226 results about "Confusion matrix" patented technology

The invention discloses an employment talent supply and demand distribution method and system based on data analysis, and belongs to the technical field of data analysis, and the method comprises the steps: obtaining talent basic information and post basic information; cleaning and sorting the talent basic information and the post basic information, identifying and deleting repeated data, and processing missing values and abnormal values; key features are extracted from the talent basic information and the post basic information by using a deep learning technology, and the key features are weighted or screened; establishing a matching model between the talent basic information and the post basic information by using a machine learning algorithm; a matching result is analyzed, model optimization is carried out according to the result, parameters or feature weights are adjusted, and the matching accuracy and efficiency are improved; the generalization ability and accuracy of the matching model are evaluated through cross validation and confusion matrix methods; and continuously optimizing the matching model, and adjusting the matching model according to feedback and experience to improve the matching effect. The method has the effect of helping to reasonably configure human resources.

The invention discloses an image analysis method and system for concrete apparent quality defect detection, and relates to the technical field of concrete apparent quality defect detection.The method comprises the steps that a camera device is used for collecting concrete surface images in a specified distance interval, and the optical axis of a camera lens is controlled to be perpendicular to the concrete surface; carrying out image preprocessing on the collected concrete surface image, and carrying out defect type image marking; carrying out key feature extraction on the preprocessed concrete surface image by adopting a convolutional neural network to obtain key feature information; constructing a multi-algorithm comparison verification framework, performing cross validation in combination with the key feature information to obtain defect detection results and evaluation results of a plurality of defect detection models, optimizing model parameters in combination with a confusion matrix, and determining a final concrete surface detection model; obtaining an apparent quality defect detection result based on the defect detection result and a preset quality defect grading threshold value; the efficiency of concrete apparent defect detection is improved.

The invention discloses a bearing fault diagnosis method based on multi-scale frequency sensing dynamic enhancement, and the method comprises the steps: collecting a vibration signal in a bearing operation state, carrying out the preprocessing of the vibration signal, obtaining a time-frequency matrix, and dividing the time-frequency matrix into a training set and a test set; building a multi-scale frequency network sensing model, inputting a time-frequency matrix in a training set into the model to realize extraction of multi-scale features, then performing pooling, time sequence compression, flattening and dimension reduction on the extracted multi-scale features, and then outputting fault category probability distribution through a classifier; and finally, testing the trained model by using a test set, and calculating evaluation indexes such as accuracy, a confusion matrix, an ROC curve and the like. The method has high accuracy while keeping light weight, breaks through double limitations of fixed frequency band and sensitive rotating speed of a traditional method, and can provide a high-precision and low-cost light-weight solution for engineering application of variable-rotating-speed mechanical fault diagnosis.

The invention relates to the field of bearing fault detection, in particular to a bearing fault diagnosis method based on time-frequency enhancement CNN-Transformer, which comprises the following steps: acquiring a bearing vibration signal, and preprocessing the bearing vibration signal; selecting a window size, an interval step length and an overlapping rate, and dividing the preprocessed bearing vibration signal data into a training data set, a verification data set and a test data set by adopting a sliding window; an FFT-CNN-Transform network model is constructed, and the FFT-CNN-Transform network model is constructed; an Adam optimizer and a cross entropy loss function are adopted to train an FFT-CNN-Transform network model, hyper-parameters of the FFT-CNN-Transform network model are adjusted through a verification data set, and the performance of the FFT-CNN-Transform network model is evaluated through the test data set; and by drawing a confusion matrix, the classification effect of the FFT-CNN-Transform network model on different fault categories is displayed. According to the method, key features of a time domain and a frequency domain can be captured at the same time, and data features are compared with fault features, so that more accurate fault prediction is realized, and the accuracy and reliability of bearing fault diagnosis under complex working conditions are remarkably improved.

The invention provides a vocal music training vowel pronunciation quality evaluation method based on auditory and visual spatial-temporal feature fusion, and the method comprises the steps: collecting vowel pronunciation audio signals and corresponding videos of a singer, and constructing a multi-modal data set; generating a fractional order Mel spectrogram for the audio signal through short-time fractional order Fourier transform of an adaptive order; extracting time sequence features and spatial features of the fractional order Mel spectrogram, and fusing the time sequence features and the spatial features through a gating mechanism to generate audio spatio-temporal features; face visual features in the video are extracted and fused with the audio spatio-temporal features through a cross attention mechanism, and the cross attention mechanism is integrated with a periodic modeling network; the fused features are input into a classifier, a dynamic weight multi-mode cosine loss function training model is adopted, the dynamic weight multi-mode cosine loss function dynamically adjusts the sample weight through a confusion matrix, and the weight is increased for the samples with classification errors based on the historical frequency mistaken division times of the samples; and outputting a pronunciation quality evaluation result.

Machine learning models, such as machine learning classifiers, are trained to make medical predictions using loss functions that include at least one term derived from a confusion matrix, such as negative predictive value (NPV). The medical predictions are based on sets of features extracted from medical images. The medical images may be either radiology images, such as CT, MRI, or PET scans, or pathology images, such as whole-slide images of tissue. The loss functions may include terms that are designed to improve the overall accuracy of the machine learning model in addition to the confusion matrix term or terms. The loss function is typically constructed such that it is continuously differentiable.

The invention relates to the field of machine learning, in particular to a domain adaptive image recognition and classification method in a passive continuous scene, a storage medium and electronic equipment, and the method comprises the steps: carrying out the enhancement processing of source domain data, and constructing a confusion matrix; performing semantic recombination on the confusion matrix to obtain a negative sample data set; training an image quality quantification model by using the negative sample data set and the source domain data; performing invariance quantization processing on the image samples of the target data set; training the target model, performing feature extraction on the target domain image data in the current incremental task at the end of each training round, and storing an extraction processing result; carrying out source tag identification; and optimizing the total loss function of the target model, training and optimizing the target model, and obtaining a final target model so as to identify the target domain image in the passive continuous adaptive scene. The problem of performance reduction caused by domain offset in passive continuous domain adaptation is solved.

The invention discloses an urban flood prediction method based on a dual-drive urban flood model, and the method comprises the steps: carrying out the early-stage preparation of model construction, constructing an urban flood hydrological and hydrodynamic coupling model frame, constructing a deep learning model frame based on a GBDT algorithm, and carrying out the prediction of the urban flood. Assimilation of predicted values and measured data of a hydrological hydrodynamic model and a deep learning model is realized through a real-time data assimilation technology, then an output result of an urban flood hydrological hydrodynamic coupling model is used as an input feature of the deep learning model, and the input feature and parameters are dynamically adjusted according to the matching degree of a confusion matrix. The TP in the confusion matrix is maximum, the TN in the confusion matrix is minimum, finally, construction of the dual-drive urban flood model is completed, and prediction is conducted through the model. According to the method, the characteristic of high calculation efficiency of the deep learning model is exerted while calculation accuracy is considered, a layered coupling architecture is provided, and the problems that the calculation efficiency of a hydrological hydrodynamic model is low and a traditional deep learning model has a black box effect are solved.

The invention belongs to the field of traffic safety management, and discloses a traffic accident severity influence factor analysis method based on local cascade integration, which comprises the following steps: acquiring an accident data set D1; processing the accident data set D1, removing part of redundant attributes, dividing the accident data set D1 into a training set and a test set according to a proportion, and balancing the number of accidents with different severity degrees in the training set through an SMOTENC algorithm to obtain an accident data set D2; importing the accident data set D2 into a local cascade integration model, adjusting hyper-parameters of a local model through a Hyperpt method, and selecting an optimal hyper-parameter combination by using k-fold cross validation; drawing a confusion matrix according to a training result, and selecting indexes to evaluate model performance; and visualizing the model by applying a machine learning output explanation tool SHAP, and analyzing accident severity influence factors according to the visualized model. By adopting the SMOTENC resampling technology, the number of various accidents in the training set is balanced and the model training effect and the classification performance are improved on the premise of considering discrete and continuous variable differences.

An example method includes: serving content to a plurality of client devices associated with a plurality of tag values; predicting, using a prediction system, a plurality of attributes respectively associated with the plurality of tag values; generating a data sketch descriptive of the plurality of predicted attributes; noising the data sketch, wherein the noised data sketch satisfies a differential privacy criterion; transmitting the noised data sketch to a reference system; and receiving, from the reference system, estimated performance data associated with the predicted attributes, wherein the estimated performance data is based on an evaluation of: reference attribute data associated with one or more of the plurality of tag values and the predicted attributes for the one or more of the plurality of tag values.

Aspects of hair simulation, hair classification, and networks therefor including aspects to train such networks are provided. A classifier model alleviate the impact of human bias where the modeling of the real label distribution and annotators' biases are separated by incorporating annotator confusion matrices into a baseline model. The model was trained using a consistency-based semi-supervised learning framework. With the use of only 1000 labeled data, the final classifier model achieved a classification accuracy that was 20% higher than a human professional annotator. The model is useful as a color classifier to train generative models for hair color translation. There is provided a generative model for hair simulation (e.g. via VTO) that is guided during training by a hair classifier model. Further provided is a color mapping network to process an input image and target hair color for the generative model to define the hair simulation.

The embodiment of the invention provides a mobile communication PCI (Peripheral Component Interconnect) decision-making method and device based on an improved ant colony algorithm, and the method comprises the steps: obtaining MR (Measurement Report) data of a base station and a real-time network state; the base station MR measurement report data comprises a conflict matrix, a confusion matrix and a modulo 3 interference matrix; extracting a network topology feature vector according to the MR measurement report data of the base station; according to the network topology feature vector and the real-time network state, obtaining a heuristic weight and a pheromone volatilization rate parameter; based on an ant colony algorithm, obtaining an improved ant colony algorithm according to the heuristic weight and the pheromone occurrence rate parameter; based on the quantum revolving door, determining a three-level objective function according to the conflict matrix, the confusion matrix and the modulo 3 interference matrix; and based on an improved ant colony algorithm, generating an optimal PCI allocation scheme according to the conflict matrix, the confusion matrix, the modulo 3 interference matrix, the three-level objective function and a preset constraint condition. The problem of optimization target imbalance is solved, and the reliability of the PCI decision method can be ensured.

The invention discloses an open set fault diagnosis method based on weighted domain self-adaption and sub-domain alignment, namely WDA-SDA, which is used for solving the fault diagnosis problem that a label space of a source domain is a subset of a label space of a target domain when working conditions of the same equipment are changed. The model input of the WDA-SDA is frequency domain data of vibration signals subjected to maximum and minimum normalization and fast Fourier transform, and the model is trained through a pre-training stage and a domain adaptive alignment stage. Experiments on three data sets of a bearing, a gear and the like prove that the H-score of the WDA-SDA method in multiple tasks reaches 100.00%, and the method has good generalization ability. According to the improvement point provided by the invention, through ablation experiments on each data set, the effectiveness of an open set discriminator, rapid nuclear norm maximization and a local alignment strategy is verified. In addition, through a confusion matrix and T-SNE visual analysis, the superiority of the WDA-SDA is further verified.

In various examples, the present disclosure relates to using temporal filters for automated real-time classification. The technology described herein improves the performance of a multiclass classifier that may be used to classify a temporal sequence of input signals—such as input signals representative of video frames. A performance improvement may be achieved, at least in part, by applying a temporal filter to an output of the multiclass classifier. For example, the temporal filter may leverage classifications associated with preceding input signals to improve the final classification given to a subsequent signal. In some embodiments, the temporal filter may also use data from a confusion matrix to correct for the probable occurrence of certain types of classification errors. The temporal filter may be a linear filter, a nonlinear filter, an adaptive filter, and / or a statistical filter.

The invention discloses a tool and model testing method based on TTS, and belongs to the technical field of intelligent terminal testing. According to the method, the TTS technology is combined with natural language generation, environment simulation and a self-adaptive algorithm, a test text set containing sentence pattern variants and synonym replacement is generated through a natural language generation model, a parameter-adjustable TTS engine is used for synthesizing voice signals with acoustic and emotional characteristics, the voice signals and an environment impulse response function are subjected to convolution superposition of scene noise and then input into a terminal, and the voice signals are converted into voice signals with the acoustic and emotional characteristics. Mapping to a semantic vector space to calculate similarity, and finally generating an iterative test strategy according to the confusion matrix. According to the scheme, test text automatic generation, complex environment simulation and test strategy adaptive optimization are realized, the test efficiency and coverage are improved, the problem is accurately positioned, the multi-scene adaptability is enhanced, and a full-process intelligent test scheme is formed.

The invention discloses a real-time network intrusion detection method based on a genetic algorithm and a bidirectional long-short term memory network, and belongs to the technical field of network security, and the method comprises the following steps: 1, extracting continuous and category network traffic features from a constructed data set, and carrying out the feature preprocessing and preliminary screening; step 2, performing feature selection optimization based on a genetic algorithm so as to screen out an optimal feature combination with high accuracy and low dimension; 3, adopting a bidirectional LSTM algorithm to construct an abnormal traffic detection model based on the feature subset selected by the genetic algorithm, wherein the abnormal traffic detection model is used for identifying normal and abnormal samples in the network traffic; 4, evaluating the performance of the abnormal traffic detection model by adopting the confusion matrix; and step 5, deployment and real-time detection of an abnormal flow detection model. According to the invention, a lightweight and traceable intrusion detection framework is constructed. The generalization ability and precision of the detection model are improved; the method gives consideration to accuracy, interpretability and system response capability.

The invention relates to a nondestructive walnut grading detection method based on a ray technology, which comprises the following steps: taking walnuts as materials, firstly carrying out CT imaging and shell breaking on different walnuts with shells, then carrying out manual grading marking to obtain a training set, carrying out feature recognition and grading on walnut kernels by using a deep residual network model, and outputting a grading detection result. Training the deep residual network model by minimizing the grading detection result and the actual grade to obtain a trained walnut grading detection model; and performing CT imaging on the walnut kernels with shells by using the trained walnut grading detection model to obtain a grading result. According to the invention, an image identification classification model is constructed through an algorithm, and a confusion matrix graph and an ROC curve are introduced to verify the feasibility and accuracy of the model in a certain range. The non-destructive testing technology is used for grading the walnuts with the shells, and the method has the advantages of cost saving, time saving and convenience.

The invention discloses an artificial intelligence-based industrial electromechanical equipment management and control method and system, and particularly relates to the technical field of equipment management and control, and the method comprises the steps: collecting the operation data of industrial electromechanical equipment through a sensor and a monitoring system, and extracting the characteristics of the operation data of the industrial electromechanical equipment; performing fault diagnosis on each industrial electromechanical device by using an artificial intelligence algorithm based on the extracted features, determining a standby device of the industrial electromechanical device having the fault according to the industrial electromechanical device having the fault, and collecting historical information of the standby device according to the relative amplitude of the operation data of the standby device and the fault record of the standby device, according to records detected by the artificial intelligence algorithm of the standby equipment, confusion matrix indexes of the standby equipment are determined, artificial intelligence training information of the standby equipment is collected, and the matching degree of the standby equipment and the fault equipment is quantified. According to the method, the starting opportunity of the standby equipment can be optimized, and the minimum downtime of equipment switching in the production process is ensured.

The invention discloses an offshore wind power low-frequency circuit breaker fault diagnosis method and a related device, electrical parameter data, environmental data and mechanical data are fused and spliced into a multi-dimensional feature vector, various fault feature information of a low-frequency circuit breaker in a complex working environment is comprehensively captured, the fault sensitivity and recognition capability are improved, and the fault diagnosis accuracy is improved. The random forest algorithm is adopted to realize high-precision diagnosis under a small sample, limited low-frequency circuit breaker data is fully utilized, the overfitting problem is effectively avoided, the number of decision trees in the random forest is optimized, the classification performance of the model is improved, the ROC curve and the confusion matrix are adopted to evaluate the performance of the low-frequency circuit breaker fault diagnosis model, and the reliability of the low-frequency circuit breaker fault diagnosis model is improved. And the effectiveness and reliability of the fault diagnosis model are improved. The technical problems that an existing offshore wind power low-frequency circuit breaker fault diagnosis method based on a single data source is high in false detection rate and an offshore wind power low-frequency circuit breaker fault diagnosis method based on multi-modal data fusion has model overfitting are solved.

The invention discloses a natural gas pipeline leakage detection method based on novel wavelet basis transformation and singular value decomposition in a two-dimensional convolutional neural network. Firstly, a sound signal collected by a sound wave sensor is converted into a digital signal; secondly, in the data preprocessing stage, singular value decomposition is carried out on the digital signals to effectively eliminate background noise interference, and then batch normalization is carried out on the processed data; then, converting the one-dimensional time sequence signal into a two-dimensional time-frequency image by adopting a self-defined Morlet wavelet basis function; and finally, based on the time-frequency images, constructing and training a 2D-CNN model for fault classification, and presenting a diagnosis result through a confusion matrix and a comparison graph. According to the method, 97.55% of fault recognition accuracy is obtained in a public data set, and compared with other competitive methods, the method shows more excellent noise robustness and classification performance, and has higher accuracy and wide application prospects in pipeline leakage diagnosis in a complex noise environment.

The invention discloses an address identification detection method, which comprises the following steps of: S1, acquiring address data expressed by different texts as a data set, and dividing the data set into training data, inspection data and test data; s2, establishing a BERT neural network model based on pre-training, inputting training data into a neural network for training in the BERT neural network model, adjusting confusion matrix values according to expression differences of address texts in the training process, and substituting the confusion matrix values into the training process of the neural network model in combination with a preset model evaluation rule to obtain a final neural network model; s3, calculating a performance index, evaluating the availability of the model, and if the requirement is met, storing the model; and S4, using the model to calculate feature similarity of a plurality of address texts, comparing the feature similarity, and identifying and obtaining the address text most similar to the specified address text. According to the method, the similarity identification of the address text can be automatically carried out, the matching accuracy is improved, and the use cost of a user is also reduced.

The invention discloses a remote sensing hyperspectral image classification method, system and device and a medium, belongs to the technical field of remote sensing image processing and mode recognition, and aims to solve the technical problem of how to improve the precision and efficiency of remote sensing hyperspectral image classification, reduce the calculation complexity, fully extract spectrum and spatial features and improve the classification efficiency of remote sensing hyperspectral images. According to the technical scheme, the method comprises the following steps: dividing a training set and a test set: dividing all pixels of a remote sensing hyperspectral image into the training set and the test set according to a set proportion; constructing a classification model: through a cross attention mechanism, constructing the classification model in combination with Transform of a spectral attention mechanism and KL divergence driven mutual learning classification; the test set is input into the classification model, a prediction result of the test set is obtained, a confusion matrix of the prediction result and a real label is calculated, and the final classification precision is obtained; and splicing a visual image.

The invention relates to the technical field of remote sensing, in particular to a remote sensing extraction method for a cultivated land boundary, which comprises the following steps of: selecting a high-resolution remote sensing image of a research area in a natural year, and making time sequence NDVI (normalized difference vegetation index) image data in the year; performing edge detection, spatial superposition, cultivated land range extraction and other land type mask processing on the intra-year time sequence NDVI data image to obtain a potential cultivated land edge accumulation frequency graph; classifying cultivated land edges in the potential cultivated land edge accumulation frequency graph into permanent edges, temporary edges and non-edges through an automatic threshold segmentation algorithm, and retaining the permanent edges to obtain a cultivated land boundary graph; by calculating an ECR value, a confusion matrix and Spearman correlation analysis are adopted, comparison with a Geo-Wiki Field Size data set is carried out, and the farmland boundary extraction precision is verified. The method is simple in operation, high in extraction efficiency and accuracy, and beneficial to farmland boundary detection and control.

The invention relates to a transferable vector quantization alignment method and device based on unsupervised domain adaptation, and the method comprises the steps: extracting the features of a source domain and a target domain of a source domain data set and a target domain data set through a feature extraction unit, and calculating an overall feature distribution loss function through searching a feature item closest to the features in a codebook; calculating a local alignment loss function through a bottleneck layer, performing classification through a classifier to obtain a source domain pseudo-label and a target domain pseudo-label, calculating a cross entropy classification loss function according to the source domain pseudo-label and a corresponding truth value label, performing normalization processing on the target domain pseudo-label, introducing mutual information to obtain a sample weight in each target domain data set, and obtaining a sample weight in each target domain data set; and calculating mutual information weighted maximization confusion matrix loss functions, and updating parameters in each module by using the loss functions until convergence to obtain a target classification architecture with cross-domain extraction, feature alignment and time sequence signal classification capabilities. By adopting the method, the identification performance of the label-free target domain can be improved.

The invention provides a large-model intelligent label synthesis and data automatic labeling integration method and system, and belongs to the technical field of label synthesis and data labeling, and the method comprises the steps: carrying out the semantic embedding and robust clustering of text data, and obtaining a stable cluster set; when new data is introduced, semantic consistency alignment of cross-round clustering results is realized through a confusion matrix matching strategy, and label drift is inhibited; maintaining an editable hierarchical label directed acyclic graph to support label system evolution; driving a large language model to generate a high-quality and interpretable cluster-level semantic tag based on the representative sample; carrying out automatic annotation and confidence evaluation by using large model context learning for clustering non-attribution or low-confidence samples; and propagating the cluster-level labels to the instances, and combining the cluster-level labels with an automatic labeling result to construct a full-process traceable label management mechanism. According to the method, the efficiency, quality and consistency of text labeling are improved, and powerful support is provided for large model training and intelligent data management.

The invention relates to a keratoconus analysis system based on dynamic cornea textures. The keratoconus analysis system comprises an image acquisition unit, an image preprocessing unit, a feature extraction unit and a feature screening unit which are connected in sequence, the multi-time-point cornea dynamic image texture analysis method is provided, in order to ensure that the screened features have the optimal diagnostic capability, a feature set with the highest diagnostic value is screened through a recursive feature elimination method, and a foundation is laid for FFKC high-precision recognition. The machine learning classification system comprises a segmentation module, a model training module, a model testing module and a performance evaluation module, and is used for obtaining a machine learning classifier. The performance of the model is evaluated through an ROC curve and a confusion matrix, and high sensitivity and high specificity are ensured. A multi-modal recognition platform fusing dynamic textures and cornea biomechanical parameters is constructed, the platform sequentially comprises a data input module, a data processing module, a model module and an output module, and efficient FFKC recognition and auxiliary diagnosis are achieved.

The invention belongs to the technical field of analytical chemistry, and relates to a method for identifying and detecting endogenous mercaptan of a high-activity oxidized nano-enzyme based on machine learning, which comprises the following steps: firstly, preparing a Mnx (DTPMP) nano-enzyme, then constructing a sensor array of one nano-enzyme, four reaction times, three targets and six parallel samples, and detecting the endogenous mercaptan in the Mnx (DTPMP) nano-enzyme. Constructing a sensor array of one nano enzyme, four reaction times, 7 concentrations and 6 parallel samples for each endogenous mercaptan; and finally, analyzing the data through hierarchical clustering analysis, linear discriminant analysis and a support vector machine to obtain an LDA score chart, an HCA map, an SVM confusion matrix and a concentration prediction map of the endogenous mercaptan. After a sample to be detected is subjected to a chromogenic reaction, the absorbance is measured, and the type and concentration of the endogenous mercaptan of the sample to be detected can be distinguished according to the obtained spectrum. The kit has good anti-interference performance, normal cells and cancer cells can be accurately identified according to the content of glutathione in the cells, and the severity of diseases can be judged according to the content of homocysteine in serum.

The invention discloses a research method for acupuncture point classification and multi-channel electroencephalogram signal correlation analysis based on a convolutional network, and belongs to the field of brain-computer interfaces and the field of signal processing. The method is specifically implemented according to the following steps: firstly, acquiring required data by using an electroencephalogram signal acquisition device, and preprocessing the acquired electroencephalogram signal: filtering and segmenting; the correlation between the channels is studied, and visual display is carried out; extracting time-frequency characteristics of the electroencephalogram signals through wavelet transformation; drawing a time-frequency graph; inputting a time-frequency diagram result into an SE-ConvNeXt network for training, and obtaining an acupoint classification result; and analyzing the result, and visually expressing the accuracy by using a confusion matrix. The method disclosed by the invention has important reference significance for deeply studying the regulation mechanism of the acupuncture point acupuncture traditional Chinese medicine therapy on the central nervous system and studying the acupuncture point acupuncture classification according to the electroencephalogram signals.

The invention provides a confusion matrix scoring method and device based on entropy coding, and relates to the technical field of machine learning. The method comprises the following steps: acquiring an original confusion matrix; setting an element in a main diagonal line of the original confusion matrix to be zero to obtain an error matrix; binary entropy coding is carried out on terms of the error matrix in an entropy coding mode, and a coding matrix is obtained; calculating the code length of each code in the coding matrix according to the coding matrix to obtain a code length matrix; multiplying each item of the code length matrix by each item of the error matrix to obtain an average code length matrix; summing each item of the average code length matrix to obtain a total average code length; and the value of the total average code length is used as a scoring index. According to the invention, the error rate and the dispersity of error conditions can be reflected.

The invention discloses an electric power gallery flood risk assessment method and system based on deep learning, and belongs to the field of flood disasters, and the method comprises the steps: achieving the water extraction and risk grade assessment of a radar image, obtaining the radar image from a Sentinel-1 satellite, removing the noise through a Speckle filter, and carrying out the geometric correction, radiation correction and normalization processing. Data preprocessing is carried out to ensure data quality; performing image segmentation on the preprocessed image by using a U-Net model, extracting and recovering image features through an encoder and decoder structure, and realizing high-precision water body segmentation through a deep learning model; the water body extraction precision is evaluated by calculating indexes such as a confusion matrix, a recall rate, accuracy and MIOU, the flood risk level of the power equipment is evaluated according to the extracted water body data, the gallery bridge distance and the gradient, and risk evaluation is divided into five levels. According to the method, the problem of how to accurately extract the water body and evaluate the flood risk of the electric power gallery is solved, and a scientific basis is provided for protection and management of electric power facilities.