ing at the problem of distortion of the decryption results in the existing optical encryption methods for color images, a new single-channel encryption method for color images that combines chaotic operation and Fresnel diffraction holography is proposed. The first encryption operation uses Fresnel diffraction to convert the RGB channel components of the color image into a real-valued computed hologram; the second encryption operation uses the modified Logistic chaotic system to replace and diffuse the pixels of the computed hologram. The results show that in addition to the traditional chaotic system keys, the Fresnel diffraction distance, the reference light wavelength, and the cosine of the incident angle direction as the key keys all increase the key space (about 10 ²⁴⁹) and have a small key volume. The fidelity of the decrypted image is high, and the evaluation indicators such as the correlation of adjacent pixels, information entropy, pixel number change rate, and normalized change intensity are all close to the ideal value. The histogram of the ciphertext image is flat, and the grayscale distribution is uniform, which completely hides the grayscale and color information of the original color image.

ing at the characteristic of the internal space limitation of the optical system of large aperture telescope, in order to realize the correction of the misalignment, a method for solving the misalignment based on the eigen coefficient is proposed. Firstly, based on the principle of wavefront curvature sensor, the light spot patterns are collected by alternately measuring the front and back defocused surfaces. Then, the wavefront is reconstructed by the eigenfunction method without partition detection, the eigen coefficients are used to characterize the system wave aberration,and the sensitivity matrix model is established according to the misalignments. Finally, the misalignments can be solved according to the eigen coefficients of the misalignment state and the ideal state. Compared with other technical approaches, this method has the characteristics of no need to add optical components, no partition detection, and simple operation. The experimental results by the telescope with the primary mirror diameter of 1.8 m show that when the eccentric distance range of the secondary mirror is from -0.9 mm to 0.9 mm and the tilt angle range is from -0.2° to 0.2°, the errors of the calculated values obtained by the eigen coefficient sensitivity matrix method are less than 10%, which has certain significance for the application of large aperture telescopes.