Patents
Literature
Patsnap Eureka AI that helps you search prior art, draft patents, and assess FTO risks, powered by patent and scientific literature data.

2results about How to "Reduce dependence on experience" patented technology

Image control point processing method and electronic equipment

The invention relates to the field of image processing, and provides an image control point processing method and electronic equipment, and the method comprises the steps: obtaining at least one to-be-detected control point in a to-be-detected image, setting a plurality of epipolar lines corresponding to the to-be-detected control point according to the relative relation between a plurality of reference images and the to-be-detected image, the to-be-inspected control points are automatically inspected as the stable control points or the unstable control points by means of the multiple epipolar lines, manual intervention is not needed, experience dependence and subjective deviation of workers are remarkably reduced, and the efficiency, accuracy and reliability of processing the image control points can be considered easily.
Owner:WUHAN TIANJIHANG INFORMATION TECH CO LTD

A reconfigurable metasurface beamforming design method based on CGAN and Gumbel-Sinkhorn

PendingCN122315358Aavoid mismatchHigh precisionMean squareAlgorithm
This invention proposes a reconfigurable metasurface beamforming design method based on CGAN and Gumbel-Sinkhorn. This method constructs a conditional generative adversarial network (GAN) as the inverse design network. The generator takes the target beam pattern as a conditional input to generate the corresponding encoding matrix. The Gumbel-Sinkhorn operator is introduced to discretize this continuous matrix, transforming it into an approximately doubly random permutation matrix. The encoding is sorted and selected by multiplying this permutation matrix with a predefined structure vector (SV), outputting a discrete 0 / 1 metasurface encoding array. A noisy forward prediction network is trained to predict the beam pattern corresponding to a given encoding array. In the training of the inverse design network, the discretized encoding matrix output by the generator is input into the noisy-trained forward prediction network. The mean square error between the predicted beam pattern and the target beam pattern is calculated as the target loss term to optimize the network parameters and accurately learn the complex mapping relationship from the beam pattern to the encoding array.
Owner:NANJING UNIV