JP2008304260A - Image processing device - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To provide an image processing device equipped with a position and attitude angle calibration means instead of GPS when a GPS link is stopped for the purpose of normal detection of the position and the attitude angle, since, when a GPS radio wave is stopped during movement of an unmanned aircraft, error estimation of INS (Inertial Navigation System) cannot be performed, and resultantly the position and the attitude angle of the unmanned aircraft cannot be detected normally. <P>SOLUTION: An image database of an image acquired during movement is created by using an image acquisition means loaded on the unmanned aircraft, and when the GPS link is stopped, the image database is referred to. A correlation value between the image in the image database and a present image is calculated, and coordinate transformation is applied by using an image having high correlation as a reference value, to thereby calculate error estimation values of the position and the attitude angle of the unmanned aircraft. The acquired error estimation values of the position and the attitude angle are inputted into GPS/INS calculation, and the position and the attitude angle are detected. <P>COPYRIGHT: (C)2009,JPO&INPIT

Description

  The present invention relates to an image processing apparatus that uses an aerial image to correct the position and posture angle of a small airplane such as a light airplane or a small helicopter to generate an image over a specific area.

Aerial surveys using aircraft and satellites flying at high altitudes can obtain a wide range of information at once. Then, by using the aerial image captured during the aerial survey, a plurality of aerial images can be combined to generate a mosaic image covering a wide surface area. For this reason, there are many demands for aerial surveying, such as the construction of a geographic information system (GIS) using mosaic images. As a geographic information system often used for urban disaster prevention management, there is a spatial information GIS (Spatial Temporal GIS).
However, aerial surveys using aircraft and satellites are generally expensive. Small helicopters and light aircraft that can fly at low altitudes with low fuel costs and maintenance costs, as well as inertial sensors and computers are installed on these aircraft to meet the demands of reducing operational costs and acquiring high-resolution images. It has been studied to use a small aircraft (hereinafter, referred to as a small aircraft) such as a UAV (Unmanned Aero Vehicle) that has become autonomous (see, for example, Patent Document 1). Since a small airplane has a low ground altitude, the imaging range shown in an aerial image is narrow, and an aerial image of a desired area cannot be acquired by one imaging. For this reason, by associating the image taken by the high-resolution digital camera mounted on the small airplane with the position and attitude angle of the small airplane at the time of shooting, and overlapping the displayed images, The topographic information of a desired area can be acquired.

Conventionally, as a technology for detecting the position and attitude angle of a small airplane, an inertial sensor having high availability and continuity and a GPS intended to estimate and eliminate measurement errors due to an inertial sensor that increases with time. A combined GPS / INS (Global Positioning System / Internal Navigation System) combined positioning is used (for example, see Non-Patent Document 1).
In addition, an apparatus that accurately measures the global coordinates of an object to be imaged under adverse conditions where GPS cannot be used has been proposed (see, for example, Patent Document 2).
Further, in a survey system using GPS and autonomous navigation, a survey system that enables position measurement with a predetermined accuracy even in a place where GPS radio waves cannot be received has been proposed (for example, see Patent Document 3).

Japanese Patent Application No. 2007-17806 JP 2005-91298 A JP 2005-62083 A National Aerospace Laboratory Report “Development of Carrier Phase DGPS / INS Combined Navigation Algorithm” NAL-1416 ISSN: 0389-4010

  However, GPS using weak radio waves from satellites has a problem in continuity. For example, in aerial surveys where GPS radio waves are difficult to block, such as mountainous areas, mountainous areas, and valleys of buildings, it is possible to estimate and remove INS errors in GPS / INS combined positioning when GPS radio waves are received. However, when GPS radio waves cannot be received, INS errors are accumulated, and there is a problem that INS error estimation and removal cannot be performed completely. Therefore, there has been a problem that the accuracy of the ground image generated by combining the aerial images based on the position and the attitude angle of the small airplane deteriorates.

  The present invention has been made in order to solve such problems, and even when a small aircraft for surveying is surveyed, even when GPS radio waves from GPS satellites are interrupted, it is based on images taken by the small aircraft. An object of the present invention is to obtain an image processing apparatus that corrects the position and the posture angle and acquires an image over a specific area.

  An image processing apparatus according to the present invention is mounted on an aircraft that flies a plurality of times over a specific area, receives position signals from a positioning satellite and measures the position of the aircraft, and outputs position information of the aircraft. Compound calculation positioning unit that calculates the position and posture angle of the aircraft by combining the inertia information output from the inertia information detection unit that measures angular velocity and acceleration, the imaging unit that images the ground, and the imaging unit An image generation unit including an image storage unit that associates the captured image with the position and orientation angle of the aircraft calculated by the combined calculation positioning unit to generate and store a panoramic image of a specific area by pasting the captured image; When the position detection unit cannot determine the position of the airframe, the image generation unit is configured to generate the captured image captured by the imaging unit and the generated image. The position of the aircraft is estimated by collating with a norama image, and the combined calculation positioning unit performs a combined process of the estimated aircraft position and the inertia information from the inertia information detection means, and the position and orientation of the aircraft. The angle was calculated.

  According to the present invention, the position and attitude angle of a small airplane can be corrected even in a situation where positioning radio waves from a positioning satellite such as a GPS satellite cannot be received, and the sky captured by an imaging device mounted on the small airplane can be corrected. By synthesizing the captured images, it is possible to acquire a high-accuracy image over a specific area.

  In the following embodiments, an unmanned aerial vehicle (UAV: Unmanned Aerial Vehicle) will be described as an example of a small airplane. This unmanned aerial vehicle is equipped with various calculation units such as an autopilot calculation unit for performing autonomous flight, but only the part that is the gist of the invention will be described here. Further, although GPS will be described as an example of the positioning system, a positioning system other than GPS such as Galileo may be used.

Embodiment 1 FIG.
Embodiment 1 of the present invention will be described below with reference to the drawings.
FIG. 1 is a configuration diagram of an image processing apparatus 30 according to the first embodiment. The image processing device 30 mounted on the unmanned airplane 50 includes a GPS / INS calculation unit 14 and an image generation unit 15 that acquires an image of a desired area.

  In FIG. 1, a GPS / INS calculation unit 14 receives a radio wave from an attitude angle detection device 1 that measures inertia information such as angular velocity and acceleration of an unmanned aircraft 50 and a positioning satellite (GPS satellite) (not shown). The position detection device 2 that acquires the position of 50, the speed update calculation unit 4 that updates the speed information using the inertia information 90 output by the attitude angle detection device 1, and the speed information and correction that the speed update calculation unit 4 outputs. The position update calculation unit 5 that updates the position of the unmanned airplane 50 using the data and outputs the updated position and velocity, and the attitude angle update calculation unit 6 that updates the attitude angle of the unmanned airplane 50 using the inertia information 90. A filter calculation unit 11 that estimates and outputs an error of the position, speed, and attitude angle of the unmanned aircraft 50 calculated by the position update calculation unit 5 and the attitude angle update calculation unit 6, and each error estimated by the filter calculation unit 11 And a bias correction unit 7 and the attitude angle correction unit 8 and a speed correction unit 9 and the position correcting unit 10 for correcting the respective values based on the distribution.

The self-position correcting unit 15 performs an ortho-transform on the image captured by the image acquisition device 3 based on the information on the position, speed, and attitude angle of the unmanned airplane 50 output from the GPS / INS calculation unit 14 to generate an ortho image. Generating and storing the ortho image in association with position information on the ground surface, and synthesizing the ortho image with the ortho image obtained in the past to generate a panorama composite image, and image data and image of the image database 12 An image correlation calculation unit 13 that performs image correlation calculation to check the degree of coincidence of the captured images taken by the acquisition device 3 and estimates the position, speed, and attitude angle of the unmanned aircraft 50 is included.
Here, the image acquisition device 3 is, for example, a general-purpose high-resolution digital camera. The ortho conversion will be described later. The GPS / INS calculation unit corresponds to a combined calculation positioning unit, and the attitude angle detection device corresponds to inertia information detection means. The image acquisition device corresponds to a photographing unit, and the image information database corresponds to an image storage unit.

Next, an operation of acquiring a panoramic image of a desired area on the ground when the unmanned aircraft 50 autonomously flies and the image acquisition device 3 appropriately captures the ground surface will be described.
FIG. 2 is a diagram illustrating a relationship between a flight path using an unmanned aerial vehicle and an aerial image of a desired area. Autonomous flight can follow the flight path more accurately than manual flight. Therefore, as shown in FIG. 2, the unmanned aerial vehicle 50 can fly a plurality of times with close flight paths, and each still image (one or two images shown in FIG. 2, two or three images, or three or more) Image). The unmanned aerial vehicle 50 can turn with good reproducibility a circle or an elliptical orbit (flight path) over a desired area set in advance while suppressing disturbances such as wind based on the output of GPS and inertial sensors. is there.

The image acquisition device 3 of the unmanned aircraft 50 acquires an aerial image. Since the unmanned aerial vehicle 50 flies at low altitude, a high-resolution still image of the ground surface can be obtained using a commercially available digital camera. The image acquisition device 3 is fixed to the body of the unmanned aircraft 50 (unmanned aircraft). By fixing the image acquisition device 3 to the airframe, the position and posture angle of the unmanned aircraft 50 can be associated with the position and posture angle of the image acquisition device 3.
Hereinafter, the position and posture angle of the image acquisition device 3 are the same as the position and posture angle of the unmanned aircraft 50, and the coordinate system corresponding to the posture angle of the image acquisition device 3 is a camera coordinate system. Here, in the camera coordinate system, the X-axis is the central axis in the longitudinal direction of the unmanned aircraft 50, the Z-axis is vertically upward with the X-axis facing the horizontal direction, and the Y-axis is the left-handed three-axis orthogonal coordinate. Set to configure the system. Further, the attitude angle of the unmanned aerial vehicle 50 is defined by Euler angles (α, β, γ) necessary for conversion from the world coordinate system (for example, the earth center coordinate system WGS-84) to the body axis. That is, the rotation angles around the camera coordinate axes (X, Y, Z) are (α, β, γ) = (roll angle, pitch angle, yaw angle), respectively.

The position detection device 2 measures the position of the aircraft in flight. For example, a GPS receiver is used as the position detection device 2.
The attitude angle detection device 1 detects the attitude angle of the aircraft in flight. For example, an inertial sensor such as a gyroscope or an accelerometer is used as the attitude angle detection device 1.

Next, the operation of the GPS / INS calculation unit 14 will be described. The GPS / INS calculation unit 14 uses the time-series position data 80 from the position detection device 2 and the time-series inertia information 90 from the attitude angle detection device 1 to perform composite processing in the filter calculation unit 11. Calculate and output the position, speed and attitude angle of the aircraft with improved accuracy. The filter calculation unit 11 performs complex processing using, for example, a Kalman filter, and calculates and corrects various error estimation values.
The posture angle detection device 1 outputs the measured acceleration information of the inertia information 90 to the speed update calculation unit 4 and outputs the angular velocity information of the inertia information 90 to the posture angle update calculation unit 6. The speed update calculation unit 4 calculates the speed of the aircraft in flight and outputs it to the position update calculation unit 5.
The position update calculation unit 5 calculates the position and speed of the aircraft based on the acceleration information, and outputs the calculated position and speed. The calculated position and speed of the aircraft are fed back to the filter calculation unit 11.
The attitude angle update calculation unit 6 calculates the attitude angle of the aircraft based on the angular velocity information, and outputs the calculated attitude angle. The calculated attitude angle of the aircraft is fed back to the filter calculation unit 11.
The filter calculation unit 11 receives the position data 80 from the position detection device 2, the position and velocity data of the aircraft from the position update calculation unit 5, and the attitude angle data of the aircraft from the attitude angle update calculation unit 6. Then, the estimated bias error value (for example, the estimated value of the sensor error of the gyro sensor), the estimated value of the position error, the estimated value of the velocity error, and the estimated value of the attitude angle error of the posture angle detection device 1 are calculated. To do. Each of the bias correction unit 7, the posture angle correction unit 8, the speed correction unit 9, and the position correction unit 10 calculates the acceleration, angular velocity, and position update calculation output from the posture angle detection device based on the error estimation value from the filter calculation unit 11. The posture angle output from the position and speed of the unit and the posture angle update calculation unit 6 is corrected.

  On the other hand, the image information database 12 includes aerial image data captured by the image acquisition device 3 and flight state information data calculated by the GPS / INS calculation unit 14 (position information and speed information of the aircraft at each imaging). (Posture angle information) is time-synchronized and stored as image information 100 in time series. As a method of time synchronization, for example, a method of recording in accordance with the acquisition timing of each data on the basis of GPS time is used.

  FIG. 3 shows an example of the image information 100. Each image (image 1, image 2,..., Image n) photographed by the image acquisition device 3 includes the position information (X, Y, Z) of the airframe at the time of each photographing and the attitude of the airframe at the time of each photographing. The angle information (roll angle α, pitch angle β, yaw angle γ) and the shooting time are associated with each other and stored in the image information database 12.

  The image information database 12 has an image processing unit 121 therein. The image processing unit 121 performs ortho transformation on each image stored in the image information database 12 and stores the image in the image information database 12 as an ortho image. An ortho image is also called an orthographic projection image, which is an image obtained by correcting a centrally projected image and converting it to an orthographic projection image, which is an image as if taken from directly above. Further, an image obtained by partially overlapping a plurality of ortho images is called an ortho mosaic image or a panorama composite image.

  The image processing unit 121 performs image conversion (ortho conversion) of the captured image using the rotational translation formula of Formula 1.

The rotational translation system shown in Equation 1 performs image translation in addition to image rotation. FIG. 4 is a diagram showing the relationship between the camera coordinate system and the world coordinate system (for example, WGS-84).
In FIG. 4, the image processing unit 121 first displays an aerial image represented by a camera coordinate system XYZ (origin 0) corresponding to the posture of the camera on the imaging surface on the ground surface along the X′-Y ′ plane. Is projected onto the world coordinate system X′-Y′-Z ′ (origin 0), and an image representing the imaging range viewed from the vertical direction is generated. Then, it is translated by Tx, Ty, Tz. The parallel movement amounts Tx, Ty, and Tz are set so that the imaging range is positioned at the center of the rectangular (or square) image range after the image projection according to Equation 1 in FIG. It is preferable to determine the camera center 0 (0x ′, 0y ′, 0z ′) after translation so that the margin between the image range and the imaging range is reduced. By performing the ortho conversion, it is possible to obtain an aerial image as if taken from an aircraft or an artificial satellite that performs level flight.

  The image processing unit 121 uses the converted ortho image and the position coordinates (center position coordinates of the terrain image captured in the ortho image) as the center of the ortho image (rectangle or square) as image information 101 as image information. Store in database 12. An example of the ortho image information 101 is shown in FIG.

  The panorama composite image is created based on the ortho image information 101 (ortho image and center position coordinates of the ortho image in the world coordinate system) so that the ortho image can be sequentially superimposed on the world coordinate system. FIG. 6 is a diagram showing the relationship between the panorama composite image and the ortho image.

  In this way, while the unmanned airplane 50 is able to receive GPS radio waves, the accumulated INS error is estimated and removed to obtain highly accurate position information and attitude angle information of the unmanned airplane 50 to obtain a desired area. High-accuracy panoramic images can be acquired.

Next, an operation until the unmanned airplane 50 corrects the position and the attitude angle and acquires a panoramic image of a desired area when the GPS radio wave is interrupted while the unmanned airplane 50 is flying will be described with reference to a flowchart. .
As described above, in a situation where GPS radio waves cannot be received while moving, the GPS / INS calculation unit 14 cannot remove the accumulated INS sensor error, and therefore, the aerial image captured by the imaging acquisition device and the ground There arises a problem that an error occurs in the correspondence with the position and the accuracy of the panorama composite image is deteriorated.

FIG. 7 shows the position and attitude angle of the unmanned airplane 50 regardless of the position information output by the position detection device 2 in a situation where the GPS radio wave from the GPS satellite cannot be received due to the positional relationship between the satellite in the sky and the unmanned airplane 50. It is a flowchart explaining the operation | movement which correct | amends.
Here, as an example of generating a panorama composite image, it is assumed that a new image of an area in the vicinity of an area where a panorama composite image has been acquired in the past as shown in FIG. 8 (area indicated by a dotted line in FIG. 8) is acquired. .

In FIG. 7, the image correlation calculation unit 13 detects the positioning result output from the position detection device 2. When the image correlation calculation unit 13 detects that there is no output of the positioning result from the position detection device 2 based on the reception status of the GPS radio wave (S101 in FIG. 7), the image generation unit 15 starts the correction operation.
The image correlation calculation unit 13 outputs an instruction signal for storing the aerial image captured by the image acquisition device 3 in the database to the image information database 12 (S102).
Next, the image correlation calculation unit 13 performs image correlation calculation between the aerial image captured by the image acquisition device 3 and the panorama composite image, and calculates a correlation value indicating the degree of matching of the images (S103). . For the image correlation calculation, a general image processing method such as an inter-image difference can be used. When calculating the correlation value, the position of the current acquired image on the ground surface is estimated based on the current absolute position / attitude angle information (position / angle data output by the GPS / INS calculation unit 14). The estimated position is set as the initial position when calculating the correlation value. With this initial position as the center, the image is biased / rotated by a preset value in the rotation direction about the north direction, the east direction, and the ground surface normal, and the correlation value is calculated each time.
Next, the largest correlation value among the obtained correlation values is compared with a preset set value. If the correlation value is greater than or equal to the set value (S104), the captured aerial image is an aerial image obtained by capturing a part of the panorama composite image that has already been synthesized and created. The position coordinates of the four corners of the area in the panorama composite image where the aerial images match are extracted as the matching points (S105).
Here, the position coordinates of the four corners of the coincidence area obtained in S105 are set as [Xi, Yi, 0] (i = 1 to 4). Also, the body position coordinates to be obtained are set as [P1, P2, P3], and the attitude angle vectors to be found are set as [D1, D2, D3]. A camera parameter that associates a unit vector in the direction of the posture angle and the four corners of the camera is [Cij] (i = 1 to 4, j = 1 to 3, i is the four corners and j is an index corresponding to the direction). The camera parameter [Cij] is a parameter that depends on the angle at which the camera is attached to the aircraft and the viewing angle of the camera, and becomes a known value by measuring in advance. The relational expression of various quantities is shown in Equation 2.

In the formula (2), [Ui] is used as a calculation unknown coefficient. If the known quantities are [Xi, Yi, Cij], the equations shown in Equation 2 can be solved for [P1, P2, P3, D1, D2, D3, Ui] by using these known quantities.
From the above, the position and attitude angle of the aircraft are calculated. (S106)
Next, the image correlation calculation unit 13 outputs the calculated position and orientation angle to the filter calculation unit 11 (S107).
The filter calculation unit 11 inputs the position and attitude angle data from the image correlation calculation unit 13 as an alternative to the position data output by the position detection device 2 when GPS radio wave reception is possible, and performs the filter calculation described above. Then, an error estimated value is calculated.

On the other hand, when the correlation value does not satisfy the condition in S105, the image acquisition device 3 considers that the area not included in the panorama composite image is captured, and the image correlation calculation unit 13 calculates the position and orientation angle as the filter calculation unit 11. The filter calculation unit 11 does not perform error estimation.
When photographing by the unmanned aircraft 50 is continued, the process returns to S101 again, and the processing steps of S101 to S108 are repeated.
If the GPS radio wave can be received in S101, the position detection device 2 outputs the measured position to the filter calculation unit 11 as usual (S108).

As described above, in the first embodiment, when GPS radio waves are received, various error estimations are performed using the filter configuration including the attitude angle detection device 1, the position detection device 2, and the filter calculation unit 11, and correction is performed by error estimation. A panorama composite image is acquired using the position data.
On the other hand, when the GPS radio wave cannot be received, the image correlation calculation unit checks the correlation between the captured aerial image and the existing panoramic composite image, and when the correlation is high, the unmanned aircraft 50 positions and posture angles were calculated and output to the filter calculation unit. Various error estimations are performed by a filter configuration including the attitude angle detection device 1, the image correlation calculation unit 13, and the filter calculation unit 11, and high-accuracy position, velocity, and attitude angle data can be obtained by error estimation. I made it.
Thereby, even when GPS radio waves cannot be received, the position and attitude angle of the aircraft can be corrected, and a highly accurate panoramic image can be acquired.

Embodiment 2. FIG.
In Embodiment 1, the correlation between the aerial image acquired by the image acquisition device and the panorama composite image is calculated. However, a landmark whose absolute position is known in advance is captured in the aerial image acquired from the image acquisition device. If so, the position and attitude angle of the unmanned aerial vehicle can be estimated based on the absolute position of the landmark. In the second embodiment, the position and attitude angle of the unmanned aircraft are estimated based on the absolute position of the landmark shown in the aerial image.

  FIG. 9 is a diagram showing a configuration of the image processing apparatus 30 according to the first embodiment of the present invention. The image generation unit 15 according to the second embodiment includes a landmark database 16 in which a landmark whose absolute position is known in advance and data in which the absolute position is associated are stored, and a landmark match calculation unit 17. Landmarks are characterized by their appearance such as high-rise buildings, bridges, airports, and grounds. Note that the same or corresponding parts as those in the first embodiment are denoted by the same reference numerals and description thereof is omitted.

FIG. 10 is a flowchart for explaining the operation of correcting the position and posture angle in a situation where GPS radio waves cannot be received in the second embodiment.
In FIG. 10, the image correlation calculation unit 13 detects the positioning result output from the position detection device 2. When the image correlation calculation unit 13 detects that there is no positioning result output from the position detection device 2 based on the reception status of the GPS radio wave (S201), the image correlation calculation unit 13 starts the correction operation by the image generation unit 15.
The image correlation calculation unit 13 outputs an instruction signal for storing the aerial image captured by the image acquisition device 3 in the database to the landmark database 16 (S202).
In FIG. 10, in the second embodiment, a landmark is searched from an aerial image (S203). If a landmark is shown, the absolute position of the landmark is extracted from the landmark database 16 (S204). .
The image generation unit 15 calculates the position and attitude angle of the unmanned aircraft 50 from the absolute position of the landmark obtained in S205 (S205).
Here, the absolute position coordinates of the landmark obtained in S205 are set as [Xi, Yi, 0] (i = 1 to 4). Also, the body position coordinates to be obtained are set as [P1, P2, P3], and the attitude angle vectors to be found are set as [D1, D2, D3]. When the position coordinates of the landmark on the image are set as pixel points [Gi, Hi], camera parameters [Cij] (i = 1 to 4, j = 1 to 3) for associating the posture angle with a unit vector in the pixel point direction. , I is a landmark number, j is an index corresponding to a direction), and is expressed as a function of the pixel point [Gi, Hi] (Equation 3). Equation 3 shows the relational expression of the above various quantities.

Further, in the calculation formula of Equation 3, [Ui] is used as an unknown coefficient for calculation. Using the known quantities [Xi, Yi, Gi, Hi] and using them, the equation shown in Equation 3 can be solved for [Cij, P1, P2, P3, D1, D2, D3, Ui].
The position and attitude angle of the aircraft are calculated from the above (S206).

As described above, in the second embodiment, when the GPS radio wave cannot be received, the landmark coincidence calculation unit 17 determines whether or not a landmark whose absolute position is known in advance is photographed in the ortho image. Is taken, the position and attitude angle of the unmanned aerial vehicle 50 are calculated from the absolute position of the landmark and output to the filter calculation unit. Various error estimations are performed by a filter configuration including the attitude angle detection device 1, the image correlation calculation unit 13, and the filter calculation unit 11, and high-accuracy position, velocity, and attitude angle data can be obtained by error estimation. I made it.
As a result, even when GPS radio waves cannot be received, the position and orientation angle can be corrected, and a highly accurate panoramic image can be acquired.

2 is a diagram illustrating a configuration of image processing according to Embodiment 1. FIG. It is a figure explaining the relationship between the flight path | route using the unmanned aircraft in Embodiment 1, and the aerial image of a desired area. 6 is a diagram illustrating an example of image information 100 according to Embodiment 1. FIG. FIG. 3 is a diagram showing a relationship between a camera coordinate system and a world coordinate system in the first embodiment. 6 is a diagram illustrating an example of ortho image information 101 according to Embodiment 1. FIG. 6 is a diagram illustrating a relationship between a panoramic composite image and an ortho image in Embodiment 1. FIG. FIG. 6 is a flowchart illustrating an operation for correcting a position and an attitude angle in a situation where GPS radio waves cannot be received in the first embodiment. 5 is a diagram illustrating an area for acquiring an image in Embodiment 1. FIG. 6 is a diagram illustrating a configuration of an image processing apparatus according to Embodiment 2. FIG. FIG. 10 is a flowchart illustrating an operation for correcting a position and an attitude angle in a situation where GPS radio waves cannot be received in the second embodiment.

Explanation of symbols

DESCRIPTION OF SYMBOLS 1 Posture angle detection apparatus, 2 Position detection apparatus, 3 Image acquisition apparatus, 4 Speed update calculation part, 5 Position update calculation part, 6 Posture angle update calculation part, 7 Bias correction part, 8 Posture angle correction part, 9 Speed correction part DESCRIPTION OF SYMBOLS 10 Position correction | amendment part, 11 Filter calculation part, 12 Image database, 14 GPS / INS calculation part, 15 Image generation part, 16 Landmark database, 17 Landmark coincidence calculation part, 30 Image processing apparatus, 50 Unmanned aircraft, 80 position Data, 90 inertia information, 100 image information, 101 ortho image information, 121 image processing unit

Claims (4)

It is mounted on the aircraft that flies multiple times over a specific area,
The position information output from the position detection means for receiving the radio wave from the positioning satellite and measuring the position of the aircraft is combined with the inertia information output from the inertia information detection means for measuring the angular velocity and acceleration of the aircraft. A compound calculation positioning unit for calculating the position and posture angle of the aircraft,
A panorama image of a specific area is generated by associating the photographing image which is photographed by the photographing means, the photographed image photographed by the photographing device and the position and posture angle of the aircraft calculated by the combined calculation positioning unit, and the photographed image. An image generation unit including an image storage unit that stores the image storage unit,
When the position detection unit cannot determine the position of the aircraft, the image generation unit estimates the position of the aircraft by comparing the captured image captured by the imaging unit with the generated panoramic image, and combines An image processing apparatus characterized in that the calculation positioning unit performs combined processing of the estimated position of the aircraft and the inertia information from the inertia information detection means to calculate the position and attitude angle of the aircraft. The image generation unit performs an image correlation calculation to check a degree of coincidence between the captured image captured by the photographing unit and the panoramic image, and an area where a correlation value indicating the degree of coincidence of the images is equal to or larger than a predetermined set value is selected. The image processing apparatus according to claim 1, wherein when extracted from an image, position coordinates of the four corners of the area are acquired, and the position of the aircraft is estimated based on the position coordinates of the four corners. 2. The position detection unit, when detecting a landmark whose position is known in advance in a photographed image photographed by the photographing unit, estimates the position of the aircraft based on the position of the landmark. Image processing apparatus. The image generation unit ortho-transforms the captured image using the position and attitude angle output by the combined calculation positioning unit to acquire an ortho image, and combines the ortho image to generate the panoramic image. The image processing apparatus according to claim 1, wherein the image processing apparatus is an image processing apparatus.

Images