156 results about "Large sample" patented technology

The invention relates to a circuit board on-line defect detection method and system, and the method comprises the steps: carrying out the synchronous collection and structural integration of multi-source technological parameters such as production line environment temperature and humidity, equipment operation states, material batches and the like, and defect detection images, and achieving the construction of large-sample original data in a production process; through standardization and de-noising preprocessing, multi-modal features are fused, and a distribution mapping model of process and defect features is established by using algorithms such as mutual information analysis and principal component analysis. Based on a feature distribution model and real-time data, a dynamic anomaly detection threshold is adaptively generated, Bayesian inference and evidence reasoning are combined, multi-level confidence levels and risk response suggestions are output, and self-learning evolution of the model and the threshold is realized through a closed-loop feedback mechanism. According to the scheme, the accuracy of anomaly detection, the response timeliness and the risk disposal intelligent level are improved, the method adapts to complex working conditions and batch changes, and the closed-loop optimization and safety control capability of the production process is remarkably enhanced.

The invention relates to the technical field of medical informatics and artificial intelligence technology cross application, in particular to a dynamic intelligent prediction system for HIV infected person immune reconstruction insufficiency. Comprising a data acquisition and management module, a data preprocessing and feature engineering module, a Bayesian joint modeling core module, a dynamic prediction calculation module, a multi-dimensional verification and evaluation module and a clinical application service module. According to the method, on the basis of large-sample multi-center longitudinal follow-up visit data, a Bayesian shared parameter dynamic joint model is adopted, the dependency relationship between dynamic trajectories of immune indexes such as cell counting and the like and IIR occurrence time is captured, real-time risk prediction is achieved through a sequential Bayesian updating mechanism, and through multi-dimensional verification, the risk prediction accuracy is improved. The method has the advantages that the performance is better than that of an expert-driven model, expert experience pre-judgment and 19 mainstream machine learning algorithms, an individualized conclusion with a confidence interval can be output, clinical decision is assisted, prediction is promoted to be clinical from scientific research, and the method has important application value and popularization prospect.

The invention discloses a method for acquiring grain orientation characteristics based on optical and height characteristics of an alloy etching structure. The method comprises the following steps: (1) acquiring grain orientation distribution and grain size of an alloy surface structure; (2) obtaining an etching structure on the surface of the alloy; (3) acquiring optical characteristics of the etched structure on the surface of the alloy; (4) obtaining height characteristics of an alloy surface etching structure; and (5) constructing a correlation model of structure grain orientation characteristics based on alloy surface optics and height characteristics, and realizing prediction. Aiming at the test of the alloy grain orientation, the optical and height characteristics of the alloy surface etching structure are synthesized, the measurement of the centimeter-level large sample size or full-area grain orientation of the alloy is easy to realize, and compared with the existing measurement aspect, the method is more efficient and convenient; and more comprehensive and reliable tissue information can be provided for alloy optimization and new material development, and the method has important significance for high-throughput characterization and intelligent manufacturing.

The small sample anti-migration prediction method suitable for the metallurgical process end point component under the zero expansion characteristic comprises the steps that smelting report data of a large sample steel grade and a small sample steel grade in the metallurgical process are collected to serve as source domain data and target domain data; adopting median to fill and restore abnormal values existing in the report data; dividing the source domain data and the target domain data into continuous feature variables and classification feature variables; inputting the continuous feature data and the classification feature data of the source domain data into a TabNet coding and decoding self-supervising network for self-supervising training, and inputting the target domain data into the trained self-supervising network for feature extraction and reconstruction; introducing an adversarial network, and gradually aligning feature distribution of a source domain and a target domain; and inputting the high-dimensional reconstruction features processed by the TabNet coding and decoding self-supervised network in the source domain into the deep table network model for training, and inputting the small sample steel grade features aligned by the adversarial network distribution into the trained deep table network model for transfer learning.

The invention provides a crack pouring glue compression rebound recovery rate test system and method, and relates to the technical field of rebound recovery rate test.The crack pouring glue compression rebound recovery rate test system comprises a tester, a high-precision injection ball and a large sample containing vessel; the tester comprises a machine body, a touch control type control display screen is mounted on an inclined panel of the machine body, a group of function buttons are mounted on the inclined panel of the machine body, the machine body is provided with a stepped circular groove, and the large sample containing vessel is matched with an upper side circular groove of the stepped circular groove; the machine body is connected with a U-shaped frame, and the U-shaped frame is connected with a first laser distance measuring sensor. Aiming at the defects in the prior art, the invention develops the system and the method for testing the compression resilience recovery rate of the crack pouring glue, and the high-precision penetration ball is adopted, so that the contact area with a sample is increased, and the detection error is reduced. By optimizing a hardware structure (replacing a high-precision sensing assembly) and control logic, automation and high precision of a test process are realized.

The invention relates to the technical field of bridge structure health monitoring and damage diagnosis, and discloses a bridge damage identification method based on a physical information neural network. According to the bridge damage identification method based on the physical information neural network, efficient and accurate identification of bridge damage is realized by optimizing a data acquisition scheme, designing a bridge exclusive physical constraint, constructing a biterm loss function and standardizing a network training process; the invention provides a damage identification method fused with a physical information neural network, aiming at solving the problems of strong manual dependence, large sample labeling demand, weak anti-interference capability and low identification precision in the existing bridge damage identification technology. The bridge structural mechanics principle and the neural network technology are deeply combined, so that the accuracy, robustness and engineering practicability of damage identification are improved while the dependence of the labeled sample is reduced, and the actual demand of real-time health monitoring in the bridge operation stage is met.

The invention discloses a large-scale composite member deformation gap distributed measurement and full-field reconstruction method, and the method comprises the following steps: setting a plurality of sets of technological parameters, and carrying out the mold pasting of a composite member based on the technological parameters, so as to enable the surface of the composite member to form a skin; determining an optimal arrangement scheme of the sensors, arranging the sensors according to the scheme, and obtaining a gap value matrix; obtaining a plurality of groups of gap value matrixes under different process parameters, and constructing a finite element model small sample training set; training the finite element model by using the finite element model small sample training set to obtain a full-field gap prediction finite element model; and generating a convolutional neural network large sample training set through the full-field gap prediction finite element model, and training the convolutional neural network by using the large sample training set to obtain a full-field gap prediction convolutional neural network model. The method has high systematicness and engineering implementability, and is more suitable for film pasting quality evaluation of a complex shell surface structure compared with a traditional method.

The invention relates to a large sample exploratory experiment conclusion macro-micro interaction verification method and system, the method is used for constructing a'macro-interaction-micro 'three-layer linkage analysis framework, and the framework comprises a macro conclusion layer, an interaction analysis layer and a micro instance layer; the interaction analysis layer is used for connecting the macroscopic conclusion layer and the microscopic instance layer; the method comprises the four steps of automatic discovery and presentation of a macroscopic conclusion, interactive deep exploration and verification based on the macroscopic conclusion, macro-micro bidirectional linkage and sample screening, and concrete redisk and causal verification of micro instances. Through the method, interactive verification of a macroscopic conclusion and a microcosmic instance is realized, and the interpretability and credibility of the conclusion are remarkably improved. According to the method, a macroscopic and microcosmic barrier is broken through, a cognitive closed loop from'knowing the course 'to'knowing the course' is constructed, and a more robust basis is provided for decision making.

Systems, methods, and tools associated with Enhanced Object Mask (SEI) messages are disclosed herein. In an example, a video device may obtain one or more syntax elements associated with video data. The one or more syntax elements may specify a parameter (e.g., a location parameter) associated with an object mask associated with the picture. The parameter may correlate with at least a sample range of object mask samples around the object mask ID value. In an example, the parameters associated with the object mask may include a maximum sample value and a minimum sample value of the object mask that define a range of the object mask samples around the object mask ID value. In an example, the parameter associated with the object mask may be a bounding box parameter associated with the object mask.

This invention discloses an efficient RNA data preprocessing method to address the problems of low processing efficiency and low information density in high-throughput sequencing data. Its core steps include: (1) introducing a parallel processing scheme for high-throughput sequence data, rapidly mapping RNA-seq data to a reference genome to generate a BAM file; (2) extracting base sequences and expression levels and storing them as compact PKL format files; (3) extracting all exon position information by parsing the genome annotation file; (4) combining multi-sample expression level data to screen for highly variable exons and constructing a high-information-density feature list based on the sample set; and (5) accurately extracting target sequences from the preprocessed file based on this list. Compared to traditional methods, this innovative approach achieves triple optimization: full-process parallel processing for accelerated computation, high-compression data storage, and adaptive feature selection. Processing speed is increased by 3-5 times, and data volume is reduced by more than 90%, making it suitable for high-throughput RNA-seq data analysis with large sample sizes.

The invention discloses a remote sensing image large sample library generation method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring a to-be-purified sample set and a remote sensing image of a target research area; performing image segmentation on the remote sensing image to obtain an initial sample set; based on the initial sample set, obtaining an optimized sample set through unsupervised sample optimization based on a residual network; and based on the optimized sample set, performing feature similarity sample optimization based on Euclidean distance measurement on the to-be-purified sample set to obtain a purified sample set, and sorting based on the purified sample set to obtain a target sample library. According to the embodiment of the invention, samples are filtered and purified layer by layer through two-stage optimization, so that the purity and accuracy of the finally generated target sample library are ensured fundamentally, a solid foundation is laid for training a high-precision deep learning model, and the method can be widely applied to the technical field of image processing.

Provided are a structure evaluation method, a structure evaluation device, and a structure evaluation program with which it is possible to distinguish between diffracted X-rays generated for each region and evaluate respective structures. A structure evaluation method performed by measuring diffracted X-rays transmitted through a sample includes a step for simultaneously irradiating a plurality of regions of a large sample S11 in a single-crystal state with a white X-ray R11, and a step for detecting a diffracted X-ray produced by the irradiation. A measurement composed of the series of steps is performed a plurality of times. In at least one of the plurality of measurements, a slit 11 that selectively passes only a diffracted X-ray produced in a specific region among a plurality of regions C11, C12 irradiated with the X-ray is arranged on the back side of the large sample S10. Diffracted X-rays produced in the plurality of regions C11, C12 of the large sample S10 are each detected by using the slit 11, and the structure is evaluated for each of the regions.

The present application provides a method for quantitatively detecting water content in concrete, a device, a medium, and a product, and relates to the technical field of detection for water content in concrete. The method includes: obtaining an X-ray image of a to-be-detected sample, where the to-be-detected sample is a to-be-detected concrete sample, and the X-ray image is an image obtained after an X-ray travels through the to-be-detected concrete sample; determining X-ray light field distribution information of the X-ray image according to the X-ray image; determining a water density of the to-be-detected sample according to the X-ray light field distribution information; and determining water content of the to-be-detected sample according to the water density. The present application can improve precision and efficiency of detection for water content in concrete, reduces detection costs, and is applicable to a large sample.

This invention discloses an automatic method for extracting ionospheric TEC anomalies based on global seismic events. The method acquires a global seismic event catalog, global ionospheric TEC data, and space weather indices, performs data preprocessing, and for each seismic event, selects those occurring during quiescent periods based on space weather conditions. Ionospheric TEC data within fixed observation windows and space windows are extracted and differiated with contemporaneous background TEC data to obtain ionospheric TEC difference values. Normalized statistical analysis is then performed on the quiescent period TEC difference values ​​and their frequency corresponding to global seismic events to obtain statistical characteristics of ionospheric TEC anomaly rates and frequencies. Random seismic events are generated globally for comparison. This method is comprehensive, programmable, and suitable for automated processing of large global samples, possessing good scientific rigor, scalability, and engineering application value.

The invention discloses a marker combination for thyroid nodule risk assessment and application thereof. Based on large-sample multi-omics sequencing data, papillary thyroid carcinoma (PTC) and benign nodules are found to have a common epithelial spectrum origin, epithelial intermediate state cells (EICs) with malignant potential are found in the benign nodules, and typical molecular markers of the EICs are further identified to comprise ZCCHC12, CDLN1 and NPC2. The three genes participate in proliferation and differentiation of cells, the expression in PTC is remarkably up-regulated, and the malignant risk of the benign thyroid nodules can be evaluated by detecting the expression level of the markers in a biological sample. According to the method, a means for recognizing the high-risk benign nodules before malignant transformation occurs is provided for the first time, risk perspectiveness is achieved, and higher specificity and sensitivity are possibly achieved for recognizing the benign nodules with malignant potential.

This invention relates to the construction and application of an exosome purification device. The purification device utilizes the differences in size and charge between exosomes and other substances, combining the tangential flow of the sample within the device with the dual effects of membrane sieving and an electric field to remove contaminants and achieve high-purity exosome enrichment. The device includes a pre-filtration section and an exosome purification section. This device is simple to operate, can handle large sample volumes, has a short processing time, is less prone to membrane clogging by contaminants, and achieves high purity exosome separation with minimal structural damage. It can meet the separation and purification needs of exosomes from different sources and can be further applied to research on exosome proteomics, drug loading, activity, and biological functions.

The invention provides a construction method of a hypertension-cerebral hypoperfusion composite cerebral small vascular disease rat model. The construction method comprises the steps of experimental animal and grouping, hypertension model induction, cerebral hypoperfusion model induction and model evaluation system. According to the model, hereditary hypertension is replaced by 'L-NAME-induced drug-induced hypertension', and the model is combined with bilateral carotid artery occlusion surgery, so that clinical common hypertension and cerebral hypoperfusion dual pathological processes are successfully simulated. Compared with an SHR model, drug-induced hypertension is adopted to replace hereditary hypertension, and the method has the remarkable advantages in the aspects of period-cost-controllability-flux, 1, the modeling period is shortened to 8 weeks, and about 2 / 3 of time is saved; (2) the cost of a single animal is less than 1 / 3 of that of SHR, and the method is suitable for large-sample and multi-batch research; (3) the dosage of the L-NAME is adjustable, and the two-way intervention research of'hypertension driving 'and'blood pressure normalization' is supported; background data of the SD rats are complete and sufficient in supply, and multi-center repeated experiments are facilitated.

The application discloses a dynamic main shaft thermal error prediction method based on a multi-view space-time graph neural network, which jointly considers multi-angle evolution characteristics of thermal errors in axial, radial and bending directions, constructs a multi-dimensional graph structure and fuses space-time characteristics to model; the method cooperatively extracts local and global space-time dependent relationships by introducing graph convolution and a bidirectional long short-term memory network, combines modeling strategies and attention mechanisms of short-term, medium-term and long-term time scales, and effectively captures long-term dynamics and multi-scale thermal response characteristics in large sample thermal information data; comparative experiments show that the model exhibits superior prediction accuracy and robustness under various working conditions, and the average thermal error prediction accuracy is maintained at 92%-97%, which is significantly better than other models and variants.

The invention belongs to the technical field of underwater sound target recognition and machine learning, and discloses a small sample ship voiceprint data enhancement and high-precision recognition method, which comprises the following steps of: dividing an effective audio slice into a small sample data set, a large sample data set and an overall data set based on sample sizes of various categories; performing short-time Fourier transform and Mel filtering processing on the effective audio slices to generate a logarithmic Mel spectrogram with a fixed size; constructing and training a convolutional variational auto-encoder to form a reconstructed small sample data set; and combining the small sample data set, the reconstructed small sample data set and the large sample data set, carrying out feature extraction on the combined data set by adopting a BEATs encoder of a frozen parameter to obtain a voiceprint feature vector, training and calling a Light GBM classifier to carry out classification identification, and outputting an identification result of a ship category. According to the method, the overfitting problem of the classifier is remarkably relieved, and the precision and stability of ship voiceprint data recognition can be improved.

The invention relates to a repairing method for fractured stone of an ancient building, which comprises the following steps: step 1, collecting stone column stumps on the site of the ancient building, numbering, surveying, surveying and analyzing the site of the ancient building, and confirming the size, engraved texture and engraved character of an original stone column; 2, drawing numbered stub bulk sample drawings, establishing a model, and performing simulation splicing; step 3, actually splicing the stumps according to the simulated splicing result, and connecting the stumps through supporting pieces; and step 4, complementing missing segments or missing gaps among the stumps by using a material. According to the method, old material utilization is maximized, minimum intervention is carried out on the ancient building, and historical information is reserved to the maximum extent. According to the method, by analyzing the shape and shape of the historic building and the surrounding building information of the historic building site, the repaired stone column keeps the original shape and shape effect and is integrally coordinated and consistent with other components of the historic building site in year, shape and shape and appearance.

The application discloses a method for digesting and separating fluorinated aluminum by program, which comprises the following steps: placing a catalyst with metal elements supported on alumina as a carrier into a reaction container of a cluster digestion device, adding hydrochloric acid and / or nitric acid, setting a digestion program, introducing inert gas for pre-pressurization, performing primary digestion, and simultaneously starting a condensation reflux atomizer; after the program is completed, adding hydrofluoric acid, setting a digestion program, and performing secondary digestion; after the digestion is completed, entering a cooling and crystallization program, performing cooling and crystallization, and separating to obtain a sample solution to be detected and a solid; the solid is dried and calcined to obtain fluorinated aluminum crystals, and the liquid is used for detecting the content of metal elements. The method is applied to the digestion treatment of the alumina sample with metal elements supported thereon, has a large sample treatment range, complete digestion, no volatile loss of elements, and can separate AlF3 in an alpha crystal form through the program digestion of the digestion device.

The invention relates to the technical field of heart rate data processing, in particular to a health state change recognition method and device based on a time sequence heart rate mode. Acquiring a night heart rate sequence of the user for continuous M days; acquiring night heart rate sequences of the user for continuous N days from the night heart rate sequences of the user for continuous M days; calculating a daily heart rate average value of the continuous N days of the user based on the night heart rate sequence of the continuous N days of the user; k-means clustering is carried out on the daily heart rate average values of the user in continuous N days to obtain a plurality of first clustering clusters; k-shape clustering is carried out on the night heart rate sequence corresponding to the daily heart rate average value in each first clustering cluster to obtain a plurality of second clustering clusters; calculating a sample ratio of each second cluster, and determining the second cluster with the maximum sample ratio as a normal mode cluster; and determining whether other second clusters are abnormal mode clusters based on the normal mode clusters. In this way, early warning of potential health risks and long-term tracking of individual health can be achieved.

A video decoder is configured to: determine a minimum sample value for one or more blocks of video data, the minimum sample value being greater than 0; determining a maximum sample value for the one or more blocks of video data, the maximum sample value being less than 2bd-1, and bd being a bit depth of a sample in the one or more blocks; in response to determining that a process applied to a block of the one or more blocks of the video data results in a sample value outside a range from the minimum sample value to the maximum sample value, clipping the sample value to one of the minimum sample value or the maximum sample value to produce a clipped sample value; a decoded version of the block is determined based on the clipped sample values.

The invention discloses a cytarabine syndrome risk prediction model, a training method thereof and a system adopting the same. The model training method comprises the following steps: constructing an artificial intelligence model; wherein the input of the artificial intelligence model comprises the weight and the heating duration of the to-be-evaluated object; the output of the artificial intelligence model is the risk probability that the to-be-evaluated object belongs to the cytarabine syndrome fever; large sample data are adopted to train the artificial intelligence model, cases with fever of the cytarabine syndrome are positive samples, and other fever cases are negative samples. According to the present invention, the identification standard of the cytarabine syndrome fever and the infected fever is established, and the risk probability of the cytarabine syndrome of the object to be evaluated can be evaluated by analyzing the difference of the two detection results in the laboratory, such that the unreasonable use of the antibacterial agent is reduced, the medical cost is reduced, and the drug resistance risk of the antibacterial agent is reduced.

The utility model discloses a universal elasticity modulus detection platform, the platform is composed of an upper platform, a lower platform and a shock absorption layer between the upper platform and the lower platform, the upper platform is composed of a working plane, a receiver guide rail and a tool groove, soft silica gel or plastic materials are laid on the working plane, and a sample is softly supported on the working plane during testing. According to the method, the requirement for a testing method in the standard can be met, the testing requirements of large samples and special-shaped samples which cannot determine node positions can be met, meanwhile, the problem that small samples shift in the testing process is solved, the samples can be directly placed on a testing platform in the testing process, operation can be conducted, and testing steps are simplified; the receiver guide rail is arranged on the test platform, the damping layer is arranged between the upper platform and the lower platform, and the supporting legs with the damping pads are arranged between the lower platform and the equipment placing platform, so that the influence of environmental vibration on a test result is reduced to the greatest extent, and measurement errors caused by environmental factors are avoided.

A face key point labeling method, characterized in that, comprising the following steps: S11: five key points are labeled on small sample data selected from a face image data set; S12: pixel-level positioning and labeling are performed on small sample data of different sizes by using inference of a multi-task model, and a first labeling model is formed; S13: labeling effect of the small sample data is checked, and parameters in the previous step are adjusted until the labeling effect reaches an expectation; S14: a second labeling model is obtained by training the first labeling model by using large sample data selected from the face image data set; S15: the second labeling model is used to pre-label the face image data set; and S16: supplementary labeling is performed on the pre-labeling to improve the labeling effect. The effect of a small model can also be obviously improved.

The invention provides an engineering jointed rock mass unloading stability evaluation method and system and a storage medium, and belongs to the technical field of underground engineering and slope engineering rock mass stability evaluation.The method comprises the steps that core parameters of a jointed rock mass are obtained through drilling and coring, then an uncertainty regression model is established, large-sample random sampling is carried out, and an uncertainty regression model is established; respectively obtaining probability distribution of rock mechanical parameters; calculating rock mass equivalent mechanical parameters, comparing a rock parameter mean value with the rock mass equivalent parameters to obtain a reduction coefficient, and correcting rock parameter probability distribution into rock mass parameter probability distribution; and selecting representative parameters of the rock mass to carry out unloading simulation, extracting displacement of a key part and depth data of a plastic zone, and fitting probability distribution to complete stability evaluation. Parameter uncertainty full transmission from rock to rock mass is realized, and limitation of a single deterministic parameter is avoided; and an efficient and reliable technical framework is provided for rock mass unloading stability evaluation under complex geological conditions.

The invention discloses a nickel-based alloy fatigue life prediction system and method based on TPOT-AE combination, and belongs to the technical field of material life prediction. Wherein the data input and preprocessing unit completes normalization of AE multi-modal data and learns features through TPOT optimization; the multi-source data processing unit completes deep optimization of large sample data in TPOT to improve the learning ability for complex fatigue behaviors; and the fatigue life prediction unit dynamically adjusts the feature weight and the model parameters according to the deviation between the predicted value and the actual fatigue life to form a closed-loop optimization mechanism. The fatigue life prediction system automatically searches an optimal model pipeline and outputs a preliminary prediction result by fusing acoustic emission time domain and frequency domain characteristics, alloy element components, loading conditions and heat treatment parameters and utilizing TPOT; the method comprises the following steps of: constructing a Basquin fatigue model, then performing parameter correction and MLP neural network compensation by combining TPOT output based on the Basquin fatigue model, finally realizing physical consistency optimization through a constructed structure enhanced correction model, and dynamically updating a feature weight through residual feedback to realize model self-learning and performance enhancement. The service life prediction precision and interpretability can be improved, and the method is suitable for service state evaluation and health management of the nickel-based single crystal alloy.

The application discloses a method for constructing a gene regulation network based on meta-analysis, and aims to solve the problem that a gene regulation network has a large error and a low accuracy because different research data are not completely homogeneous when the gene regulation network is expanded by combining data of multiple studies to enlarge a sample size, and the method comprises the following steps: performing meta-analysis on each transcriptomic gene expression dataset obtained to obtain a significant differential gene list; obtaining a transcription factor gene list according to a species to which the transcriptomic gene expression data belong, and generating a core gene list according to the transcription factor gene list and the significant differential gene; calculating a Pearson correlation coefficient of each core gene and each significant differential gene, and constructing a co-expression network according to the Pearson correlation coefficient; clustering the co-expression network to obtain a plurality of differential gene modules with high correlation of the core genes; and combining biological knowledge and a structural equation model to construct a corresponding gene regulation network according to each module. The application belongs to the field of gene regulation networks.

The present application belongs to the technical field of underwater acoustic target recognition and machine learning, and discloses a small sample ship acoustic print data enhancement and high-precision recognition method. The method divides effective audio slices into small sample data sets, large sample data sets and overall data sets based on the sample amount of each category. Then, the effective audio slices are subjected to short-time Fourier transform and Mel filtering processing to generate fixed-size log Mel spectrograms. Next, a convolutional variational autoencoder is constructed and trained to form a reconstructed small sample data set. After that, the small sample data set, the reconstructed small sample data set and the large sample data set are combined, and the merged data set is subjected to feature extraction by a BEATs encoder with frozen parameters to obtain an acoustic print feature vector. A LightGBM classifier is trained and called for classification and recognition to output the recognition result of the ship category. The method significantly alleviates the overfitting problem of the classifier and can improve the precision and stability of ship acoustic print data recognition.