JP2006352372A - Image sensor unit, and hologram reproducing apparatus - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To solve problems of a conventional image processing system that wasteful circuit scale increase and difficulty in high speed processing or the like are caused due to a huge amount of data to be processed between an image sensor 101 and a binarizing circuit 105 shown in Figure 9. <P>SOLUTION: An image sensor unit for using a plurality of light receiving elements to sample an optical image, and let a frequency of a pixel clock, at each period of which one light receiving element samples a unit pixel of an optical image be a first frequency, then one or over of a prescribed number of the light receiving elements sample each unit pixel of the optical image by using a pixel clock with a second frequency that is a prescribed multiple of the first frequency. <P>COPYRIGHT: (C)2007,JPO&INPIT

Description

  The present invention relates to an image sensor unit and a hologram reproducing apparatus having the image sensor unit.

  As a hologram medium on which digital data is recorded as a hologram, for example, there is one provided by sealing a photosensitive resin (for example, photopolymer) with a glass substrate. When digital data is recorded on the hologram medium as a hologram, first, a coherent laser beam from the laser device is separated into two laser beams by a PBS (Polarization Beam Splitter). An SLM (Spatial Light Modulator) in which one laser beam (hereinafter referred to as “reference beam”) and digital data are set as a two-dimensional array binary image pattern (1: bright point, 0: dark point). A laser beam (hereinafter referred to as “data beam”) in which the information of the binary image pattern is reflected by irradiating the other laser beam to the spatial light modulator) on the hologram medium. To enter at a predetermined angle. As a result, digital data to be recorded is recorded on the hologram medium.

  The photosensitive resin that constitutes the hologram medium has a finite number of monomers, and the internal molecular structure changes chemically based on the energy corresponding to the product of the laser beam power and irradiation time. To do. At this time, the monomer not only simply changes chemically, but also moves from a low energy density region to a high region at the same time, thereby forming a hologram as an interference fringe. That is, a hologram corresponding to digital data is recorded on the hologram medium. This region having a high energy density is called a polymer and exists as a molecule inside the medium. In addition, the amount of monomer transfer is closely related to the energy density. FIG. 6 schematically shows how the monomer changes into a polymer in the hologram medium.

  When a large amount of digital data is recorded on the hologram, so-called hologram multiplex recording can be performed in which a large number of holograms are formed by changing the incident angle of the reference beam to the medium. In the case of holographic multiplex recording, one multiplex number of holograms is generally called a page, and a multiplex number of holograms composed of a predetermined number of pages is called a book. FIG. 7 is a diagram schematically showing an example of a book and a page in the case of hologram multiplex recording.

  When reproducing digital data from a hologram medium, a reference beam is incident on an interference fringe indicating the digital data at the same incident angle as when the interference fringe is formed, and is diffracted by the interference fringe (hereinafter referred to as a reference beam). , A reproduction beam) is received by an image sensor or the like. The reproduction beam received by the image sensor or the like is an optical image corresponding to the binary image pattern at the time of recording described above. The original digital data can be reproduced by decoding the data of each unit pixel in the optical image with a decoder or the like.

  In addition, as a hologram recording / reproducing apparatus for recording / reproducing holograms on / from a hologram medium as described above, for example, it is disclosed in FIG.

  By the way, in the case of hologram reproduction, each unit pixel in an image pattern (hereinafter referred to as a capture image pattern) read out by an image sensor has a bright point (1) and a dark point (0) as shown in FIG. ) Is not a two-tone distribution completely separated. Actually, each unit pixel in the captured image pattern is not limited to a bright point and a dark point, but has a multi-gradation including an intermediate density between the bright point and the dark point. Thus, the light spot (1) and the dark spot (0) are overlapped.

  For this reason, when the light spot (1) and the dark spot (0) are separated using a constant slice level, there is a risk that the light point (1) and the dark spot (0) are erroneously determined in the overlapping range. is there. Therefore, after taking a multi-tone optical image as it is and performing digital image processing on each unit pixel of the multi-tone optical image using an image process method, a bright point (1) and a dark point (0) Need to be separated, that is, binarized. FIG. 9 shows the configuration of a conventional image processing system in this case. A conventional configuration similar to that shown in FIG. 9 is disclosed, for example, in FIG. 5 of Patent Document 2 shown below.

  The analog image signal A read out from the image sensor 101 for every line, for example, is supplied to the ADC 102 while having multi-gradation information. The ADC 102 converts the analog image signal A into a multi-bit digital signal corresponding to the multi-gradation and supplies the analog image signal A to the image processor 103 in order to leave the multi-tone information of the analog image signal A. Each data of the converted digital signal is temporarily stored in the buffer memory 104.

The image processor 103 performs predetermined digital image processing such as noise removal and pixel matching on the digital signal having multi-gradation information stored in the buffer memory 104, and then to the binarization circuit 105. Supply. The binarization circuit 105 binarizes a digital signal after predetermined image processing into a bright point (1) and a dark point (0) based on the slice level supplied from the register 106 or the like. Converted to signal D and output.
JP 2004-177958 A JP-A-6-125461

In the conventional image processing system that handles the multi-gradation data read by the image sensor as it is, for example, in the example shown in FIG. 9, the resolution of the image sensor 101 is “1280 bits × 1280 bits”, and the quantization number of the ADC 102 Is “8 bits”, the number of bits handled by the image processor 103 is “13, 107, 200 (= 1280 × 1280 × 8) bits”, which is an enormous amount of data.
For this reason, the image sensor 101 needs to have a large-capacity buffer memory 104 for its own processing. Furthermore, when adopting a pipeline processing or interleave processing mechanism in order to improve the throughput, a large-capacity buffer memory 104 is further required.

However, of the quantization number “8 bits” of the ADC 102, the data that is actually required is only “1 bit” data of the bright point (1) and the dark point (0), and the remaining “7”. "Bit" is redundant data. As described above, a large capacity is required as the buffer memory 104. However, since the number of bits finally required is "1 bit", the use efficiency of the buffer memory 104 is very low and wasteful. It was.
Thus, in the conventional image processing system that handles the multi-gradation data read by the image sensor as it is, for example, the amount of data to be processed between the image sensor 101 and the binarization circuit 105 shown in FIG. As a result, there have been problems such as a wasteful increase in circuit scale and difficulty in speeding up.

  The main present invention for solving the above-described problem is an image sensor unit that performs sampling of an optical image with a plurality of light receiving elements. In one light receiving element, sampling of unit pixels of the optical image is performed every cycle. When the frequency of the pixel clock for performing the above operation is the first frequency, in each of the predetermined number of light receiving elements exceeding one, each unit pixel of the optical image is set to the predetermined number times the first frequency. Sampling is performed by the pixel clock having the second frequency.

  According to the present invention, it is possible to provide an image sensor unit capable of appropriately processing an optical image and a hologram reproducing apparatus having the image sensor unit.

<Overall configuration of hologram reproducing apparatus>
Based on FIG. 1, a configuration of a hologram reproducing apparatus according to an embodiment of the present invention will be described.
The laser device 10 emits a coherent laser beam excellent in temporal coherence and spatial coherence. For example, helium neon laser, argon neon laser, helium cadmium laser, semiconductor laser, dye laser, ruby laser and the like are employed.
The CPU 1 performs overall control of the entire hologram reproducing apparatus. For example, the CPU 1 transmits various instruction signals for starting the galvo mirror control unit 17, the image sensor control unit 28, the decoder 30, and the like when starting hologram reproduction.

The galvo mirror 16 reflects the laser beam emitted from the laser device 10. The galvo mirror control unit 17 adjusts the angle at which a laser beam (hereinafter referred to as a reference beam) reflected by the galvo mirror 16 is incident on the hologram medium 22 via the scanner lens 20 based on an instruction signal from the CPU 1. Therefore, the inclination angle of the galvo mirror 16 is adjusted. The tilt angle of the galvo mirror 16 at the time of hologram reproduction is adjusted so that the reference beam is incident on the hologram at the same incident angle as the reference beam at the time of hologram recording.
The scanner lens 20 refracts the reference beam so as to reliably irradiate the hologram medium 22 with the reference beam from the galvo mirror 16.

  The hologram medium 22 is made of a photosensitive resin (for example, photopolymer, silver salt emulsion, dichromated gelatin, photoresist, etc.) that can store digital data as a hologram, and the photosensitive resin is sealed with a glass substrate. Configured. Similarly to the coherent data beam reflecting the information of the binary image pattern (1: bright point, 0: dark point) of the two-dimensional array set by the SLM or the like on such a hologram medium 22. By making a coherent reference beam enter / interfere at a predetermined incident angle, a hologram corresponding to the binary image pattern is recorded in the hologram medium 22.

  The Fourier transform lens 26 receives a beam diffracted by a hologram recorded on the hologram medium 22 (hereinafter referred to as a reproduction beam) when a reference beam is incident on the hologram medium 22 during hologram reproduction. The incident angle of the reference beam at the time of reproducing the hologram is required to be the same as the incident angle of the reference beam at the time of recording the hologram to be reproduced. The Fourier transform lens 26 emits the reproduction beam subjected to the inverse Fourier transform to the image sensor unit 27.

  The image sensor unit 27 corresponds to an “image sensor unit” according to the present invention. The image sensor unit 27 receives the reproduction beam subjected to inverse Fourier transform in the Fourier transform lens 26 as an analog multi-value image pattern including multi-tone light / dark (light / dark) level information in an image sensor array 271 described later. To do. The image sensor 27 converts the light / dark (light / dark) level of the received multi-value image pattern into the strength level of the electric signal based on a capture command at an appropriate timing from the CPU 1 via the image sensor control unit 28. Read. Hereinafter, the read analog amount of the image pattern is referred to as a “capture image pattern”. Further, the image sensor unit 27 converts the captured image pattern into a binary image pattern binarized into a bright point (1) and a dark point (0). This binary image pattern is a reproduction of the binary image pattern at the time of hologram recording.

  The decoder 30 is supplied with a digital image signal D corresponding to the final binary image pattern from the image sensor unit 27. The decoder 30 performs a decoding process such as error correction on the digital image signal D. A reproduction signal R which is the digital image signal D after the decoding process is output to an external host computer or the like.

<Configuration and operation of image sensor unit>
The configuration and operation of the image sensor unit 27 according to the present invention will be described with reference to FIG. 2, with reference to FIGS. 3, 4, and 5 as appropriate.

The image sensor array 271 is, for example, a CCD sensor or a CMOS sensor, and is configured by two-dimensionally arranging a plurality of light receiving elements. FIG. 3 shows a general configuration of the image sensor array 271 in the case of a CCD sensor.
In FIG. 3, a plurality of light receiving elements 2711 corresponding to a predetermined resolution of the image sensor array 271 are two-dimensionally arranged, and light incident on each light receiving element 2711 is converted into signal charges. Then, the signal charge of each light receiving element 2711 is transferred (sampled) in the order of the vertical register 2712 and the horizontal register 2713 at a timing corresponding to the frequency of the pixel clock PC. Further, after the signal charge transferred in the charge-voltage conversion circuit 2714 connected to the subsequent stage of the output of the horizontal register 2713 is converted into a voltage signal, the voltage signal (hereinafter referred to as an analog image) amplified through the amplifier 2715. (Referred to as signal A).

  By the way, in the conventional case, one light receiving element 2711 samples each unit pixel of the optical image in one light receiving element 2711 for each period of the pixel clock PC. That is, in the conventional case, one period of the pixel clock PC is associated with a unit pixel of the optical image. Here, the frequency of the pixel clock PC in this case is referred to as a “first frequency”.

  On the other hand, in the case of this embodiment, in order to perform high-precision analog filter processing and rapid binarization on the analog image signal A read by the image sensor array 271, each unit pixel of the optical image is In a predetermined number of light receiving elements 2711 exceeding one, sampling is performed with a pixel clock PC having a frequency (hereinafter referred to as “second frequency”) that is a predetermined number times the first frequency. Note that the pixel clock PC having the second frequency is generated in the timing generation circuit 276 and supplied to the image sensor array 271. That is, in the present embodiment, a plurality of periods of the pixel clock PC corresponding to the second frequency are associated with the unit pixel of the optical image.

  In the example shown in FIG. 3, since the unit pixels A to D of the optical image are sampled by a total of four light receiving elements 2711 of “2 × 2”, the second frequency is “twice” the first frequency. Will be set to. The second frequency need not be limited to “twice” of the first frequency. For example, “7/5 times” that can obtain a resolution that is approximately twice the predetermined resolution of the image sensor array 271. Can be adopted.

  Thus, by sampling each unit pixel of the optical image with the pixel clock PC of the second frequency, for example, the influence of the boundary data of each unit pixel can be reduced, and the influence of the quantization distortion can be suppressed, As a result, it is possible to appropriately separate the bright point (1) and the dark point (0), that is, binarization of the analog image signal A.

  A BPF (Band Pass Filter) 272 is an analog filter circuit that is provided in a preceding stage of the binarization circuit 273 to which the analog image signal A is supplied and filters a predetermined frequency band of the analog image signal A. More specifically, the BPF 272 functions as a noise filter that filters noise components other than the second frequency from the analog image signal A. Further, the BPF 272 also functions as an anti-aliasing filter that cuts off the high-frequency component when the second frequency exceeds the so-called Nyquist frequency. Thus, by providing an analog filter circuit such as BPF272, the analog image signal A can be binarized more appropriately.

The binarization circuit 273 binarizes the analog image signal A read out by sampling each unit pixel of the optical image with the pixel clock PC of the second frequency at the slice level S.
An LPF (Low Pass Filter) 274 corresponds to an integration circuit that integrates the binarized output C of the binarization circuit 273 and supplies the integration result as a slice level S to the binarization circuit 273.

  That is, the level of the analog image signal A generated as a result of optical image sampling performed by a plurality of adjacent light receiving elements 2711 can be regarded as actually showing a gradual AC change with a large time constant. Therefore, in order to employ the intermediate level of the analog image signal A as the slice level S, the integration result of the binarized output C corresponding to the intermediate level is employed as the slice level S of the binarization circuit 273. As a result, it is possible to follow an AC change in the analog image signal A compared to the case where the slice level S is always constant or the case where a plurality of slice levels are switched in stages, and appropriate binarization is achieved. It is more likely to be done. Further, by adopting the integration result of the binarized output C as the slice level S, there is a high possibility that the bright point (1) and the dark point (0) are separated at the appropriate boundary.

  The resampling circuit 275 has a frequency dividing circuit 2751. The frequency dividing circuit 2751 converts the pixel clock PC having the second frequency supplied from the timing generation circuit 276 into the pixel clock PC having the first frequency. The resampling circuit 275 resamples the binarized output C of the binarizing circuit 273 with the pixel clock PC having the first frequency generated by the frequency dividing circuit 2751, and the resampling result is converted into a digital image. Output as signal D. That is, the resampling circuit 275 performs pulse shaping by compressing the data for the second frequency into data for the first frequency and synchronizing with the edge of the first pixel clock PC. Note that the pixel clock PC having the first frequency may be generated in the timing generation circuit 276 and supplied to the resampling circuit 275. In this case, it is not necessary to provide the frequency dividing circuit 2751 in the resampling circuit 275.

  FIG. 4 is a diagram showing a waveform example of main signals of the image sensor unit 27 according to the present invention. The example shown in FIG. 4 is a case where the pixel clocks PC for two cycles are associated with the unit pixel of the optical image. (A) shows the waveform of the pixel clock of the second frequency, (b) shows the waveform of the output B and slice level S of the BPF 272 input to the binarization circuit 273, and (c) shows the binary value. 7 shows the waveform of the binarized output C of the digitizing circuit 273, (d) shows the waveform of the resampling clock RSC corresponding to the pixel clock of the first frequency generated in the resampling circuit 275, and (d) shows the resampling clock RSC. The waveform of the digital image signal D output from the sampling circuit 275 is shown.

  The output B of the BPF 272 exhibits an AC behavior, and the slice level S substantially follows the intermediate level of the output B of the BPF 272 (see FIG. 4B). Therefore, the binarized output C is binarized into a bright point (1) and a dark point (0) in accordance with a realistic change in the output B of the BPF 272 (see FIG. 4C). The first frequency of the resampling clock RSC is “½ times” the second frequency of the pixel clock PC (see FIGS. 4A and 4D). There is no fear of losing the state change between the point (1) and the dark point (0) (see FIGS. 4C and 4D). As a result, the binarized output C is corrected to an appropriate digital image signal D (see FIGS. 4C and 4E).

FIG. 5 is a diagram showing various image patterns handled by the image sensor unit 27 according to the present invention.
The image sensor array 271 receives the reproduction beam from the hologram medium 22 as a multi-value image pattern of “m pixels × m pixels” having multi-gradation information. In the image sensor array 271, for example, “2 m pixels × 2 m pixels” light receiving elements 2711 are two-dimensionally arranged with respect to the multi-value image pattern. That is, the captured image pattern read by the image sensor array 271 is generated by receiving the unit pixels of the multi-value image pattern by the four light receiving elements 2711.
The image sensor unit 27 separates each unit pixel in the captured image pattern into a bright point (1) and a dark point (0) by the process described above. As a result, the image sensor unit 27 finally generates a binary image pattern of “m pixels × m pixels” corresponding to the multi-value image pattern of “m pixels × m pixels”.

  As mentioned above, although embodiment of this invention was described, embodiment mentioned above is for making an understanding of this invention easy, and is not for limiting and interpreting this invention. The present invention can be changed / improved without departing from the gist thereof, and the present invention includes equivalents thereof.

It is a figure which shows the structure of the hologram reproduction apparatus which concerns on one Embodiment of this invention. It is a figure which shows the structure of the image sensor unit which concerns on one Embodiment of this invention. It is a figure which shows the structure of the image sensor array which concerns on one Embodiment of this invention. It is a figure which shows the example of a waveform of the main signal of the image sensor unit which concerns on one Embodiment of this invention. It is a figure which shows the various image patterns which the image sensor unit which concerns on one Embodiment of this invention handles. It is the figure which showed typically a mode when a monomer changes to a polymer in a hologram medium. It is a figure explaining the recording format of a hologram medium. (A) shows the appearance data distribution at the time of ideal hologram reproduction, (b) is a figure which shows the appearance data distribution at the time of realistic hologram reproduction. It is a figure which shows the structure of the conventional image processing system.

Explanation of symbols

1 CPU
DESCRIPTION OF SYMBOLS 10 Laser apparatus 16 Galvo mirror 17 Galvo mirror control part 20 Scanner lens 22 Hologram medium 26 Fourier transform lens 27 Image sensor unit 271 Image sensor array 2711 Light receiving element 2712 Vertical register 2713 Horizontal register 2714 Charge voltage conversion circuit 2715 Amplifier 272 BPF 273 Binary Circuit 274 LPF 275 resampling circuit 2751 frequency dividing circuit 276 timing generation circuit 28 image sensor control unit 30 decoder 101 image sensor 102 ADC
103 Image processor 104 Buffer memory 105 Binary circuit 106 Register

Claims (5)

An image sensor unit for sampling an optical image with a plurality of light receiving elements,
When the frequency of the pixel clock for performing sampling of the unit pixel of the optical image in one light receiving element for each cycle is the first frequency,
In the predetermined number of light receiving elements exceeding one, each unit pixel of the optical image is sampled by the pixel clock having a second frequency which is the predetermined number times the first frequency. Image sensor unit to be used. A binarization circuit that binarizes an analog image signal read out by sampling a unit pixel of the optical image with a pixel clock of the second frequency at a slice level;
An integration circuit that integrates the binarization output of the binarization circuit and supplies the binarization circuit to the binarization circuit as the slice level;
The image sensor unit according to claim 1, further comprising:   The image sensor unit according to claim 2, further comprising a resampling circuit that resamples the binarized output of the binarization circuit with the pixel clock having the first frequency.   3. The image sensor unit according to claim 2, wherein an analog filter circuit that filters a predetermined frequency band of the analog image signal is provided in a preceding stage of the binarization circuit to which the analog image signal is supplied. A hologram recorded by causing a coherent data beam reflecting data to be recorded and a coherent recording reference beam to interfere with each other by a hologram medium can be used at the same incident angle as the recording reference beam. A hologram reproducing apparatus for performing reproduction based on an optical image obtained by making a coherent reproduction reference beam incident on the hologram medium,
Consists of multiple light receiving elements,
When the frequency of the pixel clock for performing sampling of the unit pixel of the optical image in one light receiving element for each cycle is the first frequency,
An image sensor unit that samples each unit pixel of the optical image with the pixel clock having a second frequency that is the predetermined number times the first frequency in the predetermined number of light receiving elements exceeding one. A hologram reproducing apparatus.