7results about How to "Increase information entropy" patented technology

The invention relates to the technical field of computer vision, and provides a text image matching method and system based on large language model entropy increase, and the method comprises the steps: enabling data set image features to pass through a visual attention encoder to obtain original image codes; constructing a hypergraph based on original image coding and generating hypergraph image features through a hypergraph neural network; performing residual connection on the hypergraph image features and original image codes to construct a visual hypergraph adapter; inducing a large language model to generate a synonym material through a prompt learning method by utilizing a first text feature generated by the image feature description, and obtaining an entropy-increased fusion text feature; encoding the fused text feature and the first text feature; constructing a hypergraph based on multi-dimensional feature coding, and generating hypergraph text features through a hypergraph neural network; performing dimension reduction on the supergraph text features, and then combining with original corpus feature codes to construct a text supergraph adapter; and calculating a similarity matrix based on the visual hypergraph adapters and the text hypergraph adapters to obtain a similarity ranking.

The invention discloses a deep forged audio identification method based on hybrid enhancement and a contrast loss function. The method comprises the following steps: step 1, preprocessing an original audio training data set by using a sliding window technology and a hybrid enhancement technology; step 2, for the preprocessed original audio training data set, using a contrast loss function to constrain and forge feature vector spacing of real audio training data in a high-dimensional space; and step 3, using a sliding window to divide the audio to be detected into a plurality of audio clips to be detected with the total length of 64600, and if any audio clip to be detected is detected to be forged, determining that the audio to be detected is forged. According to the method, the analysis and identification capability of the algorithm on weak and local counterfeit traces is improved, the accuracy, stability and generalization of the deep counterfeit audio identification algorithm are ensured, the extraction and analysis capability of the algorithm on essential counterfeit traces is improved, and the interference of environmental noise is reduced.

The invention discloses a high-precision radar target recognition system based on orthogonal waveform design and a convolutional neural network. The high-precision radar target recognition system comprises a radar orthogonal waveform design module, a waveform generation module, a transmitting antenna module, a receiving antenna module, a waveform receiving module and a convolutional neural network module, by designing an orthogonal radar transmitting pulse string, the receiving entropy is increased, so that a radar echo pulse string received by a receiving antenna contains enough target information, and strong difference is presented between a target and clutter as well as multipath interference characteristics; therefore, the convolutional neural network can extract target key features from a radar echo pulse string received by an antenna and suppress clutter and multipath interference at the same time. According to the method, radar echoes are directly processed in the electromagnetic original domain, features are efficiently extracted, high-precision recognition is achieved, traditional multi-step signal processing is integrated into one step, the process is simplified, the hardware complexity of a receiver can be reduced possibly, and the method has great significance in improving the performance of an electronic information system.

This invention discloses a reconfigurable job shop scheduling method based on cooperative multi-agent deep reinforcement learning, belonging to the field of intelligent manufacturing and job shop scheduling technology. Addressing the reconfigurable job shop scheduling problem under constraints of limited auxiliary modules and limited extended processing functions, this invention constructs a cooperative multi-agent deep reinforcement learning framework, decoupling job scheduling and equipment function reconfiguration into two sub-problems and setting independent agents. Collaborative optimization is achieved through a prior knowledge interaction mechanism. A heterogeneous graph embedding model is designed to extract state features; a rule mining method based on gene expression programming is introduced to initialize the experience replay pool; a reward-based decision space reduction mechanism is proposed to reduce computational complexity; and a target network soft update and local search restart mechanism are adopted to improve training stability and global search capability. Experiments show that the average RPD of this invention is significantly better than existing methods on 100 test cases, and it is suitable for multi-variety, small-batch reconfigurable intelligent manufacturing scenarios such as automotive parts and aerospace structural components.

The invention discloses an infrared polarization multi-dimensional fusion imaging method based on nonlinear mapping, and belongs to the technical field of photoelectric detection and computational imaging. Performing Stokes vector solution on the input detection data to obtain a total intensity component, a polarization degree component and a polarization angle component; performing logarithmic domain mapping and gradient calculation on the total intensity component, and performing nonlinear enhancement based on local statistical characteristics on the polarization degree component to obtain an enhanced polarization degree; in an HSV color space, constructing a brightness component according to a logarithm domain mapping result, constructing a saturation component by combining the enhanced polarization degree and gradient information, constructing a hue component according to polarization angle information, and performing threshold constraint on the hue component based on the enhanced polarization degree; and a multi-dimensional fusion image is generated through color space inverse transformation from HSV to RGB. According to the method, background clutters are effectively suppressed, the image information entropy and the average gradient are remarkably improved, and high-contrast clear imaging of a weak and small target in a complex scene is realized.

This invention discloses a two-dimensional heterojunction array sensing method and device for high-precision monitoring of carbon pollution. The two-dimensional heterojunction array sensing method includes: emitting light signals with dynamically changing wavelength, power, and modulation frequency to a two-dimensional heterojunction array using a tunable light source according to a preset switching sequence; acquiring the electrical signals generated by the two-dimensional heterojunction array under the combined action of the target gas and the light signals, and converting the electrical signals into digital signals in multi-dimensional matrix form; and using a hybrid model integrating convolutional neural network branches and recurrent neural network branches to extract spatial pattern features and dynamic temporal features, ultimately outputting the type and concentration of the gas. This invention effectively solves the cross-sensitivity problem in gas identification by exciting the two-dimensional heterojunction array with optical sequence modulation and combining it with a lightweight neural network model, thus improving the identification accuracy and response speed in complex environments.

The present invention relates to the field of molecular marker technology, and specifically discloses a pig whole genome low-density 5K SNP chip and its application. The pig whole genome low-density 5K SNP chip provided by the present invention includes a SNP molecular marker combination for pig genotyping, and the SNP molecular marker combination consists of 5603 SNP molecular markers, and the SNP molecular markers are shown in Table 1. The SNP chip or molecular marker combination provided by the present invention can be applied to multiple fields such as pig whole genome breeding value evaluation, cluster analysis, molecular marker assisted breeding, etc., greatly reducing breeding costs.