13 results about "Hypercomplex number" patented technology

The invention provides a ship detection-oriented multi-source remote sensing image fusion method, a ship detection-oriented multi-source remote sensing image fusion medium and a ship detection-oriented multi-source remote sensing image fusion system, and belongs to the technical field of remote sensing images. Constructing an electromagnetic scattering constraint model based on physical optical approximation and a geometric diffraction theory to calculate a predicted value of the backscattering cross section of the target, introducing a fusion cost function, generating a fusion weight matrix by adopting an adaptive feature weighting model, and performing multi-scale fusion on the image through hypercomplex wavelet transform; and iteratively optimizing the fusion result by using an information diffusion heat conduction equation, and performing physical consistency correction to output an enhanced fusion image, thereby solving the technical problem of poor physical consistency of the fusion result caused by lack of physical model constraint when the synthetic aperture radar image and the multispectral image are fused.

This invention introduces a novel algebraic framework known as the Omnidimensional Hypercomplex System (OHS), defined over an n -dimensional vector space. It is characterized by a positive-squared real temporal axis e 0, intermediate imaginary dimensions e j satisfying (I)= -1 for 1≤ j≤ n-2, and a final dimension e n-1 satisfying (II) = + 1 if and only if n is odd. Dimensional interactions follow non-commutative multiplication rules governed by structure constants (III) and quantum deformation parameters n jk. This system enables the unification of general relativity and quantum mechanics within a single mathematical structure. It also allows for the implementation of innovative industrial applications, including quantum communication, encryption, battery technologies, navigation, multi-use processors, and more. The OHS-17 framework provides a unified model for activating hidden dimensions in a physically applicable form, representing a groundbreaking advance in the mathematical and technological description of hypercomplex dimensional spaces.

The present invention provides a convolutional quaternion-based inverse relational rotation embedding knowledge representation method and system, which relates to the field of knowledge graph technology. The method comprises: mapping the entities and relationships of triples in the acquired knowledge graph to a complex vector space, wherein the entities include a head entity and a tail entity, and each relationship is defined as a rotation from the head entity to the tail entity; generating embedding vectors for the head entity and the tail entity so that the embedding of each entity has two vectors, and using an inverse relational embedding function to enhance the mutual dependence between the two vectors; introducing a hypercomplex representation to model entities and relationships, and inputting them into a preset convolutional neural network for score prediction and loss correction to obtain an embedded knowledge representation of the triples. In this way, a rich semantic match can be achieved between the head entity and the tail entity, thereby improving the accuracy of knowledge graph link prediction.

The present disclosure relates to a system and methodology designed for the spectroscopic detection of materials found on surfaces or dispersed in air. Some embodiments utilize a multi-modal sensor suite that operates through a progressive and multi-level acquisition pipeline, escalating across different sensing modalities such as visible, thermal, and hyperspectral. Data acquired at each level can be fully integrated and fused for efficient and precise material detection. Some embodiments use a physics-constrained computational model that uses latent hypercomplex states and learnable hypercomplex algebra to accurately disentangle and reconstruct spectral signatures, enabling robust substance identification. The system is adaptable to a range of hardware platforms, such as handheld devices, UAVs, and robots.

The embodiment of the invention provides a method and device for processing hypercomplex numbers. The present application proposes a new data type that combines a number of at least two components into a local CPU data word. According to the new data type, an m-bit-to-even number or an m-bit complex number is regarded as a composite number comprising two components, rather than an m-bit number, so that an operation can be performed directly on the two components of the pair of even number or the complex number using a single instruction. In addition, a hardware implementation mode of even multiplication and a combined hardware implementation mode of even and complex multiplication are also provided, so that the processing speed and the utilization rate of hardware resources can be improved, and the power consumption is reduced.

The present invention discloses a knowledge graph representation method based on arbitrary-dimensional hypercomplex embedding. The main feature of this method is that the linear layer of quaternion embedding in the original knowledge graph representation is replaced with a linear layer of hypercomplex embedding. The specific steps are as follows: 1. Preprocess the knowledge graph data, and preprocess the traditional knowledge graph into structured data according to the model requirements; 2. Use the deep learning framework pytorch to construct a preliminary embedding and build a new linear layer, that is, a hypercomplex embedding linear layer, to learn the vector representations of entities and relationships on the graph; 3. Use the knowledge graph validation set for validation and adjust to the optimal network parameters; 4. Test the knowledge graph test set and count the results. The present invention improves an existing quaternion knowledge graph embedding method QuatE, introduces a hypercomplex strategy, reduces memory occupancy, reduces parameters, and at the same time maintains excellent embedding results.

The embodiment of the invention provides a method and a device for processing hypercomplex numbers. According to the proposed scheme, one of the operands of a first operation is negated prior to performing the first operation (e.g., a subtraction operation, a multiplication operation or a convolution operation). The scheme can be applied to any product requiring subtraction, without the need for dedicated subtraction logic or subtraction instructions, such as application to complex valued matrix (or one-dimensional vector) multiplication or convolution. By means of the proposed scheme, the operations on the complex valued matrix and the CVNN can be simulated with the maximum time and area efficiency. In addition, the operation can be simulated with the maximum time and area efficiency on a matrix of high-order algebra needing subtraction.

The invention discloses a nonlinear rotor system dynamic response and sensitivity calculation method and device, equipment and a medium, and the method comprises the steps: S1, carrying out the dynamic modeling of a nonlinear rotor system according to the structural characteristics of the nonlinear rotor system, and building a motion equation of the nonlinear rotor system; s2, establishing a hypercomplex form of the motion equation of the nonlinear rotor system to obtain a hypercomplex motion equation of the nonlinear rotor system, and performing dimension expansion conversion on the motion equation of the nonlinear rotor system of a complex matrix into an equivalent real number matrix by using a complex matrix expression; and S3, solving the established hypercomplex motion equation of the nonlinear rotor system, extracting the real part value and the imaginary part value of a calculation result, and obtaining the response of the system and the corresponding sensitivity of each order. According to the method, the jacobian matrix is automatically updated and solved in the iteration process, the accuracy of the calculation result is ensured, the response sensitivity and the dynamic response sensitivity of the nonlinear rotor system can be calculated at the same time, and the calculation efficiency is high.

Embodiments of the present application provide a method and apparatus for processing hypercomplex numbers. The present application proposes a new data type that composes at least two components of a number in one native CPU data word. Based on the new data type, a m-bit dual number or a m-bit complex number is perceived as a compound with two components, not a m-bit number, which makes it possible to operate directly on the two components of the dual number or the complex number with a single instruction. In addition, hardware implementation of multiplication for dual numbers and combined hardware implementation of multiplication for both dual numbers and complex numbers are proposed, which can improve processing speed and utilization of hardware resources, and reduce power consumption.

A neural network implementation method based on a dynamic Hamiltonian product is characterized by comprising the following steps that firstly, a dynamic kernel generation mechanism driven by input features is designed, convolution kernel parameters are adjusted in a self-adaptive mode according to sample features, fine-grained extraction of multi-scale and multi-direction features is achieved, and the spatial perception advantage of quaternion representation is fully played; secondly, an overall activation function conforming to quaternion geometric characteristics is constructed, a rotation scaling relation between channels is kept through nonlinear transformation in a hypercomplex space, and relevance fracture caused by traditional component type activation is avoided; according to the dynamic convolution provided by the invention, the convolution kernel is dynamically generated according to the characteristics of the input samples, so that the network can dynamically adjust the convolution kernel for each input sample. The flexible adjustment mechanism not only enhances the expression ability of the model, but also enables the model to adapt to different input features, and extracts features with more fine granularity.

A data processing method, apparatus, and related equipment relate to the field of artificial intelligence technology. The method involves acquiring raw data, which serves as input to an AI model including an attention layer and a feedforward layer; generating input data for the attention layer based on the raw data, wherein the values ​​of the elements in the input data are hypercomplex numbers, including real and imaginary parts; performing calculations on the input data in the attention layer to obtain a first calculation result, wherein the values ​​of the elements in the first calculation result are hypercomplex numbers; and executing the feedforward layer based on the first calculation result. Since the imaginary part of the hypercomplex number can indicate features such as signal amplitude, phase, or frequency corresponding to the raw data, hypercomplex numbers can retain more dimensional features of the raw data compared to real numbers. Therefore, performing calculations based on hypercomplex numbers in the attention layer can improve the accuracy of the calculation results output by the attention layer, thereby improving the overall training or inference accuracy of the AI ​​model.

The present disclosure relates to a system and methodology designed for the spectroscopic detection of materials found on surfaces or dispersed in air. Some embodiments utilize a multi-modal sensor suite that operates through a progressive and multi-level acquisition pipeline, escalating across different sensing modalities such as visible, thermal, and hyperspectral. Data acquired at each level can be fully integrated and fused for efficient and precise material detection. Some embodiments use a physics-constrained computational model that uses latent hypercomplex states and learnable hypercomplex algebra to accurately disentangle and reconstruct spectral signatures, enabling robust substance identification. The system is adaptable to a range of hardware platforms, such as handheld devices, UAVs, and robots.

This application discloses a method, apparatus, equipment, and medium for calculating the dynamic response and sensitivity of a nonlinear rotor system. The method includes the following steps: S1, based on the structural characteristics of the nonlinear rotor system, performing dynamic modeling of the nonlinear rotor system and establishing the equations of motion for the nonlinear rotor system; S2, establishing the hypercomplex form of the equations of motion for the nonlinear rotor system to obtain the hypercomplex equations of motion for the nonlinear rotor system, using complex matrix expression, and expanding the dimension of the complex matrix equations of motion for the nonlinear rotor system to convert them into equivalent real matrix equations; S3, solving the established hypercomplex equations of motion for the nonlinear rotor system, extracting the real and imaginary parts of the calculation results, and obtaining the system response and corresponding sensitivities of each order. This application achieves automatic updating and solving of the Jacobian matrix during the iteration process, ensuring the accuracy of the calculation results, and can simultaneously calculate the response and dynamic response sensitivity of the nonlinear rotor system, resulting in high computational efficiency.