High-precision light curtain detection method for impact point of high-speed target

Abstract

The invention provides a high-precision light curtain detection method for an impact point of a high-speed target, which comprises the following steps that: a near-infrared light source irradiates a target plate, camera shooting equipment continuously shoots near-infrared light images of the target plate and collects a before-shooting-trigger frame image, a shooting trigger frame image and an after-shooting-trigger frame image of each shooting action, and the three frame images are stored as a group of shooting bullet sequence frame images; two frame images are selected from the shooting bullet sequence frame images for frame difference processing to obtain a frame difference result image; gray threshold segmentation is performed on the frame difference result image to obtain a binary image; candidate bullet hole targets are identified from the binary image by using a bullet hole target identification search area template; and a real bullet hole target is screened out from the candidate bullet hole targets. According to the method, the near-infrared light images of the target plate are collected, the bullet hole impact point is accurately identified from the images. The method works in the near-infrared band and is not affected by immersive live ammunition shooting, and the training effect is not affected either.

Description

technical field [0001] The invention belongs to the technical field of impact point detection, in particular to a high-precision light curtain detection method for the impact point of a high-speed target. Background technique [0002] In special operations training, immersive live ammunition confrontation training is required, and real-time measurement of the coordinates of the bullet's target is required to provide a basis for rapid video switching and evaluate the strike effect. Detection equipment is required to be capable of high-resolution positioning (2mm), fast response <0.3s, and large-scale Monitoring (monitoring 54m×4m at the same time), at the same time, the monitoring system is required not to be affected by immersive live ammunition shooting, nor can it affect the training effect, especially the immersive projection video effect. [0003] There are several techniques for measuring impact indicators: thermal imaging cameras, infrared cameras, and visible light...

AI Technical Summary

Problems solved by technology

[0004] Infrared thermal imaging camera: It can quickly detect the heat generated by the target burn at the moment of live ammunition impact. The detection probability is high, but there are also false alarms. There are several problems that cannot meet the requirements of use: First, the number of pixels is low and the resolution is not high. , unable to meet high-precision positioning (required 2mm); second, the cost is too high. To realize large-scale (54m×4m) monitoring, the required equipment is huge, and the cost is too high to bear; During rapid fire training, the coordinates of each bullet cannot be determined

[0005] Infrared camera: The existing infrared camera measurement system is generally near-infrared (or short-wave infrared), or the visible light camera works in the near-infrared band. The camera cannot directly detect the coordinates of the live ammunition. The existing system is generally used to detect near-infrared lasers Simulated shooting positioning, the detection target is the near-infrared laser spot coordinates, not the actual crater, although it is also an immersive monitoring, it cannot be used for live ammunition training and cannot meet the requirements of use

[0006] Visible light array cameras use visible light-shaped spots to form a light curtain, and detect the shadow position on the screen when the bullet passes through the light curtain to calculate the coordinates of the bullet. This method can theoretically meet the monitoring requirements, accuracy, cost, and real-time performance. There are several problems in practical application: one is that the precision of the installation platform is too high (urad level), and the risk of successful debugging and testing is extremely high, especially for large-scale monitoring, which is more difficult; the other is that the visible light array laser spot is uniform The performance and stability requirements are extremely high, and the existing light sources cannot meet the requirements, otherwise the detection probability is extremely low or the false alarm rate is extremely high; the third is to meet the high resolution and high-speed target detection, camera pixels (greater than 5k line array) and frame The high frequency (greater than 50khz) has a large amount of image information, which requires extremely high real-time computer processing and real-time transmission bandwidth, and the cost is extremely high; Fourth, the positioning accuracy of different parts of the screen varies greatly, the positioning accuracy is nonlinear, and the local accuracy is too low; Fifth, it cannot handle curved screens, only flat screens; Sixth, the measurement error is related to the flight trajectory of the bullet. When the flight direction of the bullet is not perpendicular to the target board, the measurement error is relatively large

Images