Research On Target Detection Method In Sea Clutter

Target detection in the background of sea clutter plays a key role in military defense.The ocean area accounts for about three-quarters of the world's total area.Radar detection of targets in sea clutter is of great significance in maritime safety,disaster search and rescue,environmental detection,and strategic early warning.As the radar resolution increases,the sea clutter exhibits non-stationary and non-Gaussian characteristics in space and time.This paper uses the measured data published by McMaster University to analyze the characteristics of pure clutter and explores and studies the detection of sea surface targets.This paper studies the adaptive detection algorithm,and verifies the effectiveness of the algorithm for detecting target on the sea surface when the parameters such as cumulative pulse length,Doppler frequency,signal-to-noise ratio and radar range resolution are different.This paper simulates point target and distributed targets on the background of pure sea clutter measured by IPIX radar.Then,the reference samples are used to estimate the covariance matrix of the sea clutter.Finally,the estimation matrix is sent to the coherent accumulation detector to obtain the detection results under different parameters,and the effectiveness of the adaptive detection algorithm is verified.By studying the characteristic of sea clutter,a floating small target detection method based on joint fractal features in the fractional Fourier transform domain is proposed.This paper studies the pure sea clutter echo data and extracts two fractal features to form a two-dimensional decision space.On this basis,the fractional Fourier transform is used to gather the target echo energy,and the difference between the clutter and the target echo is expanded in the two-dimensional decision space.The paper introduces radial basis neural network and support vector machine as detectors,and selects the training set to train a suitable model to detect radar echoes.The results show that the detection accuracy of this algorithm is better than the comparison algorithm.The algorithm still has good detection performance when adding the affected range unit echo.This paper proposes the concept of the intrinsic mode cross correlation coefficient entropy,which is used as a feature to detect small floating targets on the sea surface.Empirical mode decomposition is ideal for analyzing non-stationary sea clutter.By studying the characteristics of each component and combining the cross correlation coefficient criterion,the new feature of the intrinsic mode correlation coefficient entropy is obtained.The eigenvalues of the IPIX radar echo data are extracted and sent to the machine learning classifier.The experimental results show that the detection performance of the algorithm based on this feature is better than the comparison feature.