KR20070002952A - Method for optimally detecting frame on ccd image in a hdds - Google Patents

Abstract

An optimum frame detection method in an HDDS(Holographic Digital Data Storage) system is provided to be capable of exactly detecting a reference frame from a data image taken by a CCD(Charge-Coupled Device), thus holographic digital data of the data image is exactly decoded. A data image of page unit stored in a storage medium of an HDDS system is taken, and the taken image is converted into electric data(S200). A sum of pixel values of each line is obtained in row or column direction from the data image of the page unit of a CCD(S202). Values that divide a pixel sum of plural lines which are peripherally adjacent by a proportional constant are subtracted from a pixel sum of the current lines(S204). A line having a maximum value among the subtracted values is detected as a reference frame(S206,S208).

Description

Optimum Frame Detection Method in Holographic Digital Data Storage System {METHOD FOR OPTIMALLY DETECTING FRAME ON CCD IMAGE IN A HDDS}

1 is a diagram illustrating a page data image ccd photographed when decoding in a conventional holographic data storage system,

2 is a line sum distribution diagram of the page data image of FIG. 1;

3 illustrates a conventional 6: 8 balanced coded page data image.

4 is a line sum distribution diagram of the page data image of FIG. 3;

5 is a line sum distribution diagram of a page data image reflecting a sum and a difference of a conventional adjacent line;

6 is a diagram illustrating an image of a page data photographed by a cmos camera during decoding in a conventional holographic data storage system.

7 is an enlarged view illustrating a part of the page data image of FIG. 6;

8 is a line sum distribution diagram of the page data image of FIG. 6;

9 is a line sum distribution diagram of a page data image reflecting a sum and a difference of adjacent neighboring lines in the page data image of FIG. 6;

10 is a schematic block diagram of a holographic data storage system to which an embodiment of the present invention is applied;

11 is a flowchart illustrating an optimal frame detection process by adjusting opt_val according to an image characteristic according to an embodiment of the present invention;

12 is a line sum distribution diagram of a ccd-photographed page data image according to an embodiment of the present invention;

FIG. 13 is an enlarged diagram illustrating line sums detected as frames in FIG. 12. FIG.

<Brief description of the major symbols in the drawings>

100: light source 102: optical separator

104, 110: shutters 106, 112: reflection mirrors

108: actuator 114: spatial light modulator

116, 122: optical lens 118: aperture

120: storage medium 124: CCD

126: data preprocessing unit 128: DSP unit

130: coding unit

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a holographic digital data storage system (HDDS), and more particularly, to detect a frame in an image photographed on a charge-coupled device (CCD) that detects data reproduced on a holographic storage medium. It is about a method.

In general, a holographic digital data storage system is a page-oriented memory input / output method using a volume hologram principle in terms of data recording / reproducing, and uses a parallel data processing method to achieve an ultra-high speed input / output speed of 1 Gbps or more. It is a next-generation memory system that can be realized very quickly with a data access time of less than 100ms because the system configuration without the mechanical driving part is possible.

Such a holographic digital data storage system uses a recording medium such as a crystal that reacts sensitively to the amplitude of an interference fringe when an interference fringe generated when the object light from a target object and a reference light interfere with each other during recording. Record it. By varying the angle of the reference light, it is possible to record the intensity and phase of the object light to display the three-dimensional image of the object, and also to display hundreds to thousands of holographic digital units composed of pages of binary data. Data can be recorded on the recording medium.

Also, when reproducing holographic digital data recorded on the storage medium, the object light separated from the light source is blocked, and only the reference light is deflected at a predetermined reproduction angle to irradiate the storage medium so that the recorded interference fringe is used to reproduce the reference light for reproduction. Diffraction is demodulated into an information image, a page of binary data consisting of the original pixel contrast, taken by the CCD, and restored by the digital signal processor (DSP) to the original holographic digital data.

On the other hand, currently developed holographic worms (H-WORM) data is not a bit based pattern (bit based pattern), but a page based pattern (page based pattern). Therefore, in order to accurately determine the data, it is necessary to know where the data of the page starts. In addition to the data, a frame is added to determine the starting position of the data. Theoretically, the frame should be covered sufficiently by the current algorithm, but the information on one page is lost because the frame is not detected properly due to the special pattern at the time of encoding and the light distribution at the time of reading. Often occurs.

The conventional frame detection method for this purpose is a line sum for a certain area (left, top), (right and bottom) of the page, and determines the line having the largest value among the line sums as a frame, The decoding is performed based on the data area of the data. This comparison may be seen as a desirable algorithm for selecting the largest value for the entire page or for a selected portion of the page, but in the real bed, due to the distribution of light, one of the lines of data rather than the frame where the center of light is located. There is a probability that is detected as a frame.

Actually, the line sum is executed to find the frame in the dotted line area on the left. In the case of the page such as the image shown in FIG. 1, since one line of the data area has the largest sum value, the H / W decoder Large data errors occur during data decoding. The reason for this is that as mentioned above, since the light distribution is located in the center, the light in the center is stronger than the upper left direction in which the frame is located. FIG. 2 is a plot of the line sum distribution of the image of FIG. 1. As shown in FIG. 2, the actual frame is A, but it is determined as B in the H / W decoder to detect an error in data decoding. Generate.

FIG. 3 is an example of an image generated by repeating data in a 2 × 2 oversampling image (in this case, a 6: 8 balanced pattern generated by repetition of '0'). As shown in FIG. With frames in, there are lines in the data part. The same pattern is repeated. When the image generated by the camera is actually decoded by the H / W decoder, as shown in FIG. 4, the actual frame is A, but all lines are detected as if the frame is not effective.

Therefore, in order to solve the problem caused by the above light distribution, it is solved by obtaining a relative maximum value of several lines around the line to be compared. That is, the relative value of the current reference line is obtained as follows.

Sum_optimal = Sum _ｉ- (Sum _ｉ-1 + Sum _{ｉ + 1} )

By comparing the line sums in the same way as above, the maximum value is eliminated, which eliminates the sum of the adjacent lines from the current line, thus solving the general data pattern for the light distribution. However, as mentioned above, if a particular pattern is repeated, it won't be solved. If both sides of the i-th line are connected by '0' (off) bits, there is a possibility that the data area will be detected as a frame in the above-mentioned comparison, and so it is.

So, to find frames more effectively, we need to look at the features of 6: 8 balanced code. FIG. 3 schematically shows how encoding 6: 8 data “00100111” appears repeatedly. As described above, this is a repetition of a pattern judged as a frame in the data area. Since the encoding data is balanced, there may be up to two '0' (off) lines on both sides of '1' (on). This also applies to repetitions such as "11001001".

Therefore, conventionally, as shown in Equation 2 below, the sum of neighboring four-line pixel adjacent to the sum of the current line pixels (Sum _i ) is summed together (Sum _i-2 + Sum _i-1 + Sum _{i +).} Sum_optimal of the line pixels having the maximum value minus ₁ + Sum η _{+ 2} ) is detected as a reference frame.

Sum_optimal = Sum _ｉ- (Sum _ｉ-2 + Sum _ｉ-1 + Sum _{ｉ + 1} + Sum _{ｉ + 2} )

Furthermore, as shown in Equation 3 below, the sum of the adjacent six-line pixel sums from the sum of the current line pixels (Sum _i ) (Sum _i-3 + Sum _i-2 + Sum _i-1) The sum (Sum_optimal) of the line pixels having the maximum value minus + Sum _{i + 1} + Sum _{i + 2} + Sum _{i + 3} ) is detected as the reference frame.

Sum_optimal = Sum _i- (Sum _i-3 + Sum _i-2 + Sum _i-1 + Sum _{i + 1} + Sum _{i + 2} + Sum _{i + 3} )

Then, in ccd, when the captured data image is 2 × 2 oversampling, the sum of the sum of line pixels (Sum_optimal) of the maximum value as shown in Equation 4 below is adjacent to the sum of the current line pixels (Sum _i ). This is obtained by subtracting the sum of the four-line pixels (2 pixels each for left and right) (Sum _{i-2 × 2} + Sum _{i-1 × 2} + Sum _{i + 1 × 2} + Sum _{i + 2 × 2} ).

Sum_optimal = Sum _i- (Sum _{i-2 × 2} + Sum _{i-1 × 2} + Sum _{i + 1 × 2} + Sum _{i + 2 × 2} )

In addition, as shown in [Equation 5], the sum of the maximum line pixels Sum_optimal is the sum of neighboring six-line pixels adjacent to each other (sum 3 at each right) from the sum of the current line pixels Sum _i . It is calculated by subtracting the sum (Sum _{i-3 × 2} + Sum _{i-2 × 2} + Sum _{i-1 × 2} + Sum _{i + 1 × 2} + Sum _{i + 2 × 2} + Sum _{i + 3 × 2} ).

Sum_optimal = Sum _i- (Sum _{i-3 × 2} + Sum _{i-2 × 2} + Sum _{i-1 × 2} + Sum _{i + 1 × 2} + Sum _{i + 2 × 2} + Sum _{i + 3 × 2} )

When comparing the sum_optimal values in the same way as above, the CCD line Sum_optimal at the position where the frame exists is the largest value regardless of light distribution or repetition of a specific pattern. That is, the above method compares based on the relative maximum value in a specific region, and thus it is possible to find the frame effectively regardless of how the light distribution is arranged. In addition, since the maximum value is obtained for the maximum area using the feature of 6: 8 encoding data, the frame can be surely found even if a specific pattern is repeated. FIG. 5 illustrates an example of a graph in which a frame is properly found when a specific pattern is repeated as described above.

However, even in the above effective method, when the camera is changed or the media is changed, unexpected variables are generated. In other words, it means a phenomenon in which the intensity is somewhat high in the portion that should not be 'on' during recording. If this happens regularly, you can find frames using the method above, but it's not. In other words, if you follow the above method, you may find the wrong frame.

6 is a photograph captured by a CMOS camera of an image reproduced from the media. FIG. 7 is an enlarged view of a portion of FIG. 6. That is, referring to FIG. 7, an unexpected pattern <FPN (fixed pattern noise, etc.)> Occurs outside the image region, and this pattern interferes with a method of finding a frame by using a difference from an adjacent line, thereby making a wrong line. Make frame find. Above the actual frame position is (153,29). A simple comparison of sums and a difference between adjacent lines yields the following results:

That is, referring to FIGS. 8 and 9, the value of the line on the other side is larger than the 29th line, which is the original frame position, to find another place as the frame. 8 is based on the light distribution, and FIG. 9 is because the values of the lines subtracting the value from the reference line are large. In addition, since the values outside the frame have almost '0' values, and the reference values thereof are near '0', the result as shown in FIG. 9 is obtained.

Accordingly, an object of the present invention is to provide a method for detecting a frame in an image photographed in a charge-coupled device (CCD) for detecting data reproduced in a holographic storage medium.

The present invention for achieving the above object in the frame detection method of the holographic digital data storage system (Holographic Digital Data Storage System), in which the image of the data stored in the page unit stored in the storage medium of the system as electrical data Converting, obtaining a pixel sum value of each line in a row or column direction in the image data of the page unit of the CCD, and calculating a pixel sum of a plurality of adjacent adjacent lines in the pixel sum of the current line. Subtracting a value divided by), and detecting a line having a maximum value among the subtracted values as a reference frame.

Hereinafter, with reference to the accompanying drawings will be described in detail the operation of the preferred embodiment according to the present invention.

10 is a system configuration diagram illustrating a holographic digital data storage system for frame detection according to the present invention.

Referring to FIG. 10, a holographic digital data storage system using frame detection according to the present invention includes a light source 100, a light separator 102, a reflection mirror 106 and 112, a spatial light modulator 114, and a storage medium ( 120, a CCD 124, a data preprocessor 126, a DSP unit 128, a microcomputer 130, a coding unit 132, and the like. Here, unexplained reference numerals 104 and 110 denote shutters and 108 denote actuators. Reference numerals 116 and 122 denote Fourier lenses and 118 denote apertures.

The light source 100 generates laser light, and the light separator 104 splits the light generated by the light source 100 into the reference light path S1 and the signal light path S2. The reflection mirror 106 reflects the light transmitted from the optical separator 102 to the storage medium 116 at a predetermined angle (the same angle as the proxy angle). The reflection mirror 112 reflects the light reflected by the light separator 102 to the spatial light modulator 114.

The spatial light modulator 114 carries input data of the coding unit 132, that is, binary data of a plurality of pixels composed of pages in units of light reflected through the reflection mirror 112 at the time of recording, and converts the signal into a signal light. 116 and aperture 118 to the storage medium 116.

The storage medium 120 records the interference fringes, which are holographic digital data generated by the interference of the signal light provided from the spatial light modulator 114 and the reference light which is optical in the reflection mirror 106 at the time of recording, and is recorded by irradiation of only the reference light during reproduction. The reproduced interference fringe is diffracted and reproduced. The CCD 124 captures a page of a binary pixel image in the form of an interference fringe reproduced in the storage medium 120 through the Fourier lens 122 and converts the image into electrical data.

The data preprocessor 126 of the present invention detects a reference frame from the image data photographed by the CCD 124, but obtains the sum of pixel values of each line in the row and column directions, and the neighboring majority The sum of the pixels of the lines is added to each other, and these differences are compared to detect a value having a large difference as a reference frame.

The DSP unit 128 extracts the value of the actual holographic digital data in the data area based on the reference frame of the data preprocessor 126, digitally processes the signal, and decodes the signal processed data through the decoding unit not shown. Restore and output the holographic digital data of.

The coding unit 130 of the present invention codes binary data of a plurality of pixels, which are configured in units of pages, as input data and inputs the input data to the spatial light modulator 114. The pixel difference, minus the spacing between the first row (or column) pixels and the reference frame, is coded at least two pixels or three pixels.

11 is a flowchart sequentially illustrating a frame detection method of a holographic digital data storage system according to the present invention. Hereinafter, a frame detection method of a holographic digital data storage system according to the present invention will be described with reference to FIGS. 10 and 11.

First, the laser light of the light source 100 branches through the light separator 104 and is transmitted through the reflection mirror 106 to the storage medium 116 at a predetermined angle (the same angle as the proxy angle) and the light separator 104. When the other light branched at is blocked by the shutter 110, the storage medium 120 reproduces the diffraction pattern recorded by irradiation of only the reference light. The CCD 124 photographs a binary pixel image of a page in the form of an interference fringe reproduced by the storage medium 120 and converts the image into electrical data (S200).

The data preprocessor 126 detects a reference frame from the image data photographed by the CCD 124, but obtains a sum of pixel values of the lines in the row and column directions (S202).

As shown in Equation 6 below, the data preprocessor 126 adds adjacent four-line pixel sums Sum _i-2 + Sum _i-1 + Sum _{i + 1} + in the sum of the current line pixels Sum _i . Sum_optimal of the line pixels having the maximum value is obtained as the reference frame by subtracting the sum _{i + 2} ) by the proportional constant (opt_val) (S204 to S206).

Sum_optimal = Sum _i- (Sum _i-2 + Sum _i-1 + Sum _{i + 1} + Sum _{i + 2} ) / (opt_val)

Further, as shown in Equation 7 below, the sum of neighboring six-line pixels (Sum _i-3 + Sum _i-2 + Sum _i-1 + Sum _{i +} ) adjacent to the sum of the current line pixels (Sum _i ). The sum (Sum_optimal) of the line pixels having the maximum value is detected as the reference frame by subtracting ₁ + Sum _{i + 2} + Sum _{i + 3} ) by the proportional constant (opt_val) (S208).

Sum_optimal = Sum _i- (Sum _i-3 + Sum _i-2 + Sum _i-1 + Sum _{i + 1} + Sum _{i + 2} + Sum _{i + 3} ) / (opt_val)

Then, in ccd, when the captured data image is 2 × 2 oversampling, the sum of the sum of the maximum line pixels (Sum_optimal) is adjacent to the sum of the current line pixels (Sum _i ), as shown in Equation 8 below. Sum of 4 line pixels (2 pixels each left and right) (Sum _{i-2 × 2} + Sum _{i-1 × 2} + Sum _{i + 1 × 2} + Sum _{i + 2 × 2} ) divided by proportional constant (opt_val) Is obtained by subtracting the value,

Sum_optimal = Sum _i- (Sum _{i-2 × 2} + Sum _{i-1 × 2} + Sum _{i + 1 × 2} + Sum _{i + 2 × 2} ) / (opt_val)

In addition, as shown in Equation 9, Sum (optimum) of the maximum line pixels (Sum_optimal) is summed from the sum of the current line pixels (Sum _i ), adjacent neighboring six-line pixels (3 pixels each of left and right). _{i-3 × 3} + Sum _{i-2 × 3} + Sum _{i-1 × 3} + Sum _{i + 1 × 3} + Sum _{i + 2 × 3} + Sum _{i + 3 × 3} ) divided by the proportional constant (opt_val) Find by subtracting

Sum_optimal = Sum _i- (Sum _{i-3 × 2} + Sum _{i-2 × 2} + Sum _{i-1 × 2} + Sum _{i + 1 × 2} + Sum _{i + 2 × 2} + Sum _{i + 3 × 2} ) / ( opt_val)

Therefore, the above-mentioned problem of the conventional method can be solved by the above formula. When finding the frame of an image, one of the above formulas is used. If the current reference line is 'i', if the adjacent line is pixel matching, i + 1, i-1, i + 2 , i-2. In addition, in the case of 2 × 2, adjacent lines become i + 1 * 2, i-1 * 2, i + 2 * 2, and i-2 * 2 to define a frame position with respect to the original image. Opt_val can be obtained experimentally. In the present invention, '2' is used as an embodiment. When the sum of the adjacent lines is divided by the opt_val value '2', the value is significantly lower than that of the reference line, so that the frame is not detected outside the frame and the frame can be accurately detected.

12 and 13 illustrate graphs of detecting frames of the image of FIG. 6 by setting opt_val to '2';

As shown in FIG. 12 and FIG. 13, it can be seen that the position of the frame is correctly found, and the case of incorrectly finding the frame as in FIG. 9 disappears. As described above, the present invention can complement the conventional method and detect the frame more effectively against the adverse effects of the image shifting caused by the characteristics of the camera or the rotation of the disk.

Meanwhile, in the above description of the present invention, specific embodiments have been described, but various modifications may be made without departing from the scope of the present invention. Therefore, the scope of the invention should be determined by the claims rather than by the described embodiments.

As described above, the present invention provides an optimal frame detection method in a holographic digital data storage system, wherein the sum of pixel values of each line in a row and column direction in an image captured by a CCD is obtained, and pixels of a plurality of adjacent lines are obtained. The sum is subtracted by a value obtained by dividing by a proportional constant (opt_val) set according to the characteristics of the image, and the line having the maximum value among them is detected as the reference frame. Accordingly, the present invention has the advantage of accurately detecting the reference frame in the data image photographed by the CCD and accurately decoding the holographic digital data of the data image.

Claims (5)

In the frame detection method of the holographic digital data storage system (Holographic Digital Data Storage System), Photographing a data image of a page unit stored in a storage medium of the system and converting the image into electrical data; Obtaining a pixel sum value of each line in a row or column direction in a data image of a page unit of the CCD; Subtracting a value obtained by dividing a pixel sum of adjacent multiple lines by a proportional constant (opt_val) from the pixel sum of the current line; Detecting a line having a maximum value among the subtracted values as a reference frame Frame detection method of a holographic digital data storage system comprising a. The method of claim 1, The sum Sum_optimal of the line pixels of the maximum value detected by the reference frame is obtained by the following equation. [Equation] Sum_optimal = Sum _i- (Sum _i-2 + Sum _i-1 + Sum _{i + 1} + Sum _{i + 2} ) / (opt_val) here, Sum _ｉ : Sum of current line pixels Sum _ｉ-2 , Sum _ii-1, Sum _{ii + 1,} Sum _{ｉ + 2} : Sum of the pixels in the surrounding 4 lines, subtracting or adding 1 pixel, 2 pixels from the current line, respectively opt_val: proportional constant The method of claim 1, The sum Sum_optimal of the line pixels of the maximum value detected by the reference frame is obtained by the following equation. [Equation] Sum_optimal = Sum _i- (Sum _i-3 + Sum _i-2 + Sum _i-1 + Sum _{i + 1} + Sum _{i + 2} + Sum _{i + 3} ) / (opt_val) here, Sum _ｉ : Sum of current line pixels Sum _i-3 , Sum _i-2, Sum _i-1 , Sum _{i + 1,} Sum _{i + 2,} Sum _{i + 3} : Each of the six surrounding lines plus or minus one pixel, two pixels, or three pixels from the current line Sum of pixels opt_val: proportional constant The method of claim 1, The sum Sum_optimal of the line pixels of the maximum value detected by the reference frame is obtained by the following equation. [Equation] Sum_optimal = Sum _i- (Sum _{i-2 × 2} + Sum _{i-1 × 2} + Sum _{i + 1 × 2} + Sum _{i + 2 × 2} ) / (opt_val) here, Sum _ｉ : Sum of current line pixels Sum _{i-2 × 2,} Sum _{i-1 × 2} , Sum _{i + 1 × 2,} Sum _{i + 2 × 2} : Subtract or add 2x2 oversampled 1 pixel, 2 pixels to the CCD on the current line, respectively Sum of pixels around 4 lines opt_val: proportional constant The method of claim 1, The sum Sum_optimal of the line pixels of the maximum value detected by the reference frame is obtained by the following equation. [Equation] Sum_optimal = Sum _i- (Sum _{i-3 × 2} + Sum _{i-2 × 2} + Sum _{i-1 × 2} + Sum _{i + 1 × 2} + Sum _{i + 2 × 2} + Sum _{i + 3 × 2} ) / ( opt_val) here, Sum _ｉ : Sum of current line pixels Sum _{i-3 × 2} , Sum _{i-2 × 2,} Sum _{i-1 × 2} , Sum _{i + 1 × 2,} Sum _{i + 2 × 2,} Sum _{i + 3 × 2,} respectively. Sum of pixels around 6 lines subtracted or added to × 2 oversampled 1 pixel, 2 pixels, or 3 pixels opt_val: proportional constant

Images