108results about How to "Improve classification effect" patented technology

The invention discloses a semiconductor defect classification and prediction method using a deep learning algorithm, and the method comprises the following steps: S1, collecting image data and manufacturing process parameters in a semiconductor manufacturing process through a multi-modal sensor, and generating a multi-dimensional data set; s2, preprocessing the data set; s3, constructing and training a deep learning model in which a convolutional neural network and a long-short-term memory network are combined, and automatically detecting and marking a defect region; s4, classifying the defects by adopting a quantum convolutional neural network in combination with a Hash embedding algorithm; s5, performing defect prediction based on historical data and real-time process parameters through a chaotic time sequence prediction network; s6, dynamically adjusting the model structure through the self-adjusting neural network, and maintaining the detection and prediction capabilities; and S7, adopting a reinforcement learning strategy to carry out adaptive optimization on the model. The method can improve the accuracy and prediction precision of semiconductor defect detection, is suitable for different manufacturing environments, and effectively reduces the defective product rate and production loss.

The invention provides a sea island reef landform identification and classification method and system and a storage medium, and relates to the technical field of remote sensing data analysis, and the method comprises the steps: obtaining remote sensing data of a sea island reef research area; performing prediction by using the remote sensing data through the water depth inversion model to obtain water depth data of the sea island reef research area; inputting the water depth data and the remote sensing data into a landform identification and classification model, and obtaining a landform classification result of the sea island reef research area according to the output of the landform identification and classification model; wherein the initial classification model is pre-trained by utilizing the preset water depth data and the preset remote sensing data through comparative learning, and the pre-trained initial classification model is trained according to the preset water depth data and the preset remote sensing data to obtain a landform identification classification model. According to the invention, high-reliability intelligent identification and classification of the topography and landform of the sea island reef are realized, and the classification effect of the topography and landform of the sea island reef is improved.

The invention provides a valve leakage diagnosis method and system based on an improved Hereum algorithm optimized SVM (Support Vector Machine), and relates to the technical field of valve leakage diagnosis. Decomposing the collected original signal to obtain a limited number of IMF components; calculating a fuzzy entropy, and constructing a feature vector; taking the feature vectors as data samples, and dividing the data samples into a training set, a verification set and a test set; using the training set to train the SVM model, and according to the classification error rate of the verification set on the trained SVM model, performing iterative optimization on the parameters of the SVM model based on an improved He-Ma algorithm to obtain an optimized SVM model; inputting the test set into the optimized SVM model to obtain a valve leakage diagnosis model; and acquiring a new acoustic emission signal, and inputting the new acoustic emission signal into the valve leakage diagnosis model to obtain a diagnosis result. The method not only can improve the classification accuracy and generalization ability of the diagnosis model, but also can enhance the stability of the model under different working conditions and environmental noise.

The invention discloses a small sample image classification method and system based on multi-scale cross-modal prompt enhancement, and relates to the technical field of small sample image classification, and the method comprises the steps: obtaining a query image and a support image small sample data set with a labeled category; inputting the acquired data into a trained small sample image classification model, extracting visual features of a query image and enhanced visual features of each support image through a feature extractor based on hierarchical multi-modal prompt enhancement in the model, and performing global average pooling on all the extracted enhanced visual features to obtain an enhanced visual feature of each support image; obtaining an enhanced visual feature prototype of each category, calculating similarity scores of the categories according to the visual features of the query image and the enhanced visual feature prototype of each category, determining the category with the highest score as a classification result of the query image, and outputting the classification result; wherein the small sample image classification model is obtained by training a global-local self-guiding supervision mechanism. According to the invention, the precision and robustness of small sample image classification can be significantly improved.

The invention belongs to the technical field of hydraulic engineering, and provides a sediment particle size distribution measuring system and method.The method comprises the steps that images of sediment particles in a water body are collected, and a first sediment particle image is obtained; processing the first sediment particle image to obtain a sediment grayscale image; a K-means clustering algorithm is adopted to carry out clustering analysis on the sediment gray level image, and an image with a first-level clustering result is obtained; carrying out clustering analysis on sediment particles with particle sizes smaller than a preset threshold value in the image with the first-level clustering result by adopting a DBSCN clustering algorithm to obtain an image with a second-level clustering result; and carrying out edge detection on the sediment particles in the image with the secondary clustering result, and then sequentially carrying out binarization processing, equivalent circle diameter transformation and statistics on the particle size distribution of the sediment particles to obtain a particle size distribution result. The problem that an existing sediment particle size measurement system is low in measurement precision in a complex underwater environment with dense sediment and more fine particles is solved.

The invention relates to a medical image processing technology, and discloses a pan cancer species histopathologic full-slice classification method and device, which is a trained multi-modal pan cancer species classification model based on histopathologic full-slice image data and full-slice diagnosis report pairing data of various tissue parts and various cancer species, and has the advantages of wide application range and high efficiency. And full slices of various pathological tissues can be classified. For classification tasks of different tissue parts, only text prompts (prompts) are used, i.e., a candidate cancer species list is set, and zero-sample full-slice classification can be realized without training. And for rare cancer species, full-slice classification can be carried out. Besides, the model has high expansibility, in order to better adapt to downstream tasks, better classification performance can be achieved by combining slice embedding with a linear classifier for use, freezing a slice embedding part and only finely adjusting the linear classifier, and the requirement for the training data size is low.

The invention relates to a noise feature vector classification model construction method, a speech enhancement method and device, computer equipment, a storage medium and a computer program product. The method comprises the following steps: coding penalty parameters and kernel function parameters of a support vector machine; taking the penalty parameter and the kernel function parameter as optimization variables of a genetic algorithm, setting an initial population, and constructing a fitness evaluation function based on classification accuracy; performing selection operation, crossover operation and mutation operation on the initial population, and updating an iterative population according to a fitness evaluation function to obtain an optimal penalty parameter and an optimal kernel function parameter; and training a support vector machine through the optimal penalty parameter and the optimal kernel function parameter, and constructing a noise feature vector classification model. In a noisy environment, the classification effect of the noise feature vector is improved under the condition that nonlinear and non-stationary noise exists, and the noise feature classification requirement in a complex sound signal processing scene is met.

The invention discloses an unmanned aerial vehicle image multi-crop plot identification and productivity prediction method based on DeeplabV3 +, and the method comprises the steps: obtaining an unmanned aerial vehicle image of a research region, and carrying out the splicing processing of the image; preprocessing and labeling the images, and dividing the images into a training set and a verification set; training, verifying and testing the improved DeeplabV < 3 + > deep learning network model by using the data set to obtain a final unmanned aerial vehicle image plot recognition and crop classification model based on DeeplabV < 3 + >; carrying out land parcel identification and crop classification on the preprocessed to-be-detected image by using the trained land parcel identification and crop classification model, and carrying out splicing processing; and the spliced prediction segmentation map is further processed by using envi and ArcGis software to obtain a productivity prediction map of various crops. According to the invention, the dependence on manual field investigation is reduced, the method is less limited by topographic factors, the method is faster and more accurate, and the efficiency of agricultural asset checking is improved.

The invention relates to a BNN-based hyperspectral image classification method, apparatus and device, and a memory. According to the method, a BNN-based hyperspectral image classification model is designed, excellent classification performance is realized while a CNN model is compressed, and a self-adaptive gradient scaling module based on a multi-scale pyramid compression attention mechanism is used as a full-precision real value bypass branch of a basic BNN network in a model training process. The method is used for flexibly adjusting the gradient in the back propagation process to reduce the gradient error, so that the binary neural network is better optimized without increasing the reasoning overhead, and meanwhile, the representation capability of the network is kept in forward propagation. Compared with an existing binaryzation or quantification method, the method has the advantages that the classification accuracy of the hyperspectral image is remarkably improved, and meanwhile, a high compression ratio is kept.

The utility model relates to the technical field of acupuncture and moxibustion storage boxes, and discloses a traditional Chinese medicine acupuncture and moxibustion portable box which comprises a box body and a connecting plate, upper rotating seats are fixedly connected to the positions, close to the front portion and the rear portion, of the two sides of the lower end face of the connecting plate, and connecting rods are rotationally connected to the lower end faces of the multiple upper rotating seats. Damping grooves are formed in the positions, close to the front portion and the rear portion, of the two sides of the inner wall of the lower end of the box body correspondingly, sliding bases are slidably arranged in the multiple damping grooves correspondingly, and damping rods are fixedly connected between the multiple sliding bases and the inner walls of one sides of the corresponding damping grooves correspondingly. According to the acupuncture and moxibustion box, the damping structure is arranged to achieve a better buffering and protecting effect on appliances in the box body, so that the purposes of better and stable placement and damage prevention are achieved, and disposable cotton swabs and acupuncture and moxibustion bags can be respectively placed through the classification and limiting structure, so that the acupuncture and moxibustion box is convenient to use. Therefore, better tidying property and practicability are achieved.

The invention relates to a hyperspectral image classification method, in particular to a hyperspectral image classification method based on a greedy strategy guided graph self-attention network. The invention provides a hyperspectral image classification method of a graph self-attention network based on greedy strategy guidance in order to solve the problems that in existing HSI classification, samples are limited, GAT tends to only consider directly-connected nodes, indirectly-related relations are insufficiently considered, and the classification effect is affected. The method comprises the following steps: step 1, constructing a graph self-attention network based on greedy strategy guidance; the greedy strategy guidance-based graph self-attention network comprises a data preprocessing stage, a super-pixel-based graph self-attention branch stage, a pixel-based multi-attention fusion branch stage, a classification stage and a greedy training strategy stage. And step 2, classifying the hyperspectral image to be measured based on the trained greedy strategy guided graph self-attention network to obtain a classification result. The method is applied to the field of hyperspectral image classification.

The invention relates to the field of text classification, in particular to a BiLSTM and mixed attention ResNet-based text classification network, a Bi-Hybrid Att-ResNet text classification network model comprises a BiLSTM, mixed attention and ResNet, the BiLSTM of the model is used for establishing a long-term dependency relationship, context representation before and after access and a mixed attention mechanism, key information in BiLSTM output can be captured more effectively, and the text classification network model can be applied to the field of text classification. The method comprises the following steps of: firstly, extracting a sentence by using ResNet, facilitating feature extraction, increasing network depth, accelerating network convergence and capturing n-gram features in the sentence, and finally, classifying the processed context information by using a full connection layer. And the Bi-Hybrid Att-ResNet can be used for capturing the local features and the global semantic information of the text.

The invention belongs to the technical field of information recommendation, and particularly relates to an intelligent image recognition and commodity classification system. Comprising an image acquisition module, an image processing and feature extraction module, a deep learning model training module, a commodity classification and identification module, a user interface module, a database module, a system integration and deployment module, an update and maintenance module and a security module. The invention provides an intelligent image recognition and commodity classification system, and a multi-layer perceptron is adopted as a feature representation unit in the intelligent image recognition and commodity classification system, so that the system can capture a more complex nonlinear relationship in an image. Therefore, abstract features in the image can be better expressed, and the recognition capability of the system on complex modes is improved.

The invention discloses an epilepsy detection method combining uncorrelated multi-linear discriminant analysis and deep metric learning, and belongs to the technical field of epilepsy detection.The epilepsy detection method comprises the steps that S1, original electroencephalogram signals are collected and preprocessed; s2, forming a third-order electroencephalogram tensor by using the electroencephalogram signals subjected to the improved S-transform time-frequency decomposition processing, and extracting electroencephalogram features through uncorrelated multi-linear discriminant analysis; s3, performing epilepsy disease classification by using a triple network metric learning classifier; s4, a Softmax function is used for outputting probability distribution of the test data, and a detection label is obtained; s5, a moving average filter is used for eliminating burst noisy points and isolated abnormal values, and a classification result and post-processing are output; according to the epilepsy detection method combining uncorrelated multi-linear discriminant analysis and deep metric learning provided by the invention, the workload of doctors can be effectively reduced, the efficiency and accuracy of epilepsy detection are improved, and the epilepsy detection method has a remarkable promotion effect on clinical diagnosis and treatment.

The present invention belongs to the field of target detection technology, and specifically is a target detection method applied to smart construction site security, the method comprising the following steps: constructing a YOLOV8 target detection original model; using historical images collected on the construction site to train the YOLOV8 target detection original model to obtain a trained target detection model; inputting the real-time to-be-detected images collected on the construction site into the trained target detection model, detecting target violations on the construction site in real time, and classifying and locating the detected behaviors, while triggering an early warning system and outputting violation images. The present invention also discloses a target detection system applied to smart construction site security. The target detection model provided by the present invention is applied to a smart construction site, and can monitor various behaviors on the construction site in real time, discover potential safety hazards in a timely manner, and issue alarms or take corresponding preventive measures in a timely manner. This method greatly improves the efficiency of construction site safety management.

The invention discloses a real-time topic tweet classification method and device based on small samples, and the method comprises the steps: S1, obtaining original tweet sample data, and marking the data; s2, dividing the original tweet data marked with the data into a training set and a verification set, and performing preprocessing; s3, establishing a word embedding model, inputting the preprocessed training set data into the model for training, and converting the labeled data into word vector representation; s4, establishing a classifier model, transmitting the training set data into the classifier model, and performing pre-training by taking a mask language modeling task as a target task; s5, obtaining tweet data containing each target topic specific keyword on each large social network site through a web crawler, and performing incremental training on the model; s6, inputting the verification set data into the text classifier to obtain a final topic tweet classification result; s7, repeatedly executing the incremental training process in the step S5 in real time by taking day as a unit; the problem that a traditional text classification method depends on large-scale marked data is solved.

The invention relates to the technical field of image recognition, in particular to a roadway crack recognition method and system based on image recognition. The method comprises the following steps: acquiring roadway structure data; performing regional camera deployment according to the roadway structure data to obtain roadway camera deployment data; performing roadway image acquisition according to the roadway camera deployment data to generate a roadway time sequence image set; performing roadway form change analysis on the roadway time sequence image set to generate roadway time sequence change data; performing surface crack area identification on the roadway time sequence image set through the roadway time sequence change data to generate a roadway surface crack area image; and performing surface crack trend analysis on the roadway surface crack area image to generate surface crack texture trend data. According to the method, through dynamic change detection, hidden crack identification, crack classification and analysis and crack age deduction, the identification accuracy and comprehensiveness are improved.

The invention discloses a flue-cured tobacco maturity classification method and device, and relates to the technical field of flue-cured tobacco classification. The method comprises the following steps: respectively manufacturing upper, middle and lower flue-cured tobacco maturity samples by a tobacco leaf grading expert, acquiring spectral data of the flue-cured tobacco maturity by an imaging system, and eliminating noise in the spectral data of the flue-cured tobacco maturity by using a moving average filtering method and a multivariate scatter correction method; the method comprises the following steps: integrating a spider bee intelligent optimization algorithm into a 1D-ResNet model to establish an SWO-ResNet model, using SWO to intelligently optimize batchsize and epochs parameters in the 1D-ResNet model, using the SWO-ResNet model to extract a spectral characteristic wave band of flue-cured tobacco maturity, and using an LDA classifier to establish a flue-cured tobacco maturity classification model. The flue-cured tobacco maturity classification method capable of being described is established, and classification of the flue-cured tobacco maturity is completed through spectral characteristics.

The invention relates to an intelligent insurance agent face visit log text repetition detection device and a method thereof.The device comprises an input module, a preprocessing module, a similarity calculation module and a post-processing module which are connected in sequence, and the method comprises the steps that a current log to be detected and a historical log are obtained, text pair data are constructed, the historical log comprises customer visit log records in N historical days; performing text cleaning processing on the data for the text to obtain a processed current log text and a historical log text; for the current log text and the historical log text, respectively calculating character similarity and semantic similarity between the current log text and the historical log text; and based on the character similarity and the semantic similarity, performing comparative judgment in combination with a preset corresponding threshold, and outputting to obtain a repetition detection result. Compared with the prior art, the method has the advantages that the character similarity and the semantic similarity are combined, efficient and accurate log repetition detection can be carried out from the aspects of character similarity and semantic repetition, and the log quality is effectively improved.

The present invention is applicable to the technical field of graph structure data, and provides a graph node classification method based on a new graph neural network, comprising the following steps: reading graph data; calculating feature similarity between nodes, and setting a similarity threshold, adding edges between node pairs when the similarity between nodes exceeds the preset threshold; dividing neighbors of different relationships for nodes and adding relationship attributes to the edges of nodes, obtaining a new graph structure after neighborhood reconstruction; inputting the reconstructed graph into MRN‑GNN, aggregating neighbors of the same type with the central node first, and during the aggregation process, calculating the attention coefficient of neighbors and central node features under each spatial relationship, and weightedly aggregating the attention coefficient with neighbor features; splicing the obtained multi-relation features with the central node features, and finally outputting the classification prediction of the node. The present invention can effectively capture the features of distant nodes without deepening the network, and effectively distinguish the relationship structures of different neighbors.

The application provides a video classification method and device, an electronic device and a storage medium, wherein the method comprises: encoding a first video to obtain a plurality of corresponding image frames, wherein the first video is any video in a plurality of videos to be classified; determining the complexity of the first video based on a target image frame in the plurality of image frames, wherein the complexity comprises spatial complexity and temporal complexity; and classifying the plurality of videos based on the complexity. Through the application, the technical problem that the prior art classifies by using the bottom features of a video and needs to rely on strong prior knowledge, resulting in poor video classification effect and weak robustness, is solved.

The invention provides a shale oil and gas reservoir multi-scale pore structure characterization method based on machine learning. The shale oil and gas reservoir multi-scale pore structure characterization method specifically comprises the following steps of obtaining a shale sample image and performing image segmentation; clustering the images by using a K-means clustering algorithm; classifying the segmented images into N classes through three features of gray vectors, variances and information entropies; and calculating a gray average value of the images in each category, and evaluating an image classification effect by using the gray average value. According to the technical scheme, the problem that in the prior art, comprehensive understanding and characterization of multi-type and multi-scale micro-pore structures of heterogeneous shale cannot be achieved at the same time is solved.

The invention relates to the field of brain image processing, in particular to an image analysis processing system and method based on thought control, and the system comprises an electroencephalogram signal collection system which is used for collecting and preprocessing electroencephalogram signals of a user; the thought analysis system is used for extracting features from the preprocessed electroencephalogram signals and then carrying out thought classification and instruction mapping; the image acquisition and processing system is used for acquiring image information according to the mind control instruction of the user and performing image processing; the VR display system is used for converting the processed image into a format suitable for VR equipment display and providing interactive feedback; and the control and communication system is used for converting the mind state into a control instruction and realizing data transmission and communication among the systems. The EEG signals and the fNIRS signals are acquired through the multi-mode signals, the defect of a single signal in the aspect of reflecting the brain activity is overcome, and therefore brain activity information corresponding to the thought is more accurately interpreted.

The invention is applicable to the technical field of text sentiment analysis, and provides a multi-granularity text sentiment analysis method and model based on sequential three-way decision, and the method comprises the following steps: preprocessing data by using a word segmentation tool; secondly, carrying out multi-granularity layer granulation by using an N-gram language model; then, decision classification is carried out on each layer by using a sequential three-way decision model of a jump order; finally, the ensemble learning thought is fused, voting is carried out after independent decision making of the classifier, and therefore the final decision making result is determined, the sequential three-way decision fusing the ensemble learning thought has the better classification effect on the sentiment analysis task, and the classification cost is reduced.

The application relates to an automatic text classification method based on a car community, which comprises the following steps: obtaining a data set of car community texts; obtaining word vectors and text feature vectors as input values of a double-layer clustering model; performing clustering calculation on the word vectors and the text feature vectors to generate word classification and text classification respectively, so as to form the double-layer clustering model. When a new text enters: calculating the word vectors and the text feature vectors of the new text; calculating the word classification of each word vector of the new text and the word frequency under each classification; calculating the text classification of the new text; performing dynamic analysis according to the word classification, the word frequency and the text classification of each word vector of the new text; when the number of words and the word frequency of the new text generated outside the existing word classification reach a threshold value, the double-layer clustering model is updated. Therefore, the application can realize automatic classification of car community texts in the whole process, improve the classification accuracy and efficiency, and form a closed-loop management.

The invention discloses a multi-extended target joint tracking and classification method based on star convex RHM and LMB filters, and belongs to the field of radar target tracking. According to the method, an LMB parameter set is initialized by using target prior information, and then a sensor measurement set is divided through a mean shift algorithm; then, state prediction is achieved by combining survival target parameter updating and new target parameter sampling, and then measurement updating is completed through LMB-to-GLMB, GLMB updating and GLMB-to-LMB; and then trimming and fusing the LMB parameter set, estimating the number of targets, extracting state information, and circularly executing until observation is finished. According to the method, a star convex RHM modeling expansion state is adopted to reduce dimensions, the low detection probability / high clutter scene performance is improved based on an LMB framework, the tracking classification effect and the real-time performance are both considered, and the engineering application value is high.

The invention discloses a noise robust voice endpoint detection method and system, and the method comprises the steps: 1, constructing a voice endpoint detection model which comprises a CNN, a BILSTM, an attention layer, a DNN layer, and a Speech and Nospeech classification layer which are cascaded; 2, a Nospeeck noise classification layer is introduced to serve as a network training model after BILSTM of the voice endpoint detection model, the introduced Nospeeck noise classification layer is a branch connected with the attention layer, the DNN layer, the Speech classification layer and the Nospeeck classification layer in parallel, iterative training is conducted on the network training model till the training stopping condition is met, and the trained voice endpoint detection model is obtained; and step 3, obtaining a frame-level dichotomy result through the trained voice endpoint detection model, and obtaining an endpoint detection result by adopting smooth merging on the basis of the dichotomy result. According to the invention, the voice detection effect is improved.

The invention relates to a lightweight wheat disease image recognition method for an intelligent terminal under a complex background. Compared with the prior art, the defect that the lightweight design of wheat disease image recognition under the complex background is difficult to realize is overcome. The method comprises the following steps: acquiring a wheat disease image data set; constructing a lightweight wheat disease identification model; training a lightweight wheat disease recognition model; obtaining a to-be-identified wheat disease image; and identifying the wheat disease image. According to the method, high-precision disease degree identification can be realized while flexible deployment is performed on mobile equipment, the model structure is compressed, the model is optimized by using Leaky ReLU and AdamW optimizers, the response speed of the model is improved, the feature extraction result of the high-complexity model is learned by adopting a knowledge distillation mode, the accuracy of the model is improved, and the accuracy of the model is improved. Finally, the wheat disease degree can be efficiently and accurately identified on the mobile equipment.

The invention relates to the technical field of medical information, and discloses a lung transplantation patient clinical bacterial infection factor prediction model and a construction method, and the method comprises the following steps: S1, selecting a hospital needing research, and determining a research object according to an inclusion standard and an exclusion standard; s2, selecting a proper number of variables as candidate predictive factors; s3, clinical bacterial infection patients are defined as typical symptoms, signs and sputum characters; laboratory inspection: inflammation indexes are increased, and pathogens can be determined by bacteria detected by sputum culture; s4, if missing values exist in part of variables, simple interpolation processing is adopted, mean interpolation is adopted for continuous variables, and mode interpolation is adopted for classified variables; s5, sample size analysis; and S6, model establishment: firstly carrying out description analysis, and then carrying out Lasso regression and model prediction. A method for respectively establishing prediction models by applying various machine learning and deep learning is adopted, and multi-aspect comparative analysis is carried out on lung transplantation patient clinical bacterial infection influence factor prediction models.

The invention relates to the technical field of data classification, in particular to an intelligent data sharing method for construction of an intelligent sharing system. According to the method, each condition attribute is used for splitting total historical service data to obtain different sub-data sets, and the index characteristics of the index attributes in the total historical service data and the sub-data sets of each condition attribute are analyzed according to the difference of confusion degrees of the index characteristics of the index attributes in the sub-data sets of the condition attributes. Obtaining an initial priority index of each condition attribute according to the change trend of the historical business data of each condition attribute and the change trend of the historical business data of the attention attribute, and adjusting the initial priority index in combination with the influence degree of each condition attribute on the attention attribute and the correlation degree of each condition attribute and other condition attributes so as to construct a decision tree; and classifying and sharing the total historical business data. According to the method, the decision tree is constructed by combining the information gain and the importance degree brought by the condition attributes during data set segmentation and the correlation degree of the different condition attributes, and the classification effect of the decision tree based on the attention attributes is improved.