63 results about "Critical regions" patented technology

The invention relates to the technical field of geological disaster monitoring and early warning, and discloses a railway tunnel portal geological disaster deformation early warning system based on an SAR radar, and the system comprises a digital twinborn body construction module which constructs a digital twinborn body with initial parameters based on basic data; the SAR deformation monitoring module is used for acquiring SAR deformation observation data; the digital twinborn body dynamic optimization module is used for inverting and updating parameters by using an optimization algorithm based on the SAR data, and generating an optimized twinborn body; a risk prediction and key area identification module which deduces a disaster scene based on the optimized twinborn body, predicts the risk and identifies a key risk area; and the intelligent early warning module is used for generating early warning information based on the prediction risk and the key risk area. According to the method, the geomechanical digital twins are dynamically optimized by adopting the SAR data, accurate prediction, key area identification and intelligent grading early warning of the geological disaster of the railway tunnel portal are realized, and the initiative and accuracy of risk cognition and early warning are remarkably improved.

The invention belongs to the technical field of plate spring performance analysis and evaluation, and discloses a high-strength two-stage gradual-change-stiffness plate spring performance analysis and evaluation system and method.The method comprises the steps that plate spring initial mechanical data collection and heterogeneous load response integrated processing are conducted on the basis of a multi-dimensional vehicle-mounted test experiment, and plate spring integrated mechanical response data are generated; performing double-rigidity gradient recognition analysis to generate rigidity transformation critical data; performing critical region stress field deconstruction analysis to generate stress field distribution characteristic data; carrying out local microstructure change characterization to generate structural deformation data; constructing a multi-scale stress transfer link, and generating critical region stress transfer path data; performing fatigue damage accumulative mapping to generate accumulative feature data; constructing a fatigue life prediction model, and generating life prediction data; carrying out comprehensive performance index quantitative evaluation to generate a performance evaluation result; executing plate spring optimization improvement scheme generation and verification; and a scientific basis is provided for performance optimization of the two-stage gradual-change-rigidity plate spring.

The invention belongs to the technical field of sound source processing, and provides a sound source positioning and detecting method and device. The method comprises the following steps: acquiring delay estimation of microphones I and II and delay estimation of microphones I and III based on a three-path linear uniform microphone array; based on the delay estimation sum, respectively carrying out positioning calculation under near-field and far-field conditions; according to the sound source distance under the near-field condition, comparing the sound source distance with a distance judgment threshold value, and determining a far-field / near-field output sound source position; and carrying out feature extraction on the signals of any microphone array, and carrying out event classification based on a pre-trained convolutional neural network. According to the method, far / near field model selection is carried out according to the distance judgment threshold, large deviation generated by a single model in a critical region is avoided, continuous and stable positioning from short distance to long distance is achieved, time classification can be achieved while position calculation is carried out, and integrated output is achieved.

A heart treatment system is disclosed capable of diagnosing one or more critical regions of interest for a biological rhythm disorder by sensing signals from biological tissue. If a critical region is not present at the current location of sensed signals, the system is capable of indicating a guidance direction in which to navigate to reach one or more critical regions. Ablation energy is delivered to treat said region of interest. Signals are again sensed and analyzed to assess the impact of treatment. This process is repeated until all critical regions of interest are treated. In some embodiments, all functionality is provided by a single sensing and treating catheter with display device and analytical software.

The invention belongs to the technical field of hydraulic test and intelligent control, and particularly discloses a hydraulic component test system integrating data acquisition and automatic data analysis functions, which comprises a physical hardware test platform, a test control software platform and a digital twinborn model of a tested component, and realizes real-time interaction through an OPC UA protocol. Model parameters are corrected on line by means of measured data, and steady state, transient state and disturbance working condition tests are automatically executed based on a working condition library; and dynamically optimizing a test sequence in a test process, skipping a redundant working condition or encrypting a critical region, and synchronously generating a structured performance report. The accuracy and consistency of test results are improved, working condition simulation signals are automatically generated, performance data are collected in real time, a comprehensive performance report is generated, manual operation is reduced, the test process optimization unit can skip redundant working conditions and refine failure boundary positioning, and the test time and cost are effectively saved.

The invention discloses a composite material welding parameter regulation and control method and device, and belongs to the technical field of intelligent welding, and the method comprises the steps of constructing a process coupling model, setting a critical positioning method, generating a parameter bifurcation atlas, constructing a chaotic boundary probability cloud atlas, associating image feature differences of different materials, and positioning a parameter critical region; obtaining a dynamic data set, setting a critical early warning method, extracting key features, analyzing abnormal changes of image features, and triggering early warning once the abnormal changes are recognized; calculating a safety offset, setting a dynamic reconstruction method, introducing a time-varying damping factor, dynamically adjusting collaborative output, and reconstructing an optimal parameter combination in combination with a welding depth and an interface fusion state fed back by a real-time image; the method comprises the steps of collecting batch welding data, setting an iterative optimization method, updating process coupling model parameters, dynamically expanding a critical region recognition range, correcting bifurcation boundary conditions in real time and enhancing the adaptability to parameter critical point drift.

The invention provides a dynamic disaster assessment method and system based on iron tower data, and relates to the technical field of GIS data processing, and the method comprises the steps: obtaining the iron tower state data of a target region, collecting multi-dimensional basic information, determining a plurality of critical regions based on the target region, carrying out the classification according to disaster levels, adding disaster level identifiers, and carrying out the clustering analysis. The method comprises the following steps: drawing up a disaster aggregation map, carrying out risk correction on the disaster aggregation map through iron tower state data and multi-dimensional basic information, determining a disaster influence level and identifying a key risk area, and carrying out resource allocation management in the disaster aggregation map according to the disaster influence level and the key risk area. The technical problems of low resource allocation decision accuracy and low rescue efficiency caused by inaccurate disaster assessment results in the prior art are solved. The technical effects of improving disaster assessment accuracy, ensuring scientificity and pertinence of resource allocation work and further improving rescue efficiency are achieved.

The invention relates to a method for determining at least one safety-critical peak region (R1, R2, R3) on the basis of an indicator value (Z) for statistical characterization of event descriptions detected by at least one vehicle in a location-related manner, wherein those inflow- and outflow-free road sections of which the section length is greater than or equal to a predetermined minimum section length are identified on the basis of a road map or navigation map. Identified road sections which adjoin one another are connected to form in each case a section chain (10) with a starting point (11) and an end point (12). Event descriptions comprising a geoposition (P1, P2) and at least one parameter of a vehicle environment model are detected by vehicles and transferred to an analysis system. The analysis system in each case assigns a nearest road network location (O1, O2) of a section chain (10) to a geoposition (P1, P2) of an event description and describes it by a longitudinal distance (L, L1, L2) in relation to the starting point (11) of the section chain (10). For at least one section chain (10), a profile of a density value (p) of event descriptions that is related to the longitudinal distance (L, L1, L2) from the starting point (11) is determined by means of a kernel density estimator of a predetermined bandwidth. The profile of the density value (p) is transformed, using local, bandwidth-related, Z standardization, into a Z profile (Z1, Z2) of a Z value (Z) which is mean-free in relation to bandwidth. In each case a region of road network locations (O1, O2) along the motorway-like road (100) which are connected along a section chain (10) between a peak beginning (Lmin) and a peak end (Lmax) and at which the Z value (Z) exceeds a predetermined threshold value (θ) is identified as a safety-critical peak region (R1, R2, R3).

The invention discloses a side slope excavation dynamic monitoring and early warning method based on digital twinning. The method comprises the following steps that terrain point cloud, rock and soil layering and construction progress are obtained and preprocessed; three-dimensional geometric twinning and physical twinning are established, finite element simulation is adopted, boundary push-back and pore pressure rebalance are completed, and a twinning base line is formed; an observation residual error and a physical conservation residual error are calculated, and a high-consistency twin and assimilation sample set is obtained through joint updating; applying a topological constraint sliding surface evolver and implementing adaptive mesh refinement and interface reconstruction of a critical zone; and constructing a displacement and osmotic pressure uncertainty conical domain based on the assimilation sample set and the critical region strengthening model, and outputting a grading early warning result. According to the method, a double-consistency assimilation and topological constraint sliding surface evolution method is adopted, high-consistency simulation and quantitative early warning of the whole slope excavation process are achieved, and the method has the advantages of being high in accuracy, high in stability and good in traceability.

The invention discloses an airborne camera anti-shake control method and system, and relates to the field of camera anti-shake control, and the method comprises the steps: building a hysteretic deformation tensor field, generating a visual hysteretic residual tensor through differential operation and lagging feature analysis, taking the visual hysteretic residual tensor as observation input, and carrying out the state estimation through a LuGre dynamic friction model. Calculating a micro bristle deformation state and an equivalent friction torque value of a blocking holder, performing visual compensation analysis based on the equivalent friction torque value of the blocking holder, judging whether the system is in a static friction zero-crossing critical region or not based on a micro bristle deformation estimated value, generating a reverse feed-forward compensation torque instruction according to the generalized all-source anti-interference torque, and determining whether the system is in a static friction zero-crossing critical region or not. The integral term of the controller is dynamically reset, vector synthesis is carried out on a reverse feed-forward compensation torque instruction and a basic instruction of a main image stabilization control loop, final driving voltage is generated and applied to a pan-tilt motor, and stepped picture lagging and jumping caused by instant release of energy in a long-focus view field are restrained.

The invention belongs to the field of deep neural networks and image classification, and particularly relates to an image classification adversarial training improvement method based on boundary sample enhancement. Comprising the following steps: step 1, preparing an image sample set; 2, designing a GAN network architecture suitable for boundary sample generation; step 3, training a GAN network suitable for boundary sample generation; and step 4, image classifier adversarial training participated by the boundary samples. The method can effectively restrain the stability of the decision boundary of the image classification model, enables the model to carry out finer feature learning in a critical region, and prevents the boundary from excessively shifting due to adversarial disturbance. And meanwhile, the model can still maintain stable and reliable feature expression when being disturbed. According to the method, the robustness of the model is improved, the damage to the precision of a natural sample is reduced, and better accuracy-robustness balance and stronger generalization ability are realized.

The embodiment of the invention provides a picture main body focusing method and device, equipment and a storage medium. The picture main body focusing method comprises the steps of obtaining a key reference object and all target main bodies in a picture; determining a focusing distance based on the first distance between each target subject and the key reference object; and taking the centers of all the target main bodies of which the first distances do not exceed the focusing distance as the focusing center of the picture. According to the method, the focusing center of the picture is intelligently judged and output by analyzing the spatial relationship between the target subject and the key reference object, and then the most critical region of interest is cut based on the focusing center. According to the method, on the premise of completely retaining key visual information, redundant backgrounds and non-related areas in the picture can be effectively eliminated, and the area of the picture is reduced as much as possible, so that the computing power is saved, and the analysis efficiency of a subsequent large model is improved.

A computer-implemented method for analyzing simulation results of a manufacturing or deformation process, comprising: obtaining simulation results, the simulation results comprising at least a geometry of a part (2) and a stress tensor (23) in the part (2) caused by a shaping process; for one or more origins (32) in a critical region (22), determining an origin line (3) by following a direction of maximum stress or a corresponding force; for each origin line (3), determining at least one line segment (31) of the origin line (3) and an origin trajectory (5) representing values of the stress or force directed along the origin line (3); and performing at least one of presenting information representing the line segment (31) and the origin trajectory (5) along the line segment (31) to a user and automatically adjusting parameters of the shaping process.

A computer-implemented method for analysing a result of a simulation of a manufacturing or deformation process, comprisesretrieving the result of the simulation, comprising at least the geometry of the part (2) and stress tensors (23) in the part (2) caused by the forming process;for one or more starting points (32) in a critical region (22), determining a cause line (3) by following the stress or a corresponding force in the direction in which it is maximal;for each cause line (3), determining at least one line section (31) of the cause line (3), and a cause trajectory (5) representing values of a stress or a force directed along the cause line (3);and performing at least one ofpresenting information representing the line section (31) and the cause trajectory (5) along the line section (31) to a user; andautomatically adapting, parameters of the forming process.

The invention relates to the technical field of bridge settlement detection, and particularly discloses a linear array scanning bridge settlement monitoring sensor and a monitoring method, and a carried light path system, a linear laser scanning structure and an FBG linear grating array form a sensing core with complementary advantages. The linear laser scanning structure guides laser to carry out large-range linear array or area array scanning on the surface of the bridge pier, continuous and dense macroscopic displacement data of a key area of the bridge pier can be obtained in a non-contact mode, the defect that a traditional single-point sensor is insufficient in coverage range is overcome, and full-section dynamic analysis of settlement distribution is achieved. And the FBG linear grating array is distributed along the critical path or region of the bridge pier to accurately sense the local microscopic strain and temperature change of the structure, so that the problem that the traditional bridge settlement detection method is difficult to comprehensively reflect the overall settlement form and uneven settlement distribution of the critical region is solved.

A method for generating a test coupon specification for predicting fatigue life of a component includes determining a load condition for the component, providing a component design, and performing a strength analysis of the component design under the load condition determining a critical area of the component and a stress-related parameter of the critical area. The method includes providing a material condition of the component at least for the critical area of the component. To assist an end-user in determining which are optimal tests to be performed in order to obtain most relevant data for fatigue prediction of a specific component, the method also includes providing a material model and providing, as an input to the material model, the stress-related parameter, and the material condition. The material model generates, as an output, a test coupon specification for being tested in a testing machine.

The invention belongs to the technical field of artificial intelligence, and relates to a dynamic controllability management method and system for long-term evolution of a large model, and the method comprises the steps: S1, obtaining the current state information of the large model; s2, calculating a controllability score of the large model based on the output distribution entropy, the distribution variable quantity and the logits drift of the large model, judging whether the current state of the large model deviates from a controllability critical zone or not based on the controllability score, if the current state deviates from the controllability critical zone, entering S3, and otherwise, directly entering S4; s3, actively correcting the offset behavior of the large model in the training process through a controllable calibration head, so that the current state of the large model does not deviate from the controllable critical zone; and S4, dynamically adjusting a training strategy of the large model based on the controllability score and the task category. According to the method, the large model has the capabilities of long-term plasticity maintenance, continuous performance improvement and rapid adaptation to new tasks, and the fundamental upgrade from a static optimization model to a self-evolution model system is realized.

This invention discloses a discrete element method, system, medium, and equipment for simulating equivalent crystalline materials, relating to the field of numerical simulation of rock mechanics. The method includes: converting granite images into binary images using a binarization method to identify various minerals within the binary images; determining the size data of various minerals in the binary images based on a particle size statistical algorithm; extracting micromechanical parameters using nanoindentation experiments combined with a trial-and-error method, and constructing multiple sets of equivalent crystalline models based on the mineral size data; dividing each set of equivalent crystalline models into multiple regions using a Brazilian splitting experiment; determining the optimal magnification for each region of each set of equivalent crystalline models by magnifying from non-critical regions to critical regions using a magnification algorithm, obtaining differentiated equivalent crystalline models; and performing Brazilian splitting numerical simulation using these differentiated equivalent crystalline models to obtain simulation results.

This application discloses a numerical simulation method for wave impact suppression of structures based on SPH (Self-Pulsating Hypothesis), relating to the field of data processing technology. The method includes the following steps: determining multiple critical regions and absorbing layer particles in a preset fluid domain; calculating the critical stability of each anchor point at different times based on the position and state update of each critical particle; updating each anchor point based on the critical stability to obtain multiple final anchor points; calculating the repulsive force adjustment coefficient of each critical particle based on the repulsive force adjustment requirement of each final anchor point and the disturbance degree of the critical particles; adjusting the repulsive force of each critical particle using the repulsive force adjustment coefficient; and performing wave suppression processing on each critical particle based on the adjusted repulsive force and the absorbing layer particles. This application achieves the technical effect of completely and effectively absorbing all transmitted waves in the wave impact simulation process.

The present application relates to the technical field of safety monitoring and industrial automation, and discloses a silo safety monitoring method and system, which synchronously acquires coal pile surface point cloud, lateral pressure, vertical load and environmental data through the arrangement of a laser scanning unit, an annular pressure sensing belt, a weighing sensor and a temperature and humidity node; a high-precision three-dimensional morphology is reconstructed based on an adaptive grid refinement algorithm, and a joint representation model is constructed by fusing mechanical data; a stability discrimination function is established in combination with coal body physical parameters to identify the instability critical region; and through pattern matching with a historical stability morphology library, hierarchical early warning of abnormal working conditions such as suspended arch, bonding and blockage and generation of blockage removal strategies are realized. The present application constructs a joint representation model of coal pile three-dimensional morphology and mechanical state by fusing laser point cloud, pressure distribution and load sensing data, breaks through the limitation of traditional single-point material level measurement which cannot reflect the overall coal pile structure, and realizes high-precision measurement with a storage calculation error of less than 2%.

The application discloses a kind of walking tape deviation correction method, device, equipment and computer readable storage medium, applied to industrial equipment field, the polar piece position data sent by the first sensor working in composite critical region is mirrored and copied and flips to obtain mirror polar piece position data;The diaphragm position data sent by the second sensor working in diaphragm critical region is mirrored and copied and flips to obtain mirror diaphragm position data;In switching region, using the mirror flipping data copied and polar piece position data or diaphragm position data in real time transmission are weighted average to calculate feedback value, to realize the smooth switching of sensor.It avoids the problem that the feedback value of the sensor data calculated in the process of switching the sensor in the prior art may change suddenly, resulting in excessive swing of the deviation corrector and poor deviation correction effect.

A computer processor implemented method and system for monitoring a condition associated with operating heavy equipment is disclosed. The method involves receiving a plurality of images at an interface of an embedded processor disposed on the heavy equipment, the images providing a view of at least an operating implement of the heavy equipment. The method also involves processing each of the plurality of images using a first neural network implemented on the embedded processor, the first neural network having been previously trained to identify regions of interest within the image. Each region of interest has an associated designation as at least one of a critical region suitable for extraction of critical operating condition information required for operation of the heavy equipment, and a non-critical region suitable for extraction of non-critical operating condition information associated with the operation of the heavy equipment. The method further involves causing the embedded processor to initiate further processing of image data associated with critical regions to generate local output operable to alert an operator of the heavy equipment of the associated critical operating condition. The method also involves transmitting image data associated with non-critical regions to a remote processor, the remote processor being operably configured for further processing of the image data and to generate output signals representing results of the further processing. The method further involves receiving the output signals generated by the remote processor at one of the embedded processor or another processor associated with a heavy equipment operations worksite, the output signals being presentable via an electronic user interface based at least in part on the output signals to indicate the results of the further processing.

This application relates to the field of UAV control technology, providing a method, system, device, and storage medium for generating visual environments based on neural radiation fields, solving the problem of poor visual environment generation effects. The method includes: semantically segmenting a corrected image to generate image data with semantic labels, which are used to label the categories of objects in an outdoor scene; updating a neural radiation field model based on the semantically labeled image data and mapping the semantic labels to the voxel feature space corresponding to the neural radiation field model to construct a semantic point cloud; identifying non-critical regions in multi-view images, performing redundant voxel clipping on the non-critical regions to generate clipped semantic point clouds; and using the clipped semantic point clouds to drive the neural radiation field model to generate arbitrary viewpoint visuals of an outdoor scene. This technical solution achieves lightweight semantic modeling and efficient free-viewpoint rendering of outdoor visual environments, improving the efficiency and quality of generating new perspectives in complex environments.

The present application relates to the technical field of marine disaster risk assessment, and particularly relates to a storm surge risk assessment method for a seawater photovoltaic field area, comprising the following steps: S1: identifying a critical region in the field area where a bearing stable state is converted into a bearing attenuation state, and generating a corresponding support state transition unit; S2: determining the transmission direction and the expansion boundary of the support state transition unit between array regions, and generating a corresponding imbalance transmission unit; S3: dividing the overall risk level based on the transmission range and the expansion degree, and generating a corresponding storm surge risk assessment result. According to the present application, by combining support state transition identification, imbalance transmission path analysis and risk space partitioning, the storm surge risk of the seawater photovoltaic field area is changed from static determination to dynamic evolution analysis, thereby improving the accuracy and spatial directivity of risk assessment.

This invention provides a method for intelligent monitoring and early warning of water droplet impact in substations during rainy weather, relating to the field of intelligent operation and maintenance and condition monitoring of substations. The method includes: acquiring dual-optical-path video streams; constructing a cross-spectral neural radiation field network; using joint high-dimensional features as input to output volume density and radiation color vectors; calculating the equivalent water film thickness from these two vectors to generate an equivalent water film thickness distribution field; mapping this field to a three-dimensional model, subdividing the mesh according to the thickness gradient modulus, and solving for the electric field to obtain the electric field distortion tensor and calibrate the high-discharge-risk critical region; generating a Lagrange particle set, solving the dynamic equations through smooth particle hydrodynamics, and simulating the water film evolution to obtain the probability of conductive water bridge connection; inputting the distribution field, distortion tensor, and connection probability into a multi-modal impedance reconstructor to generate an electrochemical impedance spectrum to obtain the dynamic insulation margin attenuation rate; and triggering an early warning based on the determined conditions. This invention significantly improves the timeliness and accuracy of early warning for external insulation faults during rainy weather.

The invention discloses an AK-MCS reliability analysis method based on a dynamic critical reinforcement learning function, and belongs to the field of mechanical equipment structure reliability analysis. According to the method, a DC learning function is provided, and the coupling influence of a sample point prediction value and a prediction variance on the failure probability estimation precision is comprehensively considered through an adaptive weight mechanism. Preferentially selecting sample points with large prediction variance in the exploration stage to quickly improve the global approximation capability of the model; in the refinement stage, a learning function is converted to be dominant by a mean item, the selection weight of sample points near a limit state curved surface is dynamically increased, local fine description of a critical region is enhanced, and a global relative uncertainty index is introduced to realize smooth transition from global exploration to local refinement. According to the method, the problems of excessive iteration and local optimization of a traditional learning function are effectively avoided, the construction efficiency of the Kriging proxy model and the robustness of failure probability estimation are remarkably improved, and the method is suitable for efficient reliability analysis of a complex mechanical equipment structure.

The application provides a process state real-time correction method based on acoustic characteristics, and relates to the technical field of process control, and comprises the following steps: collecting acoustic signals generated by interaction between a processing medium and a processing unit when the processing unit is executing a processing task; performing real-time spectrum analysis on the acoustic signals to extract characteristic frequencies and energy distribution characteristics thereof which are related to the current physical state of the processing medium; determining whether the current physical state of the processing medium is in an abnormal critical region in real time based on changes of the characteristic frequencies and the energy distribution characteristics thereof relative to a preset reference; the abnormal critical region comprises a first precursor state in which the processing medium tends to be excessively dried or a second precursor state in which the processing medium tends to be excessively accumulated; and if it is determined that the processing medium is in the abnormal critical region, the physical state of the processing medium being currently processed is corrected in real time by adjusting at least one execution parameter which influences the physical state of the processing medium before the processing medium being currently processed reaches a target region.

The invention discloses a protection action curve visualization engine based on an IED (Intelligent Electronic Device) model difference atlas, which relates to the technical field of relay protection teaching, training and testing of a power system and comprises an IED model difference atlas construction module and a WebGL (Web Graphics Library) visualization engine module. According to the technical scheme, data are analyzed and a JSON equation is constructed based on a to-be-processed SCD file through an IED model difference atlas construction module, a boundary gradient vector is calculated based on the JSON equation to obtain a difference degree matrix, and a protection action curved surface is constructed and rendered in a unified parameter space based on the difference degree matrix through a WebGL visualization engine module. And a visual result of the protection action curve is displayed, so that students can understand protection action conditions and critical regions more visually, and the teaching efficiency and the practical operation conversion rate are remarkably improved.

The present application discloses a method, system, and device for training an image classification model based on dynamic weight adjustment. The method calculates the image quality score, critical area ratio, and non-critical area ratio of each image sample in an image sample training data set; initializes the weight of each image sample based on the image quality score and determines a first initial weight for each image sample; determines a second initial weight for each image sample based on the first initial weight and critical area ratio of each image sample; calculates the confidence of a weak classifier based on the first initial weight; and dynamically updates the weight of the image samples during the image classification model training process based on the confidence, the second initial weight of each image sample, the image quality score, and the non-critical area ratio, and dynamically adjusts the learning rate during the training process until the image classification model is trained. The present application can improve the accuracy of image classification.

The invention discloses a granite texture surface layer pattern analysis method based on fractal dimension, and relates to the technical field of image recognition, and the method comprises the steps: obtaining texture images of a granite surface at a plurality of frequency bands, carrying out the multi-modal fusion, and generating an enhanced texture representation image; performing modal decomposition on the enhanced texture representation image to obtain a modal image sequence; the fractal dimension of each modal image in the modal image sequence is calculated, and a fractal heterogeneous feature set of granite texture is constructed; identifying a granite surface texture critical region based on the fractal heterogeneous feature set, and extracting a mineral crystal form feature set of the critical region; and constructing a comprehensive evaluation model based on the fractal isomerism feature set and the mineral crystal morphology feature set, predicting granite performance indexes and generating an analysis result. According to the method, a multi-band imaging and tensor decomposition combined feature fusion method is adopted, effective representation and fusion of cross-band features are realized through information entropy weight, and the integrity of the enhanced texture representation image is improved.