JPH01245231A - Parallel optical logic computing element - Google Patents

Abstract

PURPOSE:To perform a high-speed hierarchical computation by corresponding the polarizing condition of light with logical values 0 and 1, and electrically controlling the condition. CONSTITUTION:Two linearly polarized wave light intersecting orthogonally with each other are allowed to correspond with the logical values '0' and '1', respectively; optical signals equivalent to the negative of the logical values are formed by electrically controlling plural polarization plane control elements. That is, the polarization plane control elements 1, 2, and 3 are driven by switching their electric circuits with switches 7-1, 7-2, and 7-3, respectively; the electric circuits is comprised of power sources 8-1, 8-2, and 8-3 and the polarization plane control elements 1, 2, and 3, respectively. The polarizing condition of the light is correspond with the logical values 0 and 1, and the condition is electrically controlled. Therefore, the optical signal, which denies the inputted light and computed result, is readily formed in a real time, and the high-speed hierarchical computation is performed.

Description

[Detailed Description of the Invention] [Industrial Application Field] The present invention relates to an optical logic arithmetic unit, and in particular, the present invention relates to an optical logic arithmetic unit.
The present invention relates to a parallel processing type optical logic calculator having a function of processing two-dimensionally distributed optical information such as a binary image or data pattern consisting of two linearly polarized lights at once in two dimensions.

[Summary of the Invention] The present invention relates to a two-dimensional optical logic operation unit that performs digital arithmetic processing on two-dimensional information such as a binary image or a data pattern at once without scanning it. An element, half mirror or polarization beam splitter, polarizer, and writing light that uses two-dimensional optical information corresponding to binary digital information (for example, 0 and 1) as an input signal to electrically control the polarization state of the light. Using a spatial light modulator whose main component is a spatial light modulator whose polarization state of the readout light intensity changes nonlinearly depending on the intensity, 16 types of binary logical operations such as AND and OR can be executed two-dimensionally and the results can be displayed. This is what I did.

[Prior Art 1] Conventional parallel processing type optical logic arithmetic units include, for example, the following elements.

(1) Reference 1 (R0^, ^thale and S, H
, Lee :Optical Engineering,
Vol, 18. No. 5 (1979) p. 51
3-p., 517), the main components are a spatial light modulator consisting of a photoconductive layer and a liquid crystal layer, a polarizer, and an analyzer.

(2) Literature 2 (Y, Ichioka and J, T
an1da: Procee-dings of t
he IEEE, Vol. 72. No. 7 (19
84) As shown in p. 787 to p. 801), the main components are a mask for encoding an input signal, a mask for decoding, and a projection optical system.

(3) Reference 3 (F, T, S, Yu, S, Jutam
ulia and T, Lu:0ptics Comm
unications, Vol. 63. No. 4° (1987) p. 225-p. 229), a spatial light modulator using the magneto-optic effect.

The main components are a polarizer, analyzer, and mirror.

(4) Reference 4 (A1. Lohmann and J,
Weigelt: App-1ied 0ptics
, Vol, 2B, No, 1 (1987) p, 1
31-p, 135), a half-wave plate,
The main components include transparent glass, Wollaston prisms, lenses, spatial filters, and polarizers.

[Problems to be Solved by the Invention] The conventional parallel processing type optical logic arithmetic units (1) to (4) disclosed in Documents 1 to 4 have the following drawbacks.

In the spatial light modulator, which is one of the main components of the device in Document 1, as shown in FIGS. 12 are sandwiched, and transparent electrodes 9-2 and 9-3 are further arranged on the exposed amphibatic surfaces. Transparent electrodes 9-2 and 9
-3 and connect the power supply 8 between it. For this structure,
It is necessary to input one of the two input optical signals (for example, A) into the photoconductive layer lO, and the other input optical signal (for example, B) to input into the transparent electrode 9, thereby causing the difference between A-B and A-B.
can be calculated, but when performing AND-based calculations involving 100, A and B must be exchanged, and A+
When performing an OR operation such as B or A+B, a plurality of spatial light modulators are required, and the configuration needs to be changed, resulting in a lack of versatility. Furthermore, since two cells (a portion with the photoconductive layer IO and a portion without the photoconductive layer IO) are required for one operation, there is a problem that the degree of parallelism deteriorates.

The device in Reference 2 cannot spatially encode the input signal in real time, so it is not suitable for high-speed calculations, and since it uses a decoding mask, the resolution of the image and the degree of parallelism of the data pattern decrease, etc. It has the following disadvantages.

In the device of Document 3, input information is written into the spatial light modulator by scanning using electrical signals, so it is difficult to connect the devices in cascade and perform logical operations hierarchically, and it is difficult to perform logical operations hierarchically.
Since the optical systems for logical operations in the D system and the OR system are significantly different, they have drawbacks such as a lack of versatility.

Further, in the devices of these documents 1 to 3, the logical values correspond to the maximum output light intensity and the minimum output light intensity, so it is not easy to form a negative signal.

For example, when the minimum output light intensity corresponds to "0", a light emitting element and an amplifying element are required to form the negative signal "l". Therefore, these elements are essentially impenetrable to hierarchical operations.

In the device of Reference 4, the polarization state of light is divided into two values (“O” and “
1"), it is possible to form a negative signal;
is used to form a two-dimensional input signal. Therefore, real-time processing cannot be performed, and since two input signals are connected in cascade and logical operations are performed by spatial filtering, the optical system becomes extremely complex.

Furthermore, since the elements of Documents 1 to 4 do not have a signal amplification function, there is also a drawback that when performing hierarchical calculations, the optical signal decreases according to the number of layers, and the S/N ratio decreases.

SUMMARY OF THE INVENTION Therefore, an object of the present invention is to solve the above-mentioned various drawbacks, to be able to easily form a signal that negates an input signal and a calculation result in real time, to be able to amplify the signal, and to enable changes in the element configuration. It is an object of the present invention to provide a parallel optical logic operation unit that can execute AND and OR type operations without addition, and can increase the resolution of images and the degree of parallelism of data patterns.

[Means for Solving the Problems] In order to achieve such an object, the present invention receives a plurality of input lights each consisting of linearly polarized light and having a two-dimensional spread, and determines the polarization state of each output light. a plurality of polarization plane control elements rotated by 90 degrees, a first optical means for combining each output light from the plurality of polarization plane control elements, and a polarization direction parallel or orthogonal to the combined light from the first optical means. A polarizer that takes out the light, a threshold element that spatially changes the optical properties of the medium with respect to the output light from the polarizer, and a linearly polarized readout with a two-dimensional spread that is input to the threshold element. A polarization control element that rotates the polarization state of the light and readout light by 90 degrees, and an output/output destination from the polarization control element is input to the threshold element, and the output light from the threshold element is inputted to the threshold element. It is characterized by comprising a second optical means for separating and extracting the output light from the polarization control element.

Alternatively, the present invention provides a plurality of polarization plane control elements each receiving a plurality of input lights each consisting of linearly polarized light and having a two-dimensional spread and rotating the polarization state of each output light by 90°; A first unit that combines each output light from the control element.
an optical means, a polarizer for extracting a polarization direction parallel or orthogonal to the combined light from the first optical means, and a threshold at which the optical properties of the medium spatially change with respect to the output light from the polarizer. The threshold element comprises a value element, a linearly polarized readout light having a two-dimensional spread that is input to the threshold element, and a polarization plane control element that rotates the polarization state of the readout light by 90 degrees. B112SiO,. Or Bf12Ge0
It consists of a spatial light modulator in which an insulating layer and a transparent electrode are arranged on both sides of a 2° crystal, and on the input side of the spatial light modulator, blue light, which is input light, is transmitted as a second optical means, A dichroic mirror that reflects red light as readout light is disposed, and a wavelength filter that blocks blue light and transmits red light is disposed on the output side of the spatial light modulator.

Here, the polarization plane control element may be a liquid crystal cell that utilizes optical rotation power or birefringence based on a change in molecular alignment induced by an electric field in twisted nematic or cholesteric liquid crystal, or a LiNb01. LiTa0. .

Ferroelectric crystals such as KH2PO4 and KH2PO4 and PL
It can be configured with an optical shutter that utilizes the birefringence based on the electro-optic effect of a ceramic material such as ZT.

Alternatively, the threshold element can be constituted by a spatial light modulator in which a photoconductive layer 1 reflecting mirror and a liquid crystal layer are interposed between two transparent electrodes.

The first or second optical means can be composed of a half mirror-1 polarizing beam splitter or a dichroic mirror.

The first optical means is a prism-type polarizing beam splitter, the second optical means is a prism-type half mirror, and the plurality of polarization control elements and threshold elements are connected to the prism-type polarizing beam splitter and the prism-type polarizing beam splitter. It can also be configured in close contact with a half mirror.

In the present invention, two linearly polarized lights that are orthogonal to each other are made to correspond to logical values "0" and "1", and an optical signal corresponding to the negation of the logical values is electrically controlled by the plurality of polarization control elements. It can be formed by

[Function] In the parallel optical logic operator of the present invention, the polarization state of light corresponds to logical values 0 and 1, and the polarization state is electrically controlled. Light can be easily formed in real time, suitable for high-speed hierarchical calculation, and the problems of the above-mentioned documents 1 to 4 can be solved.

By using a spatial light modulator as a threshold element,
Since signal amplification is possible, the problems of References 1 to 4 can be solved.

Since AND-based and OR-based operations can be performed without changing the element configuration, the problems of Documents 1 and 3 can be solved.

Since there is no need to spatially separate or spatially encode input signal light, image resolution and parallelism of data patterns can be increased, thereby solving the problems of References 1 and 2. .

[Examples] Examples of the present invention will be described below with reference to the drawings.

FIG. 1 shows an embodiment of a parallel optical logic operator according to the present invention.

Referring to FIG. 1, the embodiment of the present invention consists of polarization plane control elements 1 to 1 which form negative signals of input signal lights A and B and output signal light E, which are composed of linearly polarized light and have a two-dimensional spread.
3 (INVI, INV2.TNV3), and a half mirror 4-1 which is a first optical means for combining input signal light.
A half mirror 4-2 which is a second optical means for separating the readout light and the output signal light E, a polarizer 5 having a polarization direction parallel or perpendicular to the polarization directions of the input signal lights A and B, and a polarizer The optical properties of the medium change spatially with respect to the output light from the threshold element 6, which controls the readout light.

The polarization control elements 1, 2.3 may be a liquid crystal cell (an element that utilizes the optical rotatory power or birefringence index induced by changes in molecular alignment caused by an electric field in twisted nematic liquid crystal or cholesteric liquid crystal) or an optical shutter (LiNbOs, L
iTaO5, o2po4. It can be constructed from an element that utilizes a change in birefringence based on the electro-optic effect of a ferroelectric crystal such as K)l, po4, or a ceramic material such as PLZT.

These polarization control elements 1, 2.3 are connected to a power source 8, respectively.
-1.8-2.8-3 and the polarization control element 1°2.3, each switch 7-1°7-2.7
It is driven by opening and closing with -3.

Further, as the threshold element 6, two transparent electrodes 9-2 are used.
and 9-3, a photoconductive layer 10 that converts the intensity of the optical input signal into a change in impedance, and an insulating reflecting mirror 1.
1 and an electro-optic layer 12 that replaces impedance change with refractive index change or optical rotation power change is suitable. This threshold element 6 is driven by an AC power source 8-4.

Here, as a typical example of such a spatial light modulator, L(:L
V (Reference 5: W, P, Bleha, L, T, L
ipton.

E., Wiener-Avnear; J., Grinbe.
rg, P, G, Re1f.

D, Ca5asent, H, B, Brown a
nd B, V, Markevitch: 0pti
cal Engineering, Vol. 17.
No. 4 (1978) 9.371-9.384) can be used, and its operation will be briefly explained here.

The liquid crystal layer of the LCLV is in a twisted nematic state when the electric field applied to the liquid crystal is zero, and exhibits birefringence when an electric field is applied, which is a so-called hybrid field effect mode type.

In 111:LV, when the input light C shown in FIG. 1 is not irradiated onto the photoconductive layer IO, most of the electric field is concentrated on the photoconductive layer IO, and the liquid crystal remains in the twisted nematic state. Therefore, the readout light and the output signal light E are in the same polarization state. Next, when the input light C is irradiated onto the photoconductive layer 10, the impedance of the photoconductive layer 10 decreases and an electric field is applied to the liquid crystal layer. This electric field changes the birefringence of the liquid crystal and rotates the plane of polarization of the output signal light E.

FIG. 2 shows the input light intensity versus output light intensity characteristics of the LCLV. Here, if the intensity of the input light C is set in the saturation region of the characteristic curve as shown in Figure 2, the polarization plane of the output light can be rotated by approximately 90 degrees depending on the presence or absence of the input light. can. Setting of pc can be performed by adjusting the applied voltage of LCLV, its frequency and input light intensity.

Based on the above, the operation of the element shown in FIG. 1 will be explained. However, an LCLV is used as the threshold element 6, and a twisted nematic liquid crystal cell is used as the polarization control element 1.2.3. This liquid crystal cell has the function of rotating the polarization direction of an input signal by 90'' or 0° depending on whether the switch is opened or closed.

Furthermore, when the polarization direction of the linearly polarized input light is within the plane of the paper, this changing direction is represented by ↓, and on the other hand, when it is perpendicular to the plane of the paper, the polarization direction is represented by ■. Here, it is assumed that the cross direction and the {circle around (2)} direction correspond to logical values 1 and 0, respectively. Furthermore, it is assumed that the polarization direction of the polarizer is within the plane of the paper.

Under these conditions, a case will be described in which input images A and B linearly polarized in the ↓ and [stock] directions are incident on this arithmetic element as shown in FIG. Here, assuming that the intensity of light in the A, In)↓ and ■ directions is all normalized to 1, the presence or absence of the applied voltage (O
The output light intensity C of the polarizer 5 is expressed as shown in Table 1, depending on the polarizer 5 (represented by N and OFF).

Table l When this optical signal C enters the threshold element 6, within the liquid crystal layer 12 constituting this threshold element 6, the light intensity is 0, 1.
．． 3 according to 2! 'A class A refractive index distribution is formed.

Therefore, the linearly polarized readout light is transferred to the half mirror 4-2.
When the light is made incident on the threshold element 6 through the refractive index distribution, an output signal E having three types of polarization planes is obtained depending on the refractive index distribution.

However, as shown in FIG. 2, when the input/output optical characteristics of the threshold element 6 exhibit strong nonlinear characteristics and the maximum intensity of the optical signal C is approximately equal to PC, the output signal light E is equal to that of the optical signal C. Depending on the intensity! It can be safely assumed that it consists of two types of linearly polarized light: (1) and (2). That is, when viewed from the output side, signals 1 and 2 of the signals C in Table 1 can be considered to be degenerate.

Here, when the readout light linearly polarized in the {circle around (2)} direction is incident on the threshold element 6 and a voltage is applied to the polarization plane control element 3, the polarization state of the output signal light E will be as shown in Table 2. Then, an OR operation is performed. However, here too, 0 is ■
Corresponds to the direction, 1 is! Corresponds to the direction.

Next, when the voltage applied to the polarization control element 3 is set to O, the polarization state of the output light E is rotated by 90 degrees, and as shown in Table 3,
ANO system calculations are performed.

Furthermore, when one of the input signals A and B is cut off and a voltage is applied to the polarization control element 3, 1. A, A, B, B, etc. signals can be formed. This situation is shown in Tables 4 and 5, respectively.

Table 4 Chapter 5 Table Furthermore, if both A and B signals are blocked, polarization planes can be separated.

Depending on the presence or absence of the applied voltage to the control element 3, a signal of T (all states are 1) or F (all states are 0) can be formed. This situation is shown in Table 6.

Table 6 and above are the logical operations obtained by driving the polarization control element once, and if the polarization control element is operated twice, IBB (
=AB+AB), AeB (=AB+AB), and other exclusive ORs and negative exclusive ORs can be obtained.

Another embodiment according to the invention is shown in FIG.

In the embodiment shown in FIG. 3, a polarizing beam splitter 13 is used as a merging element for input signal light, and the half mirror 4-1 and polarizer 5 in the embodiment shown in FIG. 1 are omitted.

In this embodiment, it is also possible to prevent the power of the input signal light from decreasing. However, 1 and 0 of input signal light B correspond to light linearly polarized in the ■ direction and ↓ direction, respectively, and 1 and 0 of input signal light A correspond to the same as before! It is assumed that the linearly polarized light corresponds to the direction and the {circle around (2)} direction, respectively.

Still another embodiment of the invention is shown in FIG.

In this embodiment, a prism-type polarizing beam splitter 1
Polarization control elements 1 and 2 are disposed in close contact with each of the two incident surfaces of 4, and one principal surface of a threshold element 6 is similarly disposed in close contact with the output surface. One entrance surface of the prism-type half mirror 15 is placed in close contact with the other main surface of the threshold element 6, and the polarization control element 3 is also placed in close contact with the other entrance surface.

In this way, the prism type polarizing beam splitter 1
4 and the half mirror 15, the polarization control element 1°2.3 and the threshold element 6 are closely attached and integrated. As a result, it is possible to construct a parallel optical logic unit that is compact and stable against vibrations and the like.

Still another embodiment of the invention is shown in FIG.

This embodiment utilizes the fact that the two output lights C' and C'' emitted from the polarization beam splitter 13 have a complementary relationship. That is, by applying voltages to the polarization control elements 1 and 2, C' = If A+8, then C' = A+B is obtained. In this embodiment, these lights are transmitted through reflecting mirrors 16-1 and 16-2, respectively, to two filters having the same configuration as the threshold element 6 described above. input to threshold elements 6-1 and 6-2.These threshold elements 6-1 and 6-2 are connected to power supplies 8-4-1 and 8-4-, respectively.
2. Here, when the voltage applied to the polarization control element 3 is set to O, the output signal light E is as follows: E = A + B + A + B = A B +
A B <1), and a negative exclusive OR output is obtained. Next, when the voltage applied to the polarization control element 1 is set to O and the voltage is applied to the polarization control element 2, C'=A
+B, C'=A+B (2). Here, when the voltage applied to the polarization control element 3 is set to 0, E=A+a+XTI=Aa+aa (3),
The output signal light E becomes an exclusive OR. With this configuration, the exclusive OR or the negative exclusive OR can be obtained by operating each of the polarization plane control elements 1, 2, and 3 once.

Still another embodiment of the invention is shown in FIG.

This embodiment includes one threshold element 6, one half mirror 4-2, and n+1 polarization control elements ao+aI+.
""van and n polarizers Pl+P2+""+Pn
Using a parallel optical logic unit consisting of This is an example in which logical operations ^2+A3+...+An are performed in parallel and at the same time. In this configuration, n polarization control elements a, ms·, an and corresponding n polarizers P1 . ...” +Pn is placed appropriately and the input optical signal A is
I +..., An, each polarizer PI + -,
The optical signal emitted from pn is guided to one threshold element 6. The output light from the threshold element 6 and the readout light from the polarization control element a0 are combined into an eight-frame mirror 4.
-2, and output signal light E is taken out from this half mirror 4-2.

In FIG. 6, 71.72, 73. =, 7n indicates whether or not to apply the voltage from each drive power source 81.82°83,..., 8n of each polarization plane control element al+a2+a3+...van to each polarization plane control element al+a2+a3+...van. It is a control switch. A voltage from a power source 8-3 is applied to the polarization control element a0 via a switch 7-3.

According to this example, E=^,・A2・A3・・・・＾.

E; 8,・A2・A3・・-A.

E=^1・^, fist A,...^.

E=AI・A2・A3・・・・・＾、 E=^1・A2-A,---A.

E=AI ・ A2 ・ ^ko ・ ...A11
(4) E=AI-A2・
A3..."An-1^.

E=A, +^, +^3・...+A.

E=A, +A2+A3・・−・+A.

E=A, +^, +A3・...+^.

E=AI+A2+A3...-+A.

E=^1+A2+^3 ” “” 10AnE=^1+^
2+A3 ・=・+An−+AnE=A+A,+・・
・+An-1^.

etc.2n+1◆ Two types of logical operations can be executed.

Furthermore, the parallel optical logic unit according to the present invention can be constructed using a spatial light modulator other than the LCLV as the threshold element 6. For example, as shown in Figure 7,
As a threshold element, Bi+25iOzo or Bi
A spatial light modulator 18 is used in which insulating layers 16-1 and 16-2 and transparent electrodes 9-2 and 9-3 are arranged on both sides of a 12GeO20 crystal 17, respectively, and a polarizer 5 and a space corresponding to the polarizer 5 are used. A dichroic mirror 19 that transmits blue light and reflects red light is disposed between the light modulator 18 and the second optical means, and further blocks blue light.
A wavelength filter 20 that transmits red light is arranged on the output side of the spatial light modulator 18. Furthermore, blue light is used as the input signal lights A and B, and red light is used as the readout light.

A parallel optical logic arithmetic unit having such a configuration can exhibit the same functions as the parallel optical logic arithmetic unit shown in FIG. Furthermore, in this arithmetic unit, by opening and closing the switch 27-1 connected in series to the DC electric wire 21 and the switch 27-2 also connected in parallel, the calculation result can be stored.
It can even be reversed.

Also, a half mirror 4- for combining the input signal lights A and B.
Even if a polarizing beam splitter is used instead of 1, it is possible to construct a parallel optical logic unit that exhibits the same function as the optical logic unit configured as shown in FIG. 7 and has less loss of input signal light. .

[Effects of the Invention J As is clear from the above explanation, the parallel optical logic arithmetic unit of the present invention has the following advantages.

(a) Since the polarization state of light corresponds to logical values O and 1 and the polarization state is electrically controlled, it is possible to easily form a signal light that negates the input signal light and the calculation result in real time. Therefore, it is suitable for high-speed hierarchical operations.

(b) Signal light can be amplified by using a spatial light modulator as a threshold element.

(C) AND system and O
R-based operations can be executed.

(d) Since there is no need to spatially separate or spatially encode input signal light, image resolution and data pattern parallelism are high.

Because of these features, the parallel optical logic unit of the present invention can be effectively applied to image processors that process extremely large amounts of information in a hierarchical manner at high speed, such as in video processing and pattern recognition, and to the calculation units of parallel processing computers. is possible.

[Brief explanation of the drawing]

FIG. 1 is a configuration diagram showing an embodiment of a parallel optical logic operator of the present invention, FIG. 2 is a characteristic diagram showing the relationship between the input light intensity and output light intensity of the LCLV, and FIGS. 3 to 7 are Configuration diagrams showing various other embodiments of the parallel optical logic unit of the present invention; FIGS. 8(a) and 8(b) show conventional configuration examples of the spatial light modulator described in Document 1, respectively, FIG. 2 is a cross-sectional view and a perspective view. 1 to 3-...Polarization control element (INVI, INV2゜I
NV3), 4-1.4-2...Half mirror-1 5...Polarizer, 6.6-1.6-2...Threshold element, 7-1.7-
2.7-3...Switch, 8-1.8-2.8-3.
...Power supply, 8-4.8-4-1.8-4-2...AC power supply, 9-
2.9-3...Transparent electrode, 10...Photoconductive layer, 11...Reflector, 12...Electro-optic layer, 13.14...Polarizing beam splitter, 15...Half mirror 1 16-1, 16-2...Reflecting mirror, 17...Crystal, 18...Spatial light modulator, 19-...Dichroic mirror, 20...Wavelength filter, 21...DC power supply, 27-1.27-2...Switch, 71.72,73. ..., 7n...switch.

Claims (1)

[Claims] 1) A plurality of polarization plane control elements each receiving a plurality of input lights each consisting of linearly polarized light and having a two-dimensional spread and rotating the polarization state of each output light by 90 degrees; a first optical means for combining each output light from the wavefront control element; a polarizer for taking out a polarization direction parallel or perpendicular to the combined light from the first optical means; and a polarizer for combining the output light from the polarizer. On the other hand, there is a threshold element in which the optical properties of the medium spatially change, a linearly polarized readout light with a two-dimensional spread input to the threshold element, and a polarization state of the readout light of 90%. A polarization control element that rotates the polarization control element, inputting the output light from the polarization control element to the threshold element, and separating the output light from the threshold element from the output light from the polarization control element. and second optical means for taking out the data. 2) A plurality of polarization control elements that each receive a plurality of input lights that are linearly polarized light and have a two-dimensional spread and rotate the polarization state of each output light by 90 degrees; a first optical means for combining each output light; a polarizer for taking out a polarization direction parallel or orthogonal to the combined light from the first optical means; and a medium optical means for the output light from the polarizer. a threshold element whose spatial properties change spatially; a linearly polarized readout light with a two-dimensional spread that is input to the threshold element; and a polarization plane that rotates the polarization state of the readout light by 90 degrees. A dichroic mirror that transmits blue light as input light and reflects red light as readout light is disposed as a second optical means on the input side of the threshold element, and A parallel optical logic operator characterized in that a wavelength filter that blocks blue light and transmits red light is arranged on the output side of a value element. 3) The first optical means is a prism-type polarizing beam splitter, the second optical means is a prism-type half mirror, and the plurality of polarization plane control elements and the threshold element are connected to the prism-type polarization beam splitter. 2. The parallel optical logic unit according to claim 1, wherein the parallel optical logic unit is configured in close contact with the polarizing beam splitter and the prism-type half mirror.