JPH1011481A - Design device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain plural inference results having varied meanings while maintaining a single image by changing the characters which are extracted from the design elements of a basic design via a sorting data base based on a prescribed standard. SOLUTION: When a user inputs an evaluation term via a keyboard 1, an inference means 17 retrieves the inputted term out of an adjective data base 11, for example. Then a category having the largest partial regression coefficient is extracted among all items to the input evaluation term via an image data base 13. That is, every category is extracted from the image data on the evaluation term, and these categories construct a basic design. Then the items where the extracted categories belong are sorted into proportions and characters by means of a proportion/character data base 15. Only these characters can be changed and therefore the inferences of various meanings is carried out.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a design apparatus for displaying a design diagram or the like corresponding to an inputted evaluation term on a display device such as a CRT.

[0002]

2. Description of the Related Art As a conventional design apparatus of this kind,
For example, JP-A-4-165144 and JP-A-1-34
For example, there is one disclosed in Japanese Patent No. 1129.

[0003]

However, although these conventional design devices make inferences accurately corresponding to the input evaluation terms, they output one optimal inference result. There is a problem that a plurality of inference results suitable for the evaluation term and having different flavors (tastes) cannot be output simultaneously. For example,
When developing a “calm” impression sedan, the “calm” impression skeleton is moderately “luxury” while sharing
The above problem occurs when it is desired to set two types: a type and a moderately “sporty” type.

Accordingly, an object of the present invention is to provide a design apparatus capable of outputting a plurality of inference results with different flavors while maintaining one image.

[0005]

According to a first aspect of the present invention, there is provided an input means capable of inputting an evaluation term and an instruction for expressing an image of an entire design object, and an input means for inputting the evaluation term. An adjective database that defines the relationship between the images, an image database that includes image data for each evaluation term that defines the relationship between the design elements that constitute the design and the evaluation terms, and the image database that is input when the evaluation terms are input. Is used to infer a basic design suitable for the input evaluation term, and a plurality of seasoning evaluation terms to be seasoned by partially modifying the basic design are selected automatically or by designating the impression of the seasoning evaluation term. Inference means for inferring each flavored inference, display means capable of displaying the result of the inference, and inference result on the display means A design database comprising a classification database for classifying and storing the design elements into proportions that greatly affect the change in the image of the design and characters that have little effect, and the inference means comprising: In performing the seasoning inference, a character is extracted from the design elements of the basic design using the classification database, and the seasoning is inferred by changing the character according to a predetermined standard.

Therefore, in addition to inferring a basic design suitable for input evaluation terms, a classification database is provided, so that a plurality of design diagrams having different flavors can be inferred by taking into account the impression of each of the seasoning evaluation terms. As a result, the range of design drawing selection is expanded, and a design drawing more suited to the image of the user can be obtained more quickly.

According to a second aspect of the present invention, in the design apparatus according to the first aspect, the inference means uses an image database to input among basic design characters when changing the characters. Those whose partial correlation coefficient with respect to the evaluation term exceeds a predetermined value are extracted as change candidates, and those of the extracted candidate characters whose partial correlation coefficient with respect to the seasoning evaluation term is below the predetermined value are set as final change targets. It is characterized in that seasoning inference is performed by extracting and changing.

Therefore, the operation according to the first aspect of the present invention
In addition to the effects, inferring means, when performing seasoning inference, among the inferred basic design characters, a character having a relatively strong correlation with the input evaluation term is set as a change candidate for seasoning, and the impression of the input evaluation term is made. The seasoning inference is made so that the seasoning evaluation terms of the candidate change characters are finally changed for the seasoning to have relatively weak correlation, so that the seasoning is not overemphasized. It is possible to obtain a plurality of design drawings that are appropriately seasoned.

According to a third aspect of the present invention, there is provided the design apparatus according to the first aspect, further comprising a case image range database for defining image ranges of a plurality of existing design cases, wherein the inference means includes the evaluation term. And the case name of the existing design are input, and the case image range database is used to search whether or not the input evaluation term is within the image range of the existing design case. In inferring, a plurality of evaluation terms weakly correlated with the input evaluation terms within the image range are selected as seasoning evaluation terms, and seasoning inference is performed in consideration of the impression of the selected seasoning evaluation terms. I do.

Therefore, the operation according to the first aspect of the present invention
In addition to the effect, since the case image range database is provided, the inference means performs a plurality of seasoning inferences by changing the image range of the designated existing design case. As a result, a seasoning inference that deviates from the image range expected by the user is not performed, so that a design diagram that meets the user's expectations can be obtained more quickly.

According to a fourth aspect of the present invention, there is provided the design apparatus according to the first aspect, further comprising a parts image database for storing, for each evaluation term, information on commercially available parts as components of a unit constituting a design object. The inference means includes: in performing a seasoning inference, using a parts image database relating to a specified evaluation term for seasoning, using the part image database for the evaluation term for seasoning according to the strength of the correlation between the input evaluation term and the evaluation term for seasoning. The present invention is characterized in that a commercially available part is selected in consideration of the degree of contribution of the commercially available part to the impression, and seasoning inference is performed.

Therefore, the operation according to the first aspect of the present invention
In addition to the effects, the inference means is provided with a parts image database, and when inferring the seasoning, when the correlation between the input evaluation terms and the evaluation terms for seasoning is strong, a commercially available part that contributes a high degree to the impression of the evaluation terms for seasoning In the case where the correlation between the two is weak, it is possible to select a commercially available part with a low degree of contribution so that the seasoning is not emphasized too much. It is possible to adjust the seasoning.

According to a fifth aspect of the present invention, there is provided the design apparatus according to the first aspect, wherein the evaluation terms are divided into a higher-level evaluation term and a lower-level evaluation term based on the strength of correlation between the evaluation terms. An adjective tree structure database that is hierarchically classified and stored, wherein the inference means performs seasoning inference for each of the lower evaluation terms with respect to the input evaluation terms using the adjective tree structure database in performing the seasoning inference. It is characterized by.

Therefore, the operation according to the first aspect of the present invention
In addition to the effects, since the adjective tree structure database is provided, the inference means can automatically extract the seasoning evaluation terms and infer the seasoning even when the seasoning evaluation terms are not specified when performing the seasoning inference. .

The invention according to claim 6 is the design apparatus according to any one of claims 1 to 5, wherein the design object to be inferred is a vehicle interior.

Therefore, the operation and effect according to any one of the first to fifth aspects of the invention can be applied to the design of a vehicle interior.

[0017]

As is clear from the above description, according to the first aspect of the present invention, in addition to inferring a basic design suitable for input evaluation terms, a classification database is provided, so that evaluation for seasoning is performed. It is possible to infer a plurality of design drawings with different flavors taking into account the impression of each term. As a result, the range of design drawing selection is expanded, and a design drawing more suited to the image of the user can be obtained more quickly.

According to the invention of claim 2, according to claim 1,
In addition to the effects of the invention, the inference means infers the seasoning so as not to emphasize too much, so that a plurality of appropriately seasoned design diagrams can be obtained.

According to the invention of claim 3, according to claim 1,
In addition to the effects of the invention, the inference means performs a plurality of seasoning inferences based on a change in the image range of the designated existing design case, so that the seasoning does not deviate from the image range expected by the user. The design drawing that meets the expectation can be obtained faster.

According to the invention described in claim 4, according to claim 1,
In addition to the effects of the invention, the inference means selects a commercially available part having a high degree of contribution to the impression of the seasoning evaluation term when the correlation between the input evaluation term and the seasoning evaluation term is strong, and emphasizes the seasoning. If the correlation is weak, it is possible to select a commercially available part with a low degree of contribution so as not to overemphasize the seasoning, so that it is possible to make an inference using commercially available parts for the seasoning and adjust the seasoning.

According to the invention described in claim 5, according to claim 1,
In addition to the effects of the invention, the inference means can automatically extract the seasoning evaluation term and infer the seasoning even if the seasoning evaluation term is not specified.

According to the invention of claim 6, according to claim 1,
The effect according to any one of the inventions of (5) to (5) can be applied to the design of the vehicle interior.

[0023]

BEST MODE FOR CARRYING OUT THE INVENTION

[First Embodiment] A first embodiment of the present invention will be described with reference to FIGS. FIG. 1 shows a schematic configuration of a vehicle interior design device of the present embodiment, and FIGS. 2 to 9 are explanatory diagrams.

First, the configuration will be described. As shown in FIG. 1, the present design apparatus includes a keyboard 1 as input means for inputting evaluation terms (adjectives) and instructions representing an image of the entire vehicle interior (design object), and a computer 3.
And a CRT 5 as display means. A computer 3 comprising a memory 3a having a knowledge database 7;
And a CPU 3b.

The knowledge database 7 provided in the memory 3a is an image database 1 used for inference.
3 as well as memorizing adjectives expressing the impression of the vehicle interior, for example, “luxurious”, “spacious”,..., Etc., and, for example, as shown in FIG. The axis is provided with an adjective database 11 that defines the strength of correlation between evaluation terms, and a proportion / character database 15 as a classification database.

As shown in FIG. 2, the image database 13 stores the contribution (partial regression coefficient) of each category of the vehicle design element (item) to each evaluation term for each evaluation term. Here, the item is
As shown in FIG. 2, the number of meters, the connection between the instrument and the meter cluster, the connection between the meter cluster and the center console, the thickness of the door, and other components of the vehicle interior design are classified. The items focused on the number of meters are further subdivided into five categories, and partial regression coefficients are given for each category. Among the plurality of categories for each item corresponding to each evaluation term, the category having the maximum partial regression coefficient (the partial regression coefficient marked with * in FIG. 2) is the design element most suitable for the evaluation term. It is shown that.

As shown in FIG. 3, the proportion / character database 15 is a database in which the above-described design elements are classified into proportions and characters and stored. The design element belonging to this proportion is a design element that has such a great influence that changing it changes the skeleton of the entire design. On the other hand, the design element belonging to the character is a seasoning design element that can be changed independently since the influence of the change is small. In the example of FIG. 3, the proportions are classified according to the thickness, shape, depth, etc. of the instrument, and the characters are classified according to the number of spokes of the steering wheel, the material, etc.

On the other hand, when the evaluation terms representing the overall image of the design object are input from the keyboard 1, the CPU 3b infers means for inferring an appropriate design diagram and output means for displaying and outputting the inference result on the CRT 5. 19 is provided.

Next, the operation of the design apparatus will be described with reference to FIGS.
This will be described with reference to FIG. FIG. 4 shows a main flow of the inference.

First, in step S1, when the user inputs an evaluation term such as "calm" from the keyboard 1, the inference means 17 searches the adjective database 11 for "calm" (step S2). Then
Using the image database 13, the category in which the partial regression coefficient in each item for the input evaluation term is the maximum value is extracted. That is, each category surrounded by an elliptical frame is extracted from the image data of "calm" shown in FIG. 5 (step S3). These categories constitute the basic design.

Next, using the proportion / character database 15, the items belonging to the category extracted in step S3 are classified into proportions and characters. That is, in the example shown in FIG. 5, the “stance of the instrument”, the “width of the center console”,... Are classified as belonging to the proportion, and the “grain on the instrument”,
"Seat upholstery", "Number of steering spokes", ...
Are classified as belonging to the character (step S
4).

In the subsequent steps, only items classified into this character are treated as changeable.
Reasoning with various changes in seasoning is performed. That is, in steps S5 to Sn, while maintaining the impression of the input evaluation term “calm”, some other typical evaluation terms in the adjective database 11 (for example, “luxury”,. , Etc.) as seasoning evaluation terms, according to the user's selection, instruction, or automatically, and seasoning inference is sequentially performed for each.

In step S5, for example, as shown in FIG.
One evaluation axis (first
Assuming that the axis of luxury is taken as the axis) and the axis of sporty feeling is selected and instructed as the second axis, FIG.
Since the coordinates (diagonal circles) of the input evaluation term “calm” are plotted on a plane composed of the second axis, the method of adding a sense of quality while maintaining the “calm” impression is “calm”. For example, a seasoning located at a middle point (cross hatched circle) on a vector (broken line) extended from the coordinates of “ta” to the high-quality axis can be adopted. for that purpose,
As shown in the upper part of FIG. 7, in the image data of the input evaluation term “calm”, the items “grain on the instrument”, “the sheet upholstery” classified as belonging to the character in the previous step S4,. Are to be changed (step S5). The detailed flow in step S5 will be described later.

The second axis, for example, "sporty"
As in step S5, an inference with a sporty feeling is made. That is, in the image data of the input evaluation term “calm” shown in the upper part of FIG. 8, items “grain on instrument”, “stretched sheet”,... Classified as belonging to the character are to be changed ( Step S6).

In this way, the seasoning up to the n-th axis is sequentially performed (step Sn), and the design means inferred with each seasoning is displayed on the CRT 5 together with the basic design diagram (inference result) inferred in step S3 (step S3). Sn +
1) to end the flow.

Here, the detailed flow of the seasoning inference of the first axis in step S5 of FIG. 4 will be described with reference to FIG. The same applies to the flow of each seasoning inference from the second axis.

In step S5, when the seasoning inference of the first axis is started based on the input evaluation term "calm", first, for each item (the total number of items is n) in the image data of "calm", For “calm”, it is classified as either a proportion or a character depending on whether it belongs to a character or not (step S51). If the item belongs to the proportion, leave the item as it is and move on to the next item classification. But,
If the item belongs to the character (for example, “grain on instrument” in the upper part of FIG. 7), the process proceeds to step S52, and the partial correlation coefficient (upper part) of the item in the image data (upper part of FIG. 7) Is the predetermined value 0.5
Check if it exceeds. If not, step S51 is executed for the next item. If the partial correlation coefficient exceeds 0.5 (since the partial correlation coefficient is 0.51 in the case of “grain on instrument”, it exceeds 0.5)
Proceeding to the next step S53, it is determined whether the partial correlation coefficient of the item in the evaluation term of the first axis, for example, the image data of "luxury" (lower diagram in FIG. 7) is less than a predetermined value 0.6. Confirm. If it is less than 0.6 (in the case of “grain on instrument”, the partial correlation coefficient is 0.57), the category (in this case, “grained”) with the maximum partial regression coefficient for the item is set Select (Step S5
4).

If the value exceeds 0.6, the process returns to step S51 for the next item without changing the item.

When the processing has been completed until all the items (total number n) have been processed similarly (step S
55), the detailed flow of the seasoning of the first axis (high-grade) ends.

Next, the seasoning inference is repeated from the second axis to the n-th axis. In the seasoning inference on the second axis, for example, as shown in FIG. 8, when the evaluation term for seasoning on the second axis is “sporty”, the item “number of steering spokes” is in a phase relation with “sporty” The number is 0.8
Since it is 8 (because it exceeds 0.6), step S51 is executed for the next item without changing the "number of steering spokes" as in the above-described seasoning inference for the first axis.

The reason is "the number of steering spokes".
If the item is set to change the character whose seasoning evaluation term exceeds the predetermined value 0.6 of the partial correlation coefficient, the "number of steering spokes" is changed to the four items shown in the upper part of FIG. This is because the spokes are changed from the spokes to the three-spokes shown in the lower diagram, and the "sporty" seasoning is emphasized too much.

As described above, according to the present embodiment, after inferring a basic design suitable for an input evaluation term (for example, “calm”), the item belonging to the character in the basic design is changed to the input evaluation term “calm”. On the other hand, an item having a relatively strong correlation (for example, a partial correlation coefficient> 0.6) is extracted as a change candidate, and a direction for weakening the impression of “calm” is set. Items having relatively weak correlations (for example, partial correlation coefficient <0.6) with respect to each (from the first axis to the n-th axis) are finally changed to infer seasoning so as not to overemphasize the seasoning. . Therefore, a plurality of design drawings appropriately seasoned can be obtained, so that a design drawing more suitable for the image of the user can be obtained more quickly.

[Second Embodiment] A second embodiment of the present invention will be described with reference to FIGS. FIG. 10 is a flowchart illustrating the inference method of the design apparatus according to the present embodiment.

The present design apparatus includes a memory 23a in addition to the databases 11, 13, and 15 of the first embodiment.
Has a vehicle image range database (case image range database) 16 for existing vehicles.

The inference means 27 of the CPU 23b includes a judgment unit 27a described later in addition to the first embodiment. The other configuration is the same as that of the first embodiment, and a duplicate description will be omitted.

The above-mentioned vehicle image range database 16
Defines an area corresponding to the image of each vehicle in the evaluation space based on the evaluation terms, as shown in FIG. 11 by enclosing an example of “bluebird” with a hatched circle.

Returning to FIG. 10, first, at step S11
When the user inputs the evaluation term "calm" from the keyboard 1, the inference means 27 searches the adjective database 11 for "calm" (step S).
12). Next, when the user inputs the vehicle name "bluebird" (step S13), the inference means 27 searches the vehicle image range database 16 for a range corresponding to the image of "bluebird" (step S14).

Next, in step S15, the inference means 2
The determination unit 27a determines whether the input evaluation term “calm” is included in the range searched in step S14. If it is included (in the case of YES), a design diagram (basic design) suitable for “calm” is inferred using the image database 13 in step S16.

In step S17, the items (design elements) of the design drawing inferred in step S16 are classified into proportions and characters by using the proportion / character database 15 as in the first embodiment.

Next, in step S18, step S
In the image range searched in 14 (see FIG. 11), two points (or n points) are set for the "calm" point and the other points so that the distance between the points becomes maximum. FIG.
When the input evaluation term "calm" and "Bluebird" are entered as the car name, "calm" is included in the image range of "bluebird" including "calm"
And a state in which three black points are set, namely, Variation 1 and Variation 2.

In the following steps S19 and S20, using the image data (image database 13) of the evaluation term existing in the vector direction from the point of the input evaluation term “calm” to each of the variations 1 and 2, the first As in the embodiment, only the items belonging to the character are changed to infer seasoning design diagrams of variation 1 and variation 2. Then, step S21
Then, the design drawings of Variation 1 and Variation 2 are output to the CRT 5 together with the basic design drawings, and the flow ends.

On the other hand, in step S15, if the input evaluation term is not included in the image range of "bluebird" searched in step S14 (in the case of NO), that is, if the input evaluation term is For example, if it is “running image”, as shown in FIG. 12, the “running image” of the input evaluation term is not included in the image range of “bluebird”. Proceed to determine another evaluation term for the impression closest to the "running image" within the "bluebird" image range. Then, in step S23, a design diagram suitable for the determined evaluation term is inferred, and output to the CRT 5 in step S21 to end the flow.

Thus, according to the present embodiment, when the existing vehicle name “bluebird” is input together with the evaluation term “calm”, the inference means uses the vehicle image range database 16 to convert the seasoning evaluation term. Since the setting and the seasoning inference are performed, the seasoning inference deviating from the image range expected by the user is not performed, and the design diagram meeting the user's expectation can be obtained more quickly.

[Third Embodiment] A third embodiment of the present invention will be described with reference to FIGS. FIG. 13 is a flowchart illustrating an inference method of the design apparatus according to the present embodiment.

The present design apparatus includes a memory 33a in addition to the databases 11, 13, and 15 of the first embodiment.
Has a parts image database 18. Therefore, the flow of the seasoning inference is different from the first embodiment,
The other parts are the same as those of the first embodiment, and the duplicate description will be omitted.

The parts image database 18 is
As shown in FIG. 14, an example of the evaluation term “sporty” is that, as shown in FIG. 14, each of the commercially available components of the unit “steering”, “meter”,. This is a database in which the degree of contribution is stored together with its commercial price.

FIG. 13 is a flow chart of the inference of this design apparatus.
This will be described below. Here, the description is based on the assumption that the input evaluation term is “calm” and “sporty” is designated as the taste (taste) of the first axis.

The processing between steps S5 and S502 is the same as the processing between steps S5 and S52 in FIG. 9 of the first embodiment. That is, in step S502, the image database 1 of the input evaluation terms
In 3, it is confirmed whether the partial correlation coefficient of each item (design element) belonging to the character exceeds 0.5. For items whose partial correlation coefficient exceeds 0.5, the process proceeds to step S503.

In step S 503, the inference means 37 uses the parts image database 18 (FIG. 14) to determine a commercially available part to be changed so as to be appropriately “sporty” while maintaining the “calm” impression. As a method of determining, for example, the part image database 1
8 (FIG. 14) is determined according to the strength of the correlation between the input evaluation term “calm” and the seasoning evaluation term “sporty”, that is, the correlation between the two terms. If is strong, a commercially available part that has a high contribution to the impression of the seasoning evaluation terms is used. Conversely, if the correlation between the two terms is weak, seasoning is adopted by using a commercially available component with a low contribution so as not to emphasize the seasoning too much.

Now, the input evaluation term “calm” and the first
There is a strong correlation opposite to the term “sporty”, so do n’t overemphasize the sporty impression.
In FIG. 14, TYPE1 having a small contribution to the sporty impression (contribution 0.48) is determined. If the correlation between the input evaluation term “calm” and the seasoning term is positive and strong, the seasoning is emphasized more. T that greatly contributes to
There is a method of determining YPE3 (contribution 0.75).

The above steps are performed for all items (total number n).
, The price of the entire changed part is calculated in step S504, and the flow of the seasoning inference of the first axis is ended.

As described above, according to the present embodiment, the inference means 37 uses the parts image database 18 relating to the designated evaluation term for seasoning to perform the seasoning inference and calculates the correlation between the input evaluation term and the evaluation term for seasoning. If it is strong, select a commercially available part that emphasizes the impression of the evaluation terms for seasoning and emphasize the seasoning.On the contrary, if the correlation between the two is weak, select a commercially available part that does not overemphasize the seasoning. It is possible to make inferences that use commercially available parts for seasoning and clarify their prices, and to adjust seasoning.

[Fourth Embodiment] A fourth embodiment of the present invention will be described with reference to FIGS. FIG. 15 is a flowchart illustrating an inference method of the design apparatus according to the present embodiment.

In the present design apparatus, the memory 43a has an adjective tree structure database in addition to the first embodiment. Therefore, this difference will be mainly described, and redundant description will be omitted.

The adjective tree structure database 20
As shown in FIG. 16, is a database in which evaluation terms are classified and stored in upper and lower layers based on correlation values (or other indices indicating similarity).

FIG. 15 is a flow chart of the inference of this design apparatus.
This will be described below.

In step S 101, when the input evaluation term “calm” is input, the inference means 47 uses the adjective tree structure database 20 to select an appropriate lower adjective for the upper calm adjective “calm”. For example, “elegant”, “adult”, “formal”, and “relaxable” (see FIG. 16) are selected as lower adjectives (step S102).

Then, in step S103, using the image database 13, step S3 of the first embodiment is performed.
A basic design diagram is inferred in the same manner as (FIG. 4).
Next, in step S104, the items of the basic design drawing are classified into proportions and characters using the proportion / character database 15.

Next, steps S105 to S1
0n (steps S105 to S108), the inference for seasoning each of the lower adjectives selected in the previous step S102 is repeated, for example, similarly to steps S5 to Sn in FIG. The output means 19 outputs each seasoned design drawing to the CRT 5
Is output (step S10n + 1), and the flow ends.

Thus, according to the present embodiment, the inference means 47 automatically uses the adjective tree structure database 20 to perform seasoning inference for each of the lower evaluation terms for the input evaluation terms when performing seasoning inference. Therefore, even if a seasoning evaluation term is not specified, a seasoning evaluation term can be automatically extracted and seasoning inference can be performed.

In each of the above embodiments, the design apparatus for the vehicle interior has been described. However, these design apparatuses are not limited to the design of the vehicle interior, but can be applied to the design of the vehicle exterior. For example, by replacing each item (design element) of the proportion / character database 15 shown in FIG. 3 with a vehicle exterior item, it is possible to make a seasoning inference about the vehicle exterior design.

[Brief description of the drawings]

FIG. 1 is a schematic configuration diagram of a first embodiment.

FIG. 2 is an explanatory diagram illustrating an image database according to the first embodiment.

FIG. 3 is an explanatory diagram showing a proportion / character database according to the first embodiment.

FIG. 4 is a main flowchart of the first embodiment.

FIG. 5 is an explanatory diagram of the first embodiment.

FIG. 6 is an explanatory diagram of the first embodiment.

FIG. 7 is an explanatory diagram of the first embodiment.

FIG. 8 is an explanatory diagram of the first embodiment.

FIG. 9 is a detailed flowchart of the first embodiment.

FIG. 10 is a flowchart of the second embodiment.

FIG. 11 is an explanatory diagram of the second embodiment.

FIG. 12 is an explanatory diagram of the second embodiment.

FIG. 13 is a flowchart of the third embodiment.

FIG. 14 is an explanatory diagram showing a part image database according to the third embodiment.

FIG. 15 is a flowchart of the fourth embodiment.

FIG. 16 is an explanatory diagram showing a tree structure database of adjectives according to the fourth embodiment.

[Explanation of symbols]

Reference Signs List 1 keyboard (input means) 5 CRT (display means) 11 adjective database 13 image database 15 proportion / character database (classification database) 16 vehicle image range database (case image range database) 17, 27, 37, 47 inference means 18 part image Database 19 output means 20 adjective tree structure database

──────────────────────────────────────────────────続 き Continued on the front page (51) Int.Cl. ⁶ Identification code Agency reference number FI Technical display location G06F 15/419 310

Claims (6)

[Claims] 1. An input means capable of inputting an evaluation term and an instruction representing an image of an entire design object, an adjective database defining a relationship between the evaluation terms, a design element constituting a design, and the evaluation terms And an image database having image data for each evaluation term that defines the relationship between the evaluation terms. When the evaluation terms are input, a basic design suitable for the input evaluation terms is inferred using the image database, and the basic design is Inference means for automatically or by designating a plurality of seasoning evaluation terms to be changed and seasoned, and inferring each seasoning in consideration of the impression of the seasoning evaluation terms, and display means capable of displaying a result of the inference A design device comprising: an output unit that outputs an inference result to the display unit. A classification database that classifies and stores proportions that greatly affect the change in the image of the design and characters that have a small effect.The inference means uses the classification database to perform basic seasoning based on the classification database when performing the seasoning inference. A design apparatus for extracting a character from design elements and changing the character according to a predetermined standard to perform seasoning inference. 2. The design apparatus according to claim 1, wherein the inference means uses an image database to change the character, and uses an image database to change a partial correlation coefficient of a basic design character with respect to an input evaluation term. Is extracted as a change candidate, and the extracted candidate characters whose partial correlation coefficient is less than a predetermined value with respect to the evaluation term for seasoning are extracted as final change targets,
A design apparatus characterized by performing seasoning inference by changing. 3. The design apparatus according to claim 1, further comprising: a case image range database that defines image ranges of a plurality of existing design cases, wherein the inference unit determines the evaluation term, a case name of the existing design, Is input, the case image range database is used to search whether or not the input evaluation term is within the image range of the existing design case. If the input evaluation term is within the range, the image is used for seasoning inference. A design apparatus characterized by selecting a plurality of evaluation terms having a weak correlation with an input evaluation term within a range as a seasoning evaluation term, and performing seasoning inference in consideration of an impression of the selected seasoning evaluation term. 4. The design apparatus according to claim 1, further comprising: a part image database that stores information on a commercial part as a component of a unit constituting a design object for each evaluation term, wherein the inference means includes: In seasoning inference, the degree of contribution of commercially available parts to the impression of the seasoning evaluation terms according to the strength of the correlation between the input evaluation terms and the seasoning evaluation terms using the parts image database for the specified seasoning evaluation terms A design device which selects commercially available parts in consideration of the above and makes seasoning inferences. 5. The design apparatus according to claim 1, wherein the evaluation terms are hierarchically classified into upper-level evaluation terms and lower-level evaluation terms based on the strength of correlation between the evaluation terms, and stored. A design apparatus comprising: an adjective tree structure database; and the inference means performs seasoning inference for each lower-level evaluation term with respect to the input evaluation term using the adjective tree structure database when performing seasoning inference. 6. The design apparatus according to claim 1, wherein a design object to be inferred is a vehicle interior.