Acoustic-Electromagnetic Scattering-Based for Covert Target Detection

Measuring and analyzing the characteristics of a target and identifying it by radar is a viable method of dealing with covert target detection in modern radar-based electronic warfare. However, existing covert target detection techniques make it difficult to capture target feature information and impossible to accurately assess the target due to the small scattering cross Section of the target and the low contrast with the environment. To this end, this article proposes a hidden target detection method based on acoustic-electromagnetic scattering, which can realize noncontact detection of targets in hidden scenes. Specifically, the method couples the acoustic energy into the hidden target through the acoustic field, induces the target to produce the micro-motion effect, and excites the target to produce the nonlinear characteristics with its physical characteristics, which effectively solves the problem of insufficient characteristics of the echo signal due to the small scattering cross Section of the hidden target. In our experiments, we simulated the boundary conditions of the acoustic field distribution with the target deformation. Through a convolutional neural network model based on multifeature fusion, we effectively differentiate between four different materials of covert targets, and giving quantitative predictions in the confusion matrix. The experimental results show that the object exhibits maximum amplitude when the acoustic frequency reaches the material's resonance frequency at 41 kHz. Using our proposed multifeature fusion convolutional neural network model can achieve 77% accuracy, which is better than the single-feature neural network model, and successfully realize the detection task of four materials in a hidden scene.