JP5713939B2 - Target detection apparatus and target detection method - Google Patents

Description

  Embodiments described herein relate generally to a target detection apparatus and a target detection method that are mounted on a flying object and detect a target from a captured image.

  Generally, in order to detect a target, a target detection device using an imaging device is often used for a guidance device mounted on a flying object. In this target detection device, a portion having a feature such as a brightness level equal to or higher than a specified value is extracted as a feature point from an image taken by the image pickup device. Further, each velocity vector is obtained from each extracted feature point, and based on the velocity vector, a feature point that moves differently from the background is selectively detected as a target, thereby detecting a small moving target.

JP 2003-67754 A

  However, in the target detection device mounted on a flying object, when the ground background of the depression angle is imaged by the imaging device to detect a low altitude target, the background image becomes complicated, and the movement due to the movement of the flying object at each feature point Parallax occurs. For this reason, the velocity vector is not uniform, and it becomes difficult to detect a small moving target using the velocity vector extracted from the complex background image in flight.

  Therefore, the present embodiment has been made in view of the above problems, and provides a target detection device and a target detection method that can reliably detect a target even from a background image captured during flight. With the goal.

  According to the present embodiment, the target detection device is mounted on a flying object and detects a target from a captured image of an imaging device that is flying. In the target detection device, image processing means, target information acquisition means, and flying object An information acquisition means, a flying object position / speed detection means, a target speed vector estimation means, and a target selection means are provided. The image processing means extracts a characteristic part from the captured image of the imaging device as a feature point, calculates a gravity center position of each of the extracted feature points, and a movement direction of the gravity center position due to the movement of the flying object And the velocity vector of each feature point is calculated from the movement interval. The target information acquisition means acquires the target position information and speed information. The flying object information acquisition means acquires position information and posture information of the flying object. The flying object position / speed detecting means calculates the position, moving direction and moving interval of the flying object from the position information and attitude information of the flying object, and detects the speed of the flying object from these information. The target speed vector estimation means estimates the target speed vector from the target position / speed information and the flying object position / speed information. The target selection means compares the speed vector calculated by the image processing means with the estimated target speed vector, and selects the feature point at which the estimated target speed vector is closest to the calculated speed vector. Then, the selected feature point is detected as a true target.

It is a block diagram which shows the structure of a guidance device provided with the target detection apparatus which concerns on this embodiment. In the embodiment of FIG. 1, it is a figure which shows an example of the background image imaged during flight. It is a flowchart which shows the target detection method of the target detection apparatus which concerns on embodiment of FIG.

  Hereinafter, embodiments will be described with reference to the drawings.

  FIG. 1 is a block diagram illustrating a configuration of a flying object-mounted guidance device including a target detection device according to the present embodiment.

  In FIG. 1, an imaging device 1 images an indication direction using an optical system 2. The captured image is sent to the image processing unit 3. The imaging device 1 and the optical system 2 are mounted on the gimbal 4.

  The gimbal 4 sends the gimbal angle detected by the gimbal angle detection unit 5 to the gimbal control unit 6 and controls the imaging direction of the imaging apparatus 1 in accordance with an instruction from the gimbal control unit 6. The gimbal angle detected by the gimbal angle detection unit 5 is sent to the target speed vector estimation unit 11.

  The image processing unit 3 extracts, as feature points, a region having a luminance level that is equal to or greater than a specified value from the captured image of the imaging device 1, calculates the centroid position of each feature point, and obtains an error angle from the center of the visual field. Furthermore, the velocity vector (including motion parallax) of each feature point is calculated from the movement direction and the movement interval of the gravity center position of each feature point due to the movement of the flying object. The barycentric position and velocity vector of each feature point are sent to the target velocity vector estimation unit 11 and the target selection unit 12, respectively.

  The target information acquisition unit 8 acquires target position information and velocity information sent from the launching device 7 that launches the flying object. The acquired target position information and speed information are sent to the target speed vector estimation unit 11 and the guidance control unit 13, respectively.

  The flying object information acquisition unit 9 acquires the position information and speed information of the flying object sent from the launching device 7. The acquired target position information and speed information are sent to the flying object position / speed detection unit 10.

  The flying object position / velocity detection unit 10 detects the position, movement direction, and movement interval of the flying object from the position information and posture information of the flying object acquired by the flying object information acquisition unit 9, and detects each of them. The speed of the flying object is detected from the result. The detected position and speed of the flying object are sent to the target speed vector estimation unit 11 and the guidance control unit 13, respectively.

  The means for sending information to the target information acquisition unit 8 and the flying object information acquisition unit 9 is not limited to the launching device 7.

  The target velocity vector estimation unit 11 is an eye that is the direction in which the flying object is actually facing based on the barycentric position of each feature point calculated by the image processing unit 3 and the gimbal angle detected by the gimbal angle detection unit 5. The gaze angle is calculated. The target velocity vector estimation unit 11 considers the parallax of each feature point at the calculated visual line angle, and is detected by the target position / velocity information acquired by the target information acquisition unit 8 and the flying object position / speed detection unit 10. The target velocity vector is estimated from the position and velocity information of the flying object. The estimated target velocity vector is sent to the target selector 12.

  The target selection unit 12 compares the velocity vector of each feature point calculated by the image processing unit 3 with the target velocity vector estimated by the target velocity vector estimation unit 11, and the estimated target velocity vector is calculated. Select the feature point closest to the velocity vector. The target selection unit 12 detects the selected feature point as a true target.

  The guidance control unit 13 generates a steering signal for guiding the flying object toward the target detected by the target selection unit 12, and controls the steering device 14 based on the generated steering signal. The propulsive force for flying is obtained from the propulsion device 15. In addition, the guidance control unit 13 issues an instruction to the gimbal control unit 6 so that the imaging direction of the imaging device 1 mounted on the gimbal 4 faces the target detected by the target selection unit 12.

  Next, a method for detecting the target of the guidance device having the above configuration will be described with reference to FIGS.

  FIG. 2 is a diagram illustrating an example of a ground background image captured during flight, and FIG. 3 is a flowchart illustrating a target detection method of the target detection device.

  In the initial guidance, the target detection device detects the target information (position, velocity vector, expected meeting position, etc.) acquired by the target information acquisition unit 8 and the flying object position / speed detection unit 10 before the target detection. The flying object position and speed information (rate and acceleration) are obtained, and based on the information, the flying object is caused to fly in the expected meeting direction shown in FIG. Further, the imaging device 1 on the gimbal 4 is directed in the target direction. Then, after letting it fly to a predetermined position by a flying program, target detection is started.

  Target detection is performed by image processing of the image processing unit 3. That is, when the image picked up by the image pickup apparatus 1 is input, the image processing unit 3 extracts feature points from the picked-up image, and calculates the centroid position and velocity vector of each feature point. (Step S1). Here, an error occurs in the directing direction of the gimbal 4. For this reason, since the target position is not necessarily the center of the image, the visual line angle is calculated from the barycentric position of each feature point and the gimbal angle detected by the gimbal angle detection unit 5 (step S2). Next, the target velocity vector is estimated from the target information and the flying object position / velocity information in consideration of the parallax of each feature point based on the calculated visual line angle (step S3). The estimated target velocity vector and the calculated velocity vector are compared to select the closest feature point (step S4), and the position / velocity information of the selected feature point is detected as a true target. Output (step S5). Thereby, the target is detected and the flying object is guided to the target.

  As described above, according to the target detection device according to the present embodiment, a target can be reliably detected even if a background image captured during flight is a complex image of the ground surface.

  As mentioned above, although embodiment was described, this embodiment is shown as an example and is not intending limiting the range of invention. This embodiment can be implemented in various other forms, and various omissions, replacements, and changes can be made without departing from the spirit of the invention. This embodiment and its modifications are included in the scope of the present invention and the gist thereof, and are also included in the invention described in the claims and the equivalent scope thereof.

  DESCRIPTION OF SYMBOLS 1 ... Imaging device, 2 ... Optical system, 3 ... Image processing part, 4 ... Gimbal, 5 ... Gimbal angle detection part, 6 ... Gimbal control part, 7 ... Launching device, 8 ... Target information acquisition part, 9 ... Flying object Information acquisition unit, 10 ... flying object position / speed detection unit, 11 ... target speed vector estimation unit, 12 ... target selection unit, 13 ... guidance control unit, 14 ... steering device, 15 ... propulsion device.

Claims (3)

In a target detection device that is mounted on a flying object and detects a target from a captured image of an imaging device in flight,
Extract a characteristic part from the captured image of the imaging device as a feature point, calculate the center of gravity position of each extracted feature point, from the movement direction and the movement interval of the center of gravity position by the movement of the flying object, Image processing means for calculating a velocity vector of each feature point;
Target information acquisition means for acquiring the target position information and speed information from the information transmission means for transmitting the position information and speed information of the target and the position information and posture information of the flying object to the flying object ;
From the information transmission means, the flying object information acquisition means for acquiring the position information and posture information of the flying object,
Calculate the position, moving direction and moving interval of the flying object from the position information and attitude information of the flying object, and a flying object position / speed detecting means for detecting the speed of the flying object from these information,
Target speed vector estimation means for estimating the target speed vector from the target position / speed information and the flying object position / speed information;
The speed vector calculated by the image processing means and the estimated target speed vector are compared, the feature point at which the estimated target speed vector is closest to the calculated speed vector is selected, and the selected feature And a target selecting means for detecting a point as a true target. Comprising the imaging device, a gimbal on which the imaging device is mounted, and a gimbal angle detecting means for detecting a gimbal angle indicating a directing direction of the gimbal,
The target velocity vector estimation means calculates the visual line angle of the imaging device from the gimbal angle and the barycentric position of each feature point, and when estimating the target velocity vector, the respective features based on the calculated visual line angle. The target detection apparatus according to claim 1, wherein a point parallax is taken into consideration. In the target detection method of the target detection device that is mounted on the flying object and detects the target from the captured image of the imaging device in flight,
Extracting a characteristic part from the captured image of the imaging device as a feature point;
Calculate the center of gravity position of each extracted feature point,
From the movement direction and movement interval of the center of gravity position due to the movement of the flying object, calculate the velocity vector of each feature point,
From the information transmitting means for transmitting the position information and speed information of the target and the position information and attitude information of the flying object to the flying object, the position information and speed information of the target are acquired,
From the information transmission means, obtain the position information and attitude information of the flying object,
Calculate the position, movement direction and movement interval of the flying object from the position information and attitude information of the flying object,
Detecting the speed of the flying object from the calculated position, moving direction and moving interval of the flying object,
From the position / velocity information of the target and the position / velocity information of the flying object, the velocity vector of the target is estimated,
Comparing the estimated target velocity vector and the calculated velocity vector to select the closest feature point;
A target detection method comprising: detecting the selected feature point as a true target.

Images