CN-121720329-B - Rocket take-off section speed measurement method and system based on event camera

CN121720329BCN 121720329 BCN121720329 BCN 121720329BCN-121720329-B

Abstract

The invention discloses a method and a system for measuring the speed of a rocket take-OFF section based ON an event camera, which belong to the technical field of image processing, wherein the method comprises the steps of carrying out space accumulation ON ON events generated by capturing rocket tail flame by the event camera to form an ON event density map, estimating a tail flame region from the ON event density map, and acquiring the edge position of the rocket bottom in the tail flame region, thereby constructing an effective region only comprising rocket bodies; and overlapping the OFF events in the effective area into a binary image, extracting a linear line segment from the binary image, taking the displacement of the linear line segment as the displacement of the rocket, forming a displacement-time curve by combining the timestamp of the binary image, and obtaining a speed-time curve by solving the gradient of the displacement-time curve. The invention can still realize high robustness and high precision speed measurement of the rocket take-off section under the complex conditions of strong light interference, tail flame disturbance and the like.

Inventors

ZHAO HAOZHI
ZHANG YUSHAN
YAN LUXIN
LIU HAOYUE
CHANG YI
XU JINGHAN
FENG LUXIN

Assignees

华中科技大学

Dates

Publication Date: 20260512
Application Date: 20260225

Claims (9)

1. The method for measuring the speed of the rocket takeoff section based on the event camera is characterized by comprising the following steps of: (1) Carrying out space accumulation ON ON events generated by capturing rocket tail flame by an event camera to form an ON event density map, estimating a tail flame region from the ON event density map, and acquiring the edge position of the rocket bottom in the tail flame region, thereby constructing an effective region only comprising rocket bodies; (2) Overlapping the OFF events in the effective area into a binary image, extracting a linear line segment from the binary image, taking the displacement of the linear line segment as the displacement of a rocket, forming a displacement-time curve by combining the time stamp of the binary image, and obtaining a speed-time curve by gradient of the displacement-time curve; The effective area is extracted by the following method: equivalent focal length parameters for width and height of combined event camera The actual width W, the actual height H of the rocket body and the distance D from the rocket center point to the event camera are calculated according to the formula Calculating to obtain the pixel width of rocket body in event frame Pixel height The edge position of the bottom of the rocket is firstly obtained in the tail flame area, and then the edge position is obtained according to the pixel width Pixel height And obtaining an effective area of the rocket body in the event frame.
2. A method for measuring velocity of a rocket takeoff section based on an event camera as recited in claim 1, wherein the specific means for estimating the tail flame area is as follows: Performing space accumulation ON ON events in a time window to form an ON event density map, constructing a histogram of the ON event density map, traversing all thresholds, calculating the inter-class variance of a tail flame high-density region and a tail flame low-density region under each threshold, selecting the threshold with the maximum inter-class variance as an optimal threshold, extracting the tail flame high-density region from the histogram of the ON event density map, performing rectangular fitting or minimum circumscribed region screening ON the tail flame high-density region, combining rocket nozzle positions, determining tail flame candidate regions, detecting the position and area change of the tail flame candidate regions in adjacent time windows, performing smoothing and weighted averaging ON the tail flame candidate regions with the change exceeding a preset value, and estimating the tail flame region.
3. A method for measuring velocity of a rocket takeoff section based on an event camera as recited in claim 1, wherein the specific means for estimating the tail flame area is as follows: Performing space accumulation ON ON events in a time window to form an ON event density map, constructing a histogram of the ON event density map, traversing all thresholds, calculating the information entropy of a tail flame high-density region and a tail flame low-density region under each threshold, selecting the threshold with the maximum information entropy as an optimal threshold, extracting the tail flame high-density region from the histogram of the ON event density map, performing rectangular fitting or minimum circumscribed region screening ON the tail flame high-density region, determining tail flame candidate regions by combining rocket nozzle positions, detecting the position and area change of the tail flame candidate regions in adjacent time windows, performing smoothing and weighted averaging ON the tail flame candidate regions with the change exceeding a preset value, and estimating the tail flame region.
4. A method for measuring the speed of a rocket takeoff section based on an event camera as claimed in any one of claims 1-3, wherein the binary image is sequentially subjected to a closing operation and an opening operation before the linear line segment is extracted from the binary image in the step (2).
5. A method for measuring the speed of a rocket takeoff segment based on an event camera according to any one of claims 1-3, wherein the step (2) extracts a straight line segment from a binary image by using a straight line segment detection method based on gradient analysis or hough transform.
6. A method for measuring the speed of a rocket takeoff section based on an event camera according to any one of claims 1-3, wherein in the step (2), a plurality of straight line segments are extracted from a binary image, the average displacement of the plurality of straight line segments is used as the displacement of the rocket, a displacement-time curve is formed by combining with the time stamp of the binary image, and the gradient of the displacement-time curve is obtained by using a first-order difference or polynomial fitting, so as to obtain the speed-time curve.
7. A rocket takeoff segment speed measurement system based on event cameras, comprising: The rocket body extraction module is used for carrying out space accumulation ON ON events generated by capturing rocket tail flame by the event camera to form an ON event density map, estimating a tail flame area from the ON event density map, and acquiring the edge position of the rocket bottom in the tail flame area, thereby constructing an effective area only comprising rocket bodies; the speed measurement module is used for superposing the OFF events in the effective area into a binary image, extracting a linear line segment from the binary image, taking the displacement of the linear line segment as the displacement of the rocket, combining the timestamp of the binary image to form a displacement-time curve, and obtaining a speed-time curve by solving a gradient of the displacement-time curve; The effective area is extracted by the following method: equivalent focal length parameters for width and height of combined event camera The actual width W, the actual height H of the rocket body and the distance D from the rocket center point to the event camera are calculated according to the formula Calculating to obtain the pixel width of rocket body in event frame Pixel height The edge position of the bottom of the rocket is firstly obtained in the tail flame area, and then the edge position is obtained according to the pixel width Pixel height And obtaining an effective area of the rocket body in the event frame.
8. An electronic device comprising a processor and a computer readable storage medium, wherein: The computer readable storage medium stores program instructions for implementing the steps of a rocket takeoff segment speed measuring method based on an event camera according to any one of claims 1 to 6, and the processor is used for implementing the whole process of rocket takeoff segment speed measurement when executing the program instructions.
9. A computer product, characterized in that it, when run, causes a computer to perform the steps of a method for measuring the speed of a rocket in the take-off section based on an event camera according to any one of claims 1 to 6.

Description

Rocket take-off section speed measurement method and system based on event camera Technical Field The invention belongs to the technical field of image processing, and particularly relates to a rocket takeoff section speed measuring method and system based on an event camera. Background The rocket take-off section belongs to a typical high-speed strong radiation scene and is accompanied with complex phenomena such as strong tail flame, smoke entrainment, high-brightness nozzle light clusters and the like. The traditional optical measurement method, such as a frame imaging method based on a high-speed camera, mainly has the problems that (1) strong light interference causes image overexposure, a large amount of high-brightness areas generated by rocket tail flame are extremely easy to cause local saturation of an image sensor, so that structural characteristics of an rocket body cannot be identified, (2) tail flame disturbance causes unstable contour detection, and tail flame transient high-speed pulsation, so that a contour extraction algorithm based on a frame image is easy to be disturbed, and a steady motion track cannot be obtained, (3) the data size is large, so that real-time measurement is difficult, and the high-speed camera needs to generate a large amount of frame data, so that a real-time measurement system is not beneficial to deployment in a transmitting field. The event camera is a novel sensor with nerve morphology, each pixel is triggered asynchronously, and the sensor has the characteristics of low delay (microsecond level), low data redundancy, high dynamic range and the like, and has wide utilization value in the field of high-speed wide dynamic imaging. Thus, the event camera may be an effective modality of rocket launch phase velocity measurement. However, the rocket tail flame has extremely high event density, and the tail flame area is mixed with the rocket body area, so that the technical problems of poor precision and poor robustness still exist in the process of directly using the event signals for speed measurement and calculation. Therefore, the existing speed measuring and calculating technology based on the event signals has the technical problems of poor precision and poor robustness. Disclosure of Invention Aiming at the defects or improvement demands of the prior art, the invention provides a rocket take-off section speed measurement method and system based on an event camera, and the technical problems of poor precision and poor robustness of the existing speed measurement technology based on event signals are solved. In order to achieve the above object, according to a first aspect of the present invention, there is provided a method for measuring a speed of a rocket takeoff section based on an event camera, comprising the steps of: (1) Carrying out space accumulation ON ON events generated by capturing rocket tail flame by an event camera to form an ON event density map, estimating a tail flame region from the ON event density map, and acquiring the edge position of the rocket bottom in the tail flame region, thereby constructing an effective region only comprising rocket bodies; (2) And overlapping the OFF events in the effective area into a binary image, extracting a linear line segment from the binary image, taking the displacement of the linear line segment as the displacement of the rocket, forming a displacement-time curve by combining the timestamp of the binary image, and obtaining a speed-time curve by solving the gradient of the displacement-time curve. Further, the specific mode of tail flame region estimation is as follows: Performing space accumulation ON ON events in a time window to form an ON event density map, constructing a histogram of the ON event density map, traversing all thresholds, calculating the inter-class variance of a tail flame high-density region and a tail flame low-density region under each threshold, selecting the threshold with the maximum inter-class variance as an optimal threshold, extracting the tail flame high-density region from the histogram of the ON event density map, performing rectangular fitting or minimum circumscribed region screening ON the tail flame high-density region, combining rocket nozzle positions, determining tail flame candidate regions, detecting the position and area change of the tail flame candidate regions in adjacent time windows, performing smoothing and weighted averaging ON the tail flame candidate regions with the change exceeding a preset value, and estimating the tail flame region. Further, the specific mode of tail flame region estimation is as follows: Performing space accumulation ON ON events in a time window to form an ON event density map, constructing a histogram of the ON event density map, traversing all thresholds, calculating the information entropy of a tail flame high-density region and a tail flame low-density region under each threshold, selecting the threshold with the maximum informa