JPH11213135A - Image retrieving device - Google Patents

Abstract

PROBLEM TO BE SOLVED: To obtain positional information of a photographing position simply by photographing a condition of a road, etc., with a video camera by comparing and judging an image held by a three-dimensional CG image holding means with an image held by a color component holding means according to a prescribed judgment standard. SOLUTION: Three-dimentional data held by a three-dimensional data holding means 11 is plotted by a three-dimensional image plotting means 2. In this case, the direction of a viewpoint is decided by a viewpoint changing means 3 and a color of an object is decided by a color changing means 4. A plotted result is held by the three-dimensional; CG image holding means 7. On the other hand, the image photographed with a camera, etc., is held by an image holding means 5, a color filtering is further performed on the held image by a color filter 6 and its result is held by the color component holding means 8. And the three-dimensional image held by the three-dimensional CG image holding means 7 is compared with the image held by the color component holding means 8 by an image comparison means 9 and the images are judged according to the judgment standard held by a judgment standard holding means 11 by a judging means 10.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a computer-aided video camera for examining the location of an image taken by a video camera in advance.
The present invention relates to a technique for knowing by comparing data created by the above method with a video drawn by computer graphics. The present invention also relates to a technology for automatically detecting dirt or breakage of a sign or the like installed on a road.

Further, the present invention relates to a technique for displaying an image indicating a traveling direction based on an actual image in a car navigation system.

[0003]

2. Description of the Related Art FIG. 26 shows a first conventional example of the present invention. In FIG. 26, reference numeral 261 denotes an image capturing unit;
262 is an image recording means, and 263 is a position measuring means.

[0004] In FIG. 26, reference numerals 264 to 269 explain these means 261 to 263 in more detail. Reference numeral 264 denotes a video camera as image photographing means.
5 is a video tape mounted on the video camera, 266 is a label of the video tape, 267 is a pen for recording on the label, and 268 is a GPS (Global Position) for measuring the position.
ing System, which detects the position of the earth's surface by satellite), 2
69 is a monitor attached to the GPS.

In FIG. 26, in order to record a position photographed by a video camera, for example, a position is measured by a GPS 268 and the position information displayed on a monitor 269 is recorded on a video tape on which a video photographed by the video camera 264 is recorded. 2
65 is recorded on the label 266 using the pen 267.

FIG. 27 shows a second conventional example of the present invention. In FIG. 27, 21 is a road sign, 271
Is a camera for photographing the sign on the road, 272 is a mount for attaching a photograph, 273 is a photograph of the sign photographed by the camera 271, 274 is a ledger for photographing the past sign, 275
Is a photograph of the sign attached to the ledger.

[0007] In Conventional Example 2 shown in FIG. 27, a photograph of the site is taken by a camera 271, developed and printed, and then a photograph 273 is pasted on a mount 272. Then, in order to compare with the past state of the sign 21, the past book 275 is found by searching the ledger 274, and the comparison 276 with the current state is performed.

FIG. 28 shows a third conventional example of the present invention. In FIG. 28, reference numeral 281 denotes a GPS (Global Positioning System) for finding a position on the ground surface; 282, an image of computer graphics for displaying the state of the road three-dimensionally; 283, a computer based on the position data of the GPS 281; Graphics Image 28
The calculation and drawing means for displaying 2 is a monitor.

FIG. 28 shows an outline of a car navigation system.
Is displayed three-dimensionally on the monitor 284 on the basis of the position data.

[0010]

However, FIG.
In the conventional example 1 shown in FIG. 1, the photographed position has to be measured each time by the GPS or the like and recorded on a label or the like of the medium on which the image is recorded. Also, when comparing images of the same place taken at different times, it was necessary to search from many video tapes using the characters recorded on the label. Furthermore, when searching for an actual image corresponding to a position on the map, it was necessary to rely on the label.

In the conventional example 2 shown in FIG. 27, when comparing with a past sign image, a large number of photographs are managed by a ledger, photographs are searched from the ledger, and the user goes to the site each time. There is a problem that a photograph must be taken, and development and printing must be performed for each sheet.

Further, in the conventional example 3 shown in FIG. 28, if much labor is required to create the image 284 displayed on the monitor, or if the road is repaired, it takes time to update the data. There was a problem.

[0013]

In order to solve this problem, an image retrieval apparatus according to the present invention comprises, first, a three-dimensional data holding means for holding an image to be drawn by computer graphics; Three-dimensional image drawing means for drawing three-dimensional data held by the three-dimensional image holding means;
3 for holding an image drawn by the two-dimensional image drawing means
Dimensional CG image holding means, viewpoint changing means for changing a position of a viewpoint when drawing on the three-dimensional image drawing means, color changing means for changing a color when drawing on the three-dimensional image drawing means, image Image image holding means for holding an image, a color filter for extracting only a specific color component for the image image held in the image image holding means,
A color component holding unit that holds a color component of an image image extracted by the color filter; an image comparison that compares an image held by the three-dimensional CG image holding unit with an image held by the color component holding unit Means, a criterion holding means for holding a criterion, a judgment means for judging a result compared by the image comparing means based on a criterion held by the judgment criterion holding means, and a result judged by the judgment means Is displayed on the display unit.

Thus, it is possible to know the position information of the photographing location only by photographing the situation such as the road with the video camera.

Second, image input means for inputting an image, position input means for inputting position information,
An image image and position holding means for holding the image image input from the image image input means and the position information input from the position input means; and an image image and position information output from the image image and position holding means. Input switching means for switching a storage destination, first and second image image storage means for storing the image image and the position information held by the position switching means and the image image switched by the input switching means, Position indicating means for indicating position information to the first and second image image storing means;
Image comparison means for comparing the images held in the image storage means, and a result display means for displaying the result of comparison by the image comparison means.

This makes it possible to automatically detect dirt or damage to a sign or the like installed on a road.

Third, position input means for continuously inputting positions, route information holding means for position information continuously input by the position input means, and area image output from the route information holding means Area image holding means for holding an image, image image input means for inputting an image image, image image holding means for holding an image image input by the image image input means, and image held by the image image holding means Area extracting means for extracting an area from an image, area comparing means for comparing the area image held by the area image holding means with the area image extracted by the area extracting means, and a position input by the position input means Route guidance information holding means for outputting route guidance information based on the information; A display image generating means for generating an image of extensive route guidance information held in the image and the route guidance information holding means is constituted by a display means for displaying the image generated by said display image generating means.

This makes it possible to guide the traveling direction in the car navigation system by using an image obtained by actually photographing the road.

[0019]

(Embodiment 1) FIG. 1 is a block diagram of Embodiment 1 of the present invention. In FIG.
Reference numeral 1 denotes a three-dimensional data holding unit which holds original data for drawing a three-dimensional image by computer graphics. Reference numeral 2 denotes a three-dimensional image drawing means for drawing the three-dimensional data held in the three-dimensional data holding means 1. Reference numeral 3 denotes a viewpoint changing unit that determines the direction of the viewpoint when the three-dimensional image drawing unit 2 draws a three-dimensional image. Reference numeral 4 denotes a color changing unit that determines the color of an object when the three-dimensional image drawing unit 2 draws a three-dimensional image. Reference numeral 7 denotes a three-dimensional CG image holding unit that holds a result of drawing by the three-dimensional image drawing unit 2. Reference numeral 5 denotes an image image holding unit which holds an image captured by a camera, a video camera, or the like. In the case of a video camera, it is held as an image image for each frame. Reference numeral 6 denotes a color filter which applies a color filter to the image image held by the image image holding means 5. This result is held in the color component holding means 8. Reference numeral 9 denotes an image comparing unit that compares the three-dimensional image held by the three-dimensional image holding unit 7 and the image held by the color component holding unit 8. Numeral 10 denotes a judging means for judging the result compared by the image comparing means 9 according to the judging standard held by the eleven judging standard holding means. The criteria will be described later. When the result determined by the determination unit 10 does not reach the determination criterion held by the determination criterion holding unit 11, the determination unit 10 uses the viewpoint changing unit 3 and the color changing unit 4 to output the three-dimensional image drawing unit 2. Change the image to be drawn. As a result, the content redrawn by the three-dimensional image drawing means 2 is a three-dimensional CG
The image is held in the image holding unit 7 and compared with the image held in the color component holding unit 8. This redrawing is performed until the judgment criterion 11 is met, and the result is displayed on the result display means 12.

FIG. 2 is an explanatory diagram showing the operation of the first embodiment of the present invention. In FIG. 2, 21 is a road sign, 22
Is a road, 23 is a car running on the road, and 24 is a video camera mounted on the car. The road is photographed by the video camera 24 mounted on the car 23.

FIG. 3 shows processing of an image photographed by a video camera in this manner. In FIG. 3, reference numeral 31 denotes a captured image image, which is held in the image image holding means 5. Numeral 32 denotes a result obtained by applying a color filter, and this processing is performed by the color filter 6, and the result is stored in the color component holding means 8.
Is held. In this figure, only the green portion of the 31 images is left.

FIG. 4 is an explanatory diagram showing a method for drawing a three-dimensional image. In FIG. 4, reference numeral 41 denotes three-dimensional data held in the three-dimensional data holding means 1. The three-dimensional data includes, for example, a road 42, a sign 43, and the like.

The sign 43 is composed of several faces. When the direction of the viewpoint at which the sign is viewed is determined by the viewpoint changing unit 3, the sign is drawn as seen from the right or left as shown in FIG. 44 or 45. . When the entire road is drawn, it becomes 46.

The color changing means 4 instructs the three-dimensional image drawing means to draw only an image having a green color or only a gray color among the images to be drawn. Tell For example, when the color changing means 4 informs the three-dimensional image drawing means 2 that only the one having the green color is drawn, an image like 47 is obtained, and the result is three-dimensional CG.
The image is held by the image holding unit 7. The image comparing unit compares the image of the three-dimensional image holding unit 7 with the image of the color component holding unit 8.

As a comparison method, for example, the number of areas,
The size of the sum total of the area of the regions in the entire image is checked for the degree of overlap between the two images.

Comparing 32 and 47, the number of regions is equal every three. However, the green area held by the color component holding means 8, that is, 32, is larger in the size of the total area of the areas in the entire image. For this reason, the determination unit issues an instruction to the viewpoint changing unit 3 to further approach the sign, and issues an instruction to draw an image approaching the sign.

FIG. 5 shows an image when the distance is further approached in this manner. Numeral 51 indicates the overall state, and numeral 52 indicates a state where a color filter is applied. In the state of 52, 32 and the green area occupying the whole image are almost equal.

Therefore, these two images, namely 32 and 5
Take the correlation of 2. This method is shown in FIG. In FIG. 6, reference numeral 61 denotes an image held by the color component holding means 8, reference numeral 62 denotes an image held by the three-dimensional CG image holding means 7, and reference numeral 63 denotes an area where these two images overlap.

As shown in FIG. 6, while moving the image 61 up, down, left, and right, the overlap between the two images is compared to check the degree of similarity. As this method, the SSDA method and the like are known.

As a result, when the similarity between the two images exceeds a certain threshold value, it is determined that the two images match.

The two images are R (red), G (green), and B
(Blue), and each color has 256 levels.

The area (the number of pixels) of the overlapping region 63 is
The coordinate axes of the overlapping area 63 are set to i and j as shown in FIG. 6, and R, G, and B of the pixels of the image 62 in the overlapping area 63.
The components are P1R (i, j), P1G (i, j), P1B (i, j), respectively.
j) The R, G, and B components of the pixels of the image 61 in the overlapping area 63 are represented by P2R (i, j), P2G (i, j), and P2B (i,
j), the similarity L between the two images is given by the equation shown in FIG. Integration is performed for all pixels in S.

L has a size of 0 ≦ L ≦ 3. Thus, for example, when L <0.5, the two images are determined to be the same.

This value is held in the criterion holding means 11. In this way, when the determination unit 10 determines that the images are the same, the shooting position is displayed on the result display unit 12. FIG. 8 shows this state, and shows that the image held by the image image holding means 5 was taken from a position 81 on the road.

This position information is stored in the image / image holding unit 5.
The image is held in the second image image holding means 13 together with the image held in. Reference numeral 82 in FIG. 8 shows this state. Reference numeral 84 denotes a correspondence table between the frame number of the moving image and the shooting position of the image. Reference numeral 85 denotes a moving image frame, in which the images held in the image image holding means are sequentially stored.

As described above, according to the present invention, the position of an image obtained by photographing a road can be known by comparing it with a three-dimensional CG image registered in advance.

(Embodiment 2) FIGS. 10, 11, and 12
The operation of the second embodiment of the present invention will now be described. In FIG. 10, reference numeral 103 denotes a photographing position for photographing a road or the like; 107, a car traveling on a road;
Denotes a camera and a position measuring device mounted on the vehicle 107. The camera uses a video camera, captures the position with the video camera, acquires the position, and attaches position information to each frame of the moving image captured by the video camera. As the position measuring device, GPS is used as in the first embodiment.
In recent years, GPS has improved accuracy, and DGPS (Differen
cial GPS) is accurate to within a few meters. 1
06 is a sign installed on the road.

This system uses a sign 1 installed on a road.
The purpose of the inspection is to automatically check whether or not the vehicle 06 is damaged or dirty. The vehicle 107 runs on the road at different times and seasons.

FIG. 11 shows images taken from the same position in different time zones among the images thus taken. The sign shows the destination on a green background, indicating that going to the right will go to Tokyo and going to the left will go to Osaka. It is assumed that 112 is photographed after 111. In this figure, the sign is stained black as 112.

FIG. 12 shows the process of finding the stain on the sign. Only 121 and 122 are obtained by cutting out only the area of the sign from 111 and 112. Next, by comparing the area of the sign, a dirt portion such as 123 can be found. As a result, 124
As shown in above, on the image 112 of the sign taken recently,
The dirt can be indicated by circled locations.

Next, the operation of the second embodiment of the present invention will be described with reference to FIGS. 9, reference numeral 91 denotes an image image input unit, 92 denotes a position input unit, 93 denotes an image image and position holding unit, 94 denotes an input switching unit, 95 denotes a first image image storage unit,
Reference numeral 96 denotes a second image storage unit, 97 denotes a position indicating unit, 98 denotes an image comparing unit, and 102 denotes a result displaying unit.

The image image input means 91 can continuously capture images like a digital video camera. The position input means 92 can continuously know a position on a road or the like by GPS or the like. The image image and position holding unit 93 holds the image input from the image image input unit 91 and the position information input from the position input unit 92 as a pair.

A more detailed explanation of this content is as follows.
FIG. In FIG. 13, reference numeral 131 denotes a synchronization unit,
32 is a table, 133 and 134 are frame numbers and positions on the road in the table 132. Reference numeral 135 denotes a video deck, in which a moving image 136 is recorded, and a frame number 137 is attached to each frame. When a moving image is input from the image image input unit 91, the moving image is recorded on the video deck 135 as a moving image 136. At this time, a frame number 137 is attached to the moving image 136. On the other hand, the position information on the road is input from the position input means 92 as shown in FIG. This position information indicates the latitude and longitude on the road. Synchronization means 131
Makes an inquiry to the VCR 135 about the number of the frame currently being recorded. Then, it is recorded in the table 132 together with the position information input from the position input means 92. Thus, table 132 contains frame number 133 and position 34
Are recorded in pairs.

Returning to FIG. When the first photographing is completed by the car, a table 132 in which the image information and the image information held in the position holding means 93 are paired and the moving image information 136 are stored in the first image image storage means. .

FIG. 14 shows this state. The first image image storing means 95 and the second image image storing means 96 have the same structure, and have a frame number 1
There is a correspondence table 132 between 33 and position 134, a moving image 136, and a frame search unit 141 is built in.

In FIG. 9, when the same place is photographed again, the input image switching means 94 stores the image image and the information recorded in the position holding means 93 in the second image image storage means 96.

In this state, when the latitude and longitude on the road are inputted from the position indicating means 97, this information is inputted to the first image image storing means 95 and the second image image storing means.

As shown in FIG. 14, based on this position information, the frame search means 141 searches the position 134 of the table 132 for the one closest to the position information specified by the position specifying means 97, and reads the corresponding frame number. . Then, an image having this frame number is
Select from 6 These pieces of image information are sent to the image comparison means 98.

FIG. 15 illustrates the details of the image comparing means 98. In FIG. 15, the first color filter 151 is used.
And the second color filter 152 have the same property,
A color filter is applied to the image input from the image image holding means 95 or the second image image holding means 96. The result is stored in the first color component holding unit 153 or the second color component holding unit 154.

In the case where the first color filter 151 and the second color filter 152 perform processing to erase only the part surrounded by green and green and to erase other parts, FIG.
In reference numerals 111 and 112, only the part of the sign remains like 121 and 122 in FIG. And means 1 for determining the difference
When 121 and 122 are sent to 55, 121 and 1
A difference of 22 is obtained and a result like 123 is obtained.

Reference numeral 123 indicates that the appearance of the sign has changed. Find a circle that surrounds the whole of 123,
The circle 124 is displayed by being superimposed on the second image storage means.

As a result, the first image / image storing means 9
The difference between the image stored in the second image storage unit 96 and the image stored in the second image storage unit 96 is displayed as a circle on the image in the second image storage unit 96, and the time You can know how the sign has changed in a targeted or seasonal manner.

(Embodiment 3) FIGS. 17 and 18 are explanatory diagrams showing the functions of Embodiment 3 of the present invention. In FIG. 17, 21 is a sign, 22 is a road, 23 is a car, 24 is a camera mounted on the car, and 171 is a GPS that measures the position of the car.
A position measuring device such as 172 is the direction to go. Car 2
3 is about to proceed in direction 172 to follow sign 21.

FIG. 18 shows the inside of the car 23. In FIG. 18, reference numeral 181 denotes a car navigation monitor;
2 is the direction to go in the sign on the monitor 181, 183 message, 184 is the steering wheel of the car 23, 1
Reference numeral 85 denotes a sign installed on a road that is actually seen from inside the company.

An actual image taken by the camera 24 is displayed on the monitor. When the car 23 approaches the actual sign 185, according to the preset route information,
A circle 182 to indicate a sign that needs attention, or a message 1
83 is superimposed on the actually visible image and
1 is displayed. Therefore, the driver can clearly know the direction to go.

FIG. 16 shows a block configuration according to the third embodiment of the present invention. In FIG. 16, reference numeral 161 denotes a position input means, which corresponds to the GPS 171 in FIG. 1
63 is route information holding means, which stores information on signs. This will be described in detail with reference to FIG. 165, an area image holding unit;
The area image selected from the route information storage unit 163 is stored in accordance with the position information input by the user. Note that, as will be described later, this image is obtained by cutting out only a region such as a sign from an actual image.

Reference numeral 162 denotes an image image input means, which corresponds to the camera 24 in FIG. Reference numeral 164 denotes an image image holding unit which holds the image input by the image image input unit 162. Reference numeral 166 denotes an area extracting unit that extracts only an area such as a sign from the actual image held by the image image holding unit 165. The area comparing means 167 compares the image held in the area image holding means 165 with the image held in the area extracting means 166, and obtains the position of the marker.

The route guidance information holding means 168 holds the direction in which the vehicle should travel, and determines the direction in which the vehicle should travel in each position input to the position input means 161.
The display image generating means 169 is provided with an image image holding means 164.
The direction to go is displayed on the actual image held in the.

FIG. 19 is a block diagram of the route information holding means 163. In FIG. 19, 191 is an area information search unit, 192 is an area image, 193 is an area image number attached to the area image, 194 is an area image number, 195 is a position, 196 is an area image number 194 and a position 195 are arranged. Table 197 is attached information and will be described in detail in FIG. When the current position of the car is input from the position input unit 161, the area image search unit 191 searches for the closest one from the positions 195 in the table 196. Then, an area image number 194 corresponding to the position 195 is obtained. Next, the region image 19
2, the same area image number 19 as the area image number 193
The area image 192 having 3 is searched. This image is sent to the region image 165.

FIG. 20 shows the inside of the attached information 197. In FIG. 20, reference numeral 201 denotes an area image number;
02 is a destination, 203 is mark information, 204 is a position on an image, 205 is a long axis / short axis, and 206 is a message. This table is one for each of the region images 192 in FIG. Position 204 on the image, major axis
The short axis 205 forms the mark information 203. The destination is a destination such as "Tokyo" and "Osaka". The position 204 on the image indicates the position where the sign appears, and the major axis / minor axis 205 indicates the size of the ellipse to be displayed.
The message 206 indicates the content to be displayed on the screen.

FIG. 21 shows the operation of the area image 192 in the area image holding means 165 and the operation of the area extracting means 166. In FIG. 21, reference numeral 211 denotes an actually photographed image, 212 denotes an image in which only the sign region is extracted, and 2 denotes an image.
Reference numeral 13 denotes a green area of a sign, reference numeral 215 denotes a white line on the road, which is white, and reference numeral 216 denotes a road surface which is gray 216. The region extracting unit 166 extracts only green and a portion surrounded by green from the actually photographed image 211 to create an image like 212.

The area image 192 is an image such as 212 obtained by extracting only the green portion from the actual image.

FIG. 22 is an explanatory diagram of the attached information 197.
The branch 225 is associated with the region 223, and the branch 226 is associated with the region 224 like 227 or 228. When displaying a circle or a message on an actual image as shown in FIG. 18, this information is used.

FIG. 23 shows how the contents of the area image holding means 165 and the area extracting means 166 are compared. In FIG. 23, 231 indicates the contents of the area image holding means 165, and 232 indicates the contents of the area extracting means 166.
The comparison of the images is the same as the image comparison means of the first embodiment.

FIG. 24 shows the contents of the route guidance information holding means 168. The position 241 is a position where the car is traveling, and indicates in which direction the sign may be set for each of the signs installed at the positions where the car passes as the car travels. The destination 242 indicates the destination in the sign.

When the vehicle approaches a certain position, the route guidance information holding means 168 compares the position input from the position input means 161 with the position 241 to find the destination 242.
At the same time, the area comparison unit 167 checks which image stored in the route information holding unit 163 is equal to the image input from the current image input unit 162. And the attached information 197 of the route information holding means 163.

Therefore, the display image generating means sets the destination 24
The destination 202 corresponding to 2 is searched for. As a result, the position 204 on the image of the mark information 203 to be displayed on the image,
As shown in FIG. 18, mark information and a message can be displayed on an actual image based on the length 202 of the major axis and the minor axis of the ellipse and the message 206.

FIG. 25 is a flowchart showing this operation. At 251, the contents of the area image holding unit 165 and the area extraction unit 166 are compared. The result is determined in step 252, as described with reference to FIGS. 6 and 7 of the first embodiment. As a result, if it is determined that they are different, nothing is performed like 253.

In the same case, information on the destination is obtained from the route information holding means 163. This information corresponds to 202 in FIG.
20 shows details of the attached information 197 in FIG.

In the comparison 255 with the destination information held in the route guidance information holding means 168, the destination 202 and the destination 24 of the route guidance information holding means 168 in FIG.
Compare with 2. The attached information 197 has all the destinations displayed in each of the area images 192. On the other hand, the destination recorded in the route guidance information holding means 168 is:
This is the destination set each time you drive a car.

If the destinations match at 256, it means that a sign indicating the destination set at the destination has appeared, so the mark information is drawn on the sign as at 258. Otherwise, do nothing like 257.

[0072]

According to the present invention, firstly, it is possible to automatically attach a photographing position to an image of a video camera by using an image of three-dimensional computer graphics.

Second, it is possible to automatically detect dirt on a sign installed on a road. Third, in a car navigation system, navigation can be performed based on an actual image.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an image search device according to a first embodiment of the present invention.

FIG. 2 is an operation explanatory diagram according to the first embodiment of the present invention.

FIG. 3 is an explanatory diagram showing image processing taken by a video camera.

FIG. 4 is an explanatory diagram showing a method for drawing a three-dimensional image.

FIG. 5 is an explanatory diagram showing an image closer than FIG. 4;

FIG. 6 is an explanatory diagram showing a method of correlating two images.

FIG. 7 is a diagram showing an equation for calculating a similarity L between two images.

FIG. 8 is a diagram showing a shooting position displayed when it is determined that the images are the same.

FIG. 9 is a block diagram showing a configuration of an image search device according to a second embodiment of the present invention.

FIG. 10 is a diagram illustrating an operation according to the second embodiment of the present invention;

FIG. 11 is a diagram showing images taken from the same position in different time zones.

FIG. 12 is an explanatory diagram showing a process of finding a stain on a sign.

FIG. 13 is an explanatory diagram showing a process of recording position information for each frame.

FIG. 14 is an explanatory diagram showing a process of searching for a frame from position information using a position designating unit.

FIG. 15 is a block diagram showing a configuration for displaying a difference between color components of two stored image images in a circle.

FIG. 16 is a block diagram showing a configuration of an image search device according to a third embodiment of the present invention.

FIG. 17 is an operation explanatory diagram according to the third embodiment of the present invention.

FIG. 18 is an explanatory view showing the inside of a car.

FIG. 19 is a block diagram of a route information holding unit.

FIG. 20 is an explanatory diagram showing the inside of attached information.

FIG. 21 is an explanatory diagram showing the operation of a region image and a region extraction unit in a region image holding unit;

FIG. 22 is an explanatory diagram of attached information.

FIG. 23 is an explanatory diagram showing a comparison between the contents of the area image holding means and the area extracting means.

FIG. 24 is an explanatory diagram showing the contents of a route guidance information holding unit.

FIG. 25 is a flowchart showing an operation indicating the destination of the result of comparison between the content of the area image holding means and the area extracting means.

FIG. 26 is a block diagram of a conventional first embodiment.

FIG. 27 is an explanatory view of a conventional second embodiment.

FIG. 28 is a block diagram of a conventional first embodiment.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 3D data holding means 2 3D image drawing means 3 View point changing means 4 Color changing means 5 Image image holding means 6 Color filter 7 3D CG image holding means 8 Color component holding means 9 Image comparing means 10 Judgment means 11 Judgment standard Holding means 12 Result display means 21 Signs 22 Roads 23 Cars 24 Video camera 31 Photographed image image 32 Result image after color filter 41 3D data 42 Road information 43 Sign information 44 Signs seen from right 45 The sign seen from the left 46 The whole road is drawn 47 The processing result of the color changing means 51 The whole state 52 The color filter is applied 61 Image 1 62 The image 2 63 The overlapping area 81 The photographing position 82 The second image Internal structure of image holding means 84 Correspondence table of frame number of moving image and shooting position of the image Le 85 moving image frame 91 image image input means 92 position input means 93 image and position holding means 94 switching means 95 first image image storage means 96 second image image storage means 97 image comparison means 102 result display means 103 shooting position 104 camera 105 position measuring means 106 sign 107 shooting position 111 past image 112 recent image 113 green 114 black 121, 122 cut out only sign area 123 dirt 124 dirt mark 131 synchronization means 132 table 133 frame number 134 position 135 VCR 136 Video frame 137 Frame number 141 Frame search means 151 First color filter 152 Second color filter 153 First color component holding means 154 Second color component holding means 155 Means 161 position input means 162 image image input means 163 route information holding means 164 image image holding means 165 area extraction means 167 area comparison means 168 route guidance information holding means 169 display image generation means 170 display means 171 position measurement means 172 proceed Power direction 191 Area information search means 192 Area image 193 Frame number 194 Area image number 195 Position 196 Table 197 Attached information 201 Area image number 202 Destination 203 Mark information 204 Position on image 205 Long axis / short axis 206 Message 211 211 is actually 213 Green image 214 Mark 215 White 216 Gray 223, 224 Area 225, 226 Branch 227, 228 Correspondence relation 231 Area image holding means 232 Contents of the area extracting means 241 Location 242 Destination 261 Image taking means 262 Image recording means 263 Position measuring means 264 Video camera 265 Video tape 266 Video tape label 267 Pen for recording on label 268 GPS 269 Monitor 271 Marking sign Camera 272 taking photo 274 Mount for attaching photo 273 Sign photo 274 Ledger 275 Sign photo 281 GPS 282 Computer graphics image 283 Operation and drawing device 284 Monitor

──────────────────────────────────────────────────の Continued on the front page (51) Int.Cl. ⁶ Identification code FIG06F 15/62 335 15/70 310

Claims (6)

[Claims] 1. A three-dimensional data holding means for holding an image to be drawn by computer graphics.
Three-dimensional image drawing means for drawing three-dimensional data held by the three-dimensional image holding means; three-dimensional CG image holding means for holding images drawn by the three-dimensional image drawing means; Viewpoint changing means for changing the position of the viewpoint at the time of drawing, color changing means for changing the color at the time of drawing, image image holding means for holding an image image, and image held by the image image holding means A color filter that extracts only a specific color component from the image, a color component holding unit that holds a color component of the image image extracted by the color filter, an image held by the three-dimensional CG image holding unit, An image comparing means for comparing the images held in the color component holding means, and a result display means for displaying a result compared by the image comparing means. Image search apparatus to be. 2. The image retrieval apparatus according to claim 1, wherein said image comparison means has a judgment means for judging a result of comparing the images based on a judgment criterion stored in advance. 3. The image search device according to claim 1, wherein the image comparison unit calculates the correlation between the images while shifting the positions of the images. 4. An image input means for inputting an image image, a position input means for inputting position information, and holding an image input from the image input means and position information input from the position input means. Image and position holding means, input image switching means for switching a storage destination of the image image and position information output from the image image and position information holding means, and image image and position holding means switched by the input switching means First image image storage means and second image image storage means for respectively storing the image images and the position information held at different times, and the first image image storage means and the second image image storage means. Position instructing means for instructing position information with respect to the first image image; An image search device comprising: an image storage unit; an image comparison unit that compares images held in a second image storage unit; and a result display unit that displays a result compared by the image comparison unit. 5. The image retrieval apparatus according to claim 4, wherein said image image and position holding means records the position where the image image was taken and the image image in association with each other. 6. A position input unit for continuously inputting a position, a route information holding unit for position information continuously input by the position input unit, and a region image output from the route information holding unit. Area image holding means for inputting, image image input means for inputting an image image,
Image image holding means for holding an image image input by the image image input means, area extracting means for extracting an area from the image image held by the image image holding means, and image area holding means for holding the image image A region comparison unit that compares a region image with the region image extracted by the region extraction unit; a route guidance information holding unit that outputs route guidance information based on the position information input by the position input unit; Display image generating means for generating an image in which the area image compared by the comparing means and the route guidance information held in the route guidance information holding means are superimposed, and display means for displaying the image generated by the display image generating means An image search device configured by: