JP2000020692A - Shade removing method in high resolution satellite data and satellite data processor using the same - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a shade removing method and a shade removing means in high resolution data to automatically recognize a shade of structure from image information of satellite data and to relieve its effects. SOLUTION: A satellite data processor 100 is provided with an image input part 110, a pre-processing part 120, a shade removing part 130, an edge extracting part 140, a post-processing part 150 and a map data output part 160. The shade removing part 130 is provided with a corrected image input part 131, an image storage part 132, an image comparison part 133 extract two kinds of image data in the same area, which is picked up in different time from the image storage part 132 and to compare luminance at the same position, a shade deciding part 134 to decide the image with lower luminance as the shade, when difference in the luminance between two kinds of image data at the same position exceeds a specified threshold, an image compositing part 135 to composite the image data the shade of which is removed, a control part 138 and a recording medium 139.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a map creation method using satellite data, and more particularly to a shadow removal method and a shadow removal apparatus in map creation using high-resolution satellite data.

[0002]

2. Description of the Related Art In recent years, in addition to aerial photography, maps have been created using ground image information taken by artificial satellites such as earth observation satellites. Conventional satellite data has a side length of 10 m on the ground surface. , And a resolution of about 20 m, it was difficult to create a precise map, and shadows of structures and the like did not greatly affect the analysis, and little consideration was given to shadow removal.

[0003]

As described above, in the conventional imaging technique of the earth observation satellite, the side length of the ground surface is 10 m or 20 m.
Since the resolution was of the order of magnitude, the shadows of buildings were ignored. However, due to recent technological developments,
m in the order of m, the shadow of the building cannot be ignored in the map making using this high-resolution satellite data, and the conventional method used for the satellite data of 10 m and 20 m resolution is 1 m, There is a portion that cannot be used for satellite data having a resolution of 2 m, and there is a problem that shadows and structures cannot be classified by detecting edges of feature points for image processing.

FIG. 5 is a schematic block diagram of image processing of satellite data image information in a conventional example, and FIG. 6 is a flowchart of image processing of satellite data image information in a conventional example.
When the process is started (step S601), the satellite data 580 is input to the image input unit 510 of the satellite data processing device 500.
Is input (S602), the input image data is subjected to radiometric correction for correcting sensitivity such as variation caused by the characteristics of the individual imaging optical elements, and (S603), and the map is obtained by taking distortions due to satellite shake and the like. The pre-processing unit 620 performs a pre-process such as geometric correction for performing geometric correction to match the coordinates of the coordinates (S604), and performs edge processing on the corrected data by the edge extracting unit 540 to create an edge line (S605). ), Post-processing such as raster conversion and vector conversion is performed on the map original data from which edges have been extracted by the post-processing unit 550 (S606), and the generated map data 5
90 is output from the map data output unit 560 (S60).
7), and the process ends (S608).

[0005] In the edge extraction processing for making a figure,
If the influence of the shadow is large, the shadow is recognized as an edge, and there is a problem that a person must visually confirm and delete the shadow in order to mistake the shadow as a structure.

An object of the present invention is to provide a shadow removing method and a shadow removing means for high-resolution data which can automatically recognize a shadow of a structure from image information of satellite data and reduce the influence thereof.

[0007]

The shadow removing method for high resolution satellite data according to the present invention is a shadow removing method in image processing for creating map data based on high-definition satellite data captured by an earth observation satellite. Then, two types of image data of the same area captured at different times are compared, and a pixel having a luminance mismatch is determined as a shadow, and the image data is corrected.

[0008] A method for correcting image data is a method in which the first corrected image data and the second corrected image data which are imaged at different times in a desired area and preprocessed are performed for each same line. The brightness profiles are compared, and if the difference between the brightness of the first corrected image data and the brightness of the second corrected image data is not greater than a preset threshold based on the result of the comparison, the first correction is performed. The luminance value of the image data is selected. If the luminance value is larger than the threshold value, the luminance of the first corrected image data is compared with the luminance of the second corrected image data.
If the luminance value of the corrected image data is large, the luminance value of the second corrected image data is selected. If the luminance value of the second corrected image data is not large, the luminance value of the first corrected image data is selected. May include a process of creating composite image data with the obtained luminance value, and based on a result of the comparison, if the luminance of the first corrected image data is higher than the luminance of the second corrected image data, the first Select the brightness of the corrected image data,
If the brightness of the first corrected image data is not higher than the brightness of the second corrected image data, the process includes selecting the brightness of the second corrected image data and creating composite image data with the selected brightness value. May be.

It is preferable that the two types of image data of the same area to be compared are image data captured at different times on the same day.

A shadow removing device for high-definition satellite data according to the present invention is a shadow removing device used in a process for creating map data based on high-definition satellite data picked up by an earth observation satellite. And a corrected image input unit for inputting image data corrected by performing geometric processing, an image storage unit for storing the input corrected image data,
An image comparison unit that retrieves two types of image data of the same area captured at different times from the image storage unit and compares the luminance at the same position, and a difference in luminance between the two types of image data at the same position. A shadow judging unit for judging a low-luminance image as a shadow when the threshold value is exceeded, an image synthesizing unit for synthesizing image data from which the shadow has been removed, a control unit for controlling each unit to execute each processing, And a recording medium on which a control program for controlling the computer to execute the respective processes is recorded.

A satellite data processing apparatus according to the present invention is a satellite data processing apparatus for creating map data based on high-resolution satellite data captured by an earth observation satellite. A pre-processing unit that creates corrected image data from high-degree-of-image satellite data input by performing geometric processing, and performs edge processing on the synthesized image data created by removing shadows with a shadow processing device to perform edge processing. An edge extraction unit for extracting lines and a post-processing unit for creating map data by performing post-processing including raster conversion and vector conversion on the map original data from which the edge has been extracted.

The present invention is characterized in that, for example, data observed at different times on the same day is used, and a difference in the direction of a shadow is used.

[0013] By using the above-described method and apparatus, it is possible to use the data observed at different times on the same day and use the difference in the direction of the shadow, so that the influence of the shadow can be reduced. Thereby, when detecting the edge of the building, the phenomenon that the shadow is regarded as the edge of the building can be reduced.

[0014]

Next, embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a block diagram of a satellite data processing apparatus for creating a map provided with a shadow removing unit according to a first embodiment of the present invention. FIG. 2 is a flowchart of image processing using the satellite data processing apparatus of FIG. It is.

The satellite data processing apparatus 100 including the shadow removing means according to the first embodiment of the present invention comprises: an image input unit 110 for externally inputting high-resolution satellite data 180 obtained by capturing an image of the ground by an earth observation satellite; Before radiometric correction, which performs sensitivity correction for variations caused by the characteristics of individual imaging optical elements, and geometric correction, which performs geometric correction to match the coordinates of the map by taking the distortion caused by satellite sway etc. The pre-processing unit 120 that performs processing on the input image data is compared with two types of image data of the same region captured at different times, and a pixel having a luminance mismatch is determined as a shadow, and the image data is corrected. A shadow removing unit 130, an edge extracting unit 140 that performs edge processing on the composite image data created by removing the shadow to extract an edge line, and a map original data from which the edge is extracted. Post-processing unit 15 to create map data by performing post-processing such as raster conversion and vector converted into data
0 and a map data output unit 160.

The shadow removing unit 130 is different from a corrected image input unit 131 for inputting image data corrected by performing a radiometric process or a geometric process, and an image storage unit 132 for storing the input corrected image data. The two types of image data of the same area captured at the time are stored in the image storage unit 13.
2 and an image comparing unit 133 for comparing the luminance at the same position extracted from the second image data, and a shadow determination for judging an image with low luminance as a shadow when the difference between the luminances of two kinds of image data at the same position exceeds a predetermined threshold value. Unit 134, an image synthesizing unit 135 that synthesizes image data from which shadows have been removed, a control unit 138 that controls each unit to execute each process, and a control program that controls each unit to execute each process. And a recording medium 139 on which is recorded.

In this embodiment, a satellite data processing apparatus which executes pre-processing, shadow removal, edge extraction, and post-processing as a series of processing has been described. As data, the data is input to a processing device having a shadow removing unit and an edge processing unit and output as map original data, and post-processing may be executed by another processing device, and edge processing may be performed by another processing device. Even if executed, the object of the present invention can be executed.

Referring now to FIG. 2, a description will be given of a procedure for processing map-forming satellite data provided with the shadow removing means according to the first embodiment of the present invention.

When the process is started (S201), the ground image data 180 captured by the artificial satellite is input to the image input unit 110.
(S202), the pre-processing unit 120 performs radiometric correction (S203), performs geometric correction (S204), and generates corrected image data in accordance with the coordinates of the map. These processes can be executed using a known technique.

The corrected image data is input to the corrected image input unit 131 (S205) and stored in the image storage unit 132 (S205).
206).

From the corrected image data stored in the image storage unit 132, two types of corrected image data A and B of a desired area captured at different times A and B are read out to the image comparison unit 133. (S207). For example, the two types of corrected image data are as similar as possible in sunshine conditions such as morning and afternoon of the same day, and image data captured at a distant time can reduce errors in comparison. So desirable.

A luminance profile is compared for each of the two lines of the read two kinds of corrected image data A and B (S208). If the difference between the luminance of A and the luminance of B is not larger than a preset threshold (S209N) based on the result of the comparison, the luminance value of A is selected (S210). If it is larger than the threshold value (S209Y), the luminance of A is compared with the luminance of B. If the luminance value of A is not large (S211N), the luminance value of B is selected (S212), and the luminance value of A is increased. BA (S21
1Y), the luminance value of A is selected (S213), and the composite image data is created with the luminance value selected by the image compositing unit 135 (S214). If the comparison of the predetermined area is not completed (S215N), the next (S216), and returns to step S208 to continue the comparison.

When the comparison of the predetermined area is completed (S215)
Y), an edge is extracted from the completed combined image data by the edge extraction unit 140 to create an edge line (S21).
7) Create map original data. The post-processing unit 150 performs post-processing such as raster conversion and vector conversion, which are well-known techniques, on the map original data to create map data 190 (S
218), the map data output unit 160 sends the map data 19
0 (S219), and the process ends (S2).
20).

All the processing procedures of the satellite data processing apparatus 100 for creating a map provided with the shadow removing means have been described above. The shadow removing step of the present invention is from step S205 to step S216 surrounded by a dashed line in FIG. Yes,
Receiving the corrected image data that has been subjected to pre-processing, which is a well-known technique, performs shadow removal processing, outputs the combined image data, and the edge extraction and post-processing, which are well-known techniques, may be processed by another device. .

Next, the principle of shadow removal will be described. FIG.
3A and 3B are schematic diagrams for explaining the principle of shadow removal in satellite data according to the present invention, wherein FIG. 3A is a top view of a structure and a shadow at time A, and FIG. , (C) is the luminance profile of line M at time A, (d) is the luminance profile of line M at time B, the difference from the luminance profile of line M at time A, and the composite luminance profile, (e) is The luminance profile at the time A of the line N, the luminance profile at the time B of the line N, the difference from the luminance profile at the time A of the line N and the composite luminance profile, and (g) the composite image.

As can be seen from the luminance profile at the time A of the line M in FIG. 3C, the upper surface of the structure 311 has a high luminance, the road 313 has a somewhat low luminance, and the structure 311 has a low luminance. The shadow portion is displayed with a lower luminance. When the difference between the luminance profile at the time A of the line M and the luminance profile at the time B of the line M is obtained, there is almost no difference under the same condition, and thus the difference is seen in the luminance difference 333 in FIG. Thus, only the shadow 312 protrudes. Threshold 3 in advance
34, a portion exceeding the threshold value 334 is determined as a shadow. If the image at time A is used as a reference, the brightness of the image at time A is used when the difference 333 in brightness is equal to or smaller than the threshold 334, and the image at time A is used when the difference 333 in brightness exceeds the threshold. Is compared with the luminance of the image at time B, and the luminance value of the image with the higher luminance is adopted, whereby the line M with almost no influence of shadow as shown in FIG. A composite luminance profile 335 is obtained. FIG. 3 shows a similar process for line N.
A line N almost free from the influence of shadow as shown in FIG.
Is obtained. By performing such a process on each line in a predetermined area, the influence of the shadow is reduced, and data of the composite image 350 as shown in FIG. 3G can be created.

Next, a processing procedure for map-forming satellite data provided with a shadow removing means according to a second embodiment of the present invention will be described with reference to FIGS. In the second embodiment, the function of the shadow determination unit 134 in FIG. 1 is different from that in the first embodiment. FIG. 4 is a flowchart of the image processing of the second embodiment using the satellite data processing device of FIG.

Steps 401 to 408 are the same as steps 201 to 208 of FIG. 2 of the first embodiment, and steps 413 to 418 are steps 215 of FIG. 2 of the first embodiment.
To step 220, and a description thereof will be omitted.

In the second embodiment, step S408
And compare the luminance profile for each line. If the luminance at time A is greater than the luminance at time B (S409Y), the luminance at time A is adopted (S410), and the luminance at time A is greater than the luminance at time B. If not large (S409N), time B
Is adopted (S410), and composite image data is created (S412).

The processing method according to the second embodiment has the effect of eliminating the need for setting a threshold value and reducing the number of processing steps. There is a possibility that the continuity of the luminance profile is lower than that of the composite image according to the first embodiment.

The control program for the shadow removal step is transmitted from the recording medium 139 to the data processing device (not shown) of the control unit 138.
And controls the operation of the data processing device. The control unit 138 executes the following processing under the control of the control program.

That is, a process of inputting and storing corrected image data, and two types of corrected image data A and B of a desired region imaged at different times from the stored corrected image data.
Is read, and the two types of read-out corrected image data A and B are compared with each other for a luminance profile for each line. Based on the comparison result, the difference between the luminance of A and the luminance of B is determined in advance. If it is not larger than the set threshold value, the luminance value of A is selected. If it is larger than the threshold value, the luminance of A is compared with the luminance of B. If the luminance value of B is larger, the luminance value of B is selected. If the luminance value of A is large, a process of selecting the luminance value of A and a process of creating composite image data with the selected luminance value are executed.

[0033]

As described above, according to the present invention, two types of image information of the same area captured at different times are compared,
Since the high-resolution satellite data is corrected by judging that the pixels whose luminances do not match are shadows, there is an effect that a phenomenon in which the shadows are captured as edges in creating a map image can be reduced.

[Brief description of the drawings]

FIG. 1 is a block diagram of a satellite data processing apparatus for creating a map including a shadow removing unit according to a first embodiment of the present invention.

FIG. 2 is a flowchart of image processing using the satellite data processing device of FIG. 1;

FIG. 3 is a schematic diagram for explaining the principle of shadow removal in satellite data according to the present invention. (A) shows a top view of the structure and the shadow at time A. (B) shows a top view of the structure and the shadow at time B. (C) shows a luminance profile at the time A of the line M. (D) shows a luminance profile at the time B of the line M, a difference from the luminance profile at the time A of the line M, and a combined luminance profile. (E) shows a luminance profile at the time A of the line N. (F) shows a luminance profile at the time B of the line N, a difference from the luminance profile at the time A of the line N, and a combined luminance profile. (G) shows a composite image.

FIG. 4 is a flowchart of image processing according to a second embodiment using the satellite data processing device of FIG. 1;

FIG. 5 is a schematic block diagram of a conventional image processing of satellite data image information.

FIG. 6 is a flowchart of image processing of satellite data image information in a conventional example.

[Explanation of symbols]

100, 500 Satellite data processing device 110, 510 Image input unit 120, 520 Preprocessing unit 130 Shadow removal unit 131 Corrected image input unit 132 Image storage unit 133 Image comparison unit 134 Shadow judgment unit 135 Image synthesis unit 138 Control unit 139 Storage medium 140, 540 Edge extraction unit 150, 550 Post-processing unit 160, 560 Map data output unit 180, 580 Satellite data 190, 590 Map data 310 Image at time A 311, 321, 351 Structure 312, 322 Shadow 313, 323, 353 Roads 314, 324, 354 Level ground 320 Image at time B 331 Luminance at time A of line M 332 Luminance at time B of line M 333, 343 Difference in luminance 335, 345 Synthetic luminance profile 341 Luminance at time A of line N 342 line Brightness at time B of N 50 synthetic image S201~S220, S401~S418, S601~
S608 Step

Claims (7)

[Claims] 1. A shadow removal method in image processing for creating map data based on high-resolution satellite data captured by an earth observation satellite, comprising: two types of image data of the same area captured at different times. And correcting the image data by judging a pixel whose luminance does not match as a shadow, and correcting the image data. 2. A method for correcting image data, comprising the steps of: first and second correction image data, which are imaged at different times in a desired area and preprocessed, are the same; The luminance profiles are compared for each line, and based on the result of the comparison, the difference between the luminance of the first corrected image data and the luminance of the second corrected image data must be larger than a preset threshold value. For example, the luminance value of the first corrected image data is selected. If the luminance value is larger than the threshold value, the luminance of the first corrected image data is compared with the luminance of the second corrected image data. If the luminance value of the corrected image data is large, the luminance value of the second corrected image data is selected. If the luminance value of the second corrected image data is not large, the luminance value of the first corrected image data is selected. At the selected brightness value 2. The method of removing shadows from high-resolution satellite data according to claim 1, comprising a process of creating composite image data. 3. The method of correcting image data, wherein the first corrected image data and the second corrected image data which are imaged at different times in a desired area and preprocessed are the same. A luminance profile is compared for each line. If the luminance of the first corrected image data is higher than the luminance of the second corrected image data, the luminance of the first corrected image data is selected based on the result of the comparison. If the luminance of the first corrected image data is not higher than the luminance of the second corrected image data, the luminance of the second corrected image data is selected, and the composite image data is created with the selected luminance value. 2. The method of removing shadows from high-resolution satellite data according to claim 1, further comprising the step of: 4. The high resolution according to claim 1, wherein the two types of image data of the same region to be compared are image data captured at different times on the same day. Shadow removal method for satellite data. 5. A shadow eliminator used in a process of creating map data based on high-resolution satellite data captured by an earth observation satellite, wherein the image is corrected by performing a radiometric process or a geometric process. A corrected image input unit for inputting data, an image storage unit for storing the input corrected image data, and two types of image data of the same area captured at different times are taken out of the image storage unit and the brightness at the same position is obtained. An image comparing unit for comparing, a shadow judging unit for judging an image having low luminance as a shadow when a difference in luminance between two types of image data at the same position exceeds a predetermined threshold value; An image synthesizing unit for synthesizing, a control unit for controlling each unit to execute each process, and a recording medium storing a control program for controlling each unit and executing each process are provided. Shadow removal device of high imaging of satellite data, wherein Rukoto. 6. A shadow data removing device according to claim 5, wherein the satellite data processing device creates map data based on the high-resolution satellite data captured by an earth observation satellite. A pre-processing unit that creates corrected image data by performing radiometric processing or geometric processing on the data; and an edge line that performs edge processing on the composite image data created by removing shadows by the shadow processing device. Satellite data comprising: an edge extraction unit for extracting map data; and a post-processing unit for generating map data by performing post-processing including raster conversion and vector conversion on the original map data on which edge extraction has been performed. Processing equipment. 7. A recording medium storing a control program for a shadow removal process in high-resolution satellite data, comprising: a step of inputting corrected image data to a corrected image input unit and storing the corrected image data in an image storage unit; From the corrected image data stored in the storage unit, two types of first corrected image data and second corrected image data of a desired region imaged at different times are read out to an image comparing unit, and the two read out two types of corrected image data are read out. A procedure for comparing a luminance profile for each of the same lines for each of the first corrected image data and the second corrected image data, and a shadow determination unit based on a result of the comparison. If the difference between the luminance of the first corrected image data and the luminance of the second corrected image data is not larger than a preset threshold value, the luminance value of the first corrected image data is selected. For example, the luminance of the first corrected image data and the second
Is compared with the brightness of the corrected image data, and if the brightness value of the second corrected image data is large, the brightness value of the second corrected image data is selected, and the brightness value of the second corrected image data is increased. If not, a recording medium storing a program for executing a procedure for selecting a luminance value of the first corrected image data and a procedure for creating composite image data in the image compositing unit from the selected luminance value.