JP2793815B2 - Exposure control device - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an exposure control device capable of performing photographing for any subject that satisfies certain conditions.

[Prior Art] As a conventional example of such an exposure control apparatus, Japanese Patent Laid-Open No.
There is a device described in 42026. This is because, in a multi-photometer that divides an object field into a plurality of areas and measures a plurality of photometric outputs corresponding to the respective areas, a maximum luminance, an intermediate luminance between the maximum luminance and the average luminance, according to the maximum luminance. Luminance,
One is selected from the average luminance, the intermediate luminance between the average luminance and the minimum luminance, and the minimum luminance, thereby controlling the exposure. Japanese Patent Application Laid-Open No. 61-173226 discloses that a backlight condition is detected from a spot photometric value at the center of a screen and an average photometric value of the entire screen, and a warning display or an exposure value is corrected. A technique for eliminating a detection error by using subject distance information for backlight detection has been disclosed.

[Problems to be Solved by the Invention] In these conventional examples, since a subject pattern is recognized by a microcomputer, it can be applied only to a limited subject pattern, or an unintended photograph can be made by performing erroneous detection. And the like. Further, if the intended exposure control is to be performed on many subject patterns, the subject pattern or the brightness cannot be formulated, so that there are disadvantages such as a large number of programs and a long calculation time. Furthermore, even if a photograph is produced as intended by the manufacturer, the evaluation of the photograph largely depends on the sensitivity of the individual, so that good results are not necessarily obtained for all users.

An object of the present invention is to solve these drawbacks, by detecting a pattern of a subject by a network that generates a desired output with respect to various inputs, and performing exposure control based on the result, so that any subject pattern can be obtained. To provide an exposure control device capable of taking a photograph with an intended exposure.

[Means and Actions for Solving the Problems] In the exposure control device according to the present invention, when the subject pattern is input, the exposure strength is learned so as to output the exposure correction amount for the average photometry according to the subject pattern. There is provided a network consisting of an input layer, a hidden layer, and an output layer that are sequentially connected. Outputs of the plurality of photoelectric conversion elements are input to the network, and exposure is controlled based on the output of the network and the outputs of the plurality of photoelectric conversion elements.

Embodiment An embodiment of an exposure control device according to the present invention will be described below with reference to the drawings. FIG. 1 shows a block diagram thereof.
As can be seen from FIG. 1, in this embodiment, exposure control and focus detection are performed using a neurocomputer. Therefore, first, the neurocomputer will be described with reference to FIGS.

FIG. 2 shows a model of the neurocomputer. This model was proposed by Rumelhart and others, and backpropagation (Back Propagati
on) model (hereinafter abbreviated as BP model). A neurocomputer is composed of many units (neurons), and the units are classified into an input layer, a hidden layer, and an output layer. Each unit is connected in the order of input layer → intermediate layer → output layer to form a network (neural network). The strength of the connection of each unit is determined by learning.
However, there is no connection between the units in each layer. FIG. 3 shows a model of each unit.

Next, the principle of this BP model learning algorithm will be described. When a certain pattern P is given to the input layer, the actual output value appearing in the output layer is defined as Opj ^0, and the desired output value (hereinafter referred to as a teacher signal) at that time is defined as tpj.
Epj is expressed as follows.

Epj = 1/2 (tpj−Opj ⁰ ) ² (1) In order to perform learning, the error Epj should be reduced as follows.
What is necessary is just to change the strength of the connection of all the units.

The strength of the connection from the i-th unit of the (K-1) layer to the j-th unit of the K layer when the pattern P is given
The change amount of Wji is defined as follows. Here, K increases as the output layer becomes 0 and the input layer becomes 0.

ΔpWji ^K ∝−∂Ep / ∂Wji ^K …… (2) ∂Ep / ∂Wji ^K = (∂Ep / ∂net pj ^K ) · (∂net pj ^K / ∂Wji ^K ) …… (3) It is.

Also, if f is a sigmoid function, Opk ^K
= F (net pk ^K ), the equation (3) is modified as follows. The sigmoid function is shown in FIG.

∂Epj / ∂Wji ^K = −δpj ^K · Opi ^{K + 1} (4) where δpj ^K is the backward propagation amount of the error in the K-th layer, and δpj ^K = −∂Epj / ∂net pj ^K is there. Therefore,
Equation (2) is modified as follows. Here, η is a constant.

ΔpWji ^K = η · δpj ^K · Opi ^{K + 1} ... (5) In the case of an output unit, Epj = 1/2 (tpj−Opj ⁰ ) ² , Op
Since j ⁰ = f (net pj ⁰ ), the backward propagation amount δpj ^{0 of the} output layer
Is as follows:

In the case of the intermediate unit, there is no unit combination in each layer, and the backward propagation amount of the error is as follows.

Equation (7) is a recursive function of δ.

The general formulation of ΔpWji ^K is as follows.

ΔpWji ^K (n + 1) = ηδpj ^K · Opi ^{K + 1} + αΔpWji ^K (n) (8) where ΔpWji ^K (0) = 0 and n represents the number of times of learning. The second term on the right side of equation (8) is added to reduce the error oscillation and speed up the convergence. From equation (8), the strength of the connection is updated as follows.

Wji ^K (n + 1) = Wji ^k (n) + ΔpWji ^K (n) (K = 0, 1, 2,...) (9) Here, the sigmoid function fi is represented by fi = 1 / (1 + e- ^neti ). Since fi ′ = fi (1-fi) defined by (10), the amount of backward propagation is simplified as in the following equation.

In the case of an output unit: δpj ⁰ = Opj ⁰ (1−Opj ⁰ ) (tpj−Opj ⁰ ) (11) In the case of an intermediate unit: As can be seen from the above, the calculation of ΔW starts from the unit of the output layer and moves to the unit of the intermediate layer. Thus, the learning proceeds in a direction opposite to the processing of the input data.

Therefore, learning using the BP model is performed as follows.
First, learning data is input, and the result is output. Next, the coupling strength is changed so as to reduce the resulting error (the difference between the actual output and the teacher signal). Then, the learning data is input again. This operation is repeated until ΔW converges.

 FIG. 5 shows the basic circuit configuration of the BP model.

A random access memory (hereinafter referred to as RAM) 1 stores the strength of connection Wji, and is composed of N pages of k = 1 to N for each layer. RAM2 is the coupling strength Wji when the pattern P is given.
, And includes N pages of k = 1 to k. RAM3 stores the backward propagation amount δpj of the error, and k = 0 to
It consists of N (N + 1) pages. The RAM 4 stores the output value Opj of each unit and includes (N + 1) pages of k = 0 to N. 5 is an operation circuit of Opj, 6 is an operation circuit of δpj, and 7 is an operation circuit of ΔpWji. Reference numeral 9 denotes a sequence controller for controlling the entire sequence.

The learning process using the BP model in FIG. 5 will be described.
Here, the operation when a BP model is simulated by a Neumann computer will be described with reference to the flowcharts of FIGS. 6 is a flowchart of the Opj operation, FIG. 7 is a flowchart of the δpj operation, FIG. 8 is a flowchart of the Wpj operation, and FIG. 9 is a flowchart of the learning level determination.

In step 1 (S1), the connection strength Wji in RAM1 is initialized to a random value. In step 2, input value Opj ^{N + 1} is stored in RAM4
Set, will calculate the order unit output value OPJ ^K from the input layer toward the output layer by computation circuit 5 in step 3 to step 9.

Next, in steps 11 to 20 of FIG.
From the output value Opj ⁰ and the teacher signal tpj indicating the desired output according to the equation (11),
^{Find 0} .

Next, in steps 21 to 24 of FIG.
ΔpWji
^{0 Find} (1). Note that the initial value ΔpWji of ΔpWji ^0
K (0) is all zero. In step 25, the connection strength Wji ⁰ (1) is obtained by the arithmetic circuit 8 according to the equation (9).
As described above, the output layers Opj ⁰ , δpj ⁰ , ΔpWji ⁰ (1), Wji ⁰
(1) is obtained. Thereafter, these are stored in the RAM1 to RAM4 in such a manner that the initial data is updated.

Next, learning of the intermediate layer is performed. Returning to the flowchart of FIG. 7, δpj ⁰ , Wji
⁰ (1) and Opj ⁰ stored in RAM4,
The backward propagation amount Δpj ^K of the error is obtained. Next, in the flowchart of FIG. 8, the change amount ΔpWji ^K (1) of the coupling strength is obtained by the arithmetic circuit 7 according to the equation (8), and the coupling strength Wji ^K (1) is calculated by the arithmetic circuit 8 as (9) ) Determined according to the formula. At the same time as the output layer, the data obtained above are RAM1 to RA
It is stored in the form updated to M4. The above flow is sequentially repeated toward the input layer (K = N + 1), and the first learning is completed.

By performing the above learning a plurality of times (n), the strength of connection Wji between the units is determined, and when an input value Opj indicating a certain input pattern P is given, a desired output value Ppj is obtained. The network will be formed automatically.

FIG. 9 is a flowchart for calculating the mean square error ▲ between the actual output value Opj and the teacher signal tpj. The smaller this value is, the closer the actual output value is to the desired output value. If Ep is smaller than a certain threshold value ε, the learning is terminated, and if Ep is larger than ε, the learning is repeated.

In the above, the learning for one input pattern P has been described. However, it is also possible to perform learning such that a plurality of input patterns are obtained and a plurality of output patterns corresponding to each pattern are obtained. It is also possible to learn to output one specific output pattern for a plurality of input patterns.

The BP model described above can be realized by a Neumann-type microcomputer that is widely used in consumer devices and the like at present, but as it is, the function of speeding up by parallel processing, which is one of the great advantages of the neurocomputer, cannot be utilized. Therefore, it is preferable that the processes in FIGS. 6 to 9 be processed in parallel by a plurality of computers.

FIG. 10 shows the configuration of a parallel processing system for this purpose.
A plurality of microprocessors Pl to Pn are the host processor 11
Connected to. The neural network shown in FIG. 2 is divided into n partial networks, each of which is assigned to the microprocessors P1 to Pn. The host processor 11 controls the timing between the microprocessors Pl to Pn, and performs processing such as pattern recognition by integrating data distributed to the microprocessors Pl to Pn. Each of the microprocessors Pl to Pn follows the operation procedure described above,
The calculation of the continuous plural columns of the output value Opj shown in FIG. 5 is executed. Therefore, the microprocessors Pl to Pn are provided with a RAM and an arithmetic circuit for respectively storing δpj, ΔWji, and Wji necessary for calculating the output value in charge. When the calculation of the output values of all the units in charge is completed, communication for updating the data is performed while synchronizing the processors Pl to Pn. The host processor 11 determines the learning achievement level and controls the timing of the microprocessors Pl to Pn.

When processing such as pattern recognition is performed based on the learned result, from the input layer to the output layer shown in FIG. By performing the above calculation, the finally required output value Ppj ⁰ is obtained. Also in this case, by executing distributed processing by a plurality of microprocessors as shown in FIG. 11, the speed can be increased by the parallelism of the neural network.

In the course of learning, the circuit shown in FIG. 5 is basically required. However, when only the learning result is applied, the configuration is greatly simplified.

FIG. 11 shows a basic circuit configuration in this case. The input data is input via an input unit 12 (for example, indicating an A / D converter or the like). Are sequentially performed to obtain output data Opj ⁰ . The coefficient memory 14 storing the coupling strength Wji ^K is ROM,
Alternatively, a rewritable ROM may be used.

FIG. 12 is a schematic block diagram of a learning system at the time of manufacturing a product to which the learning result is applied. Product 16 has a built-in ROM17 for storing the intensity Wji ^K of binding. Reference numeral 18 denotes a learning device. The combination of the ROM 17 and the learning device 18 is basically the same as the device shown in FIG. 5, but when the writing of Wji ^K to the ROM 17 is completed, the learning device 18 and the learning device 18 are read. The device 18 is separated. Since it is not necessary to learn each product of the same type each time, the ROM 17 can be copied and used.

In the above description, the learning of the BP model and the application of the result have been realized by simulation using a Neumann-type computer that is currently used.
This is mainly because learning requires a complicated algorithm, and it is very difficult to automatically organize the weights of connections between neurons automatically by hardware. However, if the coupling strength Wij is known, the BP model shown in FIG. 1 can be configured by hardware, considering only the machine to which the learning result is applied. In the case of speeding up by parallel processing or applying to inexpensive consumer products, it is meaningless unless this method is adopted. This means that each unit shown in Fig. 2 is composed of an inverter, and the coupling strength Wi
It can be realized by replacing j with a resistor network Rij,
This can be easily achieved using recent LSI technology.

Next, an exposure control apparatus to which the above-described neurocomputer is applied will be described with reference to FIG.

An aperture 19 is provided on the front surface of the photographing camera 20, and a subject image passing through the aperture 19 is incident on a light receiving unit 21 in which photoelectric conversion elements Pmn are arranged in a matrix as shown in FIG. Therefore, from the light receiving unit 21, luminance information of the subject in the narrowed state is output for each photoelectric conversion element, and the
22, is digitized via an A / D converter 23 and supplied to an arithmetic circuit (ALU) 24 as a BV 'value. Arithmetic circuit 24
Is the brightness BV (= BV'−) of the subject from the light passing through the aperture 19.
AVo) is calculated.
Is open aperture value AVo.

The BV value for each photoelectric conversion element output from the arithmetic circuit 24 is supplied to the arithmetic circuit 27 and also to the neurocomputer 25. The neurocomputer 25 inputs the luminance distribution pattern of the subject output from the arithmetic circuit 24 Op
As j ⁰ , Is finally calculated to obtain the correction signal CV. The strength Wji of this connection is stored in the coefficient memory 26. Arithmetic circuit 2
7 is an apex calculation (BV + SV =) using the average value of the luminance BV, the film sensitivity SV, the aperture value AV, the shutter speed TV, the mode signal (MOD) such as shutter priority or aperture priority, and the correction signal CV supplied from the neurocomputer 25. TV + AV
+ CV) to determine the shutter speed or aperture value. The output of the arithmetic circuit 27 is the shutter control device 2
8. The exposure is supplied to the aperture control device 29, which controls the exposure. Thus, in this embodiment, the exposure is controlled in consideration of the correction value obtained by the neurocomputer 25 for the average photometric value. 30 is a sequence controller.

Although the basic block configuration of the neurocomputer 25 may be as shown in FIG. 5, here, learning is performed by a parallel computer as shown in FIG. 10 in order to increase the speed. Here, the coupling strength Wji of each unit is learned in advance so that an exposure correction signal is output when a subject pattern is input to the network. The coupling strength Wji obtained as a result of the learning is stored in the coefficient memory 26.

This learning system is shown in FIG. Model pattern 32
Is a neurocomputer for various object patterns Opj ^{N + 1}
33 is an input section for inputting to 33, including an A / D converter and the like. As the teacher signal 35, an exposure correction signal corresponding to various object patterns Opj ^{N + 1} is input to the neurocomputer 33 as a target value tpj. Neuro-computer 33 the actual output OPJ ⁰ is obtained by learning the strength Wji engagement to conform with the teacher signal TPJ. The determined connection strength Wji is stored in the coefficient memory 34. The coefficient memory 34 is incorporated in the camera as the coefficient memory 26.

As shown in FIG. 15, the neural network performs learning by independent neural nets S11,... For each row and integrates them in the output layer S0, as shown in FIG. Neural network has input layer 37, hidden layer 3
8. The output layer 39 is composed of three layers, and the principle of learning is as described above.

FIG. 16 shows a specific example of the model pattern. (A) is an example of backlight shooting, in which case the correction amount CV is set to +1 EV,
Take bright pictures of people. (B) is an example of the sea. In this case, since the sky is too bright and the sea is dark, the correction amount is + 0.5E.
V. (C) is an example of a neon street at night, and if photographed by a conventional method, the image is bright as in the daytime and the mood does not appear, so the correction amount is set to -3EV. This correction value is a correction value for the average photometric value. The above are only three examples, but in practice hundreds of patterns are learned.

Note that the neurocomputer has an excellent property that if it learns to a certain extent, it outputs the correct output even for patterns that were not input at the time of learning, and it largely depends on human sensitivity such as identification of the main part of the subject. It is very effective for solving difficult problems. In addition, by learning the neurocomputer, the relationship between a huge variety of subject patterns and the main subject can be self-organized, which could not be programmed with a Neumann-type computer, so that exposure control can be performed as intended. it can. Furthermore, since high-speed calculations can be performed by parallel control of a neurocomputer, it is suitable for a silver halide camera requiring quickness.

The present invention is not limited to the above-described embodiment, but can be variously modified. In the above description, the case where the main part of the subject is learned by the neurocomputer and the exposure control is performed has been described. However, the position of the main part of the subject may be given as a teacher signal of the neurocomputer, and learning such as spot metering may be performed. is there. Further, by inputting not only the brightness of the subject but also the temperature and humidity as input parameters of the neurocomputer, it is also possible to perform learning to give a subtle seasonal feeling that is difficult to formulate.

〔The invention's effect〕

As described above, according to the present invention, the neurocomputer is previously trained so as to output a desired exposure correction amount for various subject pattern inputs, and exposure control is performed based on the output. Thus, an exposure control device capable of taking a photograph with an intended exposure for any subject pattern can be provided.

[Brief description of the drawings]

FIG. 1 is a block diagram of an embodiment of an exposure control device according to the present invention, FIG. 2 is a diagram showing a model of a neurocomputer used in the embodiment, and FIG. 3 is a diagram showing a model of each unit constituting a network. FIG. 4 is a diagram showing a sigmoid function, FIG. 5 is a block diagram of a neurocomputer, and FIGS. 6 to 9 are flowcharts when the neurocomputer of FIG. 5 is simulated by a Neumann computer. The figure shows a flowchart for calculating the output Opj of each unit. FIG. 7 shows the backward propagation amount δ of the error.
FIG. 8 is a flowchart for calculating pj, and FIG.
9 is a flowchart for determining the level of learning, FIG. 10 is a block diagram of a parallel processing system, FIG. 11 is a block diagram of a device that applies the learning result, and FIG. 13 is a block diagram of a system for learning the neurocomputer of the embodiment, FIG. 14 is a block diagram of an arrangement example of the photoelectric conversion elements of the embodiment, and FIG. FIGS. 16 (a) to 16 (c) are diagrams showing an example network, and are diagrams showing an example of a subject to be learned. 21 ... light receiving unit, 24, 27 ... arithmetic circuit, 25 ... neuro computer, 26 ... coefficient memory, 28 ... shutter control device, 29 ... aperture control device.

──────────────────────────────────────────────────続 き Continuation of the front page (72) Inventor Shinichi Kodama 2-43-2 Hatagaya, Shibuya-ku, Tokyo Inside Olympus Optical Co., Ltd. (56) References JP-A-1-204171 (JP, A) JP JP-A-2-33687 (JP, A) JP-A-2-81159 (JP, A) JP-A-2-83656 (JP, A) (58) Fields studied (Int. Cl. ⁶ , DB name) G03B 7 / 00-7/28 G06F 15/18

Claims (1)

(57) [Claims] A light receiving unit comprising a plurality of photoelectric conversion elements; an optical system for forming a subject image on the light receiving unit; a network connected to an output of the light receiving unit for outputting an exposure correction signal; Means for controlling the exposure of the camera based on the output of the network and the output of the network, the network comprising: an input layer consisting of a plurality of units connected to the output of the light receiving unit; and individual units of the input layer. A single-layer or multiple-layer intermediate layer composed of a plurality of units coupled with a predetermined coupling strength, and an output composed of a plurality of units coupled with the individual units of the intermediate layer with a predetermined coupling strength Exposure control device comprising a layer.