Research On Small Target/Cluster Small Target Detection In Sea Environment

With the rapid development of small targets/clusters of small targets,the interference of factors such as sea clutter and the Doppler frequency range of small targets generally coincide with clutter.In addition,the target has a small signal-to-noise ratio and low echo energy.Object recognition brings difficulties.The detection and recognition of targets can be more effectively achieved by processing radar echo data to a certain extent.Therefore,echobased signal processing methods have important research on small target recognition and cluster target detection in complex environments significance.Because traditional timefrequency analysis methods cannot meet the processing of random signals in practical engineering applications,feature-based detection methods also have certain limitations.The paper explores a more in-depth target recognition method,and conducts related research on UAV recognition.Using the data fusion method,the feature difference between the target and the background is analyzed with the help of Cameron polarization coherent decomposition,and the roughness of the Hilbert-Huang transform is further used.degree definition to identify the target.Combined with the measured data,a target detection algorithm based on the IPIX data set,data fusion method and traditional time-frequency analysis is constructed,which improves the accuracy of small target detection and effectively realizes the method of distinguishing small targets from complex backgrounds.The main work of the thesis is as follows:1.Using the traditional method to deal with the measured echo.According to the measured data of IPIX,the statistical characteristics of the sea surface are summarized,and the processing results of a series of time-frequency algorithms for linear frequency modulation(LFM)and single-frequency signals are introduced.processing,and finally compare the detection results.In addition,a special time-frequency analysis algorithm such as HilbertHuang transform is introduced,and the steps of the algorithm are introduced.First,the signal EMD empirical mode solution is obtained,and then the Hilbert transform is performed on the obtained IMF natural mode function.The modal function energy ratio,the intrinsic modal function energy entropy,and the mean Hilbert spectral ridge roughness are used to distinguish the target from the clutter.2.The Cameron polarization coherent decomposition algorithm is introduced to analyze and process the sea clutter data measured by IPIX coherent radar.Through polarization coherent decomposition,the data of the four polarization channels collected by the radar are combined from the data level to realize the conversion from the data domain to the polarization domain.identify.Combining the characteristic parameter of roughness in Hilbert-Huang transform with Cameron polarization decomposition,a new more intuitive and accurate target recognition method is obtained,which combines the data fusion of polarization decomposition and Hilbert-Huang polarization decomposition.The accuracy of the transformation on data processing.3.Apply the new algorithm to small targets.A variety of major reconnaissance UAVs are modeled,and based on the new scenario of "sea surface + UAV",the corresponding radar echoes are simulated,and the results of this simulation are subjected to Hilbert-Huang transformation and polarization decomposition targets.The application of the detection method starts from the perspective of multiple features,and analyzes the difference between the target and the background.After that,the new algorithm proposed in the paper is applied,and the difference between the new algorithm and the above two algorithms is compared,and finally the effectiveness of the algorithm is analyzed.