KR101081001B1 - Method for generating and reconstructing computer-generated holograms with both amplitude and phase variables - Google Patents

Abstract

PURPOSE: A method for generating and reconstructing computer-generated holograms with both amplitude and phase variables are provided to minimize the generation of a damaged pixel due to a DC component. CONSTITUTION: An input digital data is transformed into a 2D image(S110). The 2D image is randomized through a random mask(S120). A Fourier transform is performed for the 2D image data(S130). A quantization is performed about each amplitude and phase information of the 2D random image(S140,S150). The quantized amplitude value as phase value is mapped into a predetermined gray-level value(S160).

Description

How to create and restore amplitude phase computer holograms {METHOD FOR GENERATING AND RECONSTRUCTING COMPUTER-GENERATED HOLOGRAMS WITH BOTH AMPLITUDE AND PHASE VARIABLES}

The present invention relates to a computer generated hologram pattern, and more particularly, to a method for generating and restoring an amplitude phase type gray level computer hologram capable of minimizing generation and restoration errors.

In general, a hologram is a recording of an interference pattern of an object on a film made of a photosensitive material by interfering reflected light (object light) from an object and direct light (reference light) from an object using laser light having good coherence. When the reference light hits the film on which the hologram is recorded, the object light is reproduced, so that the object looks as if it is actually present. Such holograms are called interferometer based holograms. Applications of optical interferometers include hologram labels and three-dimensional hologram imaging.

If the definition of the interference pattern on the analog holographic film using optical equipment is analog holography, instead of the analog film, the electronic device (image acquisition device such as a CCD camera) is used to record, process and transmit the interference pattern. The method of playing back images in space is called digital holography. The method developed for the process of acquiring the interference pattern, that is, the object having the ideal characteristics that are impossible in reality by calculating the interference term generated by the interference between the object light and the reference light, or the characteristic experiment of the real object in preparation for the ideal This is a computer hologram (CGH; COMPUTTER-GENERATED HOLOGRAMS).

Hologram ID technology is a technology that records and reproduces information on hologram-based optical tags.It is used for ID tags such as anti-counterfeiting, damage recovery, and storage of large amounts of information, compared to conventional 2D barcode technology or RFID (Radio Frequency Identification) technology. It has very good characteristics below. After converting detailed information such as personal fingerprint or product origin or distribution information into hologram pattern by computer-generated hologram (CGH) technology, the hologram pattern is recorded on paper by inkjet or photo printer, Recording is performed on the recording medium by a laser-based microstructure formation technique. By attaching this CGH information to a personal ID card, passport or product in the form of a tag, it is possible to create a hologram tag that can not be restored without a specific personal ID code while protecting important information from electronic devices or others using other wireless transmission methods. have. Also, the hologram tag may be restored to original digital information after acquiring computer hologram information through an image sensor such as an image sensor.

In order to implement a hologram ID tag system utilizing this advantage, a technique for effectively generating and recording a hologram pattern and reproducing information from the recorded hologram pattern is required.

Holograms based on optical interferometers are complicated and expensive to produce and record the interference images of objects using expensive and accurate optics.

Computer holograms, on the other hand, can produce virtually no-interference images of virtual objects, resulting in clear, no-noise, clear holographic images, and easy re-editing of these holographic images and the use of inkjet or photo printers. You can also write on paper easily.

Therefore, the computer hologram method for generating and recording holograms is much easier and cheaper than the optical interferometer method, and thus high usability is expected. There has been a need for a method that can minimize errors when creating and restoring such computer holograms.

The present invention has been made in view of the above, and an object thereof is to provide a method for generating and restoring an amplitude phase type computer hologram capable of minimizing restoration error as well as an encryption function.

One aspect of the present invention for achieving the above object, in the method of generating an amplitude phase type computer hologram, converting the input digital data into a two-dimensional image, randomizing the two-dimensional image through a random mask, Fourier transforming the randomized two-dimensional image data, quantizing each of the amplitude and phase information of each pixel of the Fourier transformed two-dimensional random image, and quantizing the quantized amplitude and phase values of each pixel. Mapping to level values to generate computer holograms.

In this case, the randomizing step is preferably performed by performing an XOR (Exclusive OR) operation on each pixel value of the two-dimensional image with each corresponding pixel value of a binary random mask automatically generated by a computer. In doing so, the two-dimensional image is converted into uniform binary data in which half is mixed with 0 and 1 through randomization.

Furthermore, it is preferable to further include setting the amplitude value of the pixel where the DC tones are superimposed and localized to zero before the quantization after the Fourier transform.

According to another aspect of the present invention for achieving the above object, a method of reconstructing an amplitude phase computer hologram randomized through a random mask, the method comprising: recognizing a gray-level computer hologram pattern; Extracting quantization values of amplitude and phase information for each pixel, reducing the amplitude and phase quantization values of each extracted pixel to complex values, and performing an inverse Fourier transform on the complex values of each reduced pixel And comparing the real part value from the complex value obtained through the inverse Fourier transform with a threshold to determine one of the binary real values for each pixel, and removing the random mask from the two-dimensional image of each pixel of binary data. Thereby restoring the originally input two-dimensional image.

In this case, the recognizing of the computer hologram pattern may be performed by removing the noise included in the photographed image by photographing the hologram and performing correction on the distorted image.

In addition, the removing of the random mask may be performed by performing an XOR (Exclusive OR) operation on each pixel value of the two-dimensional image made of binary data with each pixel value of the random mask that was used during randomization in the hologram generation process. desirable.

As described above, according to the present invention, data generation by encryption is possible through randomization of digital input information during hologram generation and restoration, and minimization of damaged pixels due to DC components can be achieved. Can be minimized, and lost data is recovered with a low error rate even if a part of the hologram tag is damaged. In addition, according to the method of generating and restoring the amplitude phase type computer hologram according to the present invention, the generation and recording of hologram information is simplified by eliminating the interferometer, and thus the tag recording apparatus can be miniaturized, and it is easy to handle and improves the usability. Can be. Furthermore, the method of generating and restoring the gray-level holograms of the present invention has the advantages of high information security using computer hologram technology and high reliability to restore the original input data at a relatively low error rate from a damaged CGH pattern. It will be the key technology that leads the ID tag market in the future.

1 is a view showing, for example, the entire process of generating and restoring an amplitude phase computer hologram according to a preferred embodiment of the present invention;
2 is a flowchart illustrating a method of generating an amplitude phase computer hologram according to a preferred embodiment of the present invention;
3 is a flowchart for explaining a method of restoring an amplitude phase computer hologram according to a preferred embodiment of the present invention;
4 is an explanatory diagram of binary value determination for allocating each pixel value shown in step S260 of FIG. 3 to 0 or 1;
5 is a comparative view for explaining the effect of the method of generating and restoring an amplitude phase type computer hologram according to the present invention;
6 is an exemplary view showing a restoration result for a damaged gray-level CGH pattern when a method of generating and restoring an amplitude phase computer hologram according to the present invention;
7 is a diagram comparing amplitude distribution of each pixel according to whether randomization is performed after Fourier transform is performed.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. However, the following embodiments are provided to those skilled in the art to fully understand the present invention, and may be modified in various forms, and the scope of the present invention is limited to the embodiments described below. It doesn't happen.

1 is a view showing an example of the entire process of generating and restoring an amplitude phase computer hologram according to a preferred embodiment of the present invention. 2 is a flowchart illustrating a method of generating an amplitude phase computer hologram according to a preferred embodiment of the present invention, and FIG. 3 is a flowchart illustrating a method of restoring an amplitude phase computer hologram according to a preferred embodiment of the present invention.

In general, the (m, n) th pixel of the hologram image is expressed as a function of amplitude and phase in the two-dimensional hologram plane as shown in the following equation.

Where ξ and η are the position coordinates in the hologram plane, respectively, and A _mn and φ _mn are the amplitude and phase values of the (m, n) th pixel of the hologram image, respectively. An amplitude phase hologram is a hologram with two amplitude values and two phase values. For processing digital information in a computer, these two variables, amplitude and phase values, are quantized to M level and N level, respectively. , M and N are any natural numbers).

1 is divided into a CGH pattern generating process and a restoring process. The restoring process of the CGH pattern undergoes the opposite process of the generating process. First, in the CGH pattern generation process, the input 2D image data 10 is randomly mixed with the binary random mask 20 automatically generated inside the computer and converted into encrypted data 30 through an XOR operation. However, in order to extract the original data information in the restoration process, this means that the encryption key, that is, the binary random mask 20 used in the generation process, is necessary. The encrypted binary data 30 is given a gray-level quantization value (amplitude quantization level M, phase quantization level N) to amplitude and phase, respectively, and finally a C × pattern of M × N gray-level (gray level) ( 40) is generated. In the restoration process of the gray-level CGH pattern 40, the quantization values of amplitude and phase are extracted from the CGH data 40, and then converted into binary data 50 values of 0 and 1, which are original data formats. Finally, the data is returned to the original input data 70 through an XOR operation with the encryption key 60.

A method of generating an amplitude phase computer hologram according to a preferred embodiment of the present invention will be described in more detail with reference to FIG. 2.

First, the input digital data is converted into a two-dimensional image (S110). The two-dimensional image refers to an image composed of binary data as shown by the input data 10 of FIG. 1.

Subsequently, the converted 2D image is randomized through a random mask (S120). As shown in FIG. 1, the randomization method is to perform an XOR (Exclusive OR) operation with a binary random mask 20 automatically generated inside a computer. That is, an XOR operation is performed on each pixel value of the two-dimensional image 10 and each pixel value corresponding to the binary random mask 20.

This randomly scrambles the 2D image and converts it into encrypted data. This encryption key is necessary for the restoration process to extract the original data information. Therefore, the input two-dimensional data is transformed into uniform binary data in which 0 and 1 are mixed in half through randomization, resulting in a change in spatial frequency spectral distribution. More specifically, it reduces the unnecessary DC component in the spectral distribution. Since the number of damaged pixels is minimized. As a result of the reduced number of damaged pixels, a noticeable reduction in error rate can be confirmed. This will be described later in detail with reference to FIG. 7.

Subsequently, the randomized two-dimensional image data is Fourier transformed (S130).

Subsequently, each pixel of the two-dimensional random image to which the Fourier transform is applied is quantized for each of amplitude and phase information (S140 and S150). That is, the amplitude quantization value (level M) is determined for each pixel of the two-dimensional random image, and the phase quantization value (level N) is determined at the same time.

In the quantization of the amplitude and phase information of each Fourier transformed pixel, a linear quantization method and a log scale quantization method are generally available. In order to reduce the number of quantization levels, a logarithmic scale quantization method may be used. By the way, the range of phase quantization value is limited from 0 to 2π, so there is no difficulty in quantization. However, in the case of amplitude quantization, the range of quantization value is determined according to the input data, and since the DC components are overlapped, There is a deviation, making it difficult to quantize.

However, if the randomization process is performed as in step S120, the input data can be converted into arbitrary binary data in which 0 and 1 are mixed in half, so that a random image irrelevant to the input data can be obtained, and Fourier transform is performed on the data. This concentrates the DC component into an extremely limited number of pixels (e.g. one pixel for 100x100 pixel data). The reason is that the signal spreads widely due to randomization in the spatial signal region before the Fourier transform, which is localized by overlapping DC tones in the spatial frequency region after the Fourier transform. Thus, by removing this limited corrupted pixel before amplitude quantization, an almost constant range of amplitude quantization values can be obtained.

Specifically, the DC component included in the hologram interference corresponds to the hologram image component recorded directly without interfering with the object light and the reference light in the optical interferometer, and is a holographic interference pattern because it appears as a bright and large spot in the center of the image. It is an unnecessary part damaging it. In the computer hologram image, a portion having a large amplitude value in the form of a tone appears in the center due to the DC component. Since the pixel including the DC component is damaged hologram information, the amplitude value of the pixel is assigned to zero.

For example, if the original image contains only a few tens * s or more of pixels, the randomization and Fourier transform will localize the DC to exactly one center pixel. Such damaged pixels can be easily determined even if the number of pixels in the original image is different because the DC value is approximately tens to hundreds of times larger than the average value of the amplitudes of the entire pixels. If it is statistically several times (3 to 5 times) or more than the average value of the amplitude of all pixels, it is preferable to define the error pixel and to remove the pixel.

Therefore, to quantize the amplitude, it is inevitably required to remove the pixels damaged by the DC component, and the randomization process of the input data is important to minimize the number of pixels damaged.

Finally, the quantized amplitude and phase value of each pixel are mapped to predetermined gray-level values, thereby generating a CGH pattern 40 of M × N gray-level (S160). That is, the final hologram image is generated by allocating amplitude information (M level) and phase information (N level) of each quantized pixel to all pixels with M × N gray-level values according to a predetermined rule.

Hereinafter, a method of restoring the amplitude phase type computer hologram will be described in more detail with reference to FIG. 3.

The gray-level CGH pattern reconstruction process is performed in the reverse order of the generation process. However, the remarkable process of the reconstruction process is 0 after the inverse Fourier transform (S250), and additionally by the threshold value for each pixel. Binary value determination process (S260) of 1 is required. This is due to the quantization error generated by quantizing the amplitude and phase, and in the present invention, a hologram in which randomization of original binary data is performed by a random mask as a method for minimizing the quantization error.

First, a CGH pattern of M × N gray-level is recognized (S210). Recognition of the CGH pattern may be performed by importing an image acquired by using a digital camera or a scanner to a computer to remove noise included in the image, and correcting the distorted image to generate a corrected image. .

Subsequently, the quantization values (M level and N level) of the amplitude and phase information of each pixel are extracted through recognition of the information array for the gray-level image (S230). Subsequently, using the extracted amplitude and phase quantization values of each pixel, the pixel is reduced to a complex value for each pixel (S240). In operation S250, an inverse Fourier transform is performed on the reduced complex value.

Subsequently, the value of the real part in the complex value obtained through the inverse Fourier transform is compared with the threshold and determined as one of the binary real values (S260). For example, by taking a real part from a complex value, a negative real number of 0 and a positive real number of 1 can be determined based on the threshold value 0. This will be described later in detail with reference to FIG. 4.

Subsequently, the digital data information originally input is extracted by removing the random mask (S270). The removal of the random mask is performed by XOR operation of the encryption key on an image (50 of FIG. 1) made of binary data values. The encryption key is a binary random mask 20 that was used during randomization during hologram generation. If the random mask is removed, the original two-dimensional image is restored (S280).

FIG. 4 is an explanatory diagram of binary value determination for allocating each pixel value shown in step S260 of FIG. 3 to 0 or 1. FIG. Reference numeral 330 denotes an image histogram of a CGH pattern that is not randomized, and reference numeral 340 denotes an image histogram of a randomized CGH pattern. In the binary value determination histograms 330 and 340, the X-axis is A _mn cosφ _mn , which is a real value of the complex values of each pixel, and the Y-axis is the number of pixels.

When the input data 310 is directly converted into a gray-level CGH pattern without a randomization process, a significant error in reconstructing the CGH pattern may be caused by a threshold value as shown in the CGH image histogram 330 without randomization. This is due to the inaccuracy of the values assigned to each pixel in the value determination process.

On the other hand, when the input data 310 is converted into a gray-level CGH pattern through a randomization process as in the present invention, as in the randomized CGH image histogram 340, the threshold value is exactly equal to the pixel value 0. Since it is located at the center of 1, the error of binary value discrimination is reduced to a negligible level. Therefore, the CGH pattern recovery error rate is significantly different depending on whether the randomization process is present. Using the X-axis as A _mn cosφ _mn and the Y-axis as the number of pixels as variables of the CGH image histogram, the optimal recovery error rate is when the threshold is always zero.

FIG. 5 is a comparison diagram for explaining the effect of the method of generating and restoring an amplitude phase computer hologram according to the present invention, with and without a randomization process of gray-level CGH patterns for the same input data 410. The two restoration results are shown. Looking at the reconstructed images 420 and 430, the CGH pattern reconstructed image 420 having a randomization process using the same input data 410 has been completely reconstructed with almost no error, but the CGH has not been accompanied by a randomization process. The reconstructed image 430 of the pattern has a lot of errors, so it can be confirmed that the reconstructed images are significantly different from each other.

Figure 6 is an exemplary view showing the restoration results for the damaged gray-level CGH pattern when using the method of generating and restoring the amplitude phase computer hologram according to the present invention. For the same original data 510, the restoration error rate of the restoration data 530 for the intact CGH pattern 520 is 0.00%, 25% damaged CGH pattern 540, 50% damaged CGH pattern 560 and For the 75% damaged CGH pattern 580, the restoration error rates of the restoration data were 1.48% (550), 2.50% (570), and 19.48% (590), respectively. In other words, in the case of the CGH pattern with randomization process, even if the original data is damaged to some extent, it is confirmed that the lost image is recovered with a low error rate.

7 is a diagram comparing amplitude distribution of each pixel according to whether randomization is performed after Fourier transform is performed. When the Fourier transform is performed immediately without the randomization of the original input data 610, it can be seen from the amplitude distribution graph 620 that a large number of damaged pixels are generated because the DC component is widely distributed in the center of the image. Thus, there is a disadvantage that it is very difficult to remove the damaged pixel to quantize the amplitude value.

On the other hand, when the Fourier transform is performed on binary data 630 of 50:50 after randomization is performed on the original input data as in the present invention, DC components are concentrated on extremely limited pixels in the amplitude distribution graph 640. You can see that. Therefore, only the corresponding damaged pixels can be easily removed, which makes it easy to perform amplitude quantization since the amplitude value distribution is always in a constant range regardless of the input data.

Claims (7)

In the generation method of amplitude phase computer hologram,
Converting input digital data into a two-dimensional image;
Randomizing the two-dimensional image through a random mask;
Fourier transforming the randomized two-dimensional image data;
Quantizing each of amplitude and phase information of each pixel of a Fourier transformed two-dimensional random image;
Generating a computer hologram by mapping the quantized amplitude and phase values of each pixel to predetermined gray-level values. The method of claim 1, wherein the randomizing step,
And performing an XOR (Exclusive OR) operation on each pixel value of the two-dimensional image with each corresponding pixel value of a binary random mask automatically generated by a computer. The method of claim 2, wherein the two-dimensional image is converted into uniform binary data in which 0 and 1 are mixed in half through randomization. 2. The method of claim 1, further comprising setting the amplitude value of the superimposed localized pixel to zero prior to quantization after Fourier transform. In the method of restoring amplitude phase computer hologram randomized through a random mask,
Recognizing grey-level computer hologram patterns;
Extracting quantization values of amplitude and phase information for each pixel of the recognized gray-level image;
Reducing the amplitude and phase quantization values of each extracted pixel to complex values;
Performing inverse Fourier transform on the complex value of each reduced pixel;
Comparing a real part value from a complex value obtained through an inverse Fourier transform with a threshold value and determining it as one of binary real values for each pixel;
And restoring the originally input two-dimensional image by removing the random mask from the two-dimensional image of each pixel of binary data. The method of claim 5, wherein the recognizing of the computer hologram pattern is performed by taking a hologram to remove noise included in the photographed image and correcting the distorted image. The method of claim 5, wherein the removing of the random mask comprises: performing an XOR (Exclusive OR) with each pixel value of the random mask, which was used to randomize each pixel value of the two-dimensional image made of binary data during the generation of the hologram. A restoration method, characterized in that performed by the operation.

Images