Research On Photoelectric Tracking System And Its Automatic Focusing Technology

The optoelectronic tracking system passively detects and tracks moving objects through daylight cameras or infrared cameras, and is widely used in military and defense fields due to its good anti-jamming capability. In the process of tracking the target, the moving speed, the moving distance and the moving posture of the target are constantly changing, which causes the system to be blurred and need to be automatically focused in real time to make the image clear and meet the follow-up tracking algorithm. Therefore, the study of real-time reliable automatic focusing technology has important significance and value.This paper mainly carried out the following aspects of the work:1.This paper analyzes the principle of photoelectric tracking system, establishes the basic model of photoelectric imaging subsystem, and introduces the system imaging principle and related parameters. In this paper, various types of auto-focusing technology are studied deeply, and the characteristics of various focusing methods are summarized. According to the system requirements, automatic focusing technology based on image processing is selected.2.Aiming at the shortcomings of the existing focus window construction algorithm, this paper designs an improved dynamic window construction algorithm.This algorithm makes full use of the attention mechanism of human vision, and uses the gradient energy function to detect and extract the obvious features in the image as the focusing window, and achieves the purpose of constructing the window with accurate and suitable size through less computation.3.In this paper, the gray scale gradient operator of HVS (human visual system) is designed, and an adaptive resolution evaluation strategy based on HVS operator is proposed. The HVS operator takes full account of the grayscale changes in each direction of the image and distributes the weights according to the multi-channel characteristics of the human visual system in different directions. The evaluation function achieves high sensitivity with less computational complexity. For the image of the depth defocus region, the image sharpness evaluation value is calculated by reducing the sampling rate by dividing the sub - region, which greatly improves the anti - jamming performances of the evaluation function and widens the adjustable range of the system.4.This article designs the ARM Cortex-M4 and Cyclone IV FPGA dual-core embedded control system. Through programming on the control board embedded real-time operating .system, this article achieves the function of remote focusing and measuring the focal length. By using Verilog HDL language to program the image processing board, this paper realizes the function of collecting, decoding, displaying and clarifying the PAL analog video.5 .The algorithm and hardware control system designed in this paper are experimentally verified. Based on the fighter flight video data, the improved dynamic window construction algorithm and the accuracy evaluation function based on HVS operator are experimented respectively. The basic functions of the control board and image processing board are tested and the results are good.