KR100749100B1 - Method for detecting optical information and optical information detector - Google Patents

Abstract

The present invention relates to an optical information detecting method and an optical information detector, and more particularly, to detecting an image of a source data page as 1: n (n is a non-integer greater than 1) overpixel; Detecting a matching identification region in the image of the detected data page; Determining a matching case using the light distribution of the detected matching identification region; And performing sampling to restore an image of the source data page according to a predetermined sampling pattern corresponding to the matched state of the determined frame mark. It is about. According to the present invention, since the data is sampled by selecting an appropriate sampling pattern for each matching state, it is possible to detect optical information with higher reliability than the pixel matching method, and to use the 1: 2 oversampling method or the 1: 3 oversampling method. The storage density can be much higher than that of other integer multiples such as oversampling.

Description

Optical information detection method and optical information detector {METHOD FOR DETECTING OPTICAL INFORMATION AND OPTICAL INFORMATION DETECTOR}

1A and 1B are exemplary views illustrating an example of detecting an image of a data page using a conventional 1: 1 pixel matching method.

2 is a block diagram showing the configuration of an optical information reproducing apparatus including an optical information detector according to a preferred embodiment of the present invention.

FIG. 3 is a block diagram showing the configuration of an optical information detector according to a preferred embodiment of the present invention shown in FIG.

4 is a flowchart showing the flow of the optical information detecting method according to the preferred embodiment of the present invention.

5 is an exemplary diagram illustrating a partial area of an image of a data page detected by the light detector.

6A, 6B, and 6C are exemplary views illustrating a concept of determining horizontal matching of frame marks by using vertical frame mark regions.

7A, 7B, and 7C are exemplary views illustrating a concept of determining a vertical matching state of frame marks using a horizontal frame mark area.

8A to 8D are exemplary views illustrating unit sampling patterns of data pixels according to the number of intersection points between detection pixels existing in the data pixels, respectively.

9A to 9I are exemplary diagrams illustrating sampling methods according to matching states of frame marks determined through the frame mark state determination step, respectively.

10A to 10C are exemplary views illustrating an example in which an image of a data page is detected by sampling in the above-described method.

<Description of the symbols for the main parts of the drawings>

10: light irradiation part

20: optical information storage medium

30: signal processing unit

40: decoding unit

100: optical information detector

110: light detector

120: optical information processing unit

121: frame mark area detection module

123: frame mark status determination module

125: sampling module

130: memory

The present invention relates to an optical information detecting method and an optical information detector, and more particularly, to an optical information detecting method capable of detecting highly reliable optical information and improving a storage density, and an optical information detector capable of realizing the same. It is about.

In recent years, as the demand for a next generation storage system having a large storage capacity increases, a light processing system using holography, that is, a holographic light processing system, has attracted attention.

The holographic optical information processing system records the interference pattern on the optical information storage medium by irradiating a signal light containing data and a reference light provided from an angle different from the signal light at a predetermined position of the optical information storage medium and crossing each other. In addition, when the stored information is reproduced, the reference light is irradiated to the stored interference pattern, and the original data is restored by using diffraction generated by the interference pattern.

The holographic optical information processing system can store data at the same position of the optical information storage medium by using various multiplexing techniques, and can superimpose and reproduce the superimposed stored data. Can be implemented. In this case, the multiplexing technique includes angle multiplexing, wavelength multiplexing, phase code multiplexing, and the like.

On the other hand, the holographic optical information processing system processes digital data in predetermined page units, and this page unit data is called a data page. That is, the holographic optical information processing system processes data in units of data pages. The optical information processing procedure in units of data pages is described in detail in US Patent No. 670923, Japanese Patent Application Laid-Open No. 1998-97792, and the like.

For example, the holographic optical information processing system encodes input data in units of data pages, generates an image of a two-dimensional data page by mapping each encoded binary data to each pixel, and then projects it onto a signal light. Investigate with optical information storage media. Such optical modulation may be performed through a spatial light modulator (SLM).

In this case, a reference light irradiated from an angle different from the signal light is incident on the optical information storage medium. The signal light and the reference light interfere with each other in the optical information storage medium, so that an image of the data page contained in the signal light is recorded in the optical information storage medium in the form of an interference pattern.

By the above process, the image of the data page recorded on the optical information storage medium can be reproduced by irradiating the interference pattern with reference light. The image of the reproduced data page may be detected through a light receiving array element, for example, a complementary metal-oxide semiconductor (CMOS) or a charge coupled device (CCD). At this time, the detected image of the data page is reproduced as original data through a predetermined signal processing and decoding process.

Meanwhile, various sampling techniques, such as the following, may be used to detect an image of a data page using the light receiving array element.

1: 1 Pixel Matching Method

The 1: 1 pixel matching technique is a method of 1: 1 matching a pixel of a reproduced data page image (hereinafter, abbreviated as data pixel) and a pixel of a light receiving array element (hereinafter, abbreviated as detection pixel). In this 1: 1 pixel matching technique, since one data pixel directly corresponds to one detection pixel, the storage density at the time of image detection is high.

However, when the image of the data page is actually reproduced, the position of the reproduced image formed on the light receiving array is changed due to the shrinkage or rotation of the optical information storage medium, and as a result, misalignment occurs. As a result, the data pixel and the light receiving pixel are mismatched with each other.

However, the 1: 1 pixel matching technique causes severe degradation of an image of the data page detected by the quenching array element when mismatching of 1/2 or more of the data pixel size occurs. Therefore, if the pixel mismatching is severe, accurate information cannot be obtained.

1A and 1B illustrate an example of detecting an image of a data page using the 1: 1 pixel matching method. FIG. 1A shows an image when no misalignment has occurred. The image in the case of 0.4 pixel misalignment is shown.

Referring to FIGS. 1A and 1B, it can be seen that the image of FIG. 1B having 0.4 pixels of misalignment is much larger than the image of FIG. 1A without misalignment.

2. 1: 3 Oversampling

The 1: 3 oversampling method is a method of detecting one data pixel as nine (3 × 3) detection pixels. In the case of the 1: 3 oversampling method, even if mismatching between the data pixel and the detection pixel occurs arbitrarily, the detection pixel located at the center of the nine detection pixels can detect the light of the data pixel. Therefore, even if the image of the reproduced data page exists at any position of the light receiving array element, reliable data can be obtained from the image detected by the center detection pixel.

However, this 1: 3 oversampling method requires too many detection pixels to detect one data pixel at the time of image detection, so the storage density is too low. For example, in the case of using a light receiving arrangement consisting of 1200 × 1200 detection pixels, one data page should include only 400 × 400 data. Therefore, the stability of the system can be secured, but there is a serious disadvantage of lowering the storage capacity, which is the greatest strength of the holographic memory.

3. 1: 2 oversampling

The 1: 2 oversampling method is a method of detecting one data pixel as four (2 × 2) detection pixels. Similarly to the 1: 3 oversampling method described above, one of the four detection pixels can detect light of the data pixel so that reliable data can be obtained even if arbitrary pixel mismatch occurs. However, this 1: 2 oversampling method has a disadvantage that the information storage density is only 25% compared to the pixel matching method.

As such, the conventional pixel matching method has a good storage density, but has a disadvantage of being too vulnerable to misalignment between pixels. In the conventional 1: 3 oversampling method and the 1: 2 oversampling method, the data detection reliability is high but the storage density is high. There is a problem too low. Therefore, there is a demand for an optical information detection method capable of satisfying a high storage density while ensuring the reliability of data detection.

SUMMARY OF THE INVENTION The present invention has been made to solve this problem, and provides a method for detecting optical information that can efficiently detect optical information through 1: n (n is a non-integer greater than 1) oversampling. There is this.

Another object of the present invention is to provide an optical information detector capable of realizing the optical information detecting method.

The optical information detecting method according to the present invention for achieving the first object of the present invention comprises the steps of: detecting an image of the source data page as 1: n (n is a non-integer greater than 1) overpixel; Detecting a matching identification region in the image of the detected data page; Determining a matching case using the light distribution of the detected matching identification region; And performing sampling to restore an image of the source data page according to a preset sampling pattern corresponding to the determined matching state of the frame mark.

The matching identification area may be a frame mark area corresponding to the frame mark of the source data page. In addition, the matching case may mean a matching state of the frame mark.

The frame mark area may include a vertical frame mark area and a horizontal frame mark area. In this case, the determining may determine the horizontal matching state of the frame mark using the light distribution of the vertical frame mark area. The vertical frame mark area may be an area including a column having the largest light intensity and at least one neighboring column of the column on the detected image of the data page.

The horizontal matching state may be a 'center' state where the light intensity of the column having the largest light intensity is much larger than the light intensity of the neighboring column; A 'left' state in which the light intensity of the largest column is similar to that of the left column and is significantly greater than that of the right column; And a 'right' state in which the light intensity of the largest column is similar to that of the right column and is significantly greater than that of the left column.

In addition, the determining may determine the vertical matching state of the frame mark by using the light distribution of the horizontal frame mark area. The horizontal frame mark region may be a region including a column having the largest light intensity and at least one neighboring row of the row on the detected image of the data page.

The vertical matching state may include a 'center' state where the light intensity of the row having the largest light intensity is much larger than the light intensity of the neighboring row; An 'upper' state where the light intensity of the row having the largest light intensity is similar to that of the upper row and is significantly greater than that of the lower row; And a "lower" state in which the light intensity of the row having the largest light intensity is similar to that of the lower row and is much greater than the light intensity of the upper row.

The sampling pattern may be set to correspond to the number of cases where the number of horizontal matching states of the frame marks and the number of vertical matching states of the frame marks are combined. The sampling pattern is present in the data pixel to be detected. At least one unit sampling pattern in consideration of the intersection between the detection pixels may be included.

The unit sampling pattern may include: a pattern for sampling a value obtained by averaging values of four detection pixels in contact with the intersection when the number of intersections existing in the data pixel is one; A pattern for sampling a value obtained by averaging values of two detection pixels in contact with both intersections when the number of intersections is two; And a pattern for sampling a value of one detection pixel that contacts all four intersections when the number of intersections is four. In the sampling pattern, a plurality of unit sampling patterns may be set in block units to correspond to the frame mark matching state.

The sampling step may reconstruct the data pixel by sampling at least one detection pixel corresponding to the data pixel according to a sampling pattern corresponding to the determined matching state of the frame mark.

On the other hand, in the optical information detecting method according to the present invention for achieving the first object of the present invention described above, the image of the source data page including the plurality of data pixels of the present invention is more than an integer multiple of the number of the data pixels. Detecting through the detection pixel; Determining a matching state when the image is detected using the light distribution of the image of the detected data page; And selecting a plurality of unit sampling patterns according to the determined matching state, and performing sampling to restore an image of the source data page.

In this case, the unit sampling pattern may be a sampling pattern according to the number of intersection points between detection pixels existing in the data pixel to be restored. The unit sampling pattern may include: a pattern for sampling a value obtained by averaging values of four detection pixels in contact with the intersection when the number of intersections existing in the data pixel is one; A pattern for sampling a value obtained by averaging values of two detection pixels in contact with both intersections when the number of intersections is two; And a pattern for sampling a value of one detection pixel that contacts all four intersections when the number of intersections is four.

On the other hand, the optical information detector according to the present invention for achieving the above-described second object of the present invention comprises: a light detector for detecting an image of the source data page as 1: n (non-integer greater than 1) overpixel; And detecting a frame mark area from the detected image of the data page, detecting a state of the frame mark using the light distribution of the frame mark area, and detecting the frame mark area according to a preset sampling pattern corresponding to the state of the frame mark. It may include an optical information processing unit for performing a sampling to restore the image of the source data page. The optical information detector may further include a memory in which a sampling pattern corresponding to a state of the frame mark is stored.

The optical information processing unit may include a frame mark area detection module that detects the frame mark area corresponding to the frame mark of the source data page in the image of the detected data page, and a light distribution of the detected frame mark area. A frame mark state determination module for detecting a matching state of the frame marks; And a sampling module configured to perform sampling for restoring an image of the source data page according to a preset sampling pattern corresponding to the matched state of the determined frame mark.

The frame mark area may include a vertical frame mark area and a horizontal frame mark area. The optical information processing unit may determine a horizontal matching state of the frame mark using the light distribution of the vertical frame mark area, and determine a vertical matching state of the frame mark using the light distribution of the horizontal frame mark area. . The sampling pattern may be set to correspond to the number of cases where the number of horizontal matching states of the frame marks is combined with the number of vertical matching states of the frame marks.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those skilled in the art may easily implement the present invention. In addition, in describing the preferred embodiment of the present invention illustrated in the drawings, specific technical terms are used for clarity of content. However, it is to be understood that the invention is not limited to these specific selected terms, and that each specific term includes all technical synonyms that operate in a similar manner to achieve a similar purpose.

2 is a block diagram showing the configuration of an optical information reproducing apparatus including an optical information detector according to a preferred embodiment of the present invention.

Referring to FIG. 2, the optical information reproducing apparatus may include a light irradiator 10 that generates light such as a laser beam and irradiates the optical information storage medium 20. The light irradiated by the light irradiator 10 is incident on the optical information storage medium 20 at a predetermined angle. The optical information storage medium 20 stores a plurality of data pages in the form of interference patterns. In this case, the light incident by the light irradiator 10 may be one of a reference beam and a phase conjugation reference light for reproducing the interference pattern stored in the optical information storage medium 20.

When light is incident on an interference pattern stored in the optical information storage medium 20, an image of a data page is reproduced by diffraction by the interference pattern. At this time, the image of the reproduced data page is detected by the optical information detector 100. Subsequently, the detected image 100 of the data page is converted into binary data by the signal processor 30, and then decoded by the decoder 40 to be reproduced as original data.

3 is a block diagram showing the configuration of the optical information detector 100 according to the preferred embodiment of the present invention shown in FIG.

Referring to FIG. 3, the optical information detector 100 according to the preferred embodiment of the present invention may include the optical detector 110, the memory 130, and the optical information processor 120. In this case, the optical information processing unit 120 may include a frame mark area detection module 121, a frame mark state determination module 123, and a sampling module 125.

First, the light detector 110 detects an image of a source data page stored in the optical information storage medium 20 and reproduced by reference light as 1: n (n is a non-integer greater than 1) with an over-pixel. do. In this case, the 1: n overpixel means that an image is configured by configuring an optical system so that one data pixel corresponds to n × n detection pixels.

For example, in a 1: 1.5 overpixel, since one data pixel corresponds to 1.5 × 1.5 detection pixels, 2 × 2 data pixels may be mapped to 3 × 3 detection pixels of the light detector 110. have. As another example, in a 1: 1.33 overpixel having n of 1.33, since one data pixel corresponds to 1.33 detection pixels, a 3 × 3 data pixel may be mapped to a 4 × 4 detection pixel.

The light detector 110 may be a light receiving array in which a plurality of light receiving elements are arranged. For example, it may be a complementary metal-oxide semiconductor (CMOS) or a charge coupled device (CCD).

The optical information processing unit 120 detects the frame mark area on the image of the data page detected by the optical detection unit 110, determines the matching state of the frame mark using the light distribution of the detected frame mark area, and then determines Each data pixel image is detected by performing sampling according to a sampling pattern corresponding to the matching state of the frame mark. The optical information processor 120 may include a frame mark area detection module 121, a frame mark state determination module 123, and a sampling module 125.

The frame mark area detection module 121 detects the frame mark area in the image of the data page detected by the light detector 110. In this case, the frame mark area detection module may detect the frame mark area by dividing the frame mark area into a vertical frame mark area and a horizontal frame mark area.

The frame mark state determination module determines the matching state of the frame marks using the light distribution of the frame mark regions detected by the frame mark region detection module. In this case, the frame mark state determination module may determine the vertical matching state of the horizontal frame mark by using the light distribution of the horizontal frame mark area detected by the frame mark area detection module, and by using the light distribution of the vertical frame mark area. The horizontal matching state of the vertical frame marks can be determined.

The sampling module 125 performs sampling according to a sampling pattern according to the matched state of the determined frame mark to detect each data pixel image. That is, the image of the source data page is reconstructed by sampling the detection pixels according to sampling patterns corresponding to the vertical matching state and the horizontal matching state of the frame mark to detect each data pixel.

In this case, the data sampling patterns corresponding to the vertical matching state and the horizontal matching state may be stored in the memory 130.

4 is a flowchart illustrating a method of detecting an optical information according to an exemplary embodiment of the present invention, and shows a method of detecting optical information through 1: 1.5 oversampling. In the description of FIG. 4, the device configuration for realizing the optical information detection method uses the optical information detector 100 shown in FIG. 3. Therefore, according to the following disclosure, the functions of the respective components of the optical information detector 100 described above will also be clearer.

Referring to FIG. 4, first, the light detector 110 detects an image of a source data page stored in the optical information storage medium 20 and reproduced by reference light through a 1: 1.5 overpixel (step S1). Thus, a 2x2 data pixel will correspond to a 3x3 detection pixel.

Subsequently, the frame mark area detection module 121 detects the frame mark area on the image of the data page detected by the light detector 110 (step S2).

The frame mark area is detected by reflecting an image of a frame mark existing in the image of the source data page. The frame mark may generally exist in the form of a border of a data page, but according to an implementation environment, the frame mark may exist in the form of various pixel arrangements for identifying the horizontal and vertical regions of the data page.

This frame mark has a very high intensity because on-pixels are arranged in succession for ease of identification. Therefore, the frame mark area can be easily detected by finding a portion where the optical force is much larger than other portions on the image of the data page detected by the light detector 110.

5 is an exemplary diagram illustrating a partial region of an image of a data page detected by the light detector 110.

Referring to FIG. 5, it can be seen that frame mark regions 201 and 202 having a plurality of on-pixels arranged in a border form exist outside the information region 203 of the data page image 200.

In this case, the arrangement of the on pixels arranged in the horizontal direction may be referred to as a horizontal frame mark area 202, and the arrangement of the on pixels arranged in the vertical direction may be referred to as a vertical frame mark area 201. An identification area in which a plurality of off pixels are disposed may be detected between the frame mark areas 201 and 202 and the information area 203 so as to clearly distinguish the frame mark areas 201 and 202 and the information area 203. have.

Referring to the measurement graph (a) shown on the right side of FIG. 5, it can be seen that the light intensity of the portion where the horizontal frame mark area 202 is located is much higher than the light intensity of the other portion. In addition, looking at the measurement graph (b) shown at the bottom of Figure 5, it can be seen that the light intensity of the vertical frame mark area 201 is much higher than the light intensity of the other portion.

Since the frame mark areas 201 and 202 are detected on the image 200 of the data page detected with an overpixel of 1: 1.5, when the frame marks of the actual data page are composed of one line, The frame mark areas 201 and 202 can be detected with up to three lines.

Referring to FIG. 5, it can be seen that the vertical frame mark region 201 is relatively clearly detected in one column, but the horizontal frame mark region 202 is divided into two rows. By using the light distribution of the detection frame mark areas 201 and 202, it is possible to determine the matching state of the actual frame mark.

Accordingly, the frame mark state determination module 123 determines the matching state of the frame marks by using the light distribution of the detected frame mark region (step: S3). That is, the horizontal matching and vertical matching states of the frame marks are determined using the light distribution of the detected vertical frame mark region and the horizontal frame mark region. Accordingly, the frame mark area may be referred to as a matching case identification area for determining a matching case.

First, a process of determining a horizontal matching state of frame marks using a vertical frame mark region will be described.

6A, 6B, and 6C are exemplary views illustrating a concept of determining horizontal matching of frame marks by using vertical frame mark regions, in which pixels represented by small grids represent detection pixels, and are represented by large grids. The represented pixel means a data pixel. In addition, an array line of data pixels displayed vertically in bold on FIGS. 6A to 6C means a vertical frame mark.

First, three columns, Cmax-1, Cmax, and Cmax + 1, should be considered in the vertical frame mark area for detecting the horizontal matching state. In this case, Cmax refers to the column having the highest light intensity among the detected columns, and Cmax-1 and Cmax + 1 are defined as meaning the columns before and after the Cmax. The light distribution of Cmax, Cmax-1 and Cmax + 1 can be divided into three cases as follows.

First, when the light intensity of Cmax is much larger than the light intensity of Cmax-1 and Cmax + 1, it may mean that Cmax almost matches the actual vertical frame mark. That is, it can be seen as the state shown in Figure 6a. The state of such a frame mark is defined as 'horizontal-center', that is, 'H-C'.

Second, if the difference between the light intensity of Cmax and the light intensity of Cmax-1 is not much, and the light intensity of Cmax + 1 is much smaller than these, this may mean that the vertical frame mark has moved to the left. That is, it can be seen as the state shown in Figure 6b. The state of this vertical frame mark is defined as 'horizontal-left', that is, 'H-L'.

Third, if there is not much difference between the light intensity of Cmax and the light intensity of Cmax + 1, and the light intensity of Cmax-1 is much smaller than these, this may mean that the vertical frame mark has moved to the right. That is, it can be seen as the state shown in Figure 6c. The state of this vertical frame mark is defined as 'horizontal-right', that is, 'H-R'.

As described above, three horizontal matching states, namely, horizontal matching cases, of the frame marks are defined by the light distribution of the vertical frame mark region.

Meanwhile, the vertical matching state of the frame marks can be determined using the horizontal frame mark region.

7A, 7B, and 7C are exemplary views illustrating a concept of determining a vertical matching state of frame marks using a horizontal frame mark area. An array line of data pixels displayed horizontally in bold on FIGS. 7A to 7C means a horizontal frame mark.

In the vertical frame mark region for detecting the horizontal matching state, three rows, that is, Rmax-1, Rmax, and Rmax + 1, should be considered. In this case, Rmax means the row having the highest light intensity among the detected rows, and Rmax-1 and Rmax + 1 are defined as meaning the row before and after the Rmax. The light distribution of Rmax, Rmax-1, and Rmax + 1 can be divided into three cases as follows.

First, the case where the light intensity of Rmax is significantly larger than the light intensity of Rmax-1 and Rmax + 1 may mean that Rmax almost coincides with the actual horizontal frame mark. That is, it can be seen as the state shown in Figure 7a. The state of these frame marks is defined as 'Vertical-Center' or 'V-C'.

Secondly, if the difference between the light intensity of Rmax and the light intensity of Rmax-1 is not much, and the light intensity of Rmax + 1 is much smaller than these, this may mean that the horizontal frame mark has moved upward. That is, it can be seen as the state shown in Figure 7b. The state of such a frame mark is defined as 'vertical-upper', that is, 'V-U'.

Third, when the difference between the light intensity of Rmax and the light intensity of Rmax + 1 is not much, and the light intensity of Rmax-1 is much smaller than these, this may mean that the horizontal frame mark has moved downward. That is, it can be seen as the state shown in Figure 7c. The state of the horizontal frame mark is defined as 'vertical-lower', that is, 'V-L'.

As described above, three vertical matching states, namely, vertical matching cases, of the frame marks are defined by the light distribution of the horizontal frame mark region.

Therefore, both the horizontal matching state and the vertical matching state of the frame mark may be defined through the frame mark state determination process (step: S3).

Subsequently, the sampling module 125 performs sampling according to the sampling pattern corresponding to the matched state of the determined frame mark (step S4) to restore each data pixel, thereby restoring an image of the source data page (step : S5).

In this case, the first consideration is the unit sampling pattern according to the intersection point between the detection pixels existing in the area of the data pixel to be detected.

8A to 8D are diagrams illustrating unit sampling patterns of data pixels according to the number of intersection points between detection pixels existing in the data pixels, wherein pixels represented by a small grid represent detection pixels, and pixels represented by a large grid. Represents the data pixel to be detected. In addition, a point represented by '×' means a cross point between detection pixels.

Referring to FIG. 8A, it can be seen that four intersection points between detection pixels exist in the data pixel to be detected. In this case, since the detection pixel a completely corresponds to the data pixel, the image of the detection pixel a is sampled as it is and restored to the image of the data pixel. That is, in the case of four crossing points, the values of the detection pixels that are in contact with all four intersection points, that is, the detection pixels existing in the data pixels to be detected are sampled as they are.

Referring to FIG. 8B, two intersections between detection pixels exist vertically in the data pixel to be detected. In this case, since two detection pixels b and c may affect the data pixel, an average value of the two detection pixels b and c is sampled to restore an image of the data pixel. That is, when there are two vertical crossings (2 Cross Pts), the average value of two detection pixels in contact with the two vertical crossings is sampled.

Referring to FIG. 8C, two intersection points between the detection pixels are horizontally present in the data pixel to be detected. In this case, since two detection pixels d and f may affect the data pixel, an average value of the two detection pixels d and f is sampled to restore an image of the data pixel. That is, when two horizontal crossing points are provided (2 Cross Pts), an average value of two detection pixels in contact with the two horizontal crossing points is sampled.

Referring to FIG. 8D, there is one intersection point between detection pixels in the data pixel to be detected. In this case, since all four detection pixels g, h, I, and j may affect the data pixel, the average value of the four detection pixels g, h, I, and j is sampled to restore the image of the data pixel. do. That is, when there is one cross point (1 Cross Pts), the average value of four detection pixels in contact with the cross point is sampled.

Such a unit sampling pattern may be preset in the memory 130.

On the other hand, as mentioned earlier, the matching state of the frame mark is defined as three types of 'center (C)', 'left (L)' and 'right (R)' in the horizontal direction, and 'center (C) in the vertical direction. ) ',' Upper '(U), and' lower (L) '.

Therefore, the following nine combinations may occur in the frame mark matching.

That is, [(Horizontal-Center) and (Vertical-Center)], [(Horizontal-Left) and (Vertical-Center), [(Horizontal-Right) and (Vertical-Center)], [(Horizontal-Center) and (vertical-top)], [(horizontal-center) and (vertical-bottom)], [(horizontal-left) and (vertical-top)], [(horizontal-left) and (vertical-bottom)], Combinations such as [(horizontal-right) and (vertical-top)] and [(horizontal-right) and (vertical-bottom)] are possible. Thus, sampling patterns for these nine combinations can be defined.

9A to 9I are exemplary diagrams illustrating sampling methods according to matching states of frame marks determined through the frame mark state determination step, respectively.

The small grids shown in FIGS. 9A-9B mean detection pixels, and the large grids mean data pixels. In addition, in this embodiment, the case where data is sampled by 1: 1.5 oversampling has been described, so the 2x2 data pixel corresponds to a 3x3 detection pixel. Therefore, the image may be restored in units of 2x2 data pixel blocks.

In this case, for convenience of understanding, in each of the 2 × 2 data pixel blocks, the data pixel located at the upper left side is referred to as the first data pixel, and the data pixel located at the upper right side is referred to as the second data pixel, and located at the lower left side. The data pixel will be referred to as a third data pixel, and the data pixel located on the lower right side will be referred to as a fourth data pixel.

9A to 9I, the sampling patterns according to the combination of the matching states of the frame marks will be described, respectively.

1. (Horizontal-Center) and (Vertical-Center): Case CC

Referring to FIG. 9A, it can be seen that the detected state of the frame mark is 'horizontal-center' and 'vertical-center', respectively defined in FIGS. 6A and 7A. The sampling pattern for detecting each data pixel in the data pixel block in this case is as follows.

First, since there is one intersection point between detection pixels in the first data pixel, an average value of values of four detection pixels in contact with the intersection point is detected (1 Cross Pt-> 4 Pixel Mean).

Second, since two intersections exist in the second data pixel, an average value of values of two detection pixels in contact with the two intersections is detected (2Cross Pts-> 2 Pixel Mean).

Third, since two intersections exist in the third data pixel, an average value of values of two detection pixels in contact with the two intersections is detected (2Cross Pts-> 2 Pixel Mean).

Fourth, since four intersections exist in the fourth data pixel, the value of one detection pixel in contact with the four intersections is detected as it is (4Cross Pts-> 1 Pixel Mean).

2. (Horizontal-Left) and (Vertical-Center): Case LC

Referring to FIG. 9B, it can be seen that the detected state of the frame mark is 'horizontal-left' and 'vertical-center', respectively defined in FIGS. 6B and 7A. The sampling pattern for detecting each data pixel in the data pixel block in this case is as follows.

First, since two intersections exist in the first data pixel, an average value of values of two detection pixels in contact with the two intersections is detected (2Cross Pts-> 2 Pixel Mean).

Second, since there is one intersection point between detection pixels in the second data pixel, an average value of values of four detection pixels in contact with the intersection point is detected (1 Cross Pt-> 4 Pixel Mean).

Third, since four intersections exist in the third data pixel, the value of one detection pixel in contact with the four intersections is detected as it is (4Cross Pts-> 1 Pixel Mean).

Fourth, since two intersections exist in the fourth data pixel, an average value of values of two detection pixels in contact with the two intersections is detected (2Cross Pts-> 2 Pixel Mean).

3. (Horizontal-Right) and (Vertical-Center): Case RC

Referring to FIG. 9C, when the matching state of the frame marks is 'horizontal-right' and 'vertical-center', a sampling pattern for detecting each data pixel in the data pixel block is as follows.

First, since two intersections exist in the first data pixel, an average value of values of two detection pixels in contact with the two intersections is detected (2Cross Pts-> 2 Pixel Mean).

Second, since there is one intersection point between detection pixels in the second data pixel, an average value of values of four detection pixels in contact with the intersection point is detected (1 Cross Pt-> 4 Pixel Mean).

Third, since four intersections exist in the third data pixel, the value of one detection pixel in contact with the four intersections is detected as it is (4Cross Pts-> 1 Pixel Mean).

Fourth, since two intersections exist in the fourth data pixel, an average value of values of two detection pixels in contact with the two intersections is detected (2Cross Pts-> 2 Pixel Mean).

4. (Horizontal-Center) and (Vertical-Up): Case CU

Referring to FIG. 9D, when the matching state of the frame marks is 'horizontal-center' and 'vertical-upper', a sampling pattern for detecting each data pixel in the data pixel block is as follows.

First, since two intersections exist in the first data pixel, an average value of values of two detection pixels in contact with the two intersections is detected (2Cross Pts-> 2 Pixel Mean).

Second, since four intersections exist in the second data pixel, the value of one detection pixel in contact with the four intersections is detected as it is (4Cross Pts-> 1 Pixel Mean).

Third, since there is one intersection point between the detection pixels in the third data pixel, an average value of values of four detection pixels in contact with the intersection point is detected (1 Cross Pt-> 4 Pixel Mean).

Fourth, since two intersections exist in the fourth data pixel, an average value of values of two detection pixels in contact with the two intersections is detected (2Cross Pts-> 2 Pixel Mean).

5. (Horizontal-Center) and (Vertical-Lower): Case CL

Referring to FIG. 9E, when the matching state of the frame marks is 'horizontal-center' and 'vertical-bottom', a sampling pattern for detecting each data pixel in the data pixel block is as follows.

First, since two intersections exist in the first data pixel, an average value of values of two detection pixels in contact with the two intersections is detected (2Cross Pts-> 2 Pixel Mean).

Second, since four intersections exist in the second data pixel, the value of one detection pixel in contact with the four intersections is detected as it is (4Cross Pts-> 1 Pixel Mean).

Third, since there is one intersection point between the detection pixels in the third data pixel, an average value of values of four detection pixels in contact with the intersection point is detected (1 Cross Pt-> 4 Pixel Mean).

Fourth, since two intersections exist in the fourth data pixel, an average value of values of two detection pixels in contact with the two intersections is detected (2Cross Pts-> 2 Pixel Mean).

6. (Horizontal-Left) and (Vertical-Up): Case LU

Referring to FIG. 9F, when the matching state of the frame marks is 'horizontal-left' and 'vertical-top', a sampling pattern for detecting each data pixel in the data pixel block is as follows.

First, since four intersections exist in the first data pixel, the value of one detection pixel in contact with the four intersections is detected as it is (4Cross Pts-> 1 Pixel Mean).

Second, since two intersections exist in the second data pixel, an average value of values of two detection pixels in contact with the two intersections is detected (2Cross Pts-> 2 Pixel Mean).

Third, since two intersections exist in the third data pixel, an average value of values of two detection pixels in contact with the two intersections is detected (2Cross Pts-> 2 Pixel Mean).

Fourth, since there is one intersection point between the detection pixels in the fourth data pixel, an average value of values of four detection pixels in contact with the intersection point is detected (1 Cross Pt-> 4 Pixel Mean).

7. (Horizontal-Left) and (Vertical-Lower): Case LL

Referring to FIG. 9G, when a matching state of a frame mark is 'horizontal-left' and 'vertical-bottom', a sampling pattern for detecting each data pixel in a data pixel block is as follows.

First, since four intersections exist in the first data pixel, the value of one detection pixel in contact with the four intersections is detected as it is (4Cross Pts-> 1 Pixel Mean).

Second, since two intersections exist in the second data pixel, an average value of values of two detection pixels in contact with the two intersections is detected (2Cross Pts-> 2 Pixel Mean).

Third, since two intersections exist in the third data pixel, an average value of values of two detection pixels in contact with the two intersections is detected (2Cross Pts-> 2 Pixel Mean).

Fourth, since there is one intersection point between the detection pixels in the fourth data pixel, an average value of values of four detection pixels in contact with the intersection point is detected (1 Cross Pt-> 4 Pixel Mean).

8. (Horizontal-Right) and (Vertical-Up): Case RU

Referring to FIG. 9H, when the matching state of the frame marks is 'horizontal-right' and 'vertical-top', a sampling pattern for detecting each data pixel in the data pixel block is as follows.

First, since four intersections exist in the first data pixel, the value of one detection pixel in contact with the four intersections is detected as it is (4Cross Pts-> 1 Pixel Mean).

Second, since two intersections exist in the second data pixel, an average value of values of two detection pixels in contact with the two intersections is detected (2Cross Pts-> 2 Pixel Mean).

Third, since two intersections exist in the third data pixel, an average value of values of two detection pixels in contact with the two intersections is detected (2Cross Pts-> 2 Pixel Mean).

Fourth, since there is one intersection point between the detection pixels in the fourth data pixel, an average value of values of four detection pixels in contact with the intersection point is detected (1 Cross Pt-> 4 Pixel Mean).

9. (Horizontal-Right) and (Vertical-Lower): Case RL

Referring to FIG. 9I, when a matching state of a frame mark is 'horizontal-right' and 'vertical-bottom', a sampling pattern for detecting each data pixel in a data pixel block is as follows.

First, since four intersections exist in the first data pixel, the value of one detection pixel in contact with the four intersections is detected as it is (4Cross Pts-> 1 Pixel Mean).

Second, since two intersections exist in the second data pixel, an average value of values of two detection pixels in contact with the two intersections is detected (2Cross Pts-> 2 Pixel Mean).

Third, since two intersections exist in the third data pixel, an average value of values of two detection pixels in contact with the two intersections is detected (2Cross Pts-> 2 Pixel Mean).

Fourth, since there is one intersection point between the detection pixels in the fourth data pixel, an average value of values of four detection pixels in contact with the intersection point is detected (1 Cross Pt-> 4 Pixel Mean).

Through the nine sampling patterns as described above, the sampling module 125 may detect values of each data pixel by sampling a detection pixel corresponding to each data pixel.

10A to 10C are exemplary views illustrating an example in which an image of a data page is detected by sampling in the above-described method. In this case, 10a represents an image of the source data page by the spatial light modulator, and 10b represents a sampling pattern (Case CU) shown in FIG. 9D when the matching state of the frame marks is 'horizontal-center' and 'vertical-upper'. The detected image and the binarized image sampled through are shown. In addition, FIG. 10C shows a detection image and a binarized image in which the frame marks match states of “horizontal-center” and “vertical-upper”, but are incorrectly sampled through FIG. 9G (Case LL).

10A and 10B, it can be seen that the image of the source data page shown in FIG. 10A is completely reproduced without error as shown in FIG. 10B through reconstruction by the corresponding sampling pattern.

On the other hand, referring to FIGS. 10A and 10C, it can be seen that many errors have occurred as shown in FIG. 10C due to the selection of a wrong sampling pattern in the image of the source data page shown in FIG. 10A.

Although the present invention has been described above with reference to its preferred embodiments, those skilled in the art will variously modify the present invention without departing from the spirit and scope of the invention as set forth in the claims below. And can be practiced with modification.

In particular, in the above-described embodiment, oversampling of 1: 1.5 has been described among oversampling of 1: n (n is a non-integer greater than 1). However, depending on the implementation environment, 1: 1.33 oversampling with 3x3 data pixels mapped to 4x4 detection pixels, or 1: 1.25 over with 4x4 data pixels mapped to 5x5 detection pixels. It will be appreciated that various implementations, such as sampling, are possible.

Accordingly, modifications to future embodiments of the present invention will not depart from the technology of the present invention.

As described above, according to the present invention, a matching case is determined using a frame mark area, which is a matching identification area, and an appropriate sampling pattern is selected for each matching case to sample data, thereby providing higher reliability than the pixel matching method. The branch can detect the optical information, and can significantly increase the storage density compared to other integer multiple oversampling methods such as 1: 2 oversampling or 1: 3 oversampling.

Accordingly, by simultaneously solving the weakness of the misalignment of the pixel matching and the low storage density of the integer multiple oversampling method, both of the requirements for reliable data detection and high storage density can be satisfied. have.

Claims (24)

Detecting an image of the source data page as 1: n (n is a non-integer greater than 1) overpixel; Detecting a matching identification region in the image of the detected data page; Determining a matching case using the light distribution of the detected matching identification region; And And performing sampling to restore an image of the source data page according to a predetermined sampling pattern corresponding to the determined matching case. The method of claim 1, wherein the matching identification area is a frame mark area corresponding to a frame mark of the source data page. The method of claim 2, wherein the matching case is a matching state of the frame mark. 4. The method of claim 3, wherein the frame mark area includes at least one of a vertical frame mark area and a horizontal frame mark area. 5. The method of claim 4, wherein the determining step determines the horizontal matching state of the frame marks using the light distribution of the vertical frame mark region. The optical information detection method of claim 5, wherein the vertical frame mark area is an area including a column having the largest light intensity and at least one neighboring column of the column on an image of the detected data page. Way. The method of claim 6, wherein the horizontal matching state, A 'center' state where the light intensity of the column with the largest light intensity is significantly greater than the light intensity of the neighboring column; A 'left' state in which the light intensity of the largest column is similar to that of the left column and is significantly greater than that of the right column; And And at least one of a 'right' state in which the light intensity of the largest column is similar to the light intensity of the right column and is significantly greater than the light intensity of the left column. 5. The method of claim 4, wherein the determining step determines the vertical matching state of the frame marks using the light distribution of the horizontal frame mark region. The optical information detection method of claim 8, wherein the horizontal frame mark area is a region including a row having the largest light intensity and at least one neighboring row of the row on an image of the detected data page. Way. The method of claim 9, wherein the vertical matching state, A 'center' state where the light intensity of the row with the largest light intensity is significantly greater than the light intensity of the neighboring rows; An 'upper' state where the light intensity of the row having the largest light intensity is similar to that of the upper row and is significantly greater than that of the lower row; And And at least one of a 'lower' state where the light intensity of the row having the largest light intensity is similar to that of the lower row and is significantly greater than that of the upper row. The optical information detection method according to claim 4, wherein the sampling pattern is set to correspond to the number of cases where the number of horizontal matching states of the frame marks and the number of vertical matching states of the frame marks are combined. The method of claim 1, wherein the sampling pattern comprises at least one unit sampling pattern considering an intersection point between detection pixels existing in the data pixel to be detected. The method of claim 12, wherein the unit sampling pattern, A pattern for sampling a value obtained by averaging values of four detection pixels in contact with the intersection when the number of intersections existing in the data pixel is one; A pattern for sampling a value obtained by averaging values of two detection pixels in contact with both intersections when the number of intersections is two; And  And at least one of patterns for sampling a value of one detection pixel in contact with all four intersection points when the number of intersection points is four. The optical information detection method of claim 12, wherein the sampling pattern is set in block units to correspond to the matching case. The optical information detection method of claim 12, wherein the sampling comprises reconstructing the data pixel by sampling at least one detection pixel corresponding to the data pixel according to a sampling pattern corresponding to the determined matching case. . Detecting an image of a source data page including a plurality of data pixels through a plurality of detection pixels that are non-integer than the number of data pixels; Determining a matching state when the image is detected using the light distribution of the image of the detected data page; And And selecting a plurality of unit sampling patterns according to the determined matching state, and performing sampling to restore an image of the source data page. 18. The method of claim 16, wherein the unit sampling pattern is a sampling pattern according to the number of intersection points between detection pixels existing in the data pixel to be reconstructed. The method of claim 17, wherein the unit sampling pattern, A pattern for sampling a value obtained by averaging values of four detection pixels in contact with the intersection when the number of intersections existing in the data pixel is one; A pattern for sampling a value obtained by averaging values of two detection pixels in contact with both intersections when the number of intersections is two; And  And at least one of patterns for sampling a value of one detection pixel in contact with all four intersection points when the number of intersection points is four. A light detector for detecting an image of the source data page as 1: n (non-integer greater than 1) overpixel; And Detecting a frame mark area in the image of the detected data page, detecting the state of the frame mark using the light distribution of the frame mark area, and the source data according to a preset sampling pattern corresponding to the state of the frame mark. And an optical information processing unit for performing sampling to restore an image of a page. 20. The optical information detector of claim 19, further comprising a memory in which a sampling pattern corresponding to a state of the frame mark is stored. The method of claim 19, wherein the optical information processing unit, A frame mark area detection module detecting the frame mark area corresponding to the frame mark of the source data page in the detected image of the data page; A frame mark state determination module for detecting a matching state of the frame marks by using the light distribution of the detected frame mark regions; And And a sampling module configured to perform sampling for restoring an image of the source data page according to a predetermined sampling pattern corresponding to the matched state of the determined frame mark. 20. The optical information detector of claim 19, wherein the frame mark area comprises a vertical frame mark area and a horizontal frame mark area. 23. The apparatus of claim 22, wherein the optical information processor determines a horizontal matching state of the frame marks using the light distribution of the vertical frame mark area, and vertically matches the frame marks using the light distribution of the horizontal frame mark area. And an optical information detector for determining a state.  The optical information detector according to claim 23, wherein the sampling pattern is set to correspond to the number of cases where the number of horizontal matching states of the frame marks is combined with the number of vertical matching states of the frame marks.

Images