A four-degree-of-freedom vision guidance method based on a planar redundant target point set

Abstract

The present application relates to a kind of four degrees of freedom vision guidance method based on plane redundant target group.The steps include: installing a downward shooting camera at the end of the Z axis of cartesian coordinate robot, and installing a set of upward circular vision target group on the target side;The distribution of all target points in the target group meets certain constraints, especially all target points must be located on the same spatial plane and the center coincides with the center of the target;Internal parameters of camera are obtained by calibration, and target point template is established by shooting calibration board image and target point image in the same pose;After the establishment of template, a single target picture containing target points is shot using camera in any pose to realize the three-dimensional pose measurement of target.The target group has redundant design, and a small amount of target damage does not affect the positioning accuracy;When there is damaged target, the method can give the number and number information of damaged target;The method can quickly and stably complete the three-dimensional pose measurement of target, and provide motion guidance for other mechanical structures.

Description

Technical Field

[0001] This invention relates to a four-degree-of-freedom visual guidance method based on a planar redundant target group, belonging to the field of machine vision. Background Technology

[0002] Robots are widely used in industry, replacing manual labor for repetitive or physically demanding tasks and significantly improving production efficiency. In some robot applications, the target's position is uncertain, and high-performance vision-guided methods are a key technology for the successful application of robots in these scenarios.

[0003] Common visual guidance solutions can be divided into 2D and 3D. 2D solutions can only visually locate targets moving within a single plane. 3D solutions can locate targets in three-dimensional space, typically using 3D sensors or other 3D imaging technologies. However, common 3D sensors are expensive, such as laser-based 3D scanners; or require significant space, such as binocular cameras; or have insufficient accuracy, such as TOF-based 3D cameras. The algorithms for processing the aforementioned 3D data generally involve large computational loads and demand high processor performance.

[0004] Some 3D vision guidance methods can also achieve guidance based on a single camera and a custom target, but these targets require high precision in their fabrication, and guidance cannot be completed if the target is damaged. Summary of the Invention

[0005] This invention provides a four-degree-of-freedom visual guidance method based on a planar redundant target point set. This method can achieve high-precision four-degree-of-freedom visual guidance using a single camera and a low-cost visual target point set. When some target points are damaged, the success rate and accuracy of positioning are not affected, and a warning can be given for damaged target points.

[0006] The technical solution adopted by the present invention to achieve the above objectives is as follows:

[0007] A four-DOF visual guidance method based on planar redundant target sets includes the following steps:

[0008] 1) Calibrate the camera and obtain its intrinsic parameters;

[0009] 2) Install the camera and the circular visual target assembly on the end of the Cartesian coordinate robot and the target surface, respectively;

[0010] 3) Use tool T1 to perform hand-eye calibration on the Cartesian coordinate robot and camera to obtain the relative pose relationship between the camera coordinate system and the robot coordinate system;

[0011] 4) Perform a motion teaching exercise to obtain the pose relationship between the tool coordinate system and the robot coordinate system of the end effector T2;

[0012] 5) Use a visual calibration board to create target group image templates;

[0013] 6) During the online operation phase, the Cartesian coordinate robot moves to the working position, the camera takes a single image containing the target point group, extracts the outline of the target point group, and calculates its three-dimensional pose accordingly.

[0014] It also includes the following steps:

[0015] 7) If the target group is partially damaged, output its three-dimensional pose normally based on step 6), and identify the number and number of damaged targets.

[0016] The circular visual target group consists of four target points located on the same spatial plane, with the center of the target group coinciding with the center of the target. The target points are circular, highly reflective gray, and surrounded by a fixed-width black static area. A two-dimensional coordinate system is constructed based on the spatial plane where the target group is located. The center point of the target group is taken as the origin, and the target point directly above it is marked as target point 1. The other target points are marked as target points 2-4 in a clockwise order. The two-dimensional coordinates of target points 1-4 are (0,L), (L,0), (0,-L), and (-L,-L / 2), respectively, where L is the distance from target point 1 to the center of the target group.

[0017] Step 3) includes the following steps:

[0018] 3.1) Fix the calibration tool T1 to the end of the robot's Z-axis, place the calibration plate on the surface of the positioning target, and make the center of the calibration plate coincide with the center of the positioning target. At this time, the centers of the calibration plate, the target group, and the positioning target coincide.

[0019] 3.2) Keep the calibration plate in a fixed position, move the robot to take multiple images of the calibration plate, and record the coordinates of the robot's end effector T1 at each time the image is taken. During the shooting process, keep the calibration plate from moving.

[0020] 3.3) Substitute multiple sets of circular dot matrix calibration plate images and their corresponding coordinates into the hand-eye model, and use a nonlinear optimization algorithm to obtain the pose relationship between the camera coordinate system and the tool coordinate system.

[0021] The hand-eye model is as follows:

[0022]

[0023] Where camera_H_cal is the transition matrix from the camera coordinate system to the calibration board coordinate system, and camera_H_tool1 is the transition matrix from the camera coordinate system to the tool T1 coordinate system. is the inverse of the transition matrix from the robot's base coordinate system to the tool's T1 coordinate system, and base_H_cal is the transition matrix from the base coordinate system to the calibration board coordinate system.

[0024] Step 4) includes the following steps:

[0025] 4.1) Keeping the calibration board pose unchanged, obtain the relationship between the calibration board coordinate system and the base coordinate system base_H_cal based on camera_H_tool1, and then obtain the pose of the calibration board in the base coordinate system Pcal=(Xcal,Ycal,Zcal,Rcal).

[0026] 4.2) Translate the calibration plate pose along its z-axis by a distance L t The target group pose is obtained as Pobj = (Xobj,Yobj,Zobj,Robj), where L t To calibrate the plate thickness;

[0027] 4.3) Remove the calibration plate, move the robot and bring tool T2 to the target position so that tool T2 can complete the action normally, and record the robot's posture P at this time. T20 =(XT20,YT20,ZT20,RT20);

[0028] 4.4) Since the target group coincides with the target center, calculate the attitude P from the center point of the Z-axis end of tool T2. T2 :

[0029]

[0030] Among them, P obj For the target group pose, For robot posture P T20 The reverse.

[0031] The relationship between the calibration board coordinate system and the base coordinate system, baseH_cal, obtained based on camera_H_tool1, is as follows:

[0032]

[0033] Step 5) includes the following steps:

[0034] 5.1) Keeping the target pose unchanged, move the robot to a suitable position and take a picture of the target group I. obj This ensures that the image contains all target points of the target group;

[0035] 5.2) Place the calibration plate back on the target, ensuring that the centers of the calibration plate, the target group, and the target coincide, and take an image of the calibration plate. cal ;

[0036] 5.3) Image Recognition I cal The calibration plate in the three-dimensional pose P cal , will P cal The target reference pose P is obtained by translating a distance T along the z-axis. obj T is the thickness of the calibration plate;

[0037] 5.4) Following the target numbering order, in I obj The target area is selected sequentially, and adaptive binarization is performed on the selected area to extract the center circle region of the target point. The contour of the center circle region is then fitted with an ellipse, and the fitted contour C is... model As a template outline;

[0038] 5.5) Define the possible range of changes in the target group template contour, including rotation angle, image scaling, target contrast, and image pyramid hierarchy, and then create the target contour template, setting the reference pose of the contour template to P. obj .

[0039] Step 6) specifically refers to:

[0040] A contour-based template matching algorithm is used to search for target point templates in the image. If no target is found, the output score = 0, indicating that target localization has failed; if a target is found, let P′ = 0. obj Representing the target pose, the template contour C model Based on P′ obj Projecting onto the image plane displays the matching effect and outputs a score between 0.5 and 1. Moving the robot to P′ causes the end effector to reach P′. obj To complete the action, P′ specifically means:

[0041] P′=P′ obj ·P T2

[0042] Among them, P′ bj For the target pose, P T2 The position of tool T2 relative to the center of the end flange.

[0043] Step 7) includes the following steps:

[0044] 7.1) Based on the pose obtained from target localization, project the template contour onto the image containing the target group;

[0045] 7.2) Expand the outline of each projected target point using a circle as the core to form a ring-shaped region;

[0046] 7.3) Segment the original image within the annular region and extract the edges from the segmented image;

[0047] 7.4) Fit the edges extracted from each target point contour into an ellipse. If it cannot be fitted or the fitting error is greater than the threshold, it indicates that the target point is damaged. If it can be fitted, compare the fitted ellipse with the target point projection and calculate the distance between the contours. If the difference between the fitted ellipse and the projected template contour is greater than the threshold, it also indicates that the target point is damaged.

[0048] 7.5) Count and output the number and number of damaged target points.

[0049] The present invention has the following beneficial effects and advantages:

[0050] 1. This invention achieves four-degree-of-freedom target localization and guidance using only a single camera and a set of low-cost visual target points.

[0051] 2. This invention performs hand-eye calibration on a Cartesian coordinate robot with four degrees of freedom: X, Y, Z, and R.

[0052] 3. This invention does not affect the success rate and accuracy of positioning when some target points are damaged, and can provide reminders about the number and number of damaged target points. It has the advantages of strong adaptability, high speed, high accuracy, low cost and easy maintenance. Attached Figure Description

[0053] Figure 1 It is a visual target structure diagram;

[0054] Figure 2 This is a diagram of the visual target group structure used in this method;

[0055] Figure 3 This is a schematic diagram showing the use of tool T1 to contact the origin of the calibration plate;

[0056] Figure 4 It is a flowchart for extracting the contour of a target group and calculating its three-dimensional pose accordingly.

[0057] Figure 5 This is a flowchart that detects and outputs the number and serial number of damaged target points;

[0058] Figure 6 This is a flowchart of the method of the present invention. Detailed Implementation

[0059] The present invention will now be described in further detail with reference to the accompanying drawings and embodiments.

[0060] This invention relates to a four-degree-of-freedom visual guidance method based on a planar redundant target group. It possesses advantages such as strong adaptability, high speed, high accuracy, low cost, and ease of maintenance.

[0061] like Figure 6 As shown, a four-degree-of-freedom visual guidance method based on planar redundant target points includes the following steps:

[0062] (1) Calibrate the camera and obtain its intrinsic parameters;

[0063] (2) Install the camera on the end of the Cartesian coordinate robot with the camera facing vertically downwards, and install a set of circular visual target points on the surface of the positioning target with the camera facing upwards;

[0064] (3) Perform hand-eye calibration on the Cartesian coordinate robot and the camera to obtain the pose relationship between the camera coordinate system and the robot coordinate system; a needle-shaped tool T1 of known length was used in this process, and tool T1 was only used in this hand-eye calibration process;

[0065] (4) Perform a motion teaching to obtain the pose relationship between the tool coordinate system and the robot coordinate system of the end tool T2; the tool T2 is used for operations after visually locating the target, such as docking, grasping, and picking up.

[0066] (5) Use a visual calibration board to create a target group template;

[0067] (6) During the operation phase, the Cartesian coordinate robot moves to the working position, the camera takes a single image containing the target point group, extracts the outline of the target point group and calculates its three-dimensional pose accordingly.

[0068] (7) The target group has a redundant design. If the target group is partially damaged in step (6), its three-dimensional pose can still be output normally, and the number and number of damaged targets can be identified.

[0069] In step (1), multiple images of the calibration board at different positions and angles within the field of view are taken during the calibration period. The intrinsic parameters obtained during calibration mainly include the camera's focal length, pixel size, and distortion coefficient.

[0070] In step (2), a camera is installed at the end of the Cartesian coordinate robot with the camera facing vertically downwards, and a set of circular visual target points is installed on the surface of the positioning target with the camera facing upwards;

[0071] A Cartesian coordinate robot has four degrees of freedom: X, Y, Z, and R, which represent translation in the X direction, translation in the Y direction, translation in the Z direction, and rotation around the Z axis, respectively. Correspondingly, the robot's end-effector pose is denoted as P = (X, Y, Z, R).

[0072] The visual target is circular, with a highly reflective gray color, and surrounded by a fixed-width black static area. Figure 1 The image used is the visual target image.

[0073] The local surface of the target being located is a plane. The visual target set is mounted on this plane, and the target set structure is as follows: Figure 2 The target distribution within the target group meets the following constraints:

[0074] ① All target points in the target group are located in the same spatial plane;

[0075] ② Construct a two-dimensional coordinate system in the spatial plane where the target group is located, with the center point of the target group as the origin, the target directly above as point 1, and targets 2-4 arranged clockwise. The two-dimensional coordinates of targets 1-4 are (0,L), (L,0), (0,-L), and (-L,-L / 2), respectively.

[0076] ③ The center of the target group coincides with the center of the target;

[0077] Through the above steps, we define the local coordinate system of the target to coincide with the local coordinate system of the target group.

[0078] In step (3), hand-eye calibration is performed on the Cartesian coordinate robot and camera. The calibration process uses the same dot calibration plate as in step (1). The calibration plate is placed on the surface of the positioning target, and the center of the calibration plate coincides with the center of the positioning target. At this time, the centers of the calibration plate, the target group, and the positioning target coincide. The calibration process is as follows:

[0079] The calibration plate is kept in a fixed position, and the mobile robot takes multiple images of the calibration plate and records the coordinates of the robot's end tool at each shooting time. During the shooting, the calibration plate cannot move. The multiple sets of dot matrix calibration plate images and their corresponding coordinates are substituted into the hand-eye model, and the pose relationship between the camera coordinate system and the tool coordinate system is obtained through a nonlinear optimization algorithm.

[0080] In the hand-eye calibration process, the hand-eye model is as follows:

[0081]

[0082] Tool T1 is a needle-like tool of known length that is vertically downward and mounted at the center of the robot's Z-axis end, with the structure as follows: Figure 3 As shown. camera_H_cal is the transition matrix from the camera coordinate system to the calibration board coordinate system, and camera_H_tool1 is the transition matrix from the camera coordinate system to the tool T1 coordinate system. is the inverse of the transition matrix from the robot's base coordinate system to the tool's T1 coordinate system, and base_H_cal is the transition matrix from the base coordinate system to the calibration board coordinate system.

[0083] In the above steps, since the robot can only perform translational and rotational movements, the Z-axis translation in base_H_cal cannot be accurately calculated. The moving robot brings the end effector of tool T1 to the origin of the calibration plate, as shown... Figure 3 As shown, the Z-direction translation vector in base_H_cal is measured accordingly;

[0084] Tool T1 is only used during the calibration process; after calibration, tool T1 is removed.

[0085] In step (4), a motion teaching is performed to obtain the pose relationship between the tool coordinate system and the robot coordinate system of the end effector T2; tool T2 is the tool used by the robot during long-term operation, and its function can be docking, grasping, picking up, etc. The relationship between the end effector T2 and the robot coordinate system is determined through the following steps:

[0086] ① Keeping the calibration plate pose unchanged as described in step (3), the relationship between the calibration plate coordinate system and the base coordinate system, base_H_cal, obtained through step (3) camera_H_tool1, can be obtained at this time. The formula is as follows:

[0087]

[0088] From this, we can further determine the pose P of the calibration plate in the base coordinate system. cal =(X cal ,Y cal Z cal ,R cal )

[0089] ②Translate the calibration plate pose along its z-axis by a distance L t The target group pose P can be obtained obj =(X obj ,Y obj Z obj ,R obj ), where L t To calibrate the plate thickness;

[0090] ③ Remove the calibration plate, move the robot and position tool T2 to the target location, enabling tool T2 to perform actions such as grasping, sucking, and attaching. Record the robot's posture P at this time. T20 =(X T20 ,Y T20 Z T20 ,R T20 );

[0091] ④ Since the target point group coincides with the target center, the attitude P of the tool T2 to the center point of the Z-axis end can be obtained. T2 The method is

[0092]

[0093] Where P obj The target group pose obtained in step (2) is... The robot pose P obtained in step (3) T20 The reverse;

[0094] Step (5) uses a visual calibration board to create a target group template. The process is as follows:

[0095] ① Keeping the target pose unchanged as described in steps (3) and (4), move the robot to a suitable position and take a target group image I. obj This needs to include all targets in the target group;

[0096] ② Place the calibration plate back onto the target, with the same pose and requirements as in step (3). At this point, the centers of the calibration plate, the target group, and the target will coincide again. Take an image of the calibration plate. cal ;

[0097] ③ Recognize image I cal The three-dimensional pose of the calibration plate in the figure is denoted as P. cal The calibration plate pose and the target group pose only have a translation in the Z direction, and the translation amount is the thickness T of the calibration plate. P cal The target reference pose P can be obtained by translating a distance T along the z-axis. obj ;

[0098] ④In I obj Select the target point regions sequentially, ensuring the selection order matches the target point numbering; perform adaptive binarization on the selected regions to extract the center circle region of each target point; fit an ellipse to the contour of the center circle region, and then convert the fitted contour C... model As a template outline.

[0099] ⑤ Define the possible range of changes in the target group template contour, including the range of rotation angle, image scaling, target contrast, and image pyramid levels. Based on this, create the target contour template and set the reference pose of the contour template to P. obj ;

[0100] In step (6), the contours of the target point group are extracted and their three-dimensional poses are calculated accordingly. Figure 4 This is a flowchart for calculating the 3D pose. In this stage, the Cartesian coordinate robot moves to its working position based on external sensor data, and images taken at this position ensure that the complete target set is captured.

[0101] The steps for calculating the target pose based on the target point group image are as follows: A contour-based template matching algorithm is used to search for the target point group template in the image. If no target is found, the output score = 0 indicates target localization failure; if a target is found, let P... obj 'Represents the target pose, and the template contour C model Based on P objProjecting onto the image plane is used to display the matching effect and outputs a score with a value between 0.5 and 1; the score is used to represent the degree of contour matching, and the better the contour matching, the closer the score value is to 1.

[0102] The mobile robot moves to P' so that the end effector reaches P. obj 'To complete the action, P' is calculated as follows:

[0103] P′=P′ obj ·P T2

[0104] Among them, P′ obj For the target pose, P T2 The pose of tool T2 relative to the center of the end flange;

[0105] In step (7), the damaged target points are detected and the number and number of damaged target points are output. Figure 5 This is a flowchart of the process. Based on the pose obtained from target point localization, the template contour is projected onto the image containing the target point group. The template contour is circular, and the projected contour is usually elliptical. Each projected target point contour is expanded using the circle as the core to form a ring-shaped region; the original image within this region is segmented, and edges are extracted from the segmented image. The edges extracted from each target point contour are fitted to an ellipse. If a fit is not possible or the fitting error is too large, it indicates that the target point is damaged. If a fit is possible, the fitted ellipse is compared with the projected target point, and the distance between the contours is calculated. If the difference between the fitted ellipse and the projected template contour is too large, it also indicates that the target point is damaged. The number and ID of the damaged target points are counted and output.

Claims

1. A four-degree-of-freedom visual guidance method based on planar redundant target point groups, characterized in that, Includes the following steps: 1) Calibrate the camera and obtain its intrinsic parameters; 2) Install the camera and the circular visual target assembly on the end of the Cartesian coordinate robot and the target surface, respectively; 3) Use tool T1 to perform hand-eye calibration on the Cartesian coordinate robot and camera to obtain the relative pose relationship between the camera coordinate system and the robot coordinate system; 4) Perform a motion teaching exercise to obtain the pose relationship between the tool coordinate system and the robot coordinate system of the end effector T2; 5) Use a visual calibration board to create target group image templates; 6) During the online operation phase, the Cartesian coordinate robot moves to the working position, the camera takes a single image containing the target point group, extracts the outline of the target point group and calculates its three-dimensional pose accordingly. Step 3) includes the following steps: 3.1) Fix the calibration tool T1 to the end of the robot's Z-axis, place the calibration plate on the surface of the positioning target, and make the center of the calibration plate coincide with the center of the positioning target. At this time, the centers of the calibration plate, the target group, and the positioning target coincide. 3.2) Keep the calibration plate in a fixed position, move the robot to take multiple images of the calibration plate, and record the coordinates of the robot's end effector T1 at each time the image is taken. During the shooting process, keep the calibration plate from moving. 3.3) Substitute multiple sets of circular dot matrix calibration plate images and their corresponding coordinates into the hand-eye model, and use a nonlinear optimization algorithm to obtain the pose relationship between the camera coordinate system and the tool coordinate system; Step 4) includes the following steps: 4.1) Keep the calibration board pose unchanged, based on The relationship between the calibration plate coordinate system and the base coordinate system is obtained. Thus, the pose of the calibration plate in the base coordinate system is obtained as Pcal = (Xcal,Ycal,Zcal,Rcal); 4.2) Translate the calibration plate pose along its z-axis by a distance L t The target group pose is obtained as Pobj=(Xobj,Yobj,Zobj,Robj), where L t To calibrate the plate thickness; 4.3) Remove the calibration plate, move the robot and bring tool T2 to the target position so that tool T2 can complete the action normally, and record the robot's posture P at this time. T20 =(XT20,YT20,ZT20,RT20); 4.4) Since the target group coincides with the target center, calculate the attitude P from the center point of the Z-axis end of tool T2. T2 : ； Among them, P obj For the target group pose, For robot posture P T20 The reverse; Step 5) includes the following steps: 5.1) Keeping the target pose unchanged, move the robot to a suitable position and take a picture of the target group I. obj This ensures that the image contains all target points of the target group; 5.2) Place the calibration plate back on the target, ensuring that the centers of the calibration plate, the target group, and the target coincide, and take an image of the calibration plate. cal ; 5.3) Image Recognition I cal The calibration plate in the three-dimensional pose P cal The three-dimensional pose of the calibration plate P cal Translation distance L along the z-axis t Obtain the target reference pose P obj L t The thickness of the calibration plate; 5.4) According to the target point numbering order, in target point group image I obj The target area is selected sequentially, and adaptive binarization is performed on the selected area to extract the center circle region of the target. The contour of the center circle region is then fitted with an ellipse, and the fitted contour C is... model As a template outline; 5.5) Define the possible range of changes in the target group template contour, including rotation angle, image scaling, target contrast, and image pyramid hierarchy, and then create the target contour template, setting the reference pose of the contour template as the target reference pose P. obj .

2. The four-degree-of-freedom visual guidance method based on planar redundant target point groups according to claim 1, characterized in that, It also includes the following steps: 7) If the target group is partially damaged, output its three-dimensional pose normally based on step 6), and identify the number and number of damaged targets.

3. The four-degree-of-freedom visual guidance method based on planar redundant target point groups according to claim 1, characterized in that, The circular visual target group consists of four target points located on the same spatial plane, with the center of the target group coinciding with the center of the target. The target points are circular, highly reflective gray, and surrounded by a fixed-width black static area. A two-dimensional coordinate system is constructed based on the spatial plane where the target group is located. The center point of the target group is taken as the origin, and the target point directly above it is marked as target point 1. The other target points are marked as target points 2-4 in a clockwise order. The two-dimensional coordinates of target points 1-4 are (0,L), (L,0), (0,-L), and (-L,-L / 2), respectively, where L is the distance from target point 1 to the center of the target group.

4. The four-degree-of-freedom visual guidance method based on planar redundant target point groups according to claim 1, characterized in that, The hand-eye model is as follows: ； Where camera_H_cal is the transition matrix from the camera coordinate system to the calibration board coordinate system. Let T1 be the transition matrix from the camera coordinate system to the tool T1 coordinate system. is the inverse of the transition matrix from the robot's base coordinate system to the tool's T1 coordinate system, and base_H_cal is the transition matrix from the base coordinate system to the calibration board coordinate system.

5. A four-degree-of-freedom visual guidance method based on a planar redundant target group according to claim 1, characterized in that, The basis The relationship between the calibration plate coordinate system and the base coordinate system is obtained. Specifically: 。 6. The four-degree-of-freedom visual guidance method based on planar redundant target groups according to claim 1, characterized in that, Step 6) specifically refers to: A contour-based template matching algorithm is used to search for target point templates in the image. If no target is found, a score of 0 is output to indicate that target localization has failed; if a target is found, the output score is set to 0. Representing the target pose, the template contour C model based on Projecting onto the image plane displays the matching results and outputs a score between 0.5 and 1. The robot is then moved to... Make the end tool reach To complete the action, Specifically: ； in, For the target pose, The position of tool T2 relative to the center of the end flange.

7. A four-degree-of-freedom visual guidance method based on a planar redundant target group according to claim 2, characterized in that, Step 7) includes the following steps: 7.1) Based on the pose obtained from target localization, project the template contour onto the image containing the target group; 7.2) Expand the outline of each projected target point using a circle as the core to form a ring-shaped region; 7.3) Segment the original image within the annular region and extract the edges from the segmented image; 7.4) Fit the edges extracted from each target point contour into an ellipse. If it cannot be fitted or the fitting error is greater than the threshold, it indicates that the target point is damaged. If it can be fitted, compare the fitted ellipse with the target point projection and calculate the distance between the contours. If the difference between the fitted ellipse and the projected template contour is greater than the threshold, it also indicates that the target point is damaged. 7.5) Count and output the number and number of damaged target points.

Images