An aspherical airborne camera has excellent performance in the field of photoelectric reconnaissance equipment. However, the image plane contrast of the optical system will be reduced by stray light originating from inside or outside of the optical system. In the present work, the self-designed aspheric airborne camera is manufactured with a baffle and vanes to reduce the impact of stray light on the camera imaging quality. TracePro software is used to analyze the stray light and establish an ABg mathematical model based on the scattering intensity measurement. The incident angle of the light parallel to the optical axis is set, and ray tracing is performed on the optical machine model to verify whether it conforms to the optical system design. The results showed that when the incident angle of the light source is greater than 0.5°, the point source transmittance (PST) value drops rapidly, and when the incident angle is 30°, the PST value of the system is in the order of 10⁻⁸. Stray light analysis verifies that the self-designed aspheric surface-based camera optomechanical structure has the ability to suppress stray light. The design of the baffle and vanes further enhances the ability of the optical system to suppress stray light, which can provide a reference for the design of a stray light elimination structure.

Incomplete remote sensing image observation data require proper processing to restore the original data. Nevertheless, the lack of information, due to noise and dead pixels, along with the quality of the acquired images that may be severely degraded, limits the accuracy of later processing. Therefore, this paper proposes a remote sensing image restoration scheme that employs the low-rank structure of the Bayesian probability tensor latent space and a new tensor completion method, which has strong robustness for the model selection process of the following remote sensing image data analysis. Considering the data space, we incorporate an adaptive nuclear norm regularization into the processing of latent factors that performs singular value decomposition on a more accurate scale. The proposed model adopts two algorithms: Augmented Lagrange Multiplier (ALM) and Alternating Least Squares (ALS). Several synthetic and real-data-based experiments illustrate the inherent ability of our method in restoring original information and preventing overfitting problems, even when a vast number of items are missing. In addition, the combination of adaptive nuclear norm regularization and tensor decomposition is stable, settling a high sensitivity of rank selection and low computational efficiency in traditional tensor reconstruction methods.