A pseudo star point rejection method based on pose quaternion
By calculating the target frame displacement through the change in attitude quaternion between consecutive image frames of the star sensor, false star points are identified and eliminated, thus solving the problem of false star point interference in the field of view of the star sensor and achieving accurate false star point elimination and star identification.
Patent Information
- Application Number
- CN202611161436.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2026-08-03
- Publication Date
- 2026-11-10
- Estimated Expiration
- 2046-08-03
AI Technical Summary
False stars in the field of view of the star sensor interfere with the measurement of navigation stars, affecting the reliability of star measurement, and existing technologies are difficult to effectively eliminate them.
The target's inter-frame displacement is calculated by the change in attitude quaternions between consecutive image frames of the star sensor, and compared with the actual inter-frame displacement to determine whether the target is a star or a false star point, and the false star points are eliminated.
It accurately eliminates false star points, preventing them from interfering with navigation star measurements. It has a wide range of applications and is suitable for various star sensor field-of-view conditions.
Smart Images

Figure CN122708809B_ABST
Abstract
Description
Technical Field
[0001] This invention belongs to the field of star navigation technology, specifically relating to a pseudo-star point removal method based on attitude quaternions. Background Technology
[0002] As an attitude measurement instrument, star sensors are characterized by high precision, strong adaptability, high reliability, small size, light weight, and low power consumption. These outstanding advantages make star sensors widely used in the field of astronomical navigation.
[0003] Currently, the mainstream star sensors mainly operate in two modes: one mode outputs the measured target to the control system, which then identifies the navigation star; the other mode sends the attitude quaternion obtained through star map matching to the control system.
[0004] Star sensors often operate in environments with accompanying objects. When these objects are illuminated by sunlight, they form false star points on the detector's target surface. These false star points can interfere with the star sensor's measurement of navigation stars, affecting the reliability of the measurement. Therefore, a method for eliminating false star points is needed to remove their interference with the star sensor's measurement. Summary of the Invention
[0005] In view of this, the present invention provides a pseudo-star point removal method based on attitude quaternions. This method fully considers the complexity of the star measurement environment, calculates the target inter-frame displacement by the change in attitude quaternions between consecutive image frames of the star sensor, and determines whether the target is a star or a pseudo-star point by comparing the calculated target inter-frame displacement with the actual inter-frame displacement, and finally removes the pseudo-star points from the image.
[0006] To achieve the above objectives, the present invention adopts the following technical solution:
[0007] This invention provides a pseudo-star point removal method based on attitude quaternions, comprising:
[0008] The attitude quaternion information of the star sensor and the N-1th and Nth frame images captured by the star sensor are obtained; the attitude quaternion information is obtained by inertial measurement combination measurement; the Nth frame image contains multiple targets; N is a positive integer greater than 1;
[0009] Based on the attitude quaternion information, calculate the attitude quaternion corresponding to the exposure center time of the (N-1)th frame image and the Nth frame image respectively;
[0010] Based on the attitude quaternion corresponding to the exposure center time of the N-1th frame image and the Nth frame image, the rotation angle of the star sensor around the three coordinate axes of the Cartesian coordinate system is calculated by axis-angle decomposition.
[0011] Based on the rotation angle of the star sensor around the three coordinate axes of the Cartesian coordinate system and the actual coordinates of each target in the N-1th frame image, calculate the inter-frame displacement of each target from the N-1th frame image to the Nth frame image, and based on the inter-frame displacement of each target from the N-1th frame image to the Nth frame image and the actual coordinates of each target in the N-1th frame image, calculate the predicted coordinates of each target in the Nth frame image.
[0012] Based on the difference between the predicted coordinates and the actual coordinates of each target in the Nth frame image, and the corresponding star clustering threshold value of each target, the target type of each target is determined; the target type includes stars and pseudo-stars.
[0013] Remove targets of type pseudo-star from the Nth frame image.
[0014] The beneficial effects of this invention are as follows:
[0015] The method of this invention can accurately remove false star points on the target surface of the star sensor detector, avoiding interference from false star points in navigation star measurements. Furthermore, this invention fully considers the number of stars within the star sensor's field of view, making the false star point removal method of this invention applicable to a wider range of situations. Attached Figure Description
[0016] Figure 1 This is a schematic diagram illustrating the principle of pseudo-star point removal in this invention.
[0017] Figure 2 This is a flowchart of the pseudo-star point removal method based on attitude quaternions of the present invention.
[0018] Figure 3 This is a schematic diagram of the pseudo-star point removal method based on attitude quaternions of the present invention.
[0019] Figure 4 This is a schematic diagram showing the alignment of the system time corresponding to the attitude quaternion with the exposure center time of the image.
[0020] Figure 5 A schematic diagram for calculating attitude quaternions for interpolation. Detailed Implementation
[0021] The present invention will be further described below with reference to the accompanying drawings and embodiments.
[0022] The term “exemplary” as used herein means “serving as an example, embodiment, or illustration.” Any embodiment illustrated herein as “exemplary” is not necessarily to be construed as superior to or better than other embodiments.
[0023] The terms “first”, “second”, etc., are used to distinguish similar objects, not to describe or indicate a specific order or sequence.
[0024] The term "comprising" or any other similar term is intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus / device that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent in such process, method, article, or apparatus / device.
[0025] Furthermore, to better illustrate the present invention, numerous specific details are provided in the following detailed embodiments. Those skilled in the art should understand that the present invention can be practiced without certain specific details. In some instances, methods, means, elements, and circuits well known to those skilled in the art have not been described in detail in order to highlight the spirit of the invention.
[0026] Figure 1 This is a schematic diagram illustrating the principle of pseudo-star point removal in this invention, as shown below. Figure 1 As shown, in the continuous image frames captured by the star sensor, the displacement change Δd of the star differs significantly from the displacement change Δd1 of the pseudo-star point caused by the accompanying object, meaning that the trajectories of the star and the pseudo-star point differ considerably in the continuous image frames. This is because the stellar target moves only on the detector target surface with the rotation of the star sensor, while the pseudo-star point moves on the detector target surface under the combined effect of its own motion and the rotation of the star sensor.
[0027] The calculation principle of the pseudo-star point removal method of the present invention is explained below. The pseudo-star point removal method of the present invention is based on the target position prediction model and the same-star clustering model for calculation.
[0028] (1) Target location prediction model:
[0029] According to Euler's theorem, the rotation matrix from one coordinate system to another can be obtained by rotating around a certain vector by a finite angle, which is called the Euler angle.
[0030] By mapping the Euler rotation angle to a Cartesian coordinate system, the angles of rotation of the optical system about the x, y, and z axes of the Cartesian coordinate system can be obtained. , , .
[0031] For a target at infinity, when the optical system rotates around the z-axis, the target's coordinates within the field of view change in both the x and y directions; when the optical system rotates around the x-axis, the target's coordinates change in the y-direction; and when the optical system rotates around the y-axis, the target's coordinates change in the x-direction.
[0032] When the optical system rotates about the x-axis and about the y-axis, through , By using the angle value corresponding to each pixel of the detector (i.e., the pixel angular equivalent, representing the angular size of a single pixel in the target space), the displacement of the target on the detector surface between two adjacent frames (i.e., the (N-1)th frame and the Nth frame) can be calculated.
[0033] ;
[0034] in, It represents the displacement of the target in the x direction when the optical system rotates about the y-axis; It represents the displacement of the target in the y direction when the optical system rotates about the x-axis; , These represent the pixel angular equivalents of the detector in the x and y directions of the optical system, respectively.
[0035] When the optical system rotates around the z-axis, it utilizes The displacement of the target on the detector surface between two adjacent frames can be calculated using the target's coordinates in the (N-1)th frame image.
[0036] ;
[0037] in, It represents the displacement of the target in the x direction when the optical system rotates about the z-axis; It represents the displacement of the target in the y direction when the optical system rotates about the z-axis; Let be the coordinates of the target in the (N-1)th frame of the image.
[0038] After the optical system rotates around three axes (x-axis, y-axis, z-axis), the displacement of the target between two adjacent frames can be calculated based on the imaging projection relationship of the detector. The calculation formula is as follows:
[0039] ;
[0040] in, It represents the displacement of the target in the x-direction after the optical system rotates around three axes; It represents the displacement of the target in the y-direction after the optical system rotates around three axes.
[0041] The predicted coordinates of the target in the Nth frame image. The calculation formula is as follows:
[0042] ;
[0043] (2) Same-star clustering model:
[0044] Obtain the predicted coordinates of the target in the Nth frame image. Then, the actual coordinates of the target in the Nth frame image are... With predicted coordinates Comparing the two, if the absolute value of the difference is less than the set threshold for same-star clustering, then the two targets (with coordinates of [value missing] in the (N-1)th frame image) are considered to be [targets]. The target and the coordinates in the Nth frame image are The target (i.e., the same star) is considered to be the same star. The criterion for clustering stars is shown in the following formula:
[0045] ;
[0046] Same-star clustering threshold , The threshold for same-star clustering is determined based on the target's displacement between two adjacent frames (i.e., inter-frame displacement). For example, the threshold can be set to 10% of the target's inter-frame displacement. Since the detector's noise equivalent angle affects target positioning accuracy, the same-star clustering threshold needs to include the positioning accuracy error of the star points. .
[0047] Based on the above principles, this invention proposes a pseudo-star point removal method based on attitude quaternions. The inter-frame displacement of the target is calculated by the quaternion attitude change between consecutive image frames captured by the star sensor, and the calculated inter-frame displacement is compared with the actual inter-frame displacement to identify whether the target is a star or a pseudo-star point.
[0048] Figure 2 The flowchart of the pseudo-star point removal method based on attitude quaternions of the present invention is as follows: Figure 2 As shown, the method includes:
[0049] Step 1: Obtain the attitude quaternion information of the star sensor and the N-1th frame image (i.e., the previous frame image) and the Nth frame image (i.e., the current frame image) captured by the star sensor; the attitude quaternion information is obtained through inertial measurement combination measurement; the Nth frame image contains multiple targets; N is a positive integer greater than 1.
[0050] Step 2: Based on the attitude quaternion information, calculate the attitude quaternion corresponding to the exposure center time of the (N-1)th frame image and the Nth frame image, respectively.
[0051] For example, the pose quaternions corresponding to the exposure center times of the (N-1)th frame and the Nth frame can be calculated using linear interpolation based on the pose quaternion information.
[0052] Step 3: Based on the attitude quaternions corresponding to the exposure center times of the (N-1)th frame image and the Nth frame image, calculate the angle of rotation of the star sensor around the three coordinate axes of the Cartesian coordinate system through axis-angle decomposition.
[0053] For example, step 3 specifically includes:
[0054] (1) Calculate the change in attitude quaternion based on the attitude quaternion corresponding to the exposure center time of the (N-1)th frame image and the Nth frame image;
[0055] (2) Based on the change in the attitude quaternion, calculate the angle of rotation of the star sensor around the three coordinate axes of the rectangular coordinate system using the axis-angle decomposition algorithm and Euler's theorem.
[0056] Step 4: Based on the rotation angle of the star sensor around the three coordinate axes of the Cartesian coordinate system and the actual coordinates of each target in the (N-1)th frame image, calculate the inter-frame displacement of each target from the (N-1)th frame image to the Nth frame image (i.e., the displacement of each target in the Nth frame image relative to the (N-1)th frame image). Based on the inter-frame displacement of each target from the (N-1)th frame image to the Nth frame image and the actual coordinates of each target in the (N-1)th frame image (including the actual values of the x-axis coordinate components and the y-axis coordinate components), calculate the predicted coordinates of each target in the Nth frame image (including the predicted values of the x-axis coordinate components and the y-axis coordinate components).
[0057] For example, the process of calculating the inter-frame displacement of each target from the (N-1)th frame image to the Nth frame image includes:
[0058] (1) Based on the angle of rotation of the star sensor around the y-axis of the rectangular coordinate system, the angle of rotation of the star sensor around the z-axis of the rectangular coordinate system, the pixel angle equivalent of the star sensor detector in the x-direction, and the actual coordinates of each target in the N-1th frame image, calculate the x-axis component of the inter-frame displacement of each target from the N-1th frame image to the Nth frame image.
[0059] (2) Based on the angle of rotation of the star sensor around the x-axis of the rectangular coordinate system, the angle of rotation of the star sensor around the z-axis of the rectangular coordinate system, the pixel angle equivalent of the star sensor detector in the y-direction, and the actual coordinates of each target in the N-1 frame image, calculate the y-axis component of the inter-frame displacement of each target from the N-1 frame image to the Nth frame image.
[0060] For example, calculating the predicted coordinates of each target in the Nth frame image includes:
[0061] (1) Based on the x-axis component of the inter-frame displacement of each target from the (N-1)th frame image to the Nth frame image and the actual value of the x-axis coordinate component of each target in the (N-1)th frame image, calculate the predicted value of the x-axis coordinate component of each target in the Nth frame image;
[0062] (2) Based on the y-axis component of the inter-frame displacement of each target from the (N-1)th frame image to the Nth frame image and the actual value of the y-axis coordinate component of each target in the (N-1)th frame image, calculate the predicted value of the y-axis coordinate component of each target in the Nth frame image.
[0063] Step 5: Determine the target type of each target based on the difference between the predicted coordinates and the actual coordinates of each target in the Nth frame image, and the corresponding star clustering threshold value; the target type includes stars and pseudo-stars; the star clustering threshold value includes a first threshold value and a second threshold value.
[0064] For example, step 5 specifically includes:
[0065] (1) For the i-th target among the plurality of targets, the absolute value of the difference between the actual value and the predicted value of the x-axis coordinate component of the i-th target in the N-frame image is compared with the first threshold value corresponding to the i-th target, and the absolute value of the difference between the actual value and the predicted value of the y-axis coordinate component of the i-th target in the N-frame image is compared with the second threshold value corresponding to the i-th target.
[0066] (2) If the absolute value of the difference between the actual value and the predicted value of the x-axis coordinate component of the i-th target in the N-frame image is less than the first threshold value corresponding to the i-th target, and the absolute value of the difference between the actual value and the predicted value of the y-axis coordinate component of the i-th target in the N-frame image is less than the second threshold value corresponding to the i-th target, the target type of the i-th target is determined to be a star; otherwise, the target type of the i-th target is determined to be a pseudo-star.
[0067] For example, the first threshold value corresponding to the i-th target is determined based on the x-axis component of the inter-frame displacement of the i-th target from the (N-1)-th frame image to the N-th frame image and the star point positioning accuracy error; the second threshold value corresponding to the i-th target is determined based on the y-axis component of the inter-frame displacement of the i-th target from the (N-1)-th frame image to the N-th frame image and the star point positioning accuracy error.
[0068] Step 6: Remove targets with the target type of pseudo-star points from the Nth frame image.
[0069] For example, the method further includes: if a target present in the Nth frame image does not exist in the (N-1)th frame image, directly determine that the target is a false star point and remove it from the Nth frame image.
[0070] The pseudo-star point removal method based on attitude quaternions of this invention fully considers the difference in motion characteristics between stars at infinity and pseudo-star points near the star sensor between consecutive image frames. It uses the target coordinates extracted from the previous frame and the attitude change of the star sensor between adjacent frames to predict the target's coordinates in the current frame. The actual target coordinates extracted from the current frame are compared with the predicted target coordinates, and a reasonable threshold is set to filter out stars and eliminate pseudo-star points. This method is applicable to all target tracking scenarios where attitude quaternions can be obtained.
[0071] The pseudo-star point removal method based on attitude quaternions of the present invention will be described in detail below with reference to the accompanying drawings and embodiments.
[0072] Figure 3 This is a schematic diagram of the pseudo-star point removal method based on attitude quaternions of the present invention, as shown below. Figure 3 As shown, the method includes:
[0073] Step 1: Obtain the attitude quaternion information of the star sensor and the (N-1)th frame image and the Nth frame image captured by the star sensor; the Nth frame image contains multiple targets; N is a positive integer greater than 1.
[0074] The spacecraft's inertial measurement unit can generate attitude quaternions at a high frequency and can transmit the attitude quaternion information discretely to the star sensor via a communication interface.
[0075] Attitude quaternions consist of four attitude parameters composed of Euler axis / angle parametric formulas. The definition is as follows:
[0076] ;
[0077] in, , , , Representing attitude quaternions The amount; express The real part, express The imaginary part; , , The unit vectors representing the axes of rotation are respectively The direction cosines of the x-axis, y-axis, and z-axis (i.e., (cosine of the angle between the vector and the unit vectors of the x-axis, y-axis, and z-axis). This represents the Euler angle by which the star sensor rotates about its axis.
[0078] Posture Quaternions The algebraic form is as follows:
[0079] ;
[0080] in, , , The imaginary unit corresponds to the unit vectors of the x-axis, y-axis, and z-axis in three-dimensional space, respectively.
[0081] Step 2: Based on the attitude quaternion information, calculate the attitude quaternion corresponding to the exposure center time of the (N-1)th frame image and the Nth frame image, respectively.
[0082] To accurately calculate the attitude change of the star sensor during the exposure of two frames, it is necessary to align the system time corresponding to the attitude quaternion with the exposure center time of the image through interpolation.
[0083] For example, the pose quaternion corresponding to the exposure center time of two frames of images can be obtained by linear interpolation.
[0084] Figure 4 This is a schematic diagram showing the alignment of the system time corresponding to the attitude quaternion with the exposure center time of the image. Figure 5 A schematic diagram for calculating attitude quaternions for interpolation, as shown below. Figure 4 , Figure 5 As shown, the camera's exposure time is T. exp The transmission period of the attitude quaternion is T, and the difference between the system time corresponding to the attitude quaternion and the exposure center time is t. Among the acquired attitude quaternion information, let the nearest attitude quaternion before the exposure center time (corresponding to...) be... Figure 4 The attitude quaternion at time t1 is The nearest pose quaternion after the exposure center moment (corresponding to) Figure 4 The attitude quaternion at time t2 is The attitude quaternion corresponding to the exposure center time is obtained by linear interpolation. The calculation formula is:
[0085] ;
[0086] Based on the above formulas, the pose quaternion corresponding to the exposure center time of the Nth frame image can be calculated respectively. The pose quaternion corresponding to the exposure center time of the (N-1)th frame image .
[0087] Since the frequency of attitude quaternions generated by inertial measurement units is generally greater than 100Hz, the speed of star sensor attitude adjustment is generally less than 20° / s, and the angle change corresponding to the attitude quaternions in two adjacent frames is less than 0.2°.
[0088] Step 3: Based on the attitude quaternions corresponding to the exposure center times of the (N-1)th frame image and the Nth frame image, calculate the angle of rotation of the star sensor around the three coordinate axes of the Cartesian coordinate system through axis-angle decomposition.
[0089] For example, step 3 specifically includes:
[0090] (1) Based on the pose quaternions corresponding to the exposure center times of the (N-1)th frame image and the Nth frame image. , Calculate the change in attitude quaternion The calculation process is as follows:
[0091] and The following relationship must be satisfied:
[0092] ;
[0093] Based on attitude quaternions Definition of the inverse Multiply both sides of the above equation by the left side. ,get The expression is:
[0094] ;
[0095] in, express The inverse of. Let... The four components are respectively , , , ,but The algebraic form of it is shown below:
[0096] ;
[0097] (2) Based on the change in attitude quaternions The angles of rotation of the star sensor around the three axes of the Cartesian coordinate system are calculated based on the axis-angle decomposition algorithm and Euler's theorem.
[0098] According to Euler's theorem, a rotation matrix from one coordinate system to another can be obtained by rotating a certain vector by a finite angle (i.e., ...). To obtain it. According to... The angles of rotation of the star sensor around the x, y, and z axes of the Cartesian coordinate system can be determined from the relationship with the rotation matrix. , , :
[0099] ;
[0100] ;
[0101] ;
[0102] ;
[0103] in, It is an inverse cosine function.
[0104] Step 4: Based on the rotation angle of the star sensor around the three coordinate axes of the Cartesian coordinate system and the actual coordinates of each target in the (N-1)th frame image, calculate the inter-frame displacement of each target from the (N-1)th frame image to the Nth frame image. Based on the inter-frame displacement of each target from the (N-1)th frame image to the Nth frame image and the actual coordinates of each target in the (N-1)th frame image, calculate the predicted coordinates of each target in the Nth frame image.
[0105] For the i-th target in the image, let the actual coordinates of the i-th target in the (N-1)-th frame be... Based on the calculation principle of the target location prediction model described above, the x-axis component of the inter-frame displacement of the i-th target from the (N-1)-th frame to the N-th frame can be calculated. (i.e., the displacement of the i-th target in the x-direction) and the y-axis component (That is, the displacement of the i-th target in the y-direction):
[0106] ;
[0107] in, , These are the pixel angular equivalents of the star sensor detector in the x and y directions, respectively.
[0108] Then the predicted coordinates of the i-th target in the N-th frame image. The calculation formula is:
[0109] ;
[0110] Step 5: Determine the target type of each target based on the difference between the predicted coordinates and the actual coordinates of each target in the Nth frame image, and the corresponding star clustering threshold value of each target; the target type includes stars and pseudo-stars.
[0111] The process of determining the target type of the i-th target is as follows:
[0112] Let the actual coordinates of the i-th target in the N-th frame be... Set the clustering threshold of the i-th target to the first threshold value. Second threshold Actual coordinates And the predicted coordinates calculated in the previous step Compare the results; if both conditions are met, the following discriminant will be used:
[0113] ;
[0114] If the inter-frame displacement of the i-th target matches the inter-frame attitude change of the star sensor, the i-th target is determined to be a star and is retained in the N-th frame. If the above discriminant is not satisfied, the i-th target is determined to be a false star and is removed from the N-th frame. If the target appears only in the (N-1)-th frame and not in the N-th frame, or if the target appears only in the N-th frame and not in the (N-1)-th frame, the target is considered a false star and is removed.
[0115] Considering that the target positioning accuracy and the angular rate of the star sensor can introduce prediction errors to the target's coordinate position, for moving targets, a same-star clustering threshold is set based on the inter-frame displacement corresponding to the target, while also taking into account the positioning accuracy error of star points. (The positioning accuracy error can be obtained by statistically analyzing the coordinates of test star points under static conditions using a star sensor) The first threshold value corresponding to the i-th target. Second threshold The calculation formula is:
[0116] ;
[0117] in, This is the proportionality coefficient. Between 0 and 1, in the embodiments of the present invention .
[0118] Step 6: Remove targets with the target type of pseudo-star points from the Nth frame image.
[0119] like Figure 3 As shown, the present invention first calculates the pose quaternion corresponding to the exposure center time of the (N-1)th frame image and the Nth frame image based on the pose quaternion information. and Then according to and Calculate the change in attitude quaternion And then according to The rotation angles of the star sensor around the x, y, and z axes of the Cartesian coordinate system are calculated based on the axis-angle decomposition algorithm and Euler's theorem. , , Then according to , , The inter-frame displacement of the target is calculated based on its actual coordinates in the (N-1)th frame image. Then, the predicted coordinates of the target in the Nth frame image are calculated. Finally, the target type is determined based on the difference between the predicted coordinates and the actual coordinates, as well as the same-star clustering threshold.
[0120] The pseudo-star point removal method based on attitude quaternions of this invention fully considers the complexity of the star measurement environment and the difference in motion characteristics between stars at infinity and pseudo-star points near the star sensor between consecutive image frames. It uses the target coordinates extracted from the previous frame and the attitude change of the star sensor between adjacent frames to calculate the target's inter-frame displacement and predict the target's coordinates in the current frame. The actual target coordinates extracted from the current frame are compared with the predicted coordinates, and a reasonable threshold is set to filter out stars and remove pseudo-star points. This method can accurately remove pseudo-star points and has a wider range of applications.
[0121] Example:
[0122] The pseudo-star point removal method based on attitude quaternions in this embodiment first calculates the attitude quaternion corresponding to the exposure center moment, and then calculates the attitude change of the system between two exposure cycles using the attitude quaternion. Next, the angles of rotation of the system around the three axes of the measurement coordinate system are calculated using the attitude change. Finally, the displacement of each target on the star sensor target surface is calculated using the coordinates of the target on the star sensor target surface and the rotation angle around each axis.
[0123] (a) Attitude quaternion interpolation:
[0124] Assuming the transmission period of the attitude quaternion between the measurement coordinate system and the geocentric inertial coordinate system is 5ms, the exposure center time corresponding to the (N-1)th frame image is 16.965s, the system time corresponding to attitude quaternion Q0 is 16.9648s, and the system time corresponding to attitude quaternion Q2 is 16.9698s. The attitude quaternion Q1 corresponding to 16.965s can be calculated using the above attitude quaternion interpolation formula, as shown in Table 1.
[0125] Table 1. Example 1 of attitude quaternion interpolation calculation
[0126]
[0127] Where q0, q1, q2, and q3 represent the four components of the attitude quaternion.
[0128] Similarly, the exposure center time corresponding to the Nth frame image is 17.065s, and the pose quaternion... The corresponding system time is 16.9698s, attitude quaternion The corresponding system time is 17.0648s. The attitude quaternion corresponding to 17.065s can be interpolated and calculated. As shown in Table 2.
[0129] Table 2 Example 2 of Attitude Quaternion Interpolation Calculation
[0130]
[0131] (b) Calculation of rotation angles around three axes:
[0132] Based on the above formula for calculating the change in attitude quaternions, the star sensor's range from 16.965s (Q1) to 17.065s (Q2) can be calculated. The attitude quaternion change is given by the four components deltq0, deltq1, deltq2, and deltq3, as shown in Table 3.
[0133] Table 3. Examples of Calculation of Attitude Quaternion Changes
[0134]
[0135] By using the relationship between Euler angles and the rotation matrix, the Euler angle theta of the star sensor's rotation around the rotation axis and the angle theta of its rotation around the three axes of the Cartesian coordinate system can be calculated. x theta y theta z The unit of angle is unit radians, and the calculation results are shown in Table 4.
[0136] Table 4 Calculation of Euler angles and rotation angles for star sensors
[0137]
[0138] (c) Same-star clustering:
[0139] Table 5 shows the inter-frame displacement Δx in the x-axis direction and the inter-frame displacement Δy in the y-axis direction of target 1 from the (N-1)th frame image to the Nth frame image, the actual coordinates of target 1 in the two frames, and the predicted coordinates of target 1 in the Nth frame image.
[0140] Table 5 Example 1 of Predicted Coordinate Calculation
[0141]
[0142] Assuming the positioning accuracy error of the star point is 0.5 pixels, the absolute value of the difference between the predicted and actual coordinates of target 1 in the Nth frame image is (0.03, 0.11). The first and second threshold values calculated using the above formula for the same-star clustering threshold are (0.564, 0.809). Since the absolute value of the difference between the predicted and actual coordinates of target 1 in the Nth frame image is less than the same-star clustering threshold, target 1 in the (N-1)th frame image and target 1 in the Nth frame image can be determined to be the same target. The inter-frame motion trajectory of this target matches the inter-frame attitude change of the star sensor, proving that target 1 is a star and should be preserved.
[0143] Table 6 shows the inter-frame displacement Δx in the x-axis direction and the inter-frame displacement Δy in the y-axis direction of target 2 from the (N-1)th frame image to the Nth frame image, the actual coordinates of target 2 in the two frames, and the predicted coordinates of target 2 in the Nth frame image.
[0144] Table 6 Example 2 of Predicted Coordinate Calculation
[0145]
[0146] The absolute difference between the predicted and actual coordinates of target 2 in the Nth frame image is (7.47, 0.33). The calculated threshold value for the same-star clustering is (1.254, 0.833) using the formula mentioned above. Since the absolute difference of 7.47 between the predicted and actual x-coordinates of target 2 in the Nth frame image is greater than the first threshold value of 1.254, it proves that the inter-frame motion trajectory of target 2 in the Nth frame image does not match the inter-frame attitude change of the star sensor. Target 2 is a false star point and should be removed.
[0147] The present invention also provides an electronic device, comprising: one or more processors; and a memory for storing one or more programs; wherein, when the one or more programs are executed by the one or more processors, the one or more processors implement the above-described pseudo-star point removal method based on attitude quaternions.
[0148] The present invention also provides a computer-readable storage medium storing executable instructions thereon, which, when executed by a processor, enable the processor to implement the above-described pseudo-star point removal method based on attitude quaternions.
[0149] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code. The solutions in the embodiments of the present invention can be implemented using various computer languages, such as the object-oriented programming language Java and the interpreted scripting language JavaScript.
[0150] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.
[0151] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.
[0152] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.
[0153] The contents not described in detail in this specification are existing technologies known to those skilled in the art.
[0154] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the present invention. The above descriptions are exemplary and not exhaustive. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.
Claims
1. A pseudo-star point removal method based on attitude quaternions, characterized in that, include: Obtain the attitude quaternion information of the star sensor, as well as the N-1th and Nth frame images captured by the star sensor; The attitude quaternion information is obtained through inertial measurement combination measurement; The Nth frame image contains multiple targets; N is a positive integer greater than 1; Based on the attitude quaternion information, calculate the attitude quaternion corresponding to the exposure center time of the (N-1)th frame image and the Nth frame image respectively; Based on the attitude quaternion corresponding to the exposure center time of the (N-1)th frame image and the Nth frame image, the rotation angle of the star sensor around the three coordinate axes of the Cartesian coordinate system is calculated by axis-angle decomposition. Based on the rotation angle of the star sensor around the three coordinate axes of the Cartesian coordinate system and the actual coordinates of each target in the N-1th frame image, calculate the inter-frame displacement of each target from the N-1th frame image to the Nth frame image, and based on the inter-frame displacement of each target from the N-1th frame image to the Nth frame image and the actual coordinates of each target in the N-1th frame image, calculate the predicted coordinates of each target in the Nth frame image. Based on the difference between the predicted coordinates and the actual coordinates of each target in the Nth frame image, and the corresponding star clustering threshold value, the target type of each target is determined; the target type includes stars and pseudo-stars; the star clustering threshold value includes a first threshold value and a second threshold value. Remove targets of type pseudo-star from the Nth frame image; Based on the difference between the predicted and actual coordinates of each target in the Nth frame image, and the corresponding co-stellar clustering threshold value, the target type of each target is determined, including: For the i-th target among the plurality of targets, the actual value of the x-axis coordinate component of the i-th target in the N-th frame image is... Compared with the predicted value The absolute value of the difference The first threshold value corresponding to the i-th target The comparison is performed, and the actual value of the y-axis coordinate component of the i-th target in the N-th frame image is obtained. Compared with the predicted value The absolute value of the difference The second threshold value corresponding to the i-th target Compare; like and If the target type of the i-th target is determined to be a star, then the target type of the i-th target is determined to be a pseudo-star. The first threshold value corresponding to the i-th target is determined based on the x-axis component of the inter-frame displacement of the i-th target from the (N-1)-th frame image to the N-th frame image and the star point positioning accuracy error; the second threshold value corresponding to the i-th target is determined based on the y-axis component of the inter-frame displacement of the i-th target from the (N-1)-th frame image to the N-th frame image and the star point positioning accuracy error.
2. The pseudo-star point removal method based on attitude quaternions according to claim 1, characterized in that, Based on the attitude quaternion information, calculate the attitude quaternions corresponding to the exposure center time of the (N-1)th frame image and the Nth frame image, including: Based on the attitude quaternion information, the attitude quaternions corresponding to the exposure center times of the (N-1)th frame image and the Nth frame image are calculated using linear interpolation.
3. The pseudo-star point removal method based on attitude quaternions according to claim 1, characterized in that, Based on the attitude quaternions corresponding to the exposure center times of the (N-1)th frame image and the Nth frame image, the rotation angles of the star sensor around the three coordinate axes of the Cartesian coordinate system are calculated through axis-angle decomposition, including: The change in pose quaternion is calculated based on the pose quaternion corresponding to the exposure center time of the (N-1)th frame image and the Nth frame image. Based on the change in the attitude quaternion, the angle of rotation of the star sensor around the three coordinate axes of the Cartesian coordinate system is calculated using the axis-angle decomposition algorithm and Euler's theorem.
4. The pseudo-star point removal method based on attitude quaternions according to claim 1, characterized in that, Based on the rotation angle of the star sensor around the three coordinate axes of the Cartesian coordinate system and the actual coordinates of each target in the (N-1)th frame image, calculate the inter-frame displacement of each target from the (N-1)th frame image to the Nth frame image, including: Based on the angle of rotation of the star sensor around the y-axis of the Cartesian coordinate system, the angle of rotation of the star sensor around the z-axis of the Cartesian coordinate system, the pixel angular equivalent of the star sensor detector in the x-direction, and the actual coordinates of each target in the N-1th frame image, calculate the x-axis component of the inter-frame displacement of each target from the N-1th frame image to the Nth frame image. Based on the angle of rotation of the star sensor around the x-axis of the Cartesian coordinate system, the angle of rotation of the star sensor around the z-axis of the Cartesian coordinate system, the pixel angular equivalent of the star sensor detector in the y-direction, and the actual coordinates of each target in the (N-1)th frame image, calculate the y-axis component of the inter-frame displacement of each target from the (N-1)th frame image to the Nth frame image.
5. The pseudo-star point removal method based on attitude quaternions according to claim 4, characterized in that, Based on the inter-frame displacement of each target from frame N-1 to frame N and the actual coordinates of each target in frame N-1, calculate the predicted coordinates of each target in frame N, including: Based on the x-axis component of the inter-frame displacement of each target from the (N-1)th frame image to the Nth frame image and the actual value of the x-axis coordinate component of each target in the (N-1)th frame image, calculate the predicted value of the x-axis coordinate component of each target in the Nth frame image; Based on the y-axis component of the inter-frame displacement of each target from the (N-1)th frame to the Nth frame and the actual value of the y-axis coordinate component of each target in the (N-1)th frame, calculate the predicted value of the y-axis coordinate component of each target in the Nth frame.
6. The pseudo-star point removal method based on attitude quaternions according to claim 5, characterized in that, For the i-th target among multiple targets, the formula for calculating the inter-frame displacement from the (N-1)-th frame to the N-th frame is: ; in, Let x be the x-axis component of the inter-frame displacement of the i-th target from the (N-1)-th frame to the N-th frame. Let y be the y-axis component of the inter-frame displacement of the i-th target from the (N-1)-th frame to the N-th frame. , , These represent the angles by which the star sensor rotates around the x-axis, y-axis, and z-axis of the Cartesian coordinate system, respectively. , These are the pixel angular equivalents of the star sensor detector in the x and y directions, respectively. Let be the actual coordinates of the i-th target in the (N-1)-th frame of the image.
7. The pseudo-star point removal method based on attitude quaternions according to claim 6, characterized in that, For the i-th target among the plurality of targets, the formula for calculating its predicted coordinates in the N-th frame image is as follows: ; in, Let be the predicted coordinates of the i-th target in the N-th frame of the image.
8. The pseudo-star point removal method based on attitude quaternions according to claim 1, characterized in that, If a target present in frame N does not exist in frame N-1, the target is directly identified as a false star point.
Citation Information
Patent Citations
Star sensor based star tracking method
CN106441282A
Spatial non-cooperative multi-target capturing and tracking algorithm
CN108519083A
