JPH10214328A - Picture information analyzing device and storage medium with picture information analyzing program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To precisely extract the difference between the pictures of an aerial photograph and a satellite photograph at the same spot taken at different dates by calculating shadow information of a newly photographed picture from high level information on an old picture and extracting difference information at each point of both of the pictures based on the shadow information. SOLUTION: Among pictures at the same point taken at different date to be the object of difference analysis, a picture of an old photographing date is made an original picture and a picture of a new photographing date is made a new picture. A shadow part extracting part 101 extracts the shadow part of the original picture. Through the use of shadow information extracted from this original picture, a high level information extracting part 106 calculates high level information at each point of the original picture. A shadow information calculating part 108 calculates shadow information generated at the time of taking a new picture from high level information on this original picture. Next, a difference information extracting part 103 extracts difference information at each point of the new picture and the original picture. Then a display control part 105 displays this difference information on a display 112.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image information analyzing apparatus for analyzing image data to extract and display information according to a purpose, and a storage medium having an image information analyzing program. The present invention relates to an image information analysis apparatus that analyzes and uses an image, and a technique effective when applied to a storage medium having an image information analysis program.

[0002]

2. Description of the Related Art In recent years, the use of satellites and aerial photographs has become widespread, and in particular, they have been used for efficiently exploring vast areas in map creation and maintenance.

[0003] The use of aerial photographs and satellite photographs in the creation and maintenance of this map mainly involves acquiring altitude information by stereo vision of aerial photographs, grasping changes in land conditions by using photographs, and expanding vegetation and land in a wider area. There is grasp of the use situation.

Conventionally, as such a technique, for example, as disclosed in Japanese Patent Laid-Open Publication No. Hei 3-167678 (a method for obtaining three-dimensional numerical data from stereo image data), it is obtained from an artificial satellite or an aircraft. There is a method of obtaining corresponding points in both left and right images to obtain three-dimensional numerical data based on the image. In this method, height information is obtained by finding feature points in each of the left and right images and automatically associating them. It should be noted that for those feature points that could not be automatically associated, interactive association is performed manually.

Japanese Patent Application Laid-Open No. 8-63683 discloses that
A device that takes a difference between images in order to use an image from a television camera for a monitoring device is disclosed. In this device, since monitoring is performed on an image captured by one television camera, difference information is not simply obtained between time-series images. By performing the difference processing, the effect of the change of the sunshine at the monitoring target place can be removed.

[0006]

However, Japanese Patent Laid-Open No. 8-6 / 1996
According to the method of using images disclosed in Japanese Patent No. 3683 and Japanese Patent Application Laid-Open No. 3-167678, the images to be compared are shot at almost the same time, and therefore, are shot in a period of several months to several years. It cannot be applied as it is to grasp the fluctuation of the use situation of land comparing the images with different shooting dates and times.

[0007] This is because the image contrast differs due to weather conditions such as clouds, the shading of buildings due to the difference in the sun position at the time of photographing with the color temperature difference, and the photographing conditions of each image change. Is a problem. Among these, image contrast and color temperature can be corrected between images by adjusting the average value and variance of the gradation of each image, but the shadows appearing in the image are corrected by gradation correction. It cannot be processed by itself.

For this reason, when the technology disclosed in Japanese Patent Application Laid-Open No. Hei 3-167678 is used for grasping changes in land use conditions,
A feature portion in an image to be automatically extracted greatly differs due to a difference in shading, and automatic association cannot be performed.

In the monitoring method disclosed in Japanese Patent Application Laid-Open No. 8-63683, when the land use status and the like are grasped without taking shadow into account at the time of performing image differentiation processing, a true land The change in the sun condition, which is simply a shading, is larger than the change in the building and the change in the building, making it impossible to recognize the change in the land condition.

As described above, in the prior art, sufficient accuracy cannot be obtained if differences in images are obtained without taking into account the shadow of a building or the like due to a difference in the sun position at the time of photographing to grasp a change in situation. was there.

An object of the present invention is to provide a technique capable of extracting, with high accuracy, differences between images of aerial photographs, satellite photographs, and the like of the same spot taken at different dates and times.

[0012]

SUMMARY OF THE INVENTION Among the inventions disclosed in the present application, the outline of a representative one will be briefly described.
It is as follows.

An image information analyzing apparatus for extracting and analyzing difference information between images of aerial photographs and satellite photographs of the same spot photographed at different dates and times is provided. A shadow portion extracting means for extracting a shadow portion of an image with an old shooting date and time; an altitude information extracting means for extracting altitude information for each point of an old image using the extracted shading information; and the extracted altitude information Shadow information calculating means for calculating shadow information occurring in a newly photographed image, and difference information extracting means for extracting difference information for each point of the two images based on the calculated shadow information; and Extraction result output means for outputting a result of the obtained difference information.

[0014]

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram for explaining a configuration of an image information analyzing apparatus according to an embodiment of the present invention.
In the present embodiment, as the image information analysis, a difference analysis of images of the same spot photographed at different dates and times is used in order to grasp a land use situation and the like.

As shown in FIG. 1, the image information analyzing apparatus according to the present embodiment has an image information recording section 100 storing image information such as a satellite image and an aerial photograph, and the image information recording section 100 stores the image information. Image gradation correction unit 10 that corrects the gradation of an image in order to perform difference analysis between a plurality of images
2, a shadow portion extraction unit 101 for extracting a shadow portion of the image information recorded in the image information recording unit 100, and an altitude information extraction for extracting the altitude information of the image information recorded in the image information recording unit 100 Unit 106, an altitude information recording unit 107 in which altitude information of each image extracted by the altitude information extracting unit 106 is recorded, and altitude information of each point of the image recorded in the altitude information recording unit 107 and an analysis target. A shading information calculating unit 1 that calculates shading information generated in an image from information on a sun position at the time of capturing the image and records the shading information in a shading information recording unit 109.
08, a difference information extraction unit 103 that calculates a difference value of each pixel of two images having different photographing dates and times in which gradation correction has been performed by the image gradation correction unit 102 and stores the difference value in the difference information recording unit 104; A display control unit 105 that displays image information stored in the image information recording unit 100 and the difference information recording unit 104 on the display device 112 in accordance with an input from the input device 111.
Consists of

Next, the processing in the above-described image information analyzing apparatus will be described.

FIG. 2 is a flowchart for explaining the processing of the image information analyzing apparatus according to the present embodiment.

As shown in FIG. 2, the image information analyzing apparatus according to the present embodiment firstly converts an image of the same spot taken at a different date and time, which is to be subjected to difference analysis, with the oldest shooting date and time into an original image, Assuming that a new image at the shooting date and time is a new image, a shadow portion extraction process (step 201) for extracting a shadow portion of the original image is performed by the shadow portion extraction unit 101, and the shadow information extracted from the original image is used to perform the advanced processing. Information extraction unit 10
6, the altitude information extraction processing for calculating the altitude information of each point of the original image is performed (step 202), and from the altitude information of the original image, the shadow information calculating unit 108 calculates the shadow information generated at the time of photographing the new image. An information calculation process (step 203) is performed. Next, a difference information extraction process (step 204) for extracting difference information of each point between the new image and the original image is performed by the difference information extraction unit 103, and the display control unit 105 performs this process. An extraction result display process (step 205) for displaying the difference information on the display device 112 is performed.

Next, steps 201 to 205 described above are performed.
Will be described in detail with reference to the drawings.

FIG. 3 is a flowchart for explaining the above-described shadow part extraction processing (step 201).

This shadow portion extraction processing (step 201)
Is a process in which a user designates a point of a shaded portion in the original image to mark each point in the original image as a point in the shade or not included in the shade.

As shown in FIG. 3, first, the user designates a plurality of points on the original image having a high possibility of a shadow point as a shadow point in order to determine a threshold value serving as a reference for the shadow point. The shades are extracted from the extracted points (step 301), and the shades of the extracted shades are measured (step 302). From the measured results, the range between the brightest shaded point and the darkest shaded point is set as a reference threshold. (Step 303).

Thereafter, the user designates whether or not to input a shadow point (step 304). If there is an input of a shadow point, the user uses the input device 111 to input the point using the input device 111. One point is designated (step 305).

Next, points which are adjacent to the input point and fall within a certain threshold value difference from the pixel value of the input point are also marked as shadow points, and this marking is started from the input point, and then the mark is applied. The points are sequentially expanded to peripheral points, and all points where the pixel value falls within a certain threshold difference by connection from the original input point are marked as shadow points (step 306).

Next, the state of the marking of the shadow is displayed on the display device 111 (step 307). Based on the display result, the process returns to step 304, and the user again specifies whether or not to input another shadow point. I do.

If the user has finished inputting the designated point in step 304, a deletion process is performed when there are too many shadow points in the original image. In other words, the user designates the presence or absence of an input error (deletion of a shadow point), and if there is no input of a deletion point, the process ends (step 308).

If an input error is specified in step 308, the position of the shadow point to be deleted is input by the input device 111 (step 309).

Next, the same processing as that performed in the shadow part expansion processing in step 307 is performed on the pixels around one point in the specified original image to expand the specified non-shade part.
In other words, pixels that are adjacent to the designated non-shading point and whose pixel value falls within a certain threshold difference from the pixel value of the designated point are also marked as non-shading points. This marking is started from the input point, and then, for the adjacent pixel, and for the point newly marked as a non-shadow point in the adjacent pixel, the pixel value of the adjacent pixel is further changed to the original designated pixel. It is checked whether or not the pixel value falls within a certain threshold value with the pixel value of. If the pixel value falls within the threshold value, that point is further set as a non-shadow point. By repeating this process sequentially, all points where the difference of pixel values within the specified range from the designated non-shadow point are connected are marked as non-shaded points (step 310).

Then, the status after the marking is displayed on the display device 1.
12 and return to step 308 to input again whether or not to remove a shadow point to perform a shadow point (step 31).
1).

FIG. 4 is a diagram showing a display example of the original image in step 311. In FIG. 4, the marked shadow points are indicated by hatched portions, and the non-shaded portions are shown as they are.

FIG. 5 shows an altitude information extraction process (step 2) for extracting altitude information of each point in an image based on a shadow portion marked in the original image in the shadow extraction process (step 201).
02) is a flowchart illustrating the process of FIG.

In the altitude information extraction processing, as shown in FIG.
First, the sun azimuth at the shooting point at the time of image shooting is set as a variable θ1 (step 401), and the sun altitude angle at the shooting point at the time of shooting is set as a variable θ2 (step 40).
2).

Next, in order to set the altitude information of each point estimated from the shaded portion for all the pixels of the original image, first, 1 is set to the variable I. (Step 403) Next,
If the value of the variable I exceeds the width of the original image, the process is terminated (step 404). If not all the images have been processed, the variable J is set to 1 (step 405).

Thereafter, it is checked whether the value of the variable J has exceeded the height of the image (step 406). If the value of the variable J has exceeded the height of the image, the flow advances to step 413 to proceed to step 413. , The value of the variable J is returned to 1, and the process of step 405 is performed again.

If the value of the variable J is smaller than the height of the image, the horizontal position in the image in step 406 is I and the vertical position is J
The following processing is performed on the pixels (hereinafter, abbreviated as points (I, J)) in the image of FIG.

First, it is determined whether or not the point (I, J) is a shadow point marked in the shadow extraction processing in step 201 (step 407), and the point (I, J) is included in the shadow portion. If not, 1 is added to the variable J in step 412, and the process proceeds to the next point. If the point (I, J) is included in the shaded portion, the following process is performed.

First, as shown in FIG. 6, for example, the number of pixels from the point (I, J) to the first non-shadow point pixel in the sun direction (θ1 direction) is stored in a variable L1 (step 408). Point (I, J) to anti-sun direction (-θ1 direction)
Is stored in a variable L2 (step 409).

Thereafter, the variable H is set to (L1 + L2) · tan
The value of θ2 is stored (step 411), and this value is used as altitude information estimated from the shadow (in the example of FIG. 6, the altitude of the building 2).

The value of the variable H is set as the height information of the point obtained from the point (I, J) to the point from the first non-shading point to the next shading point in the sun direction (θ1 direction) from the point (I, J). (Step 411).

Next, 1 is added to the value of the variable J, and the process proceeds to the pixel processing in step 406 (step 412).

FIG. 7 shows a shadow information calculation process (step 20).
It is a flowchart for demonstrating 3).

The shadow information calculation processing (step 203)
FIG. 7 is a process for calculating the state of shadow based on the altitude information of the original image extracted by the altitude information extraction process (step 202) and the sun azimuth and solar altitude at the time of capturing the new image.
As shown in (1), first, the sun azimuth at the time of photographing a new image is set as a variable θ1 (step 501), and the solar altitude angle at the time of photographing a new image is set as a variable θ2 (step 502).

Next, in order to calculate the shading occurring in the image from the altitude information for all the pixels of the image, first,
A value 1 is set to the variable I (step 503), and it is checked whether or not the value of the variable I exceeds the width of the image. If the value of the variable I exceeds the width of the image, processing for all pixels is performed. Is completed, and the process is terminated (step 504).

If the value of the variable I does not exceed the width of the image, a value 1 is set to the variable J (step 505).

Next, if the value of the variable J is larger than the height of the image, it is determined that the processing for one row of pixels in the vertical direction has been completed, and the routine proceeds to step 512, where 1 is added to the value of the variable I and the next pixel is added. Move on to column processing.

If the value of the variable J is smaller than the height of the image, the height of the point (I, J) in the image must first be calculated in order to calculate the shadow caused by the point (I, J) in the image. It is determined whether the information is "0" (step 507).

If the altitude information of the point (I, J) is "0", it is determined that no shadow is generated at that point, and the process proceeds to step 511 to proceed to the processing of the next pixel. And returns to step 506.

If the altitude information of the point (I, J) is not "0", the altitude information of the point (I, J) is set to a variable H (step 508), and the shadow generated by the height of the H is set. To determine the length, the value of H · cotθ2 is set to the variable L (step 509).

Next, a shadow mark is applied to a pixel whose distance from the point (I, J) is smaller than the value of the variable L in the direction opposite to the sun (anti-θ1 direction) (step 510). .

Thereafter, 1 is added to the value of the variable J (step 511), and the process returns to step 506 to proceed to the processing of the next pixel.

This shadow information calculation processing (step 203)
As a result, it is possible to calculate the prediction of the state of occurrence of the shadow at the time of capturing a new image based on the altitude information extracted from the original image.

FIG. 8 is a diagram showing the prediction of the state of occurrence of shadows in the calculated new image. It can be seen that the positions where the shadings of buildings 1, 2, and 3 shown in FIG. 8 occur are different from those shown in FIG.

FIG. 9 shows a difference information extraction process (step 20).
It is a flowchart for demonstrating the process of 4).

The difference information extraction process (step 204)
The difference information between the original image and the new image is calculated by using the above-described shading occurrence state shown in FIG. 8, the absolute value of the difference between the pixel values of the corresponding pixels of the original image and the new image is calculated, and the difference extraction of the shadow portion is performed. As shown in FIG. 9, first, all values of the difference information recording area corresponding to each pixel of the image of the difference information recording unit 104 for recording the difference information are cleared to 0 (step 601), and the variable The value of I is set to 1 (step 60)
2).

If the value of the variable I is larger than the horizontal width of the image, the process ends (step 603).

If the value of the variable I is smaller than the horizontal width of the image, the value 1 is set to the variable J in order to perform processing on the image column specified by the value of the variable I (step 60).
4).

The value of the variable J is compared with the height of the image, and the
If the value of the variable J is large, it is determined that the processing for that column has been completed, and the process proceeds to step 609, where 1 is added to the value of the variable I, and the process proceeds to the next column. If the value of the variable J is small, the next process is performed. (Step 605).

If the point (I, J) is included in either the shaded part marked in the shade extraction processing in step 201 or the shaded part extracted on the original image, Since the pixel at the point (I, J) in the new image is a shadow portion, the process proceeds to step 608 without performing difference extraction, and proceeds to the process for the next pixel (step 606). In particular,
The processing for obtaining the difference is not performed for the shaded portions shown in FIGS.

When the point (I, J) is not included in the shaded portion, the absolute value of the difference between the pixel values of the point (I, J) in the original image and the new image is recorded in the difference information recording unit 104. (Step 607). Specifically, the difference information is recorded only for the building 4 shown in FIG. 8, which is not shown in FIG.

Next, 1 is added to the value of the variable J, and the flow returns to step 605 to proceed to the processing of the next pixel (step 60).
8).

FIG. 10 shows an extraction result display process (step 2).
05) is a flowchart for explaining the present invention.

The extraction result display processing (step 205)
This is a process of displaying the difference information recorded in the difference information recording unit 104 on the display device 112. As shown in FIG. 10, first, the user inputs the display parameter of the difference information as a command from the input device 111 (step 701).
Examples of this parameter include a threshold value for ignoring a difference area of a certain number of pixels or less, and a display when the absolute value of the density difference of the pixel at each point recorded in the difference information recording unit 104 is less than a certain threshold value. It has functions such as no.

If the end of the process is designated by this command input, the process is terminated (step 702).

If the processing is not completed, image processing corresponding to the input command is performed on the difference image (step 703).

Then, the result is displayed on the display device 112, and after displaying the image, the process proceeds to step 701, where the user is again requested to input a command for correcting various display parameters, and the process is repeated (step 704).

Therefore, as described above, the shadow portion of the original image is extracted, and the altitude information of each point of the original image is extracted by using the shadow information extracted from the original image.
Shading information generated at the time of photographing a new image is calculated from the altitude information of the original image, and the difference information between each point of the new image and the original image is extracted and displayed. Since it is not a target, it is possible to prevent a shadow portion from being extracted as a difference, and to extract difference information with high accuracy.

For this reason, when comparing images such as aerial photographs and satellite photographs taken at different dates and times to understand the temporal changes in the form of use of land, such as roads and buildings, and changes in the state of buildings and roads, the influence of shadows is important. It becomes possible to prevent extraction of an erroneous change.

As described above, in the image information analysis apparatus of the present invention, when comparing images such as aerial photographs and satellite photographs and taking a difference between them, as in the shadows described in the present embodiment, differences in photographing date and time and the like are obtained. By removing those which do not become differences due to the difference in the imaging conditions, the difference information can be extracted with high accuracy, and the temporal change can be grasped with high accuracy.

Each of the units 100 to 109 constituting the image information analyzing apparatus of this embodiment may be realized by a program executable by a computer, and the program at that time may be a floppy disk, a CD / ROM, a mask, or the like. It is provided to general users on a storage medium such as a ROM. In this case, in addition to these processes, the program may be provided to the user in combination with another program such as a GUI program.

Further, as an alternative means provided by the above-mentioned storage medium, it may be provided through a network such as the Internet for a fee.

As described above, the invention made by the present inventor
Although specifically described based on the embodiment, the present invention
It is needless to say that the present invention is not limited to the above-described embodiment, but can be variously modified without departing from the scope of the invention.

[0073]

The effects obtained by the representative ones of the inventions disclosed in the present application will be briefly described as follows.

It is possible to extract differences between images such as aerial photographs and satellite photographs of the same spot taken at different dates and times with high accuracy.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration of an image information analysis device according to an embodiment of the present invention.

FIG. 2 is a flowchart for explaining processing of the image information analysis device.

FIG. 3 is a flowchart illustrating a shadow part extraction process (step 201).

FIG. 4 is a diagram showing a display example of an original image in step 308.

FIG. 5 is a flowchart illustrating an altitude information extraction process (step 202).

FIG. 6 is a diagram for explaining a step of estimating altitude information from a shadow.

FIG. 7 is a flowchart for explaining shadow information calculation processing (step 203).

FIG. 8 is a diagram illustrating prediction of a state of occurrence of a shadow in a calculated new image.

FIG. 9 is a flowchart illustrating a difference information extraction process (step 204).

FIG. 10 is a flowchart illustrating an extraction result display process (step 205).

[Explanation of symbols]

100: image information recording unit, 101: image gradation correction unit, 1
02: shadow part extraction unit, 103: difference information extraction unit, 10
4 ... Difference information recording unit, 105 ... Display control unit, 106 ... Altitude information extraction unit, 107 ... Altitude information recording unit, 108 ... Shadow information calculation unit, 109 ... Shadow information recording unit, 111 ... Input device, 112 ... Display device .

Claims (2)

[Claims] 1. An image information analyzing apparatus for extracting and analyzing difference information of aerial photographs and satellite images of the same spot taken at different dates and times, wherein the aerial photographs and satellites of the same spot taken at different dates and times are provided. Shadow portion extraction means for extracting a shadow portion of an image with an old photographing date and time; and altitude information extraction means for extracting altitude information for each point of the old image using the shadow information extracted by the shadow portion extraction means. A shading information calculating means for calculating shading information occurring in a newly shot image from the altitude information extracted by the altitude information extracting means; and a shading information calculated by the shading information calculating means, based on the shading information calculated by the shading information calculating means. A difference information extracting unit that extracts difference information for each point; and an extraction result output unit that outputs a result of the difference information extracted by the difference information extracting unit. Image information analyzing apparatus characterized by a. 2. A storage medium having an image information analysis program for extracting and analyzing difference information of an aerial photograph or a satellite photograph image of the same spot photographed at different dates and times, wherein the image information analysis program comprises: Extracting a shaded portion of an old image of an aerial photograph or a satellite photograph at the same date and time taken at the same point, and extracting altitude information for each point of the old image using the extracted shaded information. Calculating shadow information occurring in a newly captured image from the extracted altitude information; extracting difference information for each point of the two images based on the calculated shadow information; Outputting a result of the extracted difference information.