CN-121985193-A - Video axis stable video generation method and device based on dynamic fuzzy integration

CN121985193ACN 121985193 ACN121985193 ACN 121985193ACN-121985193-A

Abstract

The invention provides a video axis stable video generation method and device based on dynamic fuzzy integration, which are characterized by collecting inertial pointing data of a servo system under an inertial stable model, reading gray values of an initial scene picture as reference frames, respectively carrying out interpolation processing on collected azimuth angle, pitch angle and roll angle data to generate a high-density pointing angle sequence, calculating azimuth stability precision and pitching stability precision, carrying out reverse rotation on the reference frames to generate imaging slices, overlapping all the imaging slices corresponding to single frame exposure time pixel by pixel, carrying out result normalization processing, and overlapping character information comprising a field angle, stability precision and imaging parameters on each frame of image to generate a video axis stable effect video file. According to the embodiment of the invention, the visual axis stable precision effect video is generated according to the pointing data and the imaging sensor parameters, so that the intuitiveness of the stable precision effect display in the design stage can be improved, and the dependence of the servo system in the physical test stage on the image link of the photoelectric equipment can be reduced.

Inventors

LI ZONGYA
TU BIAO
WANG LIANG
GAO BIN
YAO MENG
ZHOU JIANZHONG

Assignees

华中光电技术研究所(中国船舶集团有限公司第七一七研究所)

Dates

Publication Date: 20260505
Application Date: 20251229

Claims (10)

1. A method for generating a video-axis stable video based on dynamic fuzzy integration, the method comprising: S101, collecting inertial pointing data of a servo system under an inertial stability model in a servo system simulation or actual measurement process, wherein the inertial pointing data comprises azimuth angle, pitch angle and roll angle of a visual axis, data updating frequency is not lower than 1kHz, and total duration of data is not lower than 20 seconds; S102, setting parameters, namely setting parameters of exposure time, frame frequency, image resolution and angle of view of an imaging sensor, and reading gray values of an initial scene picture as a reference frame; s103, data processing, namely respectively carrying out interpolation processing on the collected azimuth angle, pitch angle and roll angle data according to the exposure time to generate a high-density pointing angle sequence, and calculating azimuth stability precision and pitch stability precision; s104, generating an image, namely reversely rotating the reference frame according to the rolling angle corresponding to each unit time by taking the time interval of the interpolated pointing angle sequence as a unit to generate imaging slices; s105, generating a video, namely superposing character information comprising a view angle, stability precision and imaging parameters on each frame of image generated in the step S104, and combining all single frame images in time sequence to generate a video file with a stable view axis effect.
2. The method for generating the stable video of the visual axis based on the dynamic fuzzy integration according to claim 1, wherein the interpolation processing adopts a cubic spline interpolation, piecewise linear interpolation or radial basis function interpolation method, and the interpolated data time interval is more than 20 times of the interpolated data time interval.
3. The method for generating the video stabilization based on the dynamic fuzzy integration according to claim 1, wherein the inertial pointing data is derived from data output by simulation software of a servo system or data obtained by actual measurement of a gyroscope or an inertial measurement unit IMU in a physical photoelectric device.
4. The method for generating a stable video of a visual axis based on motion blur integration according to claim 1, wherein the generating an image comprises: Establishing a time axis, aligning the interpolated angle data sequence with imaging time, and determining an angle data sequence segment corresponding to the exposure time of each frame of image; According to the reference frame and the rolling angle corresponding to each angle data time unit, obtaining an imaging slice gray matrix corresponding to the time unit through coordinate transformation calculation; Summing imaging slice gray matrixes corresponding to all time units in the exposure time of a frame of image pixel by pixel to obtain an accumulated gray matrix; and normalizing the values of each element in the accumulated gray matrix to a preset image gray range to obtain the image data of the frame.
5. The method of claim 1, wherein the superimposed character information includes at least a horizontal field angle, a vertical field angle, an azimuth stabilization precision, a pitch stabilization precision, a frame rate, an exposure time, and a current frame number.
6. The method for generating a dual motion blur integration based video on stable video on visual axis of claim 1, wherein a fixed reference mark is superimposed on the center of each generated frame of image. The evaluation result G i of the subsystem health state evaluation is a degradation degree evaluation weight Z i and a subsystem performance evaluation weight.
7. A video axis stabilized video generating apparatus based on dynamic fuzzy integration, based on a video axis stabilized video generating method based on dynamic fuzzy integration as claimed in claims 1 to 6, characterized in that said apparatus comprises: the data acquisition module is used for acquiring inertial pointing data of the servo system under an inertial stabilization model in the simulation or actual measurement process of the servo system, wherein the data comprise an azimuth angle, a pitch angle and a roll angle; the parameter configuration module is used for setting exposure time, frame frequency, image resolution and angle of view parameters of the imaging sensor and loading an initial scene reference frame image; the data processing module is used for interpolating the collected pointing angle data to generate a high-density sequence and calculating azimuth and pitching stability precision; the image synthesis module is used for carrying out time unit-by-time unit rotation on the reference frame according to the interpolated roll angle sequence to generate imaging slices, and superposing and normalizing all slices within the single frame exposure time to synthesize a single frame image; And the video generation module is used for superposing parameter characters on the synthesized single-frame image and sequentially combining the parameter characters to generate a video file.
8. The device of claim 7, wherein the interpolation algorithm performed by the data processing module is cubic spline interpolation, and ensures that the interpolated data time interval satisfies the exposure time by more than 20 times of the interval.
9. The method of claim 7, wherein the operations performed by the image synthesis module include coordinate rotation, pixel translation, matrix summation, and normalization.
10. A computer readable storage medium having stored thereon a computer program, wherein the computer program when executed by a processor implements a dual motion blur integration based visual axis stabilized video generation method according to any one of claims 1 to 6.

Description

Video axis stable video generation method and device based on dynamic fuzzy integration Technical Field The invention belongs to the technical field of motion control of photoelectric equipment, and particularly relates to a method and a device for generating a video-axis stable video based on dynamic fuzzy integration. Background The visual axis stability precision of the photoelectric equipment refers to the stability of an optical axis under the conditions of a movable platform such as a vehicle, an airborne vehicle and a ship, and the visual axis stability precision reflects the capability of isolating carrier disturbance and is a core index of a servo system of the photoelectric equipment of the movable platform. At present, two methods for evaluating the stabilizing effect in the design stage of a servo system are mainly adopted: The first method is based on servo system simulation or test, gyro data or geographic system pointing angle data of the photoelectric equipment are obtained, stability accuracy and jitter peak value are calculated based on the data, an angle curve is drawn, and stability effect is evaluated based on the stability accuracy, the jitter peak value and the angle curve. The method is used in the servo system design stage, and has the defects that a professional with more experience is required to accurately understand the stabilizing effect based on the numerical value, and the method is not as visual and easy to understand as a video. And secondly, based on physical testing of photoelectric equipment, inertial pointing angle data and a real imaging video are obtained, frame-by-frame image processing is carried out on the video, a pointing angle sequence of a fixed reference target in the video is extracted, stability precision and a jitter peak value are calculated based on the angle sequence, an angle curve is drawn, and a stability effect is evaluated based on the stability precision, the jitter peak value, the angle curve and the actual video. The method has the defects of large consumption of manpower and material resources for constructing a test environment, high requirements on meteorological conditions and fixed reference targets, and more software such as data storage, video acquisition, data analysis and video analysis. Therefore, how to provide a method and a device for generating a stable video of a visual axis based on dynamic fuzzy integration, which generate a stable video of an effect of the visual axis according to pointing data and imaging sensor parameters, restore the phenomenon of intra-frame blurring, improve the intuitiveness of stable effect display of the precision in a design stage, and reduce the dependence of a servo system in a physical test stage on an image link of a photoelectric device has become a technical problem to be solved urgently. Disclosure of Invention The embodiment of the invention provides a video axis stable video generation method and device based on dynamic fuzzy integration, which are used for generating video axis stable precision effect videos according to pointing data and imaging sensor parameters, restoring intra-frame fuzzy phenomenon, improving intuitiveness of stable precision effect display in a design stage and reducing dependence of a servo system in a physical test stage on an image link of photoelectric equipment. In one embodiment of the present invention, there is provided a video generating method based on dynamic blur integration, the method including: S101, collecting inertial pointing data of a servo system under an inertial stability model in a servo system simulation or actual measurement process, wherein the inertial pointing data comprises azimuth angle, pitch angle and roll angle of a visual axis, data updating frequency is not lower than 1kHz, and total duration of data is not lower than 20 seconds; S102, setting parameters, namely setting parameters of exposure time, frame frequency, image resolution and angle of view of an imaging sensor, and reading gray values of an initial scene picture as a reference frame; s103, data processing, namely respectively carrying out interpolation processing on the collected azimuth angle, pitch angle and roll angle data according to the exposure time to generate a high-density pointing angle sequence, and calculating azimuth stability precision and pitch stability precision; s104, generating an image, namely reversely rotating the reference frame according to the rolling angle corresponding to each unit time by taking the time interval of the interpolated pointing angle sequence as a unit to generate imaging slices; s105, generating a video, namely superposing character information comprising a view angle, stability precision and imaging parameters on each frame of image generated in the step S104, and combining all single frame images in time sequence to generate a video file with a stable view axis effect. Further, the interpolation processing adopts a cubic spline int