in FIG. 7 the Z-axis direction of the recording medium 11
It is located at a remote position with an accuracy of ± 10 μm. 300
The bias voltage of mV is applied between the probe electrode 16 and the recording medium 1.
1 is applied from the voltage application device 19. A current flowing between the probe electrode 16 and the recording medium 11 is detected and amplified by the probe current amplifier 20 and then converted into a voltage. In the arithmetic unit 21, the probe electrode 16 and the recording medium 11 are calculated based on the voltage sent from the probe current amplifier 20.
The coarse movement control mechanism 12 is driven by the arithmetic unit 21 until the detected current value becomes 1 pA. As a result, the tip of the probe electrode 16 is brought closer to the recording surface of the recording medium 11 until the value of the current flowing between the probe electrode 16 and the recording medium 11 becomes 1 pA. Fine movement control mechanism 17
Is similarly driven by the arithmetic unit 21 so that the tip of the probe electrode 16 approaches the recording surface of the recording medium 11 until the value of the current flowing between the probe electrode 16 and the recording medium 11 becomes 1 nA. Can be Thereafter, in the area of the recording medium 11 surrounded by the dotted line in FIG.

When the probe electrode 16 is scanned to the recording position of the _first position reference pattern area 311, +10 V which is larger than the threshold value of the on / off information of the recording medium 11.
Is applied from the voltage applying device 19 between the probe electrode 16 and the recording medium 11. This allows
This portion of the recording medium 11 is set to a recording state (ON state), and the position reference pattern is recorded. When recording the position reference pattern, the probe electrode 16 and the recording medium 11 are used.
The fine movement control mechanism 17 is controlled by the arithmetic unit 21 so that the average value of the current flowing between the probe electrode 16 and the recording medium 11 is 0.9 nA in order to eliminate the influence on the distance control between the probe electrode 16 and the recording medium 11. Accordingly, the average distance between the probe electrode 16 and the recording medium 11 is made constant.

By repeating the above operation, the
First to fourth position reference pattern areas 31₁~ 31_FourTo the rank
The placement reference patterns are sequentially recorded. In addition,
The first to fourth position reference pattern areas 31₁~ 31_Four
The direction pattern and the position pattern 36₁~ 36_FourWhen
Are respectively recorded. Here, the first position reference pattern
Area 31₁ Direction patterns 35 recorded in the₁₁
~ 35₁₃And the third position reference pattern area 31₁, 31_Three
Direction patterns 35 recorded in the₃₁~ 35 ₃₃When
Are recorded at positions corresponding to each other. Also,
Second position reference pattern area 31_Two Three recorded in
Direction pattern 35_{twenty one}~ 35_{twenty three}And the fourth position reference pattern
Region 31_FourDirection patterns 35 recorded in the
₄₁~ 35₄₃Are recorded at the positions corresponding to each other.
It is. Further, each general recording area 30₁~ 30_FourAnd each position group
Quasi-pattern 31₁~ 31_FourBetween the first and fourth
Period pattern 37₁~ 37_FourAre also recorded.

Next, description each operation for reproducing information from the general recording area 30 ₁ to 30 information recording operation and the general recording area 30 ₁ to 30 ₄ to ₄ in a recording and reproducing apparatus 10.

First, according to the following procedure, the probe electrode 1
6 is performed. A bias voltage of +5 V which is smaller than the threshold value of the ON / OFF state (recording / non-recording state) of the recording medium 11 is applied from the voltage applying device 19 between the probe electrode 16 and the recording medium 11. In this state, the probe electrode 16 is two-dimensionally scanned over the entire recording surface of the recording medium 11. At this time, the value of the current flowing between the probe electrode 16 and the recording medium 11 is detected, and the recording medium 1 is detected.
The position reference pattern is detected by detecting the on / off state of No. 1.

The position reference pattern corresponds to the general recording area 30 ₁
30 ₄ because it is recorded in a wide recording unit than the recording unit of information recorded in the respective position reference pattern area 31
_{1-31 4} current detected in each general recording area 3
0 becomes lower frequency than the current detected in ₁ - 30 _4. Therefore, by passing the detected current through the band-pass filter, the position reference pattern and the information can be approximately separated. Based on the two-dimensional information of each separated position reference pattern, the probe electrode 16
One of the four position reference pattern areas 31 _{1 to} 31 ₄
Are moved around the two position reference pattern areas. continue,
The probe electrode 16 is two-dimensionally scanned over a slightly wider range than the position reference pattern area.
Is stored in the memory as two-dimensional information.

[0070] As described above, the first and third position reference pattern area 31 _1, 31 to one another corresponding position of the _3, three directions patterns 35 _{11-35 13,} 35 _31-3
5 ₃₃ are recorded respectively, and the second and fourth
The position reference pattern area 31 _2, 31 mutually corresponding positions of _4, three directions patterns 35 _{21-35 23,} 35
_{Since 41 to} 35 ₄₃ are recorded respectively, the scanning direction of the probe electrode 16 is changed to the pattern 35 _{11 to} 3 for each direction.
5 ₁₃ , 35 _{31 to} 35 ₃₃ , 35 _{21 to} 35 ₂₃ , 35 _{41 to} 35
It can be seen by detecting ₄₃ . Further, in the first to fourth position reference pattern area 31 _{1-31 4,} the position patterns 36 ₁ to 36 ₄ are respectively recorded, by detecting the respective position patterns 36 ₁ to 36 ₄ And the relative position of the probe electrode 16 can be understood. That is, since each position reference pattern includes the direction information and the position information, the information on the scanning direction of the probe electrode 16 and the position information can be obtained by using a generally used image processing technique. Further, information on the approximate position of the position reference pattern in the general recording area to be scanned next can be obtained from the positional relationship between the position reference patterns.

Recording of information is performed according to the following procedure. The probe electrode 16 is moved to, for example, the periphery of the _first general recording area 301 based on the obtained position information. Then, over a range wider than the first general recording area 30 _1, the probe electrode 16 is scanned two-dimensionally. First
After detecting the first synchronization pattern 37 ₁ of the sync signal recorded on the position reference pattern area 36 _1, based on the position information obtained, the first predetermined recording position of the general recording area 30 _1, Information is recorded with the accuracy of the piezoelectric element.

Further, for example, the first general recording area 30₁ 
Of the information recorded in the first synchronization pattern 37₁ 
Over the entire area after the synchronization signal of
Two-dimensional scanning is performed, and the two-dimensional information obtained at this time is separately
Hysteresis of input piezoelectric element and recording medium
Considering information such as the temperature and humidity expansion and contraction of FIG.
The recording unit (unit information area) as shown in FIG.
Area), and image processing is performed for each recording unit.
It is done by doing. First general recording area 30 ₁ Written in
When playback of the recorded information is completed, based on the obtained location information,
Then, the probe electrode 16 is connected to the second general recording area 30._Two 
Moved around.

[0073]

Since the present invention is configured as described above, the following effects can be obtained.

According to the first aspect of the invention (the image processing apparatus of the present invention), when it is known in advance that an elliptical point on a curved surface that is convex with respect to a recording surface indicates necessary information, the total curvature is calculated. By calculating the logical product of the output signal of the numerator of the numerator and the output signal of the numerator of the average curvature, it is possible to eliminate erroneous information included in the image information and extract only necessary information. In addition, by inputting the output signal of the AND operation means to the median filter, it is possible to eliminate erroneous information due to errors due to discretization of spatial coordinates, thereby providing high parallelism independent of the number of pixels. It is sufficient to obtain the value of the numerator of the total curvature and the value of the numerator of the average curvature, so that the calculation speed can be improved.

The invention according to the second and third aspects (the recording / reproducing apparatus of the present invention) is characterized in that a parabolic point, a hyperbolic point included in the image information and a curved surface concave with respect to the recording surface are provided. By removing the elliptical points, it is possible to obtain new image information by extracting only the recording bits that are elliptical points on the curved surface that is convex with respect to the recording surface. It is possible to reduce a decrease in the error rate during reproduction.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating an embodiment of an image processing apparatus according to the present invention.

FIG. 2 is a diagram showing shapes of ridges, steps, and recording bits in atomic order existing on a recording surface of a recording medium;
(A) is a diagram showing a shape of a ridge, (B) is a diagram showing a shape of a step, and (C) is a diagram showing a shape of a recording bit.

FIG. 3 is a block diagram showing a specific configuration of a numerator calculator of all curvatures shown in FIG. 1;

FIG. 4 is a block diagram showing a specific configuration of a numerator calculator for average curvature shown in FIG. 1;

FIG. 5 is a diagram for explaining an experimental result using the image processing apparatus shown in FIG. 1;

FIG. 6 is a graph showing an experimental example of a recording bit recognition map by the image processing apparatus shown in FIG. 1;

FIG. 7 is a schematic configuration diagram showing one embodiment of a recording / reproducing apparatus of the present invention.

FIG. 8 is a diagram for explaining a position reference pattern and a recording signal area on a recording medium in the recording / reproducing apparatus shown in FIG. 7;

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 10 Recording / reproducing apparatus 11 Recording medium 12 Coarse movement control mechanism 13 Structure 14 Large coarse movement mechanism 15 Seismic isolation table 16 Probe electrode 17 Fine movement control mechanism 18 XY controller 19 Voltage application circuit 20 Probe current amplifier 21 Arithmetic unit 30 ₁ -30 ₄ General recording areas 31 _{1 to} 31 ₄ First to fourth position reference pattern areas 35 _{11 to} 35 ₁₃ , 35 _{21 to} 35 ₂₃ , 35 _{31 to} 35 ₃₃ , 3
5 _{41 to} 35 ₄₃ direction pattern 36 _{1 to} 36 ₄ position pattern 37 _{1 to} 37 ₄ synchronous pattern 100 Image processing unit 101 Image information input unit 102 Molecular calculator for full curvature 103 Molecular calculator for average curvature 104 Logical product operation Part 105 median filter 111 _{1 to} 111 ₉ shift register 112 _{1 to} 112 ₂₂ multiplication circuit 113 _{1 to} 113 ₁₅ addition circuit 114 _{1 to} 114 ₉ division circuit 115 _{1 to} 115 ₉ sign inversion circuit 116 ₁ , 116 ₂ comparator 200 original image 201 Elliptical point area 202 Parabolic point area 203 Hyperbolic point area 210 Output image of numerator calculation unit of full curvature 220 Output image of numerator calculation unit of average curvature 230 Output image of AND operation unit 240 Elliptical point extraction image K ₁ The value of the numerator of the total curvature H ₁ The value of the numerator of the average curvature K _{11 to} K ₁₅ , H _{11 to} H ₁₃ Variables f _{i−1, j−1} , f _{i, j−1} , f _{i + 1, j−1} , F _{i-1, j} , f
_{i, j} , fi _{+ 1, j} , fi _{-1, j + 1} , fi _{, j + 1} fi _{+ 1, j + 1}
Pixels X, Y, Z axis H _P home position

──────────────────────────────────────────────────続 き Continuing from the front page (72) Inventor Shunichi Shito 3-30-2 Shimomaruko, Ota-ku, Tokyo Inside Canon Inc. (72) Inventor Kunihiro Sakai 3-30-2 Shimomaruko, Ota-ku, Tokyo Canon (56) References JP-A-6-290269 (JP, A) JP-A-6-259546 (JP, A) JP-A-4-155475 (JP, A) JP-A-63-161552 (JP, A) A) JP-A-63-161553 (JP, A) (58) Fields investigated (Int. Cl. ⁷ , DB name) G06T 5/20

Claims (3)

(57) [Claims] 1. A numerator of total curvature for calculating a value of a numerator of total curvature at each pixel of image information, and a numerator of average curvature for calculating a value of a numerator of average curvature at each pixel of the image information. AND operation means for calculating the logical product of the output signal of the numerator calculation means of the total curvature and the output signal of the numerator calculation means of the average curvature, and a median filter to which the output signal of the AND operation means is inputted An image processing apparatus including: 2. A recording medium using a signal caused by a physical interaction generated between the probe electrode and the recording medium while the probe electrode is two-dimensionally scanned relative to a recording surface of the recording medium. In the recording / reproducing apparatus for recording information in the image information and reproducing the recorded information as image information, the numerator of the total curvature and the numerator of the average curvature in each pixel of the image information reproduced from the signal by the physical interaction A recording / reproducing apparatus including image information creating means for creating new image information by calculating the respective values of 3. The image information generating means, wherein: a total curvature calculating means for calculating a value of a numerator of a total curvature in each pixel of the image information reproduced from the signal due to the physical interaction; An average curvature calculating means for calculating a value of a numerator of an average curvature; an AND operation means for calculating an AND of an output signal of the numerator calculating means of the total curvature and an output signal of the numerator calculating means of the average curvature; 3. The recording / reproducing apparatus according to claim 2, further comprising a median filter to which an output signal of the AND operation means is input.