Chaotic Analysis And Suppression Of Sea Clutter In Image Sequence

In comparison with ground clutter and sea clutter, sea clutter dynamically changesin the time and space domain constantly, which seriously influences the detectability ofsea surface targets detection. Therefore, the technology of sea clutter suppression is theemphasis and difficulty in the field of sea surface target detection and it has significantvalue both in military area and in civil area.The intensity distribution of sea clutter in image sequence mainly depends onmarine dynamics, solar altitude angle and the position of station. Moreover, sea clutterin image sequence has the characteristic of nonlinear due to the nonlinearity of marinedynamics. In this paper, the theory of chaos in nonlinear dynamics is used to analyze thecharacteristic of sea clutter in image sequence, and the technology of sea cluttersuppression is deeply studied. The main contents are as follows:1. The theory of phase space reconstruction is studied. No matter the calculation ofcharacteristic values of chaotic system or the prediction based on chaos model, both areconducted in the phase space. The phase space reconstruction is a key step in theprocess of nonlinear time series. The calculation and selection of two importantparameters in the phase space reconstruction which are called embedding dimensionand delay time is discussed with emphasis. The method of autocorrelation is used toestimate the time delay and Cao’s method is used to estimate the embedding dimension.2. The method of chaotic characteristics identification is studied. Two importantcharacteristic values of chaotic system which are called correlation dimension and thelargest Lyapunov exponent are emphatically analyzed. The correlation dimension andthe largest Lyapunov exponent of the real sea clutter data are calculated by G-P methodand small-data method respectively. Experimental results show that the sea clutterimage sequence has finite correlation dimension and positive largest Lyapunovexponent which identify the chaotic characteristics of sea clutter in image sequence.3. A novel method of sea clutter suppression in the image sequence is proposed.This algorithm is based on chaos theory and phase space reconstruction theory. TheRadial Basis Function (RBF) neural network is used to reconstruct the nonlineardynamic model of sea clutter and predict the sea clutter time series, which finally realizethe suppression of sea clutter.4. A contrastive analysis of three clutter suppression method which are the Radial Basis Function (RBF) neural network method, the nonlinear median filtering, spatialhigh pass filter method and wavelet method is given according to the real sea clutter inimage sequence. And experimental results show that the Radial Basis Function (RBF)neural network method can effectively approach the nonlinear chaos dynamic model ofsea clutter by the use of the strong nonlinear mapping and learning ability of artificialneural networks (ANN).The study of the chaotic characteristics of sea clutter in image sequence will helpimprove the detectability of small dim target in sea clutter and it has certain theoreticand practical value.