JP2954769B2 - Learning type video camera - Google Patents

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a video camera, and more particularly to an intelligent learning type video camera having a learning ability.

[0002]

2. Description of the Related Art A video camera or a camera-integrated VT
With the rapid spread of R, the number of users has also increased, and a user-friendly video camera that is easy to operate and that can capture beautiful images has been demanded. For this reason,
Various automatic functions such as auto white balance and auto focus are mounted on the video camera. Such an automatic function is described in, for example, Television Society Report Vol. 14, No. 16 (1990); pp. 13-18, Proceedings of the Annual Conference of the Institute of Television Engineers of Japan (1
990); pages 21-23.

[0003] In addition, video works are also generally made, and therefore, a function for freely expressing images has been required. An example of such a function is a professional AE system using a variable electronic shutter. In this system, the depth of field can be selected like a single-lens reflex camera, so that only the subject is focused and the background is blurred, a portlet effect is obtained, or the depth of field is increased. Focus from a near subject to a distant subject. The professional AE in a video camera is described in Television Society Report Vol. 14,
No. 16 (1990); pp. 1-16.

[0004]

In the current video camera, although the automatic functions such as the white balance described above have been improved to some extent,
A novice user will be able to use video expression and video-adding effects to the same degree as if a veteran user who was proficient in photography technology manually adjusted and photographed according to the subject being photographed and the surrounding conditions. It was difficult to improve. And, in order for anyone to perform advanced shooting like a professional photographer, it is necessary to improve the performance of the automatic function of the camera to such a level, but for this purpose the camera It is necessary to have a function of judging a subject, and to learn judgment criteria and the like in order to improve the judging ability, but such a video camera did not exist conventionally.

For free image expression, a specific subject is extracted, and the extracted subject portion and the other portions are individually controlled to portray the background of the subject or to make the subject beautiful. Although functions such as correcting the color of the subject to show it can be realized, the ability to extract the subject is limited.

SUMMARY OF THE INVENTION It is an object of the present invention to provide a video camera with learning ability to improve the ability to determine a shooting environment and a shooting subject.

[0007]

The above object can be achieved by the following means. An imaging device, a signal processing circuit for converting an output signal of the imaging device into a television signal, learning type control means including, for example, a microcomputer and an EEPROM (an example of an electrically rewritable nonvolatile memory), and a signal processing circuit. A determination means for determining a shooting environment based on a signal to be output and information provided by learning type control means, an input means for inputting a signal prompting learning from outside, and an output means for outputting a determination result, Based on the input of the learning signal, the learning type control means learns the judgment parameters of the judgment means,
Configure the video camera so that new parameters can be written to the EEPROM.

The above-mentioned object can also be achieved by the following means. An image sensor, a signal processing circuit for converting an output signal thereof into a television signal, and a microcomputer and an EE, for example.
Learning type control means composed of a PROM, subject extraction means for extracting a specific subject from a signal output from the signal processing circuit, and a signal output from the signal processing circuit and information provided by the learning type control means. A learning type control unit configured to determine a shooting environment or a type of a subject extracted by an extraction unit, an input unit for inputting a learning signal from outside, and an output unit for outputting a determination result. The means learns the parameters of the judgment by the judgment means and configures the video camera so that new parameters can be written to the EEPROM.

Further, the above-mentioned object can be achieved by the following means. An image sensor, a signal processing circuit for converting an output signal thereof to a television signal, learning type control means including, for example, a microcomputer and an EEPROM, and a subject extracting means for extracting a specific subject from a signal output from the signal processing circuit Determining means for determining the shooting environment or the type of subject extracted by the extracting means based on a signal output from the signal processing circuit and information provided by the learning type control means; and input means for inputting a learning signal from outside , An output means for outputting a judgment result, a mode selection means for selecting a mode, a reset means, and a removable external storage means to constitute a video camera.

[0010]

The image pickup device photoelectrically converts an optical signal formed on the light receiving surface, and the signal processing circuit generates a television signal from the output signal of the image pickup device. The subject extracting means extracts a subject such as a person from a video signal output from the signal processing circuit,
This extraction result is output to the learning type control means. The determining means determines the shooting environment or the type of the subject using the output signal of the signal processing circuit and the information given from the learning type control means, and outputs the result of the determination to the learning type control means. Then, the learning type control means controls the parameters of the processing of the signal processing circuit so as to be optimal for the photographing environment and the subject according to the judgment result.

The input means is a means by which a user inputs an answer to the result determined by the determining means. The input signal from the input means allows the learning type control means to correct the determination result by the determining means. Recognize Then, the learning type control means controls the judgment parameter so that the judgment result approaches a correct value, and records the new parameter in the EEPROM. By repeating such learning, the judgment parameters become optimal for the user, and the judgment ability of the judgment means is improved. The output means outputs the result of the judgment by the appropriate means such as video and audio such as characters / symbols, and the user can confirm the result of the judgment.

The mode selecting means activates the above-mentioned learning function only when the learning mode is selected. On the other hand, when a reset signal is inputted by the resetting means, learning is performed and stored in the EEPROM. All the learning results are cleared and returned to the initial setting state.

The above-described learning result is transferred and recorded to the removable external recording means as necessary, whereby the learning result can be stored outside the apparatus main body. By having the recording means individually, it is possible to obtain an image suitable for personal preference by automatic photographing. Alternatively, by taking in data from external recording means that has learned the photographing technique of a professional photographer, it becomes possible for a general user to take a picture with a photographing technique comparable to that of a professional photographer.

[0014]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below with reference to the illustrated embodiments. FIG. 1 is a block diagram showing a configuration of a video camera having a learning function according to a first embodiment of the present invention. As shown in FIG. 1, the video camera according to the present embodiment includes a lens 6, an aperture 7, an image sensor 1, a signal processing circuit 2, a determination unit 8, a microcomputer 12, and an EEPROM (an example of an electrically rewritable nonvolatile memory). The learning type control means 3, input means 5, and output means 4 are provided.

The image sensor 1 photoelectrically converts the optical signal passing through the lens 6 and the aperture 7 to generate and output a photoelectrically converted signal 100. The signal processing circuit 2 generates a luminance signal and a color difference signal from the photoelectric conversion signal 100, generates a signal 101 such as an average value of the luminance signal and the color difference signal from the generated signals, and outputs the signal 101 to the determination unit 8. On the other hand, known photoelectric processing such as gamma processing is performed on the photoelectric conversion signal 100 to generate a television signal 105 and output it to the output means 4.

The judging means 8 operates according to a predetermined judging algorithm.
Based, a signal 101, such as average or maximum value of the luminance signal and color difference signals obtained from the luminance signal and color difference signals, aperture value provided by the microcomputer 12 and countrymen - spin - the current photographing control condition information 110, such as de , Judgment parameter information
Information and shooting environment type such as shooting location and lighting type
(E.g. indoors with spotlights, etc.)
The type of the shadow environment is determined) , and the determination result 10
2 is output to the microcomputer 12. The criterion of the judgment means 8 is the judgment control signal 10 generated by the microcomputer 12.
3 is controlled. That is, the judgment control signal (judgment parameter
The data information 103 is generated by the microcomputer 12 based on the judgment parameters 109 read for each judgment item from the EEPROM 13 and sent to the judgment means 8. In addition, EEPROM1
The contents of No. 3 are rewritten each time learning is performed as described later.

The microcomputer 12 adjusts the white balance, the exposure state, the enhancement amount, and the like based on the judgment result 102 obtained by the judgment means 8 according to the determined photographing environment.
Generate a control signal 104 for controlling the operation to be appropriate ;
This is output to the signal processing circuit 2. The microcomputer 12
Encodes the judgment result 102 and encodes the judgment result 106
Is output to the output means 4. The output means 4 converts the encoding determination result 106 into characters, symbols, and the like, and superimposes the result on the television signal 105.
08 is output, and the determination result is displayed on an electronic viewfinder (not shown) of the video camera, for example. The output unit 4 may output the determination result by video other than characters and symbols, and in some cases, by voice. The input means 5 is a means by which a user inputs an answer to the determination result 102 from outside, and in this embodiment, is constituted by a push button switch.
The answer signal 107 output from the input means 5 is input to the microcomputer 12.

Next, the structure and operation of the judgment means 8 will be described with reference to FIGS. FIG. 2 shows the judgment means 8 of FIG.
FIG. 3 is a flowchart showing a processing flow by the judging means 8. As shown in FIG. 2, the judging means 8 outputs a signal 10 of the average value or the maximum value of the luminance signal / color difference signal output from the signal processing circuit 2.
1 and a switch 56 for distributing information 110 such as an aperture value and a shutter speed provided by the microcomputer 12 to a data input terminal of the comparison circuit 52, and a comparison for individually judging a photographing environment based on these information. Circuits 52-1 and 52-
, 52-n and multipliers 53-1, 53-2,.
53-n and a determination circuit 54 for comprehensively determining the shooting environment.

In the above-described configuration, information 101 and 110 provided from the signal processing circuit 2 and the microcomputer 12 are input as input data D1 to Dn to the respective comparison circuits 52-1 to 52-n via the switch 56. You. Then, the comparison circuit 52-1 compares the input data D1 with the comparison condition value (threshold value) Th11, and outputs a value 1 (true) if the input data D1 exceeds the comparison condition value Th11. Otherwise, a comparison result X1 having a value of 0 (false) is output. The comparison circuit 52-2 compares the input data D2 with the comparison condition value Th12, and the input data D2 is compared with the comparison condition value Th1.
Outputs a value of 1 (true) if it exceeds 2, and a value of 0 otherwise
The comparison result X2 is output as (false). Similarly, the comparison circuit 52-n compares the input data Dn with the comparison condition value Th1n, outputs a value 1 (true) if the input data Dn exceeds the comparison condition value Th1n, and outputs a value 1 otherwise. The comparison result Xn that is set to 0 (false) is output (step ST1 in FIG. 5).

The comparison results X1 to Xn are calculated by each multiplier 53
Weights W1 to Wn are assigned (multiplied) by −1 to 53-n, respectively, and weighted comparison results Y1 to Yn are obtained (step ST2 in FIG. 5). Then, the judgment circuit 54
Sum of these weighted comparison results Y1 to YnΣ (Yn)
Exceeds the judgment condition value (threshold value) Th2, the value 1
(True), otherwise the value 0 (false) is output as the above-mentioned determination result 102 (step ST3 in FIG. 5).
That is, the determination circuit 54 of the determination means 8 determines each determination item.
By making a comprehensive judgment taking all of them into consideration, a certain judgment condition value Th
2 (for example, a judgment setting value of “likeness of a certain shooting environment”)
Outputs a value 1 (true) or a value 0 (false) for
The "photography environment" is judged. Here, "Aru
"The likeness of the shooting environment"
If the value is 1 (true), it is determined that the room is indoors, and the value is 0 (false).
If so, it is determined that it is outdoors. Here, the explanation is simple
For the purpose of realization, the judgment circuit 54 and the judgment condition value Th2 are
However, if a plurality of these are provided, more complex species
Those skilled in the art will be able to judge another “likelihood”.
It is clear. Then, according to the determination result of the determination means 8,
As described above, the shooting control
Control conditions are controlled. Here, these comparison condition values Th
11 to Th1n, weights W1 to Wn, determination condition value Th2
Is the data transferred from the EEPROM 13. It is obvious to those skilled in the art that the judging means 8 having such a function can be constructed by hardware or software.

FIGS. 4 and 5 show examples of the learning algorithm.
This will be described with reference to the flowchart of FIG. FIG. 4 shows a processing flow in a scolding method of inputting a signal from the input means 5 when the judgment result 102 is wrong. In this scolding method, when a signal is input from the input unit 5, the microcomputer 12 determines that the above-described determination result 102 of the shooting environment is incorrect, and executes a process based on the determination. That is, whether or not a learning signal has arrived from the input means 5 in step ST11 of FIG.
Then, the process proceeds to step ST13, where it is determined that the determination result 102 is incorrect. Then, the process proceeds to step ST13.
If it is determined to be o, the determination result 102 in step ST18
Is recognized as correct, and this flow ends. In step ST13, the microcomputer 12 determines the determination result 102
A flag is set to start the change processing of the weight (Wn) so that is closer to the correct value, and the process proceeds to step ST14. In step ST14, it is determined whether or not the sum Σ (Yn) of the weighted comparison results Y1 to Yn is equal to or smaller than the determination condition value (threshold) th2.
The process proceeds to step ST16 if no. Step S
At T15 (when the sum of the weighted comparison results Σ (Yn) is equal to or less than the determination condition value th2), the weight Wn of the comparison result Xn of value 1 (true) is increased by 1, and the weight Wn of the value 0 (false) is increased.
Is to keep the value or decrease it by 1,
The weight Wn is increased so that (Yn) exceeds the determination condition value th2, and the process proceeds to step ST17. Step S
At T16 (when the sum of the weighted comparison results Σ (Yn) exceeds the determination condition value th2), the comparison result X
The weight Wn having a value of 1 (true) is decreased by 1 and the weight Wn of 0 (false) is maintained at a value or increased by 1 so that the sum Σ (Yn) becomes equal to or less than a determination condition value th2. Then, the process proceeds to step ST17. In step ST17, each new weight Wn changed in step ST15 or ST16 is stored in the EEPROM1.
3 and the series of processing ends.

FIG. 5 shows a processing flow in a method of complimenting a signal input from the input means 5 when the judgment result 102 is correct. In this praise method, when a signal is input from the input unit 5, the microcomputer 12 determines that the determination result 102 of the shooting environment is correct, and executes a process based on the determination result. That is, it is asked in step 21 of FIG. 5 whether a learning signal has arrived from the input means. If yes, the process proceeds to step ST22, in which it is determined that the determination result is correct, and then the process proceeds to step ST23.
If it is determined to be o, the determination result 102 in step ST28
Is recognized as an error (it has been instructed that the change of the weight (Wn) is unnecessary), and this flow is ended. In step ST23, the microcomputer 12 determines the weight (Wn) for setting the determination result 102 to a value that is more likely to be determined to be correct.
Is set, and the process proceeds to step ST24. In step ST24, it is asked whether the sum Σ (Yn) of the weighted comparison results Y1 to Yn is equal to or smaller than the determination condition value (threshold) th2. If yes, the process proceeds to step ST25; Proceed to ST26. In step ST25, (sum of weighted comparison results Σ
(When (Yn) is equal to or less than the judgment condition value th2), the weight Wn of which the comparison result Xn is 1 (true) is reduced by 1 and
The value of the (false) weight Wn is held or increased by one, and the weight Wn is reduced so that the sum Σ (Yn) exceeds the determination condition value th2, and the process proceeds to step ST27. In step ST26 (when the sum of the weighted comparison results Σ (Yn) exceeds the determination condition value th2), the weight Wn whose comparison result Xn is 1 (true) is increased by 1 and the value 0 (false). The value of the weight 値 (Yn) is kept or reduced by 1 so that the sum Σ (Yn) is equal to the determination condition value th2.
The weight Wn is increased as follows, and the process proceeds to step ST27. In step ST27, step S
The new weights Wn changed in T25 or in step ST26 are written in the EEPROM 13, and a series of processing ends.

The above-described learning is performed when the photographing environment changes and the judgment result 102 changes. The learning described above is realized by software of the microcomputer 12 by a learning processing program created in advance. Further, the learning method of FIG. 4 or FIG. 5 can be arbitrarily selected by the user.

In this embodiment, by repeating such learning, the ability to judge the shooting environment is improved, and automatic control such as white balance and exposure control more suitable for the environment can be performed. High performance is realized.

Next, a second embodiment of the present invention will be described with reference to FIG. FIG. 6 is a block diagram showing the configuration of a video camera having a learning function in which a subject extracting means 9 is added to the configuration of the first embodiment of FIG. In this figure, components equivalent to those in FIG. 1 are denoted by the same reference numerals, and the operation of each block is almost the same as in the first embodiment of FIG. 1, and therefore is different from that of the first embodiment to avoid duplication. moiety, i.e., the object extraction unit 9, the decision means 4, the learning type controller 3, the input means 5 _1,
5 ₂ only will be described.

The subject extracting means 9 comprises a signal processing circuit 2
A specific subject (for example, a person) is extracted from the television signal 105 output from the
Is output to the microcomputer 12. For example, subject extracting means 9
Is the center of the screen based on the predetermined subject extraction algorithm.
Processing for binarizing nearby color information using a predetermined threshold
To identify and extract the subject,
The result 200 is output to the microcomputer 12. The microcomputer 12
Subject information 203 such as the color, shape, size, and position of the subject extracted based on the subject extraction result 200 is generated, and is output to the determination means 8. The judging means 4 has a predetermined judgment
Based on the algorithm, the subject information 203, data 101 such as a luminance signal and a color difference signal output from the signal processing circuit 2 and judgment parameter information from the microcomputer 12 are output.
And the type of the extracted subject (for example, a person,
The type of the subject such as the sea and the mountain) is determined, and the determination result 102 is output to the microcomputer 12. Then, the microcomputer 12 generates a control signal 104 for controlling the white balance, the exposure state, the enhancement amount and the like to be optimal for the subject based on the sent determination result 102, and outputs the control signal 104 to the signal processing circuit 2. I do.

Further, in this second embodiment, is to have two input means 5 _1, 5 _2, for example from the input means 5 ₁ when there is an input of the learning signal 107 ₁ scolding described above scheme, such as the compliment system when the input means 5 ₂ has been input learning signal 107 ₂ is configured so as to be able to learn in selectively two learning method. Note that the configuration and operation of the determination means 4 and the learning algorithm are the same as those described above, and thus description thereof is omitted here.

Also in the second embodiment, the video camera repeats learning, thereby improving the ability to determine what the subject is, and realizing high performance of the automatic control function as in the first embodiment. .

Next, a third embodiment of the present invention will be described with reference to FIG. FIG. 7 is a block diagram showing a configuration of a video camera having a learning function in which a mode selection unit 30, a reset unit 31, and a removable external memory 32 are added to the configuration of the first embodiment of FIG. Also in this figure, components equivalent to those in FIG. 1 are denoted by the same reference numerals, and the operation of each block is almost the same as in the first embodiment in FIG. Only the different parts, that is, the mode selection means 30, the reset means 31, and the external memory 32 will be described.

The mode selecting means 30 is a means for selecting an operation mode. The video camera performs learning only when the learning mode is selected by the mode selecting means 30. When the mode is not selected, the video camera performs normal shooting. The reset means 31 is a means for inputting a reset signal. When a reset signal is input by the reset means 31, all the learning results stored so far in the EEPROM 13 are cleared and the EEPROM 13 is reset. Is returned to the initial setting state. The external memory 32 is configured by a removable external storage medium such as an IC card, for example.
The learning result stored in the EEPROM 13 is transferred and stored in the external memory 32, and the user can individually store the learning result as individual learning result data, for example. Then, the learning result data stored in the external memory 32 is read into the video camera as needed, and the photographing and recording control according to the preference is performed based on the data.

As described above, in the third embodiment, since the learning result can be stored in the external memory 32, it is possible to perform control according to personal preference by using the personal-use external memory 32. Thus, a video camera corresponding to each user can be realized. Also, an external memory 32 storing advanced learning results by a professional photographer or the like.
With the use of a video camera, a general user can realize a video camera capable of performing advanced shooting by automatic shooting.

[0032]

As described above, according to the present invention, in a video camera, the camera itself learns the judgment parameters for judging the shooting environment or the type of the subject, so that the judgment ability is improved. Thus, it is possible to realize an intelligent camera capable of performing advanced shooting by automatic shooting. Further, by accumulating the learning results, it is possible to realize a video camera in which the judging ability grows, and to give a user the joy of growing the camera.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a main configuration of a learning video camera according to a first embodiment of the present invention.

FIG. 2 is a block diagram illustrating a configuration of a determination unit in FIG. 1;

FIG. 3 is a flowchart illustrating a processing flow by a determination unit in FIG. 1;

FIG. 4 is a flowchart illustrating an example of a learning algorithm according to the first embodiment of the present invention.

FIG. 5 is a flowchart showing another example of the learning algorithm according to the first embodiment of the present invention.

FIG. 6 is a block diagram showing a main configuration of a learning video camera according to a second embodiment of the present invention.

FIG. 7 is a block diagram showing a main configuration of a learning video camera according to a third embodiment of the present invention.

[Explanation of symbols]

First image sensor 2 signal processing circuit 3 learning type controller 4 output means 5, 5 _1, 5 ₂ input section 6 lens 7 diaphragm 8 determination means 9 subject extraction unit 12 microcomputer 13 EEPROM 30 mode - de setting means 31 reset means 32 outside memory

──────────────────────────────────────────────────続 き Continuing from the front page (72) Inventor Akihito Nishizawa 292 Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa Prefecture Inside Hitachi, Ltd. Video Media Research Laboratory (72) Inventor Takuya Imade 292, Yoshida-cho, Totsuka-ku, Yokohama-shi, Kanagawa Address: Hitachi, Ltd. Visual Media Research Laboratory (58) Field surveyed (Int.Cl. ⁶ , DB name) H04N 5/232 H04N 5/225

Claims (5)

(57) [Claims] And 1. A imaging device 1, a signal processing circuit 2 for converting processing the output signal of the image sensor 1, a control unit 12 for controlling the signal processing circuit 2, to output the signal processing circuit 2
Extracting means 9 for extracting a specific subject from a signal
The type of environment or the type of subject extracted by the extraction means 9
Determining means 8 for determining at least one of the types ;
Parameters used in the determination by said determining means 8 - a storage means 13 for storing data, an output unit 4 for outputting a determination result of said determining means 8, for inputting an answer to the determination result of said determination means 8 from the outside The extracting means 9 is based on an object extracting algorithm.
From the signal output from the signal processing circuit 2,
Identifying the body and outputting the subject extraction information, the determination means is based on a predetermined determination algorithm,
Of the luminance signal / color difference signal output from the signal processing circuit.
It is obtained by appropriately processing the imaging signal such as the average value and the maximum value.
Signal and the squeeze value or shutter output from the control means.
Current shooting control condition information such as
Using the judgment parameter information stored in the storage means 13
To determine the type of shooting environment, such as shooting location and lighting type
Or, based on a predetermined judgment algorithm,
Object extraction information output from the extraction means 9 and the signal
A luminance signal / color difference signal output from the signal processing circuit;
Using the judgment parameter information stored in the storage means 13
The type of the subject such as a person is determined, the result of the determination by the determining means is learned from the input signal from the input means 5 by the operator, and the learning result is stored in the storage means 13. Learning type video camera characterized by the following. 2. The method according to claim 1, wherein said determining means comprises:
When the judgment result is wrong, an electric signal is inputted by the input means 5 or when the judgment result of the judgment means 8 is correct, the electric signal is inputted by the input means 5 so that the learning is performed. Learning type video camera characterized by the following. 3. The input means 5 according to claim 1, wherein
Is a learning type video camera comprising one or more push-button switches, which are used to input an answer signal. And wherein the imaging device 1, a signal processing circuit 2 for converting processing the output signal of the image sensor 1, a control unit 12 for controlling the signal processing circuit 2, a signal that the signal processing circuit 2 outputs the extraction means 9 for extracting a specific object, shooting
The type of environment or the type of subject extracted by the extraction means 9
Determining means 8 for determining at least one of the types ;
Parameters used in the determination by said determining means 8 - a storage means 13 for storing data, an output unit 4 for outputting a determination result of said determining means 8, for inputting an answer to the determination result of said determination means 8 from the outside It comprises input means 5, mode selecting means 30 for selecting a mode, and resetting means 31, and said extracting means 9 is based on a subject extracting algorithm.
From the signal output from the signal processing circuit 2,
Identifying the body and outputting the subject extraction information, the determination means is based on a predetermined determination algorithm,
Of the luminance signal / color difference signal output from the signal processing circuit.
It is obtained by appropriately processing the imaging signal such as the average value and the maximum value.
Signal and the squeeze value or shutter output from the control means.
Current shooting control condition information such as
Using the judgment parameter information stored in the storage means 13
To determine the type of shooting environment, such as shooting location and lighting type
Or, based on a predetermined judgment algorithm,
Object extraction information output from the extraction means 9 and the signal
A luminance signal / color difference signal output from the signal processing circuit;
Using the judgment parameter information stored in the storage means 13
The type of the subject such as a person is determined, and only when the learning mode is selected by the mode selecting means 30, the result of the determination by the determining means is determined by the operator.
Learning by an input signal from the input means 5
The learning result is stored in the storage means 13, and when a reset signal is inputted by the reset means , the storage result is stored.
A learning-type video camera, wherein the learning result of the means 13 is reset to return to an initial state. And wherein the imaging device 1, a signal processing circuit 2 for converting processing the output signal of the image sensor 1, a control unit 12 for controlling the signal processing circuit 2, a signal that the signal processing circuit 2 outputs the extraction means 9 for extracting a specific object, shooting
The type of environment or the type of subject extracted by the extraction means 9
Determining means 8 for determining at least one of the types ;
Parameters used in the determination by said determining means 8 - a storage means 13 for storing data, an output unit 4 for outputting a determination result of said determining means 8, for inputting an answer to the determination result of said determination means 8 from the outside The extracting means 9 comprises an input means 5 and a removable external storage means 32, wherein the extracting means 9 is based on a subject extracting algorithm.
From the signal output from the signal processing circuit 2,
Identifying the body and outputting the subject extraction information, the determination means is based on a predetermined determination algorithm,
Of the luminance signal / color difference signal output from the signal processing circuit.
It is obtained by appropriately processing the imaging signal such as the average value and the maximum value.
Signal and the squeeze value or shutter output from the control means.
Current shooting control condition information such as
Using the judgment parameter information stored in the storage means 13
To determine the type of shooting environment, such as shooting location and lighting type
Or, based on a predetermined judgment algorithm,
Object extraction information output from the extraction means 9 and the signal
A luminance signal / color difference signal output from the signal processing circuit;
Using the judgment parameter information stored in the storage means 13
To determine the type of the subject such as a person , learn the result of the determination by the input signal from the input means 5 by the operator, and write the learning result to the external storage means 32, or A learning type video camera wherein data previously written in the external storage means 32 is taken in.