Search

Owing to the disparity between the computing power and hardware development in electronic neural networks, optical diffraction networks have emerged as crucial technologies for various applications, including target recognition, because of their high speed, low power consumption, and large bandwidth. However, traditional optical diffraction networks and electronic neural networks are limited by long training durations and hardware requirements for complex applications. To overcome these constraints, this paper proposes an innovative opto-electronic hybrid system that combines optical diffraction networks with electronic neural networks. Using scattering layers to replace the diffraction layers in traditional optical diffraction networks, this hybrid system circumvents the challenging training process associated with diffraction layers. Spectral outputs of the optical diffraction network were processed using a simple backpropagation neural network, forming an opto-electronic hybrid network exhibiting exceptional performance with minimal data. For three-class target recognition, this network attains a classification accuracy of 93.3% within a substantially short training time of 9.2 s using only 100 data samples (training: 70 and testing: 30). Furthermore, it demonstrates exceptional insensitivity to position errors in scattering elements, enhancing its robustness. Therefore, the proposed opto-electronic hybrid network presents substantial application prospects in the fields of machine vision, face recognition, and remote sensing. [ABSTRACT FROM AUTHOR]

The performance of an active control system, crucial for the co-phase maintenance of segmented mirrors, is closely related to the spatial layout of sensors and actuators. This article compares two types of edge sensor layouts, vertical and horizontal, and proposes a novel tandem differential sensor layout that saves layout space and reduces the number of positioning references. The control performance of this scheme is analyzed in terms of error propagation, mode representation, and the scalable construction of the control matrix. Finally, we constructed a tandem differential-based sensor detection system to examine the performance of edge sensors and the effect of laboratory environmental variables on sensor measurements. Simulations and experiments demonstrate that this scheme has the same ability to fully characterize actuator modification modes as the Keck edge sensor layout. Although the total error multiplier is slightly larger than the latter, it has fewer scalable control matrix types and stronger spatial and segmental shape adaptation capabilities. Actual measurements show that the sensor's own noise in a tandem differential layout is less than 20 nm, which meets the requirements for future segmented co-phase maintenance. This layout type can potentially be applied to future small and medium-sized segmented splices. [ABSTRACT FROM AUTHOR]