Anti-smoke Interference Of Ship Target Detection Algorithm

China is a maritime power,and maritime security is directly related to the national development strategy in the future.As an important maritime military target,surveillance and strikes against ships are of great significance for maintaining maritime safety.Infrared imaging technology has been widely used in ship detection and strike system because of its unique advantages.However,with the development of interference technology,various interference methods have been widely used on ships,among which smoke screen interference is an effective Infrared interference mode.Under the interference of the smoke screen,the traditional infrared imaging detection system will be severely affected,even completely invalid.Therefore,how to effectively detect the ship target under the interference of the smoke screen is an urgent problem to be solved.In response to this problem,the paper uses the correlation characteristics of wakes,smoke screens and ship positions to carry out related researches on infrared smoke screens and wake characteristics,wake detection,smoke detection,ship detection,and fusion detection.The main work of this paper is as follows:Target characteristics are the basis for the design of target detection algorithms.The paper studies the infrared characteristics of smoke and ship wakes.Firstly,the physical mechanism of smoke screen formation is analyzed,and the spectral characteristics,texture characteristics and movement characteristics of the smoke screen are summarized.Then the edge features and geometric features of the ship’s wake on the infrared image are analyzed and compared with the interference characteristics of sea clutter.In order to solve the problem of difficult detection of the wake of the ship with dark and wide turbulence,a wake detection algorithm based on differential pyramid and multi-directional Gabor texture fusion is proposed.This method uses the construction of a differential pyramid to effectively enhance the characteristics of dark and wide turbulent wakes on multiple scales.Experimental results show that this method significantly improves the detection effect of turbulent wakes.Aiming at the problem of infrared smoke interference detection,a single-frame smoke detection algorithm based on the saliency of texture is proposed.This method can effectively distinguish the smoke screen from the background,the false alarm is very low,and the smoke screen detection effect is good,but the detection effect is not good for the smoke screen with high dissipation transmittance.To solve this problem,combining the dynamic characteristics of the smoke screen,a sequential smoke screen detection algorithm based on the saliency of texture optical flow is proposed.In order to solve the problem of large amount of optical flow calculation,the algorithm adopts the strategy of down-sampling the image and expanding the significant moving points,which significantly reduces the calculation amount while maintaining the performance of the algorithm.Experimental results show that this method significantly improves the detection effect of smoke screens.Aiming at the problem that it is difficult to effectively detect ship targets under the interference of smoke screen,a multi-attribute fusion decision-making method for ship detection is proposed.Through experiments,the influence of smoke screen interference on ship detection and wake detection is analyzed,and the wake detection algorithm is improved to reduce the image of smoke screen to wake detection.In order to improve the accuracy of ship target detection under interference conditions,a decision-level fusion ship target detection algorithm is constructed by comprehensively using smoke screen detection information,wake detection information and ship detection information.Experiments show that the decision fusion method can be effective Locate the ship target after the smoke interference.