Fig. 1. Application of a real-time all-directional 3D recognition multiple distortion correction system in automatic driving. The metasurface array supported by the deep knowledge prior diffraction neural network can recognize obstacles or people with different poses in a large field of view.

Fig. 2. Design of prior diffraction neural network. (a) Working process of the correction–recognition diffraction neural network based on deep knowledge prior. (b) Detailed network model of prior diffraction neural network.

Fig. 3. Dynamic and static three-dimensional recognition in the actual scene. (a) The camera lens is not parallel to the picture, resulting in perspective distortion. Moreover, the object has a different degree of plane distortion due to the inappropriate shooting angle. (b) Specific plane distortions caused by subjective and objective causes. Under nonideal shooting conditions, the picture has single or double plane random variation. (c) Different attitudes of the aircraft from the same perspective. The superimposed random distortion of the plane and three-dimensional space caused by the shooting angle. (d) Aircraft at different attitudes observed from different viewing angles. (e) To simulate the identification of aircraft by observers in the actual scene, the aircraft with different attitudes are recognized by the diffraction system in the experiment. (f) The training results of the deep phase generation network show that the phase loss of each metasurface is close to 0. (g) The loss of the training and test sets versus epoch. After thousands of epochs, the correction–recognition integrated system achieves an accuracy rate of 97.7% and 94.6% for the training and test sets, respectively. (h) The training results for the conventional diffraction neural network over the epoch. It is in an underfitting state, and the accuracy is only 55.2% for the test set.

Fig. 4. Experimental results of multiple distortion correction and static aircraft real-time recognition. (a) Experimental setup for planar and three-dimensional distortion correction and three-dimensional recognition. The aircraft is suspended on a graduated rack to simulate different flight attitudes. (b), (c) Experimental results of image restoration with double random distortion, including stereoscopic distortion. (d) The experimental recognition results are identified according to the diffraction distribution under different attitudes. The orientation of the aircraft is demonstrated and labeled with (θ,φ) in the left coordinate system.

Fig. 5. Experimental demonstration of dynamic three-dimensional recognition. (a) Measured electric field distribution with dynamic left-right steering (φ change) (Video 1, MP4, 1.92 MB [URL: https://doi.org/10.1117/1.AP.7.5.056005.s1]). (b) Measured electric field distribution with dynamic forward and backward tilt (θ change) (Video 2, MP4, 1.70 MB [URL: https://doi.org/10.1117/1.AP.7.5.056005.s2]). (c) Measured electric field distribution with dynamic movement in all directions (Video 3, MP4, 1.84 MB [URL: https://doi.org/10.1117/1.AP.7.5.056005.s3]).