427 results about "Data expansion" patented technology

The invention relates to the technical field of road maintenance intelligence, and discloses a road intelligent maintenance decision-making system based on multi-dimensional evaluation and deep reinforcement learning, and the system comprises a geographic space data preprocessing module, a fuzzy TOPSI S evaluation module, a context awareness DQN strategy module, a credibility driving recommendation module, and a security strategy library module. The method comprises the following steps: generating a road health degree and a clustering label through multi-source data geographical weighted preprocessing and fuzzy TOPSI S dynamic weight evaluation; a reinforcement learning reward function is configured based on label differentiation, and a strategy space is explored in combination with noise; the confidence is verified through multiple models, a historical security policy is matched, and model optimization is driven through priority sampling injection samples. According to the method, the evaluation precision is improved through entropy weight-clustering dynamic weight, and flexible decision is realized in combination with reinforcement learning; and three-level security verification and closed-loop optimization are constructed to ensure that the risk is controllable, historical experience migration and multi-source data expansion are supported, and the cross-scene adaptive capacity is enhanced.

The invention discloses a text travel sentiment analysis response method and system based on multi-modal fusion and incremental learning, and the method comprises the steps: obtaining a text, an image, an audio or a video input by a user, carrying out the query in a text and multi-modal knowledge base through employing a multi-modal retrieval technology, and optimizing a retrieval result through combining sentiment analysis, thereby achieving the purpose of improving the user experience. And finally generating a personalized tourism information response. The system integrates a large language model, text and multi-modal knowledge base construction, and a data vectorization processing and sentiment analysis technology, supports multi-modal input and output, can dynamically adjust retrieval and answer contents, and realizes personalized recommendation according to user feedback. The system also has the capabilities of multi-modal data expansion, dynamic knowledge base updating and voice interaction, can significantly improve the response speed and accuracy of tourism information service, is widely applicable to intelligent and personalized tourism information service scenes, and has relatively high innovativeness and practical value.

The invention belongs to the technical field of high-performance composite materials, and discloses a carbon fiber reinforced thermoplastic composite material performance database, a construction method and application thereof, and the method comprises the following steps: S1, database structure construction; s2, experimental sample collection and data standardization; s3, feature engineering and variable reduction; s4, training a machine learning model; s5, constructing and verifying an adaptive model; and S6, data expansion and feedback optimization. According to the method, material performance prediction and formula parameter reverse design under target performance are realized through systematic acquisition and normalization processing of three types of data of material components, preparation process and performance characterization and building of a nonlinear mapping model among a material structure, a process and performance through a machine learning method. The database can be used for intelligently recommending a high-performance composite material combination scheme, is suitable for rapid screening and customized development of various thermoplastic composite materials, effectively reduces the research and development cost and development cycle, and improves the material design efficiency.

The invention relates to the technical field of image processing, in particular to a tobacco leaf disease spot segmentation method and system based on multi-scale residual cavity convolution, and the method comprises the steps: collecting a plurality of tobacco leaf disease images as original images, and carrying out the data expansion of each collected original image through an image enhancement method, establishing a training set, a verification set and a test set based on the original image and the extended data; a semantic segmentation model is constructed, the semantic segmentation model is formed by stacking two sub-networks, and the two sub-networks are connected through an ROIE + module; in combination with the training set, training the semantic segmentation model by using the constructed loss function; verifying the trained semantic segmentation model by adopting a verification set, and selecting an optimal semantic segmentation model; the test set is adopted to test the optimal semantic segmentation model, and the performance of the optimal semantic segmentation model is evaluated; and inputting a to-be-segmented tobacco leaf disease image into the trained semantic segmentation model to obtain a tobacco leaf disease spot segmentation map. According to the method, the tobacco leaf scab area can be accurately and efficiently segmented.

The invention discloses an extremely-short-term ship motion attitude prediction method and system based on data enhancement, and relates to the field of ship motion attitude prediction. The method is used for solving the problem that data quality and model generalization performance are affected due to insufficient completeness of input data of a ship motion attitude extremely-short-term prediction model. The method comprises the following steps: performing data expansion on original observation ship attitude data by adopting a Slim-based generative adversarial filling network to obtain an expanded data set; performing data splicing and feature zooming on the original observation ship attitude data and the extended data set; performing time delay correlation analysis on the data after feature scaling to determine optimal sample data, then dividing the optimal sample data into a training set and a test set according to a proportion, and inputting the training set and the test set into a prediction model for training; optimizing parameters of the prediction model according to the prediction value of the training set; and outputting a predicted value by adopting the parameter-optimized prediction model, and carrying out reverse normalization processing to obtain a final prediction result. The method is suitable for extremely short-term ship motion attitude prediction.

The invention provides a power device defect detection method and system, and relates to the technical field of semiconductors, and the method comprises the steps: obtaining a power device defect image set with defect tags; expanding a defect image set of the power device through geometric modeling in combination with material attributes of the power device; constructing a decoupling detection model used for identifying defect types and positioning defect positions; in combination with a bimodal positioning loss function, training a decoupling detection model by using the expanded power device defect image set; collecting a real-time image of the power device; and inputting the real-time image of the power device into the trained decoupling detection model, and outputting the defect category and the defect position of the real-time image of the power device. Through accurate model training and data expansion, the accuracy and robustness of detection are improved, the problem of missing detection can be effectively solved, and the production quality and efficiency of power devices are improved.

The invention specifically discloses a method for evaluating the health state of a cigarette comprehensive test board, and the method comprises the steps: obtaining the fault type of the cigarette comprehensive test board, and carrying out the analysis of the fault type, so as to obtain the self degradation of equipment and the influence factors of the fault; carrying out fault influence factor analysis on influence factors of equipment degradation and faults, and carrying out numeralization on the influence factors; carrying out fault data acquisition and processing on the cigarette comprehensive test bench, structuring data of equipment faults, and further carrying out fault data expansion through a generative model GAN network; and a DNN deep neural network is constructed, training is performed by using the expanded fault data, and then fault probability evaluation is performed on the cigarette comprehensive test bench through the DNN deep neural network. According to the invention, the convenience and accuracy of health assessment of the cigarette comprehensive test board can be improved.

The invention relates to the technical field of remote sensing image processing, in particular to an intelligent landslide identification method based on deep learning of a multi-source remote sensing image, which comprises the following steps of: firstly, acquiring high-resolution optical and radar data of the multi-source remote sensing image by adopting a spatio-temporal data expansion method; the problem that an existing landslide identification method is insufficient in data type and data quantity is effectively solved, more comprehensive and accurate landslide data are provided, the generalization ability of the model is enhanced, then a deep learning model with semi-supervised learning and adaptive convolution multi-scale feature fusion is adopted, the size of a convolution kernel can be dynamically adjusted, and the robustness of the landslide identification method is improved. Landslide features of different scales are accurately captured, understanding of landslide morphology and structure is enhanced, robustness of the model in a complex terrain is improved, finally, a landslide boundary and morphology are corrected through a positioning correction method, inaccurate or discontinuous boundaries caused by prediction errors are eliminated, and stability of the model in a complex environment is further improved.

The invention discloses a teaching implementation method based on combination of motion perception and AI driving, and relates to the technical field of education and teaching. The implementation of the teaching method comprises the steps of multi-modal data expansion and acquisition, cross-dimensional dynamic evaluation and cognitive diagnosis, emotion adaptive feedback and intervention, cross-scene learning map construction and capability migration, and whole-process data closed loop and systematic evolution. And the multi-modal data expansion acquisition comprises the steps of dynamic data acquisition, physiological data acquisition, environmental behavior data acquisition and data preprocessing. The method has the advantages that a multi-modal data acquisition scheme of a common mobile phone camera and a built-in sensor is adopted; the system replaces the traditional dance teaching which depends on special three-dimensional modeling equipment and sensor gloves for boxing teaching, reduces the hardware cost, can realize full scene coverage of families, classrooms, outdoors and the like without professional site deployment, and solves the core problems of high hardware cost and limited scenes in the prior art.

The invention relates to a multi-spectral wavefront aberration restoration method and system based on double roof prisms, and the method comprises the following steps: taking a sub-pupil image as the input of a classification model, obtaining classification feature data, and carrying out the data expansion of the classification feature data, the sub pupil image is obtained by inputting incident wavefront aberration into a simulation light path of the double-ridge prism wavefront sensor, the double-ridge prism wavefront sensor comprises two groups of double-ridge prisms, each group of double-ridge prisms comprises two ridge prisms which are vertically and oppositely arranged, the characteristics of the ridge prisms are equal, and the characteristics of the double-ridge prisms are equal. The two groups of double-roof prisms are made of a material combination for eliminating chromatic aberration; splicing the sub pupil images and the classification feature data after data expansion to serve as input of a regression model, and obtaining a Zernike coefficient matrix corresponding to incident wavefront aberration; and according to the Zernike coefficient matrix corresponding to the incident wavefront aberration and the Zernike polynomial, obtaining a recovered wavefront aberration. Compared with the prior art, the method has the advantage that the wavefront aberration recovery accuracy is improved.

The invention provides a bearing small sample data expansion method and system based on a variational auto-encoder, and belongs to the field of deep learning and data enhancement. The problems that a traditional generation model has limitation in bearing small sample data, feature fuzziness and distortion are prone to occurring, and the data set quality is poor are solved. According to the method, a deep VAE framework is constructed, and a dimension reduction module, a data expansion module and a dimension raising module are used in a potential space; dimensionality reduction is performed on high-dimensional data by adopting a UMAP algorithm, so that the topological structure of the data is effectively reserved, and the extraction efficiency and quality of data features are improved; a Gaussian mixture model combining regularization and particle swarm optimization optimization is used for fitting distribution of scattered small sample data, new data with fusion features are expanded through sampling, and data diversity is increased; a radial basis function is used for nonlinear data dimension raising, new data can be ensured to be accurately mapped back to a high-dimensional space, meanwhile, the relation between features is reserved, and defect data with fusion features is reconstructed through a decoder.

The embodiment of the invention relates to the technical field of petroleum and natural gas development, and provides an oil pipe make-up quality evaluation model construction and evaluation method and an intelligent make-up torquemeter, and the method comprises the steps: obtaining an original oil pipe make-up torque curve data set; carrying out expansion processing on the data set by utilizing a pre-trained data expansion model to obtain an expanded oil pipe make-up torque curve data set; constructing a make-up torque curve classification model based on a convolutional neural network; and training the make-up torque curve classification model by using the expanded oil pipe make-up torque curve data set to obtain a trained make-up torque curve classification model. According to the embodiment of the invention, the accuracy and efficiency of oil pipe screwing-on quality evaluation can be improved.

The invention provides a flood session division method and device based on deep learning, and belongs to the technical field of flood data analysis, and the method comprises the steps: dividing historical flood data into flood sequence data and non-flood sequence data; and performing data expansion on the flood sequence data to obtain target flood sequence data. And performing data splicing on the target flood sequence data and the non-flood sequence data according to a time sequence to obtain a flow-rainfall time sequence sample. And based on the flow-rainfall time sequence sample, training the multi-layer long and short-term memory network by using the target loss function to obtain a flood prediction model. And performing flood session division on the flow-rainfall time sequence data to be divided by using the flood prediction model to obtain a flood session division result. According to the flood session division method and device based on deep learning provided by the invention, the accuracy of flood session identification and division in a complex flood scene can be improved.

The invention discloses a multi-unmanned aerial vehicle confrontation task execution method and device based on model reinforcement learning, and a medium, and the method comprises the following steps: inputting the current local observation into a reinforcement learning control network for each airframe in an own unmanned aerial vehicle group, and outputting meta-actions (including movement and attack), performing confrontation interaction with the enemy unmanned aerial vehicle group to obtain finite confrontation experience data; the collected historical interaction experience is used for supervised learning of a world model, so that the world model can approach probability distribution of an interaction rule of a real environment; data expansion is carried out on real interaction data by using a world model, an action network and an evaluation network are trained on expanded empirical data, and strategy optimization and parameter updating of a multi-unmanned aerial vehicle system are realized; repeating the process, and sequentially carrying out experience collection, world model training and enhancement strategy updating; and finally, the trained action network is reserved and is used for outputting a specific action strategy in an actual confrontation task.

The invention discloses a vehicle test boundary scene generation method and system based on a pre-boundary scene. The method comprises the following steps: acquiring a first feature data set; obtaining and discriminating a real boundary scene and a real pre-boundary scene based on the first feature data set and a preset risk discrimination criterion; constructing a second feature data set based on the real pre-boundary scene data and the discriminated real boundary scene data; based on the real pre-boundary scene data, utilizing the generation model to obtain generated pre-boundary scene data; training a prediction model based on the real pre-boundary scene data and the discriminated real boundary scene data; and obtaining test prediction boundary scene data based on the generated pre-boundary scene data and the prediction model. According to the method provided by the invention, the test prediction boundary scene data is acquired based on the real pre-boundary scene data, and data expansion is performed on the real boundary scene data by using the generation model and the prediction model, so that the technical problem of sparse generated test scene data caused by limited scene generation data space in the prior art is solved.

The invention discloses a power grid state evaluation method and system applying multi-source data fusion, and relates to the technical field of data analysis, and the method comprises the steps: carrying out the expansion of a to-be-evaluated shaft segment through combining the dynamic characteristics of a power grid and the correlation degree, and enabling the state of the power grid under the to-be-evaluated shaft segment to be always related to the data at the previous time, a reasonable forward time node is determined by combining the dynamic characteristics of the power grid and the correlation degree, and data expansion is performed on a to-be-evaluated shaft segment, so that the comprehensiveness and rationality of data are ensured, and a reliable basis is provided for subsequent multi-source information feature extraction. Uncertain information is fused through a combination rule, and the problem that the boundary state is misjudged by a traditional threshold method is solved. The evidence theory can process conflicts among multi-source information, and a single data source is prevented from leading an evaluation result through a discount coefficient and a dynamic evidence weight. The accuracy and adaptability of power grid state evaluation are improved, and the reasonability of power grid state monitoring and the stability and safety of the power grid operation state are ensured.

The invention discloses a drug combination patent intelligent identification method and device based on a large language model, and the method comprises the steps: obtaining a standard data set built based on drug combination labeling information of patent data, carrying out the data expansion of the standard data set, and generating an initial training set; a parameter efficient fine tuning technology is adopted, the initial training set is utilized to train the base large language model, and an initial recognition model is generated; screening and identifying difficult samples in the unlabeled patent data by using the initial identification model, and adding the difficult samples into the initial training set to form an enhanced training set; performing iterative training on the initial recognition model by using the enhanced training set and adopting a multi-task learning framework in combination with a comparative learning mechanism to obtain a drug combination patent recognition model; the to-be-recognized patent data is input to the drug combination patent recognition model, and a drug combination information recognition result is obtained.The accuracy, data efficiency and model interpretability of drug combination patent recognition can be remarkably improved, and the computing resource requirement and the manual review cost are reduced.

The invention discloses a new coronavirus variant prediction method and system with higher immune escape ability, a conditional generative adversarial network model and a variational self-encoding model are used for data expansion, the similarity between an expanded data set and a constructed original data set is compared, the expanded data set with the highest quality is incorporated into the original data set, and the new coronavirus variant prediction method and system with the higher immune escape ability are obtained. Obtaining an expanded data set; inputting the RBD amino acid sequence in the expanded data set and label data for determining whether the RBD amino acid sequence is combined with a human receptor ACE2 and eight antibodies or not into a constructed multi-task classification deep network model for training and verification to obtain an immune escape score of a virus variant corresponding to the RBD amino acid sequence; a new RBD sequence is generated by using the fitness in an immune escape score quantification genetic algorithm framework, a sequence generation strategy is optimized based on an adversarial generation method to obtain an RBD sequence with higher immune escape ability and an immune escape score thereof, and prediction of a new coronavirus variant is realized.

The invention relates to the technical field of traffic sample expansion, in particular to an industrial internet traffic sample expansion method and system based on VAE-GAN. The method comprises the following steps: firstly, obtaining tensor data according to original training data, extracting a local feature vector and a time sequence feature vector according to a VAE encoder, generating a context feature vector, generating potential variable data according to the context feature vector, and putting the potential variable data into a VAE decoder for data expansion to obtain reconstructed data; the reconstruction data is input into a discriminator of the GAN model, the reconstruction data can be evaluated, whether the reconstruction data is close to real flow data or not can be judged, feedback information is generated, the VAE decoder adjusts parameters of the VAE decoder according to the feedback information, and by continuously optimizing the parameters of the VAE decoder, the real flow data of the VAE decoder can be obtained. According to the technical scheme of the invention, the traffic sample which is more vivid and closer to real data can be generated, so that the real data of the industrial internet can be reflected more accurately and objectively.

The invention relates to the field of fault diagnosis, and discloses a bearing fault diagnosis method and device based on a dynamic domain adaptation network, and a medium, and the method comprises the steps: obtaining the vibration signal data of bearings of different models under different working conditions, and taking the data as the data of a source domain and a target domain; a generative adversarial network based on time-frequency feature structure similarity is adopted to carry out data expansion, and a 1-D CNN network containing a CBAM module and a dynamic balance domain adaptation module is utilized to carry out feature extraction and classification. Source domain data are input into the network for feature extraction, and meanwhile, the cross entropy loss and the cross-domain feature are combined to adapt to the loss optimization model. And after model training is completed, target domain test data is input into the trained fault diagnosis network, a verification result shows that the model has high diagnosis accuracy and generalization ability on different data sets, and the reliability of rolling bearing fault diagnosis is remarkably improved.

The invention discloses a compressed air foam fire extinguishing assessment method and system under the influence of multiple environmental factors. The method comprises the following steps: collecting environmental parameters of an extra-high voltage station in a fire scene to form a fire environment parameter library; constructing a three-dimensional model of the extra-high voltage station, performing environmental parameter analogue simulation on the extra-high voltage station, and selecting a simulated nested physical model to simulate a fire scene to form a simulated extra-high voltage station; respectively simulating the influence of a single environment parameter and multiple environment parameters on the fire extinguishing performance index in the simulation extra-high voltage station, obtaining simulation data associated with the fire extinguishing performance index, and recording extra-high voltage station fire test data; constructing a data set by using the simulation data and the extra-high voltage station fire test data, performing data expansion to obtain a virtual sample, and using the virtual sample and the data set as a database; a fire extinguishing performance prediction model is constructed and trained, environmental parameters of the extra-high voltage station are collected in real time, and fire extinguishing performance indexes are predicted; the method has the advantages of being accurate in evaluation result, short in calculation time and sufficient in data sample.

The invention relates to the field of acoustic signal processing, and discloses an industrial equipment abnormal sound data expansion method based on an improved generative adversarial network, and the method comprises the steps: constructing an improved generative adversarial network I-WaveGAN model; the I-WaveGAN model is trained, and a trained model is obtained; and generating the abnormal sound audio data of the industrial equipment by using the trained model. The method has the beneficial effects that high-fidelity modeling of audio features of industrial equipment is realized, the quality, diversity and discrimination of data expansion are remarkably improved, the modeling problem caused by abnormal sample deficiency is effectively relieved, and an efficient, stable and generalizable training data enhancement method is provided for an abnormal sound detection system.

The invention discloses an expressway bridge construction state real-time monitoring method and system, and relates to the technical field of bridge construction monitoring. An expressway bridge construction state real-time monitoring system comprises a data acquisition and processing module, a data expansion and analysis module, a multi-point data fusion module, a state comprehensive evaluation module and a construction scheme adjustment module. According to the method, data interpolation and expansion are carried out on the preprocessed vibration monitoring data through the Kriging interpolation method, the spatial autocorrelation between the vibration monitoring data can be fully utilized, and the monitoring data are expanded, so that the problem of insufficient coverage of monitoring points can be effectively solved, and the integrity and precision of the monitoring data are improved; the change trend and the distribution rule of vibration in space can be reflected more accurately, a more comprehensive and reliable basis can be provided for evaluation of the highway bridge construction state, and the state evaluation accuracy of the highway bridge construction state real-time monitoring method and system is improved.

The invention discloses a Mongolian handwritten text recognition method based on improved OfficientNet, a generative adversarial network is utilized to generate a Mongolian handwritten text image with a human handwriting style as a data set, model optimization is performed according to the characteristics that Mongolian handwritten text characters are easy to distort, deform and the like, targeted preprocessing and data expansion are performed on data, and the recognition accuracy of the Mongolian handwritten text image is improved. The diversity of the data set is improved; a backbone network of a feature extraction module of the recognition model is an improved OfficientNet network, and an SE module in an MBConv structure of the OfficientNet network is replaced with an ECA module; and training the recognition model by using the data set, and performing Mongolian handwriting recognition by using the trained recognition model. According to the recognition model, the recognition precision and efficiency can be optimized under limited resources.

The invention provides a double-current comparison and DHI combined equipment fault detection method and system, and belongs to the technical field of power equipment state monitoring and fault diagnosis. The method comprises a time sequence perception adversarial enhancement module, a generative adversarial network (GAN) data expansion module, a double-flow contrast attention network (D-CAN) feature extraction module and a dynamic health index (DHI) calculation module. The core lies in that a fault sample with time sequence correlation is supplemented through a physically constrained GAN, robust features are extracted by using a ResNet1D and Transform fused double-flow network, and a health state is quantified in combination with unsupervised clustering and mahalanobis distance. Through simulation and experimental verification, zero-delay detection of the early fault of the transformer can be realized, the output health index and the 3D visualization result can provide an accurate basis for operation and maintenance of the transformer, and the diagnosis accuracy, the data utilization rate and the dynamic adaptability of state evaluation are remarkably improved.

The invention relates to the technical field of data expansion, in particular to a malicious prompt data set expansion method based on Mongolian. According to the method, high-frequency roots and affixes in a Mongolian basic malicious prompt corpus and a universal corpus are extracted, morphological analysis is combined, targeted malicious prompt samples can be constructed, the diversity and authenticity of a Mongolian safety evaluation data set are enhanced, candidate malicious prompt samples are optimized by adopting a genetic algorithm, and the safety evaluation accuracy of the Mongolian safety evaluation data set is improved. According to the method, the expansion sample is enabled to better conform to syntax and semantic rules of the Mongolian, effectiveness of the attack sample in the Mongolian scene is ensured, the expansion data set is enhanced by using the antagonism generation strategy, robustness and antagonism of the data set are improved, attack behaviors possibly occurring in a real scene can be simulated, and the robustness and the antagonism of the data set are improved. Therefore, an effective sample generation tool is provided for safety evaluation in a Mongolian scene, and the anti-attack capability of the Mongolian model is improved.

This application discloses a design method and grouting material for deep coal-bearing strata, comprising the following steps: S1 Data acquisition based on material pre-tests, correlation analysis, and orthogonal experiments; S2 Data expansion based on orthogonal experimental data to establish a dataset; S3 Creating a prediction model based on the RBF algorithm; S4 Optimizing the prediction model based on the PSO-RBF algorithm; S5 Proportion design based on the RBF-PSO-EWM algorithm; S6 Obtaining the proportion of grouting material for deep coal-bearing strata. This application proposes a PSO-RBF-EWM multi-objective optimization grouting material proportion design method and applies this method to design the comprehensive optimal proportion of grouting material for deep coal-bearing strata. The grouting material with the obtained optimal proportion has excellent characteristics such as low water-cement ratio, green environmental protection, and economic efficiency, and is suitable for the reinforcement characteristics of deep coal-bearing strata.

The invention discloses a Kriging agent model-based engineering structure reliability analysis method, which comprises the following steps of: sampling from random variables and probability distribution of an object to be analyzed, and respectively obtaining a candidate sample set and an initial sample set; obtaining an initial experimental design set based on the initial sample set and the corresponding real response value, and constructing an initial Kriging agent model about the to-be-analyzed object; obtaining an optimal sample from the candidate sample set through a random weight learning function, and expanding the optimal sample and a corresponding real response value to an initial experimental design set; before the initial experimental design set enters the next round of data expansion, deleting an optimal sample in the current round and sample data in a preset redundancy distance range of the optimal sample from the candidate sample set; updating and iterating the initial Kriging agent model about the to-be-analyzed object through the experimental design set; and calculating the reliability index of the to-be-analyzed object according to the failure probability calculated by the current Kriging agent model.

The invention provides an intelligent detection method for circumferential weld defects of an oil and gas transportation pipeline based on Transform, and aims to improve the precision and efficiency of weld defect detection. In order to solve the problem that small targets are difficult to recognize in petroleum pipeline weld defect detection, targeted improvement is carried out on the basis of a traditional DETR model. By designing a small target feature enhancement encoder, an attention mechanism is dynamically adjusted to focus a small defect area, and the perception and recognition capability of small-size defects is improved; and meanwhile, a multi-scale feature fusion module is introduced, and shallow fine-grained features and deep semantic features are integrated, so that the detection precision of micro-size defects is remarkably improved. In order to further relieve insufficient training data, a generative sample expansion strategy is supplemented, and the diversity of defect samples is enriched. Compared with the prior art, the method has the advantages that the detection performance is synergistically improved from the two aspects of detection structure optimization and data expansion, the accuracy and processing efficiency of weld defect detection are improved, good robustness and engineering adaptability are achieved, and the method can be widely applied to large-scale inspection tasks of oil and gas pipelines and has wide application prospects. The method has important application value and popularization significance for guaranteeing safe operation of the pipeline and reducing manual detection cost.

The invention provides a data compression method, apparatus and device, and a storage medium. The method comprises the steps of obtaining an original data stream transmitted through a serial port; determining the data type of the original data flow according to the original data flow; determining a corresponding target compression algorithm according to the data type; compressing the original data stream according to the target compression algorithm to obtain compressed data; the defects of low efficiency and even data expansion during data abrupt change in a fixed compression mode are effectively overcome, and the data transmission efficiency and the bandwidth utilization rate of serial port communication in a resource limited environment are remarkably improved.