Neuromorphic visual calculation method and system for dynamic scene rapid identification

Abstract

The invention relates to the technical field of computer vision, and discloses a neuromorphic visual calculation method and system for rapid recognition of a dynamic scene, and the method comprises the steps: obtaining a quantum dynamic image of the dynamic scene, and recognizing the quantum polarization characteristics, so as to carry out the interference feature separation of the quantum dynamic image, and obtaining a clean information sequence; performing photomagnetic conversion on the clean information sequence to obtain a pulse characteristic pattern, identifying quantum correlation characteristics of the pulse characteristic pattern, and performing quantum characteristic fusion on the pulse characteristic pattern to obtain a fusion characteristic tensor; carrying out topological mapping on the fusion feature tensor to obtain mapping features, calculating scene attention weights of the dynamic scene, and carrying out region separation on the quantum dynamic image to obtain a target analysis region; and acquiring a quantum space-time pulse sequence of the target analysis area, performing multi-level decomposition to obtain a decomposed pulse sequence, constructing a comprehensive association map of the dynamic scene, and performing target identification on the dynamic scene. According to the invention, flexible identification of a high-complexity dynamic scene can be realized.

Description

Technical Field

[0001] The present invention relates to a neuromorphic visual computing method and system for rapid recognition of dynamic scenes, and belongs to the technical field of computer vision. Background Art

[0002] In today's intelligent vision application field, rapid recognition of dynamic scenes is crucial. From real-time capture of abnormal behavior in intelligent security monitoring to timely response of autonomous driving systems to rapidly changing conditions on the road, the level of dynamic scene rapid recognition technology directly affects the reliability and safety of these applications, and in turn is related to social operation efficiency and people's quality of life. Accurate and rapid recognition of dynamic scenes can respond to various emergencies in a timely manner, avoid potential dangers, and improve the overall performance of the system.

[0003] At present, traditional visual computing technology is mostly used for the rapid recognition of dynamic scenes. This technology usually uses image acquisition devices such as cameras to obtain scene image sequences, and uses image processing algorithms to extract and analyze features of the images. Then, a model (such as a convolution model) is built based on the analyzed feature information, and some fixed recognition rules and processes are set based on past experience to perform the recognition task. However, dynamic scenes are often highly complex and dynamic in real time. For example, in actual application scenarios, target objects have various motion patterns, complex lighting changes, and background dynamic interference. When faced with such scenes, due to the fixed recognition mode of traditional visual computing technology, the recognition system cannot be adjusted adaptively, making this method less effective for complex dynamic scene recognition. Summary of the invention

[0004] The present invention provides a neuromorphic visual computing method and system for rapid recognition of dynamic scenes, the main purpose of which is to flexibly recognize highly complex dynamic scenes.

[0005] To achieve the above objectives, the present invention provides a neuromorphic visual computing method for rapid recognition of dynamic scenes, comprising: Deploy a visual camera array in a dynamic scene to be quickly identified to obtain a quantum dynamic image of the dynamic scene, identify the quantum polarization characteristics of the quantum dynamic image, and perform interference feature separation on the quantum dynamic image based on the quantum polarization characteristics to obtain a clean information sequence; Constructing an optical signal coding matrix of the clean information sequence, performing optical-magnetic conversion on the clean information sequence using the optical signal coding matrix to obtain a pulse characteristic diagram, identifying quantum correlation features of the pulse characteristic diagram, and performing quantum feature fusion on the pulse characteristic diagram based on the quantum correlation features to obtain a fused feature tensor; Performing topological mapping on the fused feature tensor to obtain mapping features, using the mapping features to calculate the scene attention weight of the dynamic scene, and performing regional separation on the quantum dynamic image based on the scene attention weight to obtain a target analysis area; The quantum space-time pulse sequence of the target analysis area is collected, and the quantum space-time pulse sequence is multi-level decomposed to obtain a decomposed pulse sequence, and the decomposed pulse sequence is used to construct a comprehensive association map of the dynamic scene, and reinforcement learning is performed on the comprehensive association map to obtain an optimized association map, and the optimized association map is used to perform target recognition for the dynamic scene.

[0006] Optionally, the performing interference feature separation on the quantum dynamic image based on the quantum polarization characteristics to obtain a clean information sequence includes: Based on the quantum polarization characteristics, performing quantum base drift correction on the quantum dynamic image to obtain a polarization corrected image; Performing quantum polarization dimension projection on the polarization-corrected image to obtain projection features; Performing adaptive quantum filtering on the projection features to obtain a filtering information sequence; Calculating the information purity of the filtered information sequence; Based on the information purity, the reference polarization distribution of the quantum dynamic image is reconstructed to obtain a clean information sequence.

[0007] Optionally, the using the optical signal encoding matrix to perform optical-magnetic conversion on the clean information sequence to obtain a pulse characteristic diagram includes: Performing element matching on the optical signal encoding matrix and the clean information sequence to determine an optical signal property of the clean information sequence; Performing vector encoding on the optical signal attribute to obtain an optical signal vector; Performing optical-magnetic conversion on the optical signal vector to obtain a magnetic signal vector; The magnetic signal vector is spatially mapped using a preconfigured mapping function to obtain a pulse characteristic diagram.

[0008] Optionally, the identifying the quantum correlation feature of the pulse characteristic graph includes: Performing three-dimensional quantum grid division on the pulse characteristic diagram to obtain quantum division units; Calculating the density matrix of the quantum partition unit; Based on the density matrix, identifying quantum entanglement points of the pulse characteristic graph; Based on the quantum entanglement points, constructing a quantum distribution histogram of the pulse characteristic graph; Based on the quantum distribution histogram, a quantum correlation feature of the pulse signature is identified.

[0009] Optionally, performing topological mapping on the fused feature tensor to obtain mapping features includes: Performing dimensionality reduction processing on the fused feature tensor to obtain a reduced-dimensional feature tensor; Identifying the topological structure characteristics of the reduced dimension feature tensor; Based on the topological structure characteristics, the dimension reduction feature tensor is subjected to multi-level mapping to obtain a multi-level mapping tensor; Extracting multi-layer dimensional features of the multi-level mapping tensor; Adaptively scaling the multi-layer dimensional features to obtain adjusted features; The adjustment features are encoded and integrated to obtain mapping features.

[0010] Optionally, the calculating the scene attention weight of the dynamic scene by using the mapping feature includes: Calculating a Pearson correlation coefficient matrix between different features in the mapping features; Based on the Pearson correlation coefficient matrix, an important score function of the dynamic scene is constructed, wherein the important score function can be expressed in the following form: Where F represents the important score function, represents the balance parameter of the importance score function, represents the i-th feature in the mapping feature, represents the jth feature in the mapping feature, n represents the number of dimensions of the mapping feature, represents the kth feature of the mapping feature, Represents the Pearson correlation coefficient matrix of the i-th feature and the j-th feature in the mapping feature, express The maximum query function of express The maximum query function of ; Configuring a spatial position weight function for the importance score function to obtain a scene attention weight function; The scene attention weight function is used to calculate the scene attention weight of the dynamic scene: in, represents the scene attention weight, represents the importance score of the i-th mapping feature, represents the spatial position of the i-th mapping feature, represents the importance score of the j-th mapping feature, represents the spatial position of the jth mapping feature, and n represents the dimension of the mapping feature.

[0011] Optionally, the performing region separation on the quantum dynamic image based on the scene attention weight to obtain a target analysis region includes: Performing a comprehensive characteristic analysis on the quantum dynamic image to obtain comprehensive characteristics of the quantum image; Constructing a characteristic correlation matrix of the quantum dynamic image, so as to analyze the characteristic correlation of the quantum dynamic image by using the characteristic correlation matrix; Based on the comprehensive features of the quantum image and the feature correlation, the parameters of the scene attention weight are modified to obtain an optimized weight; identifying regional characteristics of the quantum dynamic image; Based on the regional characteristics and using the optimized weights, the quantum dynamic image is subjected to regional separation to obtain a target analysis region.

[0012] Optionally, performing multi-level decomposition on the quantum space-time pulse sequence to obtain a decomposed pulse sequence includes: Extracting information features of the quantum space-time pulse sequence; Based on the information characteristics, configuring the wavelet basis function of the quantum space-time pulse sequence; Utilizing the wavelet basis function, calculating detail coefficients and approximate coefficients of the quantum space-time pulse sequence at different scales; Based on the detail coefficients and the approximate coefficients, the quantum space-time pulse sequence is decomposed at multiple levels to obtain a decomposed pulse sequence.

[0013] Optionally, constructing a comprehensive correlation map of the dynamic scene using the decomposed pulse sequence includes: Quantizing the decomposed pulse sequence to obtain a quantized sequence; Performing tensor representation on the quantized sequence to obtain a pulse tensor; Performing feature extraction on the pulse tensor to obtain tensor features; Based on the tensor features, construct associated network nodes of the dynamic scene; Analyze the association relationship between the associated network nodes to obtain association information; Based on the association information, construct an initial network of the dynamic scene; defining a cell rule of the initial network; Based on the cellular rule, iteratively evolve the initial network to obtain an evolved network; The evolutionary network is visualized to obtain a comprehensive correlation map.

[0014] In order to solve the above problems, the present invention also provides a neuromorphic visual computing system for rapid recognition of dynamic scenes, the system comprising: An interference feature separation module is used to deploy a visual camera array in a dynamic scene to be quickly identified to obtain a quantum dynamic image of the dynamic scene, identify the quantum polarization characteristics of the quantum dynamic image, and perform interference feature separation on the quantum dynamic image based on the quantum polarization characteristics to obtain a clean information sequence; A feature fusion module is used to construct an optical signal coding matrix of the clean information sequence, perform optical-magnetic conversion on the clean information sequence using the optical signal coding matrix to obtain a pulse feature diagram, identify quantum correlation features of the pulse feature diagram, and perform quantum feature fusion on the pulse feature diagram based on the quantum correlation features to obtain a fused feature tensor; A target region separation module is used to perform topological mapping on the fused feature tensor to obtain mapping features, calculate the scene attention weight of the dynamic scene using the mapping features, and perform region separation on the quantum dynamic image based on the scene attention weight to obtain a target analysis region; The target recognition module is used to collect the quantum space-time pulse sequence of the target analysis area, perform multi-level decomposition on the quantum space-time pulse sequence to obtain a decomposed pulse sequence, use the decomposed pulse sequence to construct a comprehensive association map of the dynamic scene, perform reinforcement learning on the comprehensive association map to obtain an optimized association map, and use the optimized association map to perform target recognition on the dynamic scene.

[0015] Compared with the problems described in the background technology, the present invention first deploys a visual camera array in a dynamic scene to be quickly identified, acquires quantum dynamic images, and identifies its quantum polarization characteristics, and uses quantum state measurement technology to acquire information such as polarization direction and intensity, so as to separate interference features, improve data quality, and reduce interference; further, the present invention constructs an optical signal coding matrix of a clean information sequence, and then uses wavelength division multiplexing technology to establish a mapping relationship between information and optical signal properties to enhance the signal's anti-interference ability, and performs optical-magnetic conversion on the clean information sequence, so that the information can be converted between optical and magnetic media, which is convenient for in-depth analysis of the scene; further, the present invention mines the hidden information of dynamic scenes through operations such as three-dimensional quantum grid division, calculation of density matrix, identification of quantum entanglement points, and construction of quantum distribution histograms, and based on quantum correlation characteristics, performs quantum feature fusion on the pulse feature map, and uses quantum entanglement gate operations and tensor network collection to obtain quantum information. The fused feature tensor is obtained by compression calculation to realize the integrated features, thereby improving the efficiency of image analysis; the present invention simplifies the data processing steps by performing dimensionality reduction processing on the fused feature tensor, identifying topological structure characteristics, multi-level mapping, extracting multi-layer dimensional features, adaptive scale adjustment and encoding integration to obtain mapping features, and then based on the scene attention weight, the quantum dynamic image is regionally separated to more deeply explore the detailed information of key areas in the dynamic scene, so as to have a more accurate understanding and cognition of the entire dynamic scene. For example, in applications such as target recognition, the target area that needs to be focused on can be quickly located; further, the present invention constructs a comprehensive correlation map of the dynamic scene and performs reinforcement learning on the comprehensive correlation map, so that the map can more accurately reflect the scene rules and target characteristics, and finally uses the optimized correlation map to identify the target of the dynamic scene, so as to realize flexible recognition of highly complex dynamic scenes. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 A schematic diagram of a flow chart of a neuromorphic visual computing method for rapid recognition of dynamic scenes provided by an embodiment of the present invention; Figure 2 A schematic diagram of the modules of a neuromorphic visual computing system for realizing rapid recognition of dynamic scenes provided by an embodiment of the present invention.

[0017] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0018] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0019] The embodiment of the present application provides a neuromorphic visual computing method for rapid identification of dynamic scenes. The execution subject of the neuromorphic visual computing method for rapid identification of dynamic scenes includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the neuromorphic visual computing method for rapid identification of dynamic scenes can be executed by software or hardware installed on a terminal device or a server device. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0020] Embodiment 1: Reference Figure 1 FIG. 1 is a flow chart of a neuromorphic visual computing method for rapid recognition of dynamic scenes provided by an embodiment of the present invention. In this embodiment, the neuromorphic visual computing method for rapid recognition of dynamic scenes includes: S1. Deploy a visual camera array in a dynamic scene to be quickly identified to obtain a quantum dynamic image of the dynamic scene, identify the quantum polarization characteristics of the quantum dynamic image, and perform interference feature separation on the quantum dynamic image based on the quantum polarization characteristics to obtain a clean information sequence.

[0021] The embodiment of the present invention deploys a visual camera array in the dynamic scene to be quickly identified to obtain a quantum dynamic image of the dynamic scene. The dynamic scene can be fully photographed from different angles and positions to record the dynamic scene in the form of a quantum-related image.

[0022] The dynamic scene refers to a complex scene that integrates the movement of objects, environmental changes, and dynamic changes in quantum properties, such as urban traffic intersections, automobile manufacturing production lines, and drone reconnaissance scenarios.

[0023] The visual camera array refers to a system composed of multiple visual cameras combined in a specific layout and manner. The quantum dynamic image refers to a special form of image that contains the information about the shape, color, position, etc. of objects in the scene that traditional images have, and incorporates quantum-related characteristics. In dynamic scenes, images will change over time, and quantum dynamic images emphasize the distribution and evolution of the quantum state of photons (such as quantum polarization, quantum entanglement, etc.) over time and space during this change process.

[0024] Furthermore, the embodiments of the present invention can understand information such as changes in polarization direction and polarization state of light during propagation by identifying the quantum polarization characteristics of the quantum dynamic image, which facilitates analysis of specific change patterns of dynamic scenes under different materials, environments and events.

[0025] The quantum polarization characteristics refer to the unique properties exhibited by quantum in terms of polarization state, such as quantum state superposition, quantum entanglement and other characteristics.

[0026] Optionally, the quantum polarization characteristics can be obtained by using quantum state measurement technology to directly measure the quantum polarization state in the quantum dynamic image to obtain information such as its polarization direction and polarization intensity.

[0027] Furthermore, the embodiment of the present invention separates interference features of the quantum dynamic image based on the quantum polarization characteristics to obtain a clean information sequence, which can improve data quality and reduce the impact of interference information on subsequent analysis.

[0028] As an embodiment of the present invention, the interference feature separation of the quantum dynamic image based on the quantum polarization characteristics to obtain a clean information sequence includes: based on the quantum polarization characteristics, quantum basis drift correction of the quantum dynamic image to obtain a polarization corrected image, quantum polarization dimension projection of the polarization corrected image to obtain projection features, adaptive quantum filtering of the projection features to obtain a filtered information sequence, calculating the information purity of the filtered information sequence, and based on the information purity, reconstructing the baseline polarization distribution of the quantum dynamic image to obtain a clean information sequence.

[0029] The polarization-corrected image refers to an image obtained after quantum basis drift correction is performed on the quantum dynamic image, and the information purity refers to an indicator used to measure the proportion of effective information in the filtered information sequence.

[0030] Optionally, the polarization correction image can be obtained by analyzing the quantum dynamic image frame by frame to determine the basic state of quantum polarization in each frame of the image, then using quantum state measurement technology to detect the drift of the quantum basis of each frame of the image in the time dimension, and then reversely correcting the quantum polarization data of each frame of the image according to the drift. The projection feature can first identify the dimensional space of the quantum polarization of the polarization correction image, and then use the quantum projection algorithm to project the quantum polarization information of each pixel in the polarization correction image into the dimensional space. The filtered information sequence can first identify the statistical characteristics of the projection feature, such as mean, variance, etc., and then filter the projection feature using the minimum mean square error algorithm. The information purity can be obtained by mathematically modeling the quantum state of the filtered information sequence using the von Neumann entropy formula. The clean information sequence can determine the parameters of the reference polarization distribution according to the information purity, such as the weight of the polarization direction and the polarization intensity, and then use an image reconstruction algorithm, such as a model-based reconstruction method, to adjust the image according to the reference polarization distribution model.

[0031] S2. Construct an optical signal coding matrix for the clean information sequence, use the optical signal coding matrix to perform optical-magnetic conversion on the clean information sequence to obtain a pulse characteristic diagram, identify the quantum correlation characteristics of the pulse characteristic diagram, and based on the quantum correlation characteristics, perform quantum feature fusion on the pulse characteristic diagram to obtain a fused feature tensor.

[0032] The embodiment of the present invention can establish a correspondence between the information elements in the clean information sequence and the specific attributes of the optical signal (such as wavelength, phase and amplitude, etc.) by constructing the optical signal coding matrix of the clean information sequence, thereby facilitating the optical signal characteristic processing of the clean information sequence and improving the anti-interference ability of the signal.

[0033] The optical signal coding matrix refers to a mathematical matrix used to convert a clean information sequence into an optical signal representation, which defines the mapping relationship between each element in the clean information sequence and the specific properties of the optical signal (such as wavelength, phase, amplitude and polarization, etc.).

[0034] Optionally, the optical signal coding matrix can use wavelength division multiplexing technology to select a suitable number of wavelength channels according to the length and complexity of the clean information sequence. According to the length and complexity of the clean information sequence, a suitable number of wavelength channels is selected. The information sequence is divided into multiple subsequences (each subsequence corresponds to a specific wavelength channel), and the intensity, phase and other parameters of the optical signal on each wavelength channel are set, and the clean information sequence is encoded to obtain an optical signal coding matrix.

[0035] The embodiment of the present invention utilizes the optical signal encoding matrix to perform optical-magnetic conversion on the clean information sequence to obtain a pulse characteristic diagram, which can enable information to be flexibly converted between two physical media, optical and magnetic, and is conducive to using magnetic information processing technology to deeply understand and analyze complex dynamic scenes.

[0036] The pulse characteristic diagram refers to an image with specific information generated by quantum processing of a dynamic scene.

[0037] As an embodiment of the present invention, the optical signal coding matrix is ​​used to perform optical-magnetic conversion on the clean information sequence to obtain a pulse characteristic diagram, including: element matching between the optical signal coding matrix and the clean information sequence to determine the optical signal properties of the clean information sequence, vector encoding the optical signal properties to obtain an optical signal vector, optical-magnetic conversion on the optical signal vector to obtain a magnetic signal vector, and spatially mapping the magnetic signal vector using a preconfigured mapping function to obtain a pulse characteristic diagram.

[0038] Among them, the optical signal vector refers to the vector obtained by quantizing and vectorizing the properties of the optical signal, the magnetic signal vector refers to the vector obtained after the optical signal vector is subjected to optical-magnetic conversion, and the preconfigured mapping function refers to a pre-set mathematical function used to map each component in the magnetic signal vector to the signal space, which can be configured using the feature extraction algorithm in the machine learning algorithm.

[0039] Optionally, the optical signal attribute can be obtained by accurately matching and locating each element in the clean information sequence according to the mapping relationship set by the coding matrix using a data matching algorithm. For example, if the coding matrix is ​​encoded in binary form, it can be determined based on the corresponding relationship of the binary bits. The optical signal vector can use vector construction technology to determine the vector dimension and component order according to the physical characteristics of the optical signal attributes, such as the wavelength, phase and other attributes of the light, and arrange them into a multidimensional vector (such as the length of the vector) in a certain order. The magnetic signal vector can be obtained by directly converting the optical signal vector using an optical-magnetic conversion device. The pulse characteristic diagram can use a spatial mapping algorithm and a preconfigured mapping function to map each component in the magnetic signal vector to the signal space to form a pulse distribution, and is constructed according to the pulse distribution.

[0040] Furthermore, the embodiments of the present invention are helpful in identifying important information such as the microstructure of objects in highly complex dynamic scenes, material interactions, etc. by identifying the quantum correlation characteristics of the pulse characteristic graph, thereby mining hidden information in the target scene.

[0041] Among them, the quantum correlation characteristics refer to a comprehensive description of the quantum-level relationships and characteristics in the pulse characteristic diagram, such as entanglement characteristics, correlation strength and distribution characteristics.

[0042] As an embodiment of the present invention, the identifying of the quantum correlation characteristics of the pulse characteristic diagram includes: performing three-dimensional quantum grid division on the pulse characteristic diagram to obtain quantum division units, calculating the density matrix of the quantum division units, identifying the quantum entanglement points of the pulse characteristic diagram based on the density matrix, constructing a quantum distribution histogram of the pulse characteristic diagram based on the quantum entanglement points, and identifying the quantum correlation characteristics of the pulse characteristic diagram based on the quantum distribution histogram.

[0043] Among them, the density matrix refers to a mathematical tool for describing the state of a quantum system. The quantum entanglement point refers to the concept of describing the special correlation state of a quantum system in quantum mechanics. For example, in the analysis context of a pulse characteristic diagram, it refers to the points in the diagram that are closely related to quantum correlations, determined by quantum measurement technology after the pulse characteristic diagram is regionalized to obtain a regionalized diagram. There is a quantum entanglement phenomenon between the quantum states represented by these points, that is, when one of the quantum states changes, the other quantum states entangled with it will change accordingly instantly, no matter how far apart they are in space. The quantum distribution histogram refers to a chart used to intuitively display the distribution of quantum entanglement points in a pulse characteristic diagram.

[0044] Optionally, the quantum division unit can construct a three-dimensional grid framework based on the Heisenberg uncertainty principle, according to the dimension of the pulse characteristic diagram and the requirements of quantum characteristic analysis, and then determine the grid nodes at specific intervals to divide the diagram into many small three-dimensional regions. The density matrix can be calculated using a compressed sensing tomography algorithm. The quantum entangled points can be identified by using a quantum entanglement detection algorithm combined with the eigenvalues ​​and eigenvectors of the density matrix. The quantum distribution histogram can count the spatial distribution of quantum entangled points in the pulse characteristic diagram, classify them according to different spatial positions or quantum entanglement intensity intervals, and then present the number distribution of quantum entangled points in each interval in the form of a histogram. The quantum correlation feature can be identified by observing the shape of the quantum distribution histogram, the peak position, and the proportion of different intervals.

[0045] Furthermore, the embodiment of the present invention performs quantum feature fusion on the pulse feature map based on the quantum correlation feature to obtain a fused feature tensor that can integrate different aspects of quantum features in the pulse feature map with quantum correlation features to form a more comprehensive and representative feature set, thereby improving image analysis efficiency.

[0046] Optionally, the fused feature tensor can encode the spatiotemporal information of the pulse feature graph into a quantum state through a quantum entanglement gate operation, and then perform a tensor network contraction calculation on the pulse feature graph of the quantum state in the Hilbert space, and obtain it after optimizing the entanglement weights between each feature in the pulse feature graph after the tensor network contraction using a variational quantum eigensolver.

[0047] S3. Topological mapping is performed on the fused feature tensor to obtain mapping features, and the scene attention weight of the dynamic scene is calculated using the mapping features. Based on the scene attention weight, the quantum dynamic image is region-separated to obtain a target analysis region.

[0048] The embodiment of the present invention performs topological mapping on the fused feature tensor to obtain mapping features, which can convert high-dimensional and complex fused feature tensors into low-dimensional features that are easier to understand and process through topological mapping, thereby simplifying the difficulty of data processing.

[0049] As an embodiment of the present invention, the topological mapping of the fused feature tensor to obtain mapping features includes: performing dimensionality reduction processing on the fused feature tensor to obtain a reduced-dimensionality feature tensor, identifying the topological structure characteristics of the reduced-dimensionality feature tensor, performing multi-level mapping on the reduced-dimensionality feature tensor based on the topological structure characteristics to obtain a multi-level mapping tensor, extracting multi-layer dimensional features of the multi-level mapping tensor, adaptively scaling the multi-layer dimensional features to obtain adjustment features, encoding and integrating the adjustment features to obtain mapping features.

[0050] The topological structure characteristics refer to the description of the inherent properties of the reduced-dimensional feature tensor at the topological level, such as critical point characteristics, connection relationships, and overall topological morphology.

[0051] Optionally, the reduced dimension feature tensor can be obtained by calculating the covariance matrix between the dimensions of the fused feature tensor using a principal component analysis algorithm, and then decomposing the eigenvalues ​​of the covariance matrix to obtain the principal components. The topological structure characteristics can be obtained by using Morse theory to calculate the gradient vector field of each data point in the reduced dimension feature tensor, find the critical point where the gradient is zero, and then determine the topological structure of the tensor based on the type of critical point (such as saddle point, maximum point, minimum point) and the connection relationship between them. The multi-level mapping tensor can be obtained by using a local linear embedding algorithm to calculate the embedding coordinates of the reduced dimension feature tensor in different dimensions in a low-dimensional space. The multi-layer dimensional features can be obtained by calculating the geometric features (such as distance, angle), statistical features (such as mean, variance) and distribution features (such as clustering degree) of the multi-level mapping tensor. The adjustment features can be obtained by using an adaptive weight allocation algorithm to perform feature scaling on the multi-layer dimensional features and extracting key features (extracting according to the weight allocation size, the weight can be set to 0.3, and the features with weights greater than 0.3 are retained, which needs to be set in combination with the actual application). The mapping features can be encoded using a Huffman coding method to encode the adjusted features, and the features of different layers are combined in a specific order.

[0052] Furthermore, the embodiment of the present invention can identify the difference in importance of different regions or features in the dynamic scene by calculating the scene attention weight of the dynamic scene using the mapping features, which is helpful for identifying key information in the dynamic scene.

[0053] The scene attention weight refers to a quantitative indicator used to indicate the relative importance of different areas or features in a dynamic scene.

[0054] As an embodiment of the present invention, the use of the mapping features to calculate the scene attention weight of the dynamic scene includes: calculating the Pearson correlation coefficient matrix between different features in the mapping features, and constructing an important score function of the dynamic scene based on the Pearson correlation coefficient matrix, wherein the important score function can be expressed in the following form: Where F represents the important score function, represents the balance parameter of the importance score function, represents the i-th feature in the mapping feature, represents the jth feature in the mapping feature, n represents the number of dimensions of the mapping feature, represents the kth feature of the mapping feature, Represents the Pearson correlation coefficient matrix of the i-th feature and the j-th feature in the mapping feature, express The maximum query function of express The maximum query function of ; A spatial position weight function is configured for the importance score function to obtain a scene attention weight function, and the scene attention weight of the dynamic scene is calculated using the scene attention weight function: in, represents the scene attention weight, represents the importance score of the i-th mapping feature, represents the spatial position of the i-th mapping feature, represents the importance score of the j-th mapping feature, represents the spatial position of the jth mapping feature, and n represents the dimension of the mapping feature.

[0055] Among them, the Pearson correlation coefficient matrix refers to a matrix used to measure the degree and direction of linear correlation between different features in the mapping features, the important score function refers to a function that comprehensively considers the value size of the mapping feature itself and the correlation with other features, the spatial position weight function refers to a mathematical formula used to measure the influence of the spatial position of the mapping feature in the dynamic scene on its importance, and the balance parameter of the important score function refers to a parameter used to adjust the relative influence of the feature's own value size and the correlation between features in the important score function.

[0056] Optionally, the Pearson correlation coefficient matrix can be obtained by calculating the mapping features using the Pearson correlation coefficient method. The spatial position weight function can select a suitable function according to the specific application scenario and requirements. For example, in image or scene analysis, if the features of the central area are considered to be more important, and the importance of the area away from the center gradually decreases, a Gaussian function can be used to configure the spatial position weight function. For example, when processing a picture, the central part of the image may contain the main target object, so the Gaussian function can be used to give the central area a higher weight, while the edge area has a lower weight.

[0057] It should be further explained that the importance score function comprehensively considers the value of the feature itself and the correlation between features to evaluate the importance of each feature in the dynamic scene. To adjust the relative importance of these two factors, When it is close to 1, the feature value itself is emphasized; when When it is close to 0, the correlation between features is emphasized, so that the importance of different features in dynamic scenes can be more comprehensively and flexibly reflected. The value range of the balance parameter is [0-1].

[0058] Furthermore, the embodiment of the present invention can concentrate the analysis resources on the most valuable part by performing regional separation on the quantum dynamic image based on the scene attention weight to obtain the target analysis area, thereby improving the analysis efficiency. At the same time, it helps to more deeply explore the detailed information of the key areas in the dynamic scene, so as to have a more accurate understanding and cognition of the entire dynamic scene. For example, in applications such as target recognition, it can quickly locate the target area that needs to be focused on.

[0059] As an embodiment of the present invention, the method of performing regional separation on the quantum dynamic image based on the scene attention weight to obtain a target analysis area includes: performing a comprehensive characteristic analysis on the quantum dynamic image to obtain comprehensive characteristics of the quantum image, constructing a feature association matrix of the quantum dynamic image to analyze the feature correlation of the quantum dynamic image using the feature association matrix, performing parameter correction on the scene attention weight based on the comprehensive characteristics of the quantum image and the feature correlation to obtain an optimized weight, identifying regional characteristics of the quantum dynamic image, and performing regional separation on the quantum dynamic image based on the regional characteristics using the optimized weight to obtain a target analysis area.

[0060] Among them, the comprehensive characteristics of the quantum image refer to a set of descriptions of the various characteristics of the quantum dynamic image, the feature association matrix refers to a matrix used to quantify the degree of association between different features in the quantum dynamic image, and the regional characteristics refer to the relevant attributes of different regions in the quantum dynamic image, such as the type of quantum state, distribution uniformity and other attributes.

[0061] Optionally, the quantum image comprehensive feature can use quantum Fourier transform to convert the quantum dynamic image from the time domain to the frequency domain, obtain the frequency distribution characteristics of the image, and use quantum state tomography technology to accurately measure the specific parameters of each quantum state in the image, such as amplitude, phase, etc., and then determine it from the dimensions of quantum state distribution, frequency characteristics, etc. The feature association matrix can use the quantum correlation function to calculate the quantum entanglement points, quantum bits and other key quantum features of the quantum dynamic image, and then arrange the calculated correlation intensity values ​​according to the corresponding relationship of the features to construct a matrix, and then analyze the matrix elements in the feature association matrix, such as calculating the eigenvalues ​​and eigenvectors of the matrix to identify the feature association. The optimization weight can use the back propagation algorithm in machine learning, with the quantum image comprehensive features and feature association as input, the initial scene attention weight as a parameter, and by continuously adjusting the weight parameters, the simulated area separation result output by the algorithm is more matched with the ideal image content distribution, and then after multiple rounds of iterative training, the weight that is highly consistent with the actual characteristics of the image is obtained. The regional characteristics can be obtained by using a quantum edge detection algorithm, using the sensitivity of quantum state changes, detecting edge contours of different regions in the image, and then combining a quantum clustering algorithm, such as quantum K-means clustering, to cluster pixels in the image according to the similarity of quantum features, thereby determining the range, shape, and uniformity of internal quantum features of different regions. The target analysis area can be obtained by applying optimized weights to the image segmentation algorithm based on the identified regional characteristics. For areas with high weights, a sophisticated quantum image segmentation method, such as segmentation based on quantum Monte Carlo simulation, is used to ensure accurate division of key areas. Finally, the quantum dynamic image is divided into different areas, and the analysis area with high weights is extracted, and areas with weight values ​​greater than 0.3 can be extracted.

[0062] S4. Collect the quantum space-time pulse sequence in the target analysis area, perform multi-level decomposition on the quantum space-time pulse sequence to obtain a decomposed pulse sequence, use the decomposed pulse sequence to construct a comprehensive association map of the dynamic scene, perform reinforcement learning on the comprehensive association map to obtain an optimized association map, and use the optimized association map to perform target recognition on the dynamic scene.

[0063] The embodiment of the present invention can provide the original data basis for the subsequent in-depth analysis of dynamic scenes by collecting the quantum space-time pulse sequence of the target analysis area. These pulse sequences contain the changing laws of quantum states in the space-time dimension, which helps to reveal the quantum characteristics of the target in the dynamic scene and its evolution process over time, and plays a key role in accurately understanding and analyzing the dynamic scene.

[0064] The quantum space-time pulse sequence refers to a signal sequence that reflects the dynamic changes of a quantum system in time and space dimensions.

[0065] Optionally, the quantum space-time pulse sequence can be collected by a quantum sensor array.

[0066] The embodiment of the present invention performs multi-level decomposition on the quantum space-time pulse sequence to obtain a decomposed pulse sequence, which can separate the information mixed together in the original pulse sequence, so that various features and patterns hidden therein can be revealed.

[0067] As an embodiment of the present invention, the multi-level decomposition of the quantum space-time pulse sequence to obtain a decomposed pulse sequence includes: extracting information features of the quantum space-time pulse sequence, configuring a wavelet basis function of the quantum space-time pulse sequence based on the information features, calculating detail coefficients and approximate coefficients of the quantum space-time pulse sequence at different scales using the wavelet basis function, and performing a multi-level decomposition of the quantum space-time pulse sequence based on the detail coefficients and the approximate coefficients to obtain a decomposed pulse sequence.

[0068] Among them, the information feature refers to the key information extracted from the quantum space-time pulse sequence that can reflect its essential characteristics, such as energy distribution, phase and amplitude, etc. The wavelet basis function refers to the basic function used to perform wavelet transform on the quantum space-time pulse sequence. The detail coefficient refers to the coefficient obtained when the quantum space-time pulse sequence is subjected to wavelet transform using the wavelet basis function, which reflects the local change characteristics of the pulse sequence in the high-frequency part. The approximate coefficient refers to the coefficient obtained when the quantum space-time pulse sequence is subjected to wavelet transform using the wavelet basis function, which reflects the overall trend of the quantum space-time pulse sequence in the low-frequency part.

[0069] Optionally, the information feature can be identified by time domain conversion of the quantum space-time pulse sequence using quantum Fourier transform. The wavelet basis function can be configured by quantum Haar wavelet or Daubechies wavelet. If the pulse sequence has important information in the low-frequency part, the Haar wavelet can be used for configuration. The detail coefficient and the approximate coefficient can be obtained by performing inner product operations on the configured wavelet basis function and the quantum space-time pulse sequence at different scales. The decomposed pulse sequence can be decomposed into detail coefficients and coarser approximate coefficients of the next layer by using a quantum recursive algorithm starting from the approximate coefficients of the coarsest scale, and the recursive operation is continued to continuously decompose the approximate coefficients of each layer to obtain details and approximate coefficients at different scales, which are obtained by stacking these coefficients layer by layer.

[0070] Furthermore, the embodiments of the present invention can integrate various information in the quantum space-time pulse sequence by using the decomposed pulse sequence to construct a comprehensive correlation map of the dynamic scene, forming a visual or computable structure, which helps the computer to grasp the structure and evolution of the dynamic scene as a whole.

[0071] The comprehensive association map refers to a visual map used to present the complex relationships between various elements in a dynamic scene.

[0072] As an embodiment of the present invention, the use of the decomposed pulse sequence to construct a comprehensive association map of the dynamic scene includes: quantizing the decomposed pulse sequence to obtain a quantized sequence, performing tensor representation on the quantized sequence to obtain a pulse tensor, extracting features from the pulse tensor to obtain tensor features, constructing association network nodes of the dynamic scene based on the tensor features, analyzing the association relationship between the association network nodes to obtain association information, constructing an initial network of the dynamic scene based on the association information, defining cellular rules of the initial network, iteratively evolving the initial network based on the cellular rules to obtain an evolved network, and visually composing the evolved network to obtain a comprehensive association map.

[0073] The cellular rule refers to the core concept in the cellular automaton model, which is a mechanism that defines how cells are updated and evolved according to their own states and the states of surrounding neighboring cells.

[0074] Optionally, the quantization sequence can be obtained by converting the continuous value of the decomposed pulse sequence into a discrete value using a non-uniform quantization algorithm. The pulse tensor can be obtained by analyzing the information of the quantization sequence in dimensions such as time, space, and quantum features to determine the order and dimension of the tensor, and then arranging the data of the quantization sequence according to the information of each dimension. The tensor feature can be obtained by decomposing the pulse tensor into multiple low-order tensors using a tensor decomposition method, such as CP decomposition. The associated network nodes can be divided into a group based on the similarity and correlation of the tensor features using a clustering algorithm, such as K-means clustering, and each group corresponds to a node in the associated network (the attributes of the nodes can be represented by the feature mean of the group). The nodes are sorted and screened by calculating the degree centrality of the nodes to determine. The associated information can be determined by using a quantum correlation analysis method to calculate the association strength between different associated network nodes, and then analyzing the time sequence, spatial position relationship, etc. between the associated network nodes to determine the association type of the nodes, and determining the association information based on the association strength and the association type. The initial network can be constructed based on the associated network nodes and the connection relationship between the nodes is determined according to the association information. The cell rules defining the initial network can be determined according to the physical characteristics and quantum laws of the dynamic scene, and the state update rules of the cellular automaton can be determined, and then the scope and influence weight of the neighbor cells can be set according to the influence mode and degree of the neighbor cells on the current cell. The comprehensive association map can be obtained by drawing according to the node and edge information of the evolution network using the Graphviz tool.

[0075] Furthermore, the embodiment of the present invention performs reinforcement learning on the comprehensive correlation map to obtain an optimized correlation map, which can discover the deficiencies and unreasonableness in the comprehensive correlation map, and through iterative optimization, make the map more accurately reflect the inherent laws and target characteristics of the dynamic scene.

[0076] Optionally, to optimize the association graph, you can first define the comprehensive association graph state, action and reward function, where the state covers information such as graph structure and nodes, and the action is to adjust edge weights, etc. Then use the reinforcement learning algorithm to select actions according to the state, execute to obtain new states and rewards, and continuously iterate and update model parameters until the optimized association graph is obtained.

[0077] The embodiment of the present invention can more accurately identify the target in the dynamic scene by using the optimized association map to identify the target in the dynamic scene, especially in a complex and changeable quantum dynamic scene.

[0078] Embodiment 2: like Figure 2 , which is a functional module diagram of the neuromorphic visual computing system for rapid recognition of dynamic scenes of the present invention.

[0079] The neuromorphic visual computing system 200 for rapid recognition of dynamic scenes of the present invention can be installed in an electronic device. According to the functions implemented, the neuromorphic visual computing system for rapid recognition of dynamic scenes can include an interference feature separation module 201, a feature fusion module 202, a target region separation module 203 and a target recognition module 204. The module of the present invention can also be referred to as a unit, which refers to a series of computer program segments that can be executed by a processor of an electronic device and can complete fixed functions, which are stored in the memory of the electronic device.

[0080] In the embodiment of the present invention, the functions of each module / unit are as follows: The interference feature separation module 201 is used to deploy a visual camera array in a dynamic scene to be quickly identified to obtain a quantum dynamic image of the dynamic scene, identify the quantum polarization characteristics of the quantum dynamic image, and perform interference feature separation on the quantum dynamic image based on the quantum polarization characteristics to obtain a clean information sequence; The feature fusion module 202 is used to construct an optical signal coding matrix of the clean information sequence, use the optical signal coding matrix to perform optical-magnetic conversion on the clean information sequence to obtain a pulse feature diagram, identify quantum correlation features of the pulse feature diagram, and perform quantum feature fusion on the pulse feature diagram based on the quantum correlation features to obtain a fused feature tensor; The target region separation module 203 is used to perform topological mapping on the fused feature tensor to obtain mapping features, calculate the scene attention weight of the dynamic scene using the mapping features, and perform region separation on the quantum dynamic image based on the scene attention weight to obtain a target analysis region; The target identification module 204 is used to collect the quantum space-time pulse sequence of the target analysis area, perform multi-level decomposition on the quantum space-time pulse sequence to obtain a decomposed pulse sequence, use the decomposed pulse sequence to construct a comprehensive association map of the dynamic scene, perform reinforcement learning on the comprehensive association map to obtain an optimized association map, and use the optimized association map to perform target identification on the dynamic scene.

[0081] In detail, the modules in the neuromorphic visual computing system 200 for rapid recognition of dynamic scenes in the embodiment of the present invention are used in the same manner as described above. Figure 1 The technical means are the same as the neuromorphic visual computing method for rapid recognition of dynamic scenes described in, and can produce the same technical effects, so they will not be repeated here.

[0082] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0083] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A neuromorphic visual computing method for fast recognition of dynamic scenes, characterized in that: The method comprises: Deploy a visual camera array in a dynamic scene to be quickly identified to obtain a quantum dynamic image of the dynamic scene, identify the quantum polarization characteristics of the quantum dynamic image, and perform interference feature separation on the quantum dynamic image based on the quantum polarization characteristics to obtain a clean information sequence; Constructing an optical signal coding matrix of the clean information sequence, performing optical-magnetic conversion on the clean information sequence using the optical signal coding matrix to obtain a pulse characteristic diagram, identifying quantum correlation features of the pulse characteristic diagram, and performing quantum feature fusion on the pulse characteristic diagram based on the quantum correlation features to obtain a fused feature tensor; Performing topological mapping on the fused feature tensor to obtain mapping features, using the mapping features to calculate the scene attention weight of the dynamic scene, and performing regional separation on the quantum dynamic image based on the scene attention weight to obtain a target analysis area; The quantum space-time pulse sequence of the target analysis area is collected, and the quantum space-time pulse sequence is multi-level decomposed to obtain a decomposed pulse sequence, and the decomposed pulse sequence is used to construct a comprehensive association map of the dynamic scene, and reinforcement learning is performed on the comprehensive association map to obtain an optimized association map, and the optimized association map is used to perform target recognition for the dynamic scene.

2. The neuromorphic visual computing method for rapid recognition of dynamic scenes as claimed in claim 1, characterized in that: The step of performing interference feature separation on the quantum dynamic image based on the quantum polarization characteristics to obtain a clean information sequence includes: Based on the quantum polarization characteristics, performing quantum base drift correction on the quantum dynamic image to obtain a polarization corrected image; Performing quantum polarization dimension projection on the polarization-corrected image to obtain projection features; Performing adaptive quantum filtering on the projection features to obtain a filtering information sequence; Calculating the information purity of the filtered information sequence; Based on the information purity, the reference polarization distribution of the quantum dynamic image is reconstructed to obtain a clean information sequence.

3. The neuromorphic visual computing method for rapid recognition of dynamic scenes as claimed in claim 1, characterized in that: The method of using the optical signal encoding matrix to perform optical-magnetic conversion on the clean information sequence to obtain a pulse characteristic diagram includes: Performing element matching on the optical signal encoding matrix and the clean information sequence to determine an optical signal property of the clean information sequence; Performing vector encoding on the optical signal attribute to obtain an optical signal vector; Performing optical-magnetic conversion on the optical signal vector to obtain a magnetic signal vector; The magnetic signal vector is spatially mapped using a preconfigured mapping function to obtain a pulse characteristic diagram.

4. The neuromorphic visual computing method for rapid recognition of dynamic scenes as claimed in claim 1, characterized in that: The step of identifying the quantum correlation characteristics of the pulse characteristic graph comprises: Performing three-dimensional quantum grid division on the pulse characteristic diagram to obtain quantum division units; Calculating the density matrix of the quantum partition unit; Based on the density matrix, identifying quantum entanglement points of the pulse characteristic graph; Based on the quantum entanglement points, constructing a quantum distribution histogram of the pulse characteristic graph; Based on the quantum distribution histogram, a quantum correlation feature of the pulse signature is identified.

5. The neuromorphic visual computing method for rapid recognition of dynamic scenes as claimed in claim 1, characterized in that: The performing topological mapping on the fused feature tensor to obtain mapping features includes: Performing dimensionality reduction processing on the fused feature tensor to obtain a reduced-dimensional feature tensor; Identifying the topological structure characteristics of the reduced dimension feature tensor; Based on the topological structure characteristics, the dimension reduction feature tensor is subjected to multi-level mapping to obtain a multi-level mapping tensor; Extracting multi-layer dimensional features of the multi-level mapping tensor; Adaptively scaling the multi-layer dimensional features to obtain adjusted features; The adjustment features are encoded and integrated to obtain mapping features.

6. The neuromorphic visual computing method for rapid recognition of dynamic scenes as claimed in claim 1, characterized in that: The calculating the scene attention weight of the dynamic scene by using the mapping feature includes: Calculating a Pearson correlation coefficient matrix between different features in the mapping features; Based on the Pearson correlation coefficient matrix, an important score function of the dynamic scene is constructed, wherein the important score function can be expressed in the following form: Where F represents the important score function, represents the balance parameter of the importance score function, represents the i-th feature in the mapping feature, represents the jth feature in the mapping feature, n represents the number of dimensions of the mapping feature, represents the kth feature of the mapping feature, Represents the Pearson correlation coefficient matrix of the i-th feature and the j-th feature in the mapping feature, express The maximum query function of express The maximum query function of ; Configuring a spatial position weight function for the importance score function to obtain a scene attention weight function; The scene attention weight function is used to calculate the scene attention weight of the dynamic scene: in, represents the scene attention weight, represents the importance score of the i-th mapping feature, represents the spatial position of the i-th mapping feature, represents the importance score of the j-th mapping feature, represents the spatial position of the jth mapping feature, and n represents the dimension of the mapping feature.

7. The neuromorphic visual computing method for rapid recognition of dynamic scenes as claimed in claim 1, characterized in that: The step of performing region separation on the quantum dynamic image based on the scene attention weight to obtain a target analysis region includes: Performing a comprehensive characteristic analysis on the quantum dynamic image to obtain comprehensive characteristics of the quantum image; Constructing a characteristic correlation matrix of the quantum dynamic image, so as to analyze the characteristic correlation of the quantum dynamic image by using the characteristic correlation matrix; Based on the comprehensive features of the quantum image and the feature correlation, the parameters of the scene attention weight are modified to obtain an optimized weight; identifying regional characteristics of the quantum dynamic image; Based on the regional characteristics and using the optimized weights, the quantum dynamic image is subjected to regional separation to obtain a target analysis region.

8. The neuromorphic visual computing method for rapid recognition of dynamic scenes as claimed in claim 1, characterized in that: The step of performing multi-level decomposition on the quantum space-time pulse sequence to obtain a decomposed pulse sequence comprises: Extracting information features of the quantum space-time pulse sequence; Based on the information characteristics, configuring the wavelet basis function of the quantum space-time pulse sequence; Utilizing the wavelet basis function, calculating detail coefficients and approximate coefficients of the quantum space-time pulse sequence at different scales; Based on the detail coefficients and the approximate coefficients, the quantum space-time pulse sequence is decomposed at multiple levels to obtain a decomposed pulse sequence.

9. The neuromorphic visual computing method for rapid recognition of dynamic scenes as claimed in claim 1, characterized in that: The method of constructing a comprehensive correlation map of the dynamic scene using the decomposed pulse sequence comprises: Quantizing the decomposed pulse sequence to obtain a quantized sequence; Performing tensor representation on the quantized sequence to obtain a pulse tensor; Performing feature extraction on the pulse tensor to obtain tensor features; Based on the tensor features, construct associated network nodes of the dynamic scene; Analyze the association relationship between the associated network nodes to obtain association information; Based on the association information, construct an initial network of the dynamic scene; defining a cell rule of the initial network; Based on the cellular rule, iteratively evolve the initial network to obtain an evolved network; The evolutionary network is visualized to obtain a comprehensive correlation map.

10. A neuromorphic visual computing system for fast recognition of dynamic scenes, characterized in that: The system comprises: An interference feature separation module is used to deploy a visual camera array in a dynamic scene to be quickly identified to obtain a quantum dynamic image of the dynamic scene, identify the quantum polarization characteristics of the quantum dynamic image, and perform interference feature separation on the quantum dynamic image based on the quantum polarization characteristics to obtain a clean information sequence; A feature fusion module is used to construct an optical signal coding matrix of the clean information sequence, perform optical-magnetic conversion on the clean information sequence using the optical signal coding matrix to obtain a pulse feature diagram, identify quantum correlation features of the pulse feature diagram, and perform quantum feature fusion on the pulse feature diagram based on the quantum correlation features to obtain a fused feature tensor; A target region separation module is used to perform topological mapping on the fused feature tensor to obtain mapping features, calculate the scene attention weight of the dynamic scene using the mapping features, and perform region separation on the quantum dynamic image based on the scene attention weight to obtain a target analysis region; The target recognition module is used to collect the quantum space-time pulse sequence of the target analysis area, perform multi-level decomposition on the quantum space-time pulse sequence to obtain a decomposed pulse sequence, use the decomposed pulse sequence to construct a comprehensive association map of the dynamic scene, perform reinforcement learning on the comprehensive association map to obtain an optimized association map, and use the optimized association map to perform target recognition on the dynamic scene.

Images