86 results about "Pulse noise" patented technology

The invention relates to the technical field of rolling bearing fault diagnosis, in particular to a rolling bearing fault diagnosis method based on data fusion and a neural network, and the method comprises the steps: S1, collecting vibration data of a plurality of sensors, superposing Gaussian white noise and pulse noise, and simulating a strong noise industrial environment; s2, extracting time domain and frequency domain features of vibration signals of each sensor, and constructing a preliminary feature matrix; s3, performing noise reduction processing and depth feature extraction on the feature matrix by using SDAE, fusing to generate a bearing data set, and dividing a training set and a test set; and S4, inputting the training set into an Adam optimization-based fractional order GRU model for training, and verifying the trained model by using the test set. According to the rolling bearing fault diagnosis method, the precision and robustness of rolling bearing fault diagnosis in a strong noise environment are remarkably improved by fusing the multi-source sensing data and the fractional order neural network model.

The invention relates to a direct current insulation resistance detection method and a detection device. The method comprises the following steps: establishing a mathematical model of insulation resistance and bus voltage of a direct current bus to be measured, acquiring voltage to ground values of positive and negative buses when a positive bridge arm and a negative bridge arm of an unbalanced bridge are not connected and when the negative bridge arm of the unbalanced bridge is connected, introducing noise influence, and establishing a calculation method for eliminating the noise influence based on the acquired measurement values. And performing iterative calculation to obtain a to-be-identified parameter theta. And theta is substituted into the mathematical model to solve the insulation resistance value of the direct current system to be measured. According to the method, environment noise generally existing in the test process and pulse noise caused by circuit switch switching are fully considered, so that the insulation resistance detection error of the direct current system to be detected is reduced, and the reliability of a detection result is improved.

The invention discloses a seismic signal denoising method and device and a storage medium, and relates to the technical field of signal processing, and the method is characterized in that the method comprises the steps: obtaining an originally collected seismic signal; performing wavelet transform threshold denoising on the seismic signal to obtain a first signal sequence without white noise; detecting the first signal sequence, determining the position of each impulse noise in the first signal sequence, and marking as an impulse noise point; and performing median filtering processing on each pulse noise point in the first signal sequence to obtain a second signal sequence with pulse noise removed as a denoised seismic signal sequence. Through the denoising method, white noise can be efficiently removed, pulse noise is specifically removed, the original information of the signal is reserved to the maximum extent, and the preprocessing quality of the seismic signal is improved.

The invention particularly relates to a voice noise reduction method suitable for different noise environments, and relates to the technical field of voice signal processing. Analyzing a noise environment and characteristics; adaptive selection of a noise reduction algorithm; adaptive noise reduction processing is executed; and noise reduction effect evaluation and iterative optimization are carried out. According to the invention, refined noise classification is matched with the algorithm, so that the limitation of a traditional single algorithm in a complex environment is solved; the steady-state noise is dynamically counteracted by adopting an adaptive filter, the frequency domain gain of the unsteady-state noise is adjusted in real time through Wiener filtering, and the pulse noise is subjected to dynamic threshold processing by combining median filtering and wavelet transform; especially, the design of cooperation factors in wavelet transform can accurately distinguish signal details and noise: intensively suppress pulse noise, loosely retain details such as voices and consonants, and realize dynamic balance of noise reduction intensity and signal distortion.

The invention discloses an audio pulse noise processing method, a related device and a storage medium. The method comprises the following steps: separating audio stream data from a video source of an input player and decoding the audio stream data; when a frame of to-be-played audio data of the audio stream data is read, according to the type of the to-be-played audio data, analyzing parameters of the autoregression model and a corresponding threshold value; wherein the type of the audio data to be played comprises a first frame, a middle frame and a tail frame; comparing the audio data to be played with the analyzed threshold value, and determining each segment of noise data; according to the type of the to-be-played audio data and the interval of each segment of noise data in the to-be-played audio data, predicting each segment of noise data through an autoregression model and corresponding logic, and repairing the noise data by using predicted data; and providing the repaired audio data to be played to a player for playing.

The invention provides a power line carrier communication system pulse suppression method for detecting pulse noise, which comprises the following steps of: 1, calculating a rough estimation threshold value according to the energy of a received signal; step 2, calculating the position of impulse noise according to the roughly estimated threshold value; step 3, according to the position of the pulse noise obtained in the step 2 and the received signal in the step 1, calculating a time domain signal of the pulse noise by adopting a pulse suppression method, and performing burr elimination processing; and step 4, according to the time domain signal of the pulse noise, performing pulse suppression operation on the received signal to complete pulse suppression of the power line carrier communication system.

The invention discloses a navigation signal blind separation method, system and equipment in an alpha pulse interference environment based on Gaussian kernel filtering, and a medium, and aims to improve the signal processing performance in a complex interference environment. According to the method, alpha pulse interference in aliasing signals is suppressed by adopting a Gaussian kernel filtering technology, and effective suppression of the alpha pulse interference is realized by reasonably designing filtering parameters; two correlation matrixes are subjected to joint diagonalization, and a joint diagonal matrix is used as an initial value of a left aliasing matrix and a right aliasing matrix in a blind separation algorithm, so that the convergence speed of aliasing matrix estimation is effectively improved; then, separating the signals after the pulse interference is suppressed by using a blind separation algorithm of joint diagonalization and an initial value of a left and right aliasing matrix; experimental results show that the method has excellent performance in the aspects of suppressing impulse noise influence and signal separation, and the problems that a traditional method is insufficient in alpha impulse interference suppression and poor in signal separation performance are effectively solved; systems, devices, and media are used to implement the methods.

To provide a PLC noise filtering method and system based on a clustering theory, and a medium, that have the advantages of good noise filtering performance, very simple calculation, good filtering effect and high filtering efficiency.SOLUTION: A method comprises the steps: obtaining a to-be-processed PLC signal sequence; and calculating a noise determination threshold according to a clustering property of a PLC signal and noise, and filtering the noise in the PLC signal sequence according to the noise determination threshold. According to the method, In order to distinguish among a PLC modulation signal, pulse noise and background noise through the clustering property of the signals and the noise, a difference among the PLC modulation signal, the pulse noise and the background noise in an aspect of amplitude continuity is utilized.SELECTED DRAWING: Figure 1

The invention discloses a frequency self-adaptive pulse noise filter system and a self-adaptive adjustment method, and relates to the field of filter self-adaptive adjustment, the frequency self-adaptive pulse noise filter system detects and identifies EMI (Electro-Magnetic Interference) frequency in real time by utilizing the super-sensitivity of a quantum entanglement state, and dynamically suppresses interference of specific frequency through a self-adaptive filter adjustment module. The response speed and the adaptability of the system are improved, and accurate suppression of EMI can be realized on the microscopic level based on the quantum mechanics principle, so that the stability and the precision of the quantum calculation and measurement technology are remarkably improved.

The invention discloses a signal noise suppression method and device, electronic equipment and a storage medium, and relates to the field of noise suppression, and the suppression method comprises the steps: collecting a target signal, carrying out the Fourier transform of the target signal, obtaining a signal frequency spectrum of the target signal, determining a frequency spectrum threshold value based on the standard deviation of a background noise frequency spectrum, and carrying out the calculation of the frequency spectrum threshold value. The method comprises the steps of obtaining a frequency spectrum threshold value, removing a noise component in the signal frequency spectrum based on the frequency spectrum threshold value, performing Fourier inversion on the signal frequency spectrum with the noise component removed to obtain an initial noise reduction signal, positioning a pulse noise signal in the target signal based on the initial noise reduction signal and the target signal, and performing smoothing processing on the pulse noise signal to obtain a target noise reduction signal. According to the invention, the technical problem that signal noise cannot be effectively suppressed in the prior art is solved.

The invention discloses a laser echo pulse noise suppression method based on dynamic double thresholds and an embedded system, belongs to the technical field of laser signal processing, and aims to solve the problem that strong pulse noise suppression and effective signal protection are difficult to balance in the prior art. The method comprises the following steps: acquiring a frame of digital signal sample; calculating a median value M and an absolute deviation mean value MADM of the sample; determining a dynamic high threshold TH and a low threshold TL based on M and MADM; and traversing each sampling point, and performing grading processing according to the relationship between the amplitude and the double thresholds: applying first noise suppression processing to the sampling point with the amplitude higher than TH, applying second noise suppression processing to the sampling point with the amplitude between TL and TH, and keeping the original value of the sampling point with the amplitude not higher than TL. The invention also discloses an embedded system for realizing the method. According to the invention, the pulse noise can be adaptively and accurately suppressed, the signal integrity is effectively protected, and the signal-to-noise ratio and the system robustness are improved.

The invention provides a one-dimensional single-port phononic crystal structure for enhancing acoustic signals, which comprises a waveguide, a first plate and a second plate, the first end of the waveguide is open, the second end of the waveguide is closed, the first plate and the second plate are both made of acoustic rigid materials, the first plate and the second plate are periodically arranged at intervals at the position close to the second end of the waveguide, and the first plate and the second plate are arranged at intervals at intervals. And a band gap exists in an energy band structure of the formed one-dimensional photonic crystal, so that a local state of energy concentration occurs between the second end of the waveguide and the one-dimensional photonic crystal. Based on a local state, the one-dimensional single-port phonon crystal structure can realize acoustic signal enhancement in a range near a specific frequency. Meanwhile, the harmonic noise signal, the periodic pulse noise signal and the modulation noise signal are enhanced by the acoustic signal of the one-dimensional single-port phonon crystal structure, so that the detection of weak acoustic signals in strong noise can be realized. The structure is relatively simple, the manufacturing cost is relatively low, and the engineering application prospect is relatively good.

The invention discloses a data fusion method and processing system based on the industrial internet, and relates to the technical field of industrial internet and edge intelligence. Stream segmentation processing is used for replacing global calculation, mutual information analysis and resampling are only carried out on latest data segments, time consumed by alignment of hundreds of data streams is compressed to a millisecond level, and the data fusion efficiency is improved. The real-time control requirement of the flexible production line is met; the dynamic offset generation is combined with the equipment sampling rate, the communication protocol and the type weight, the compensation intensity is adaptively adjusted, and alignment distortion caused by equipment heterogeneity is avoided; a noise perception mechanism is introduced, a gradient median and absolute deviation are utilized to construct a self-adaptive threshold value, pulse noise points are accurately marked, weight penalty is applied, the integrity of fault features is kept while interference is suppressed, and signal smoothness caused by traditional filtering is prevented; tensor decomposition under a federated learning framework supports distributed data privacy protection, and a noise suppression module preposed design avoids pollution to a core feature extraction process.

The invention discloses a constant-temperature water bath kettle fractional order system parameter identification method with measurement noise, which comprises the following steps: acquiring input data and output data of a constant-temperature water bath kettle, judging whether pulse noise exists in the output data or not, and constructing a wavelet operation matrix based on a pulse noise judgment result; obtaining a non-periodic time-delay integral operation matrix according to the wavelet operation matrix, and converting the input data and the output data into an input data operation matrix and an output data operation matrix based on the wavelet operation matrix and the non-periodic time-delay integral operation matrix; judging whether Gaussian noise and local noise exist in the output data or not, adjusting a matrix coefficient in an operation matrix corresponding to the output data based on a judgment result, constructing a system model of the constant-temperature water bath kettle, and solving the adjusted system model to obtain an identification result of system model parameters of the constant-temperature water bath kettle. And reflecting the correlation between the input data and the output data of the constant-temperature water bath kettle through the system model of the constant-temperature water bath kettle with the identification result.

The present invention discloses a method, device and helicopter for suppressing high-speed pulse noise of a helicopter, and relates to the technical field of helicopter noise control. At least one synthetic jet actuator is embedded in the upper surface of the blade tip, and the synthetic jet actuator can generate an upward jet, and the jet can interact with the main flow of the blade to form an aerodynamic shape. When high-speed pulse noise is generated, the shock wave on the blade tip is a noise shock wave. The opening state and control parameters of the synthetic jet actuator are adjusted to change the shape of the aerodynamic shape to weaken the intensity of the noise shock wave. The method, device and helicopter for suppressing high-speed pulse noise of a helicopter can effectively suppress the high-speed pulse noise of the rotor of the helicopter in various flight states, and have good applicability.

The invention provides an inlet silencing device of a turbocharger compressor, a compressor inlet is provided with a plurality of silencing cavities, a vibrating plate of a cantilever is arranged in at least one silencing cavity, at least one electrostatic driver is arranged on a vibrating path of the vibrating plate, local deformation is generated by controlling the electrostatic driver, and the compressor inlet silencing device is used for silencing the compressor inlet. The electrostatic actuator is caused to interfere with the vibration of the vibration plate. The electrostatic driver is arranged in the resonant cavity, the electrostatic driver makes contact with or is separated from the vibrating plate, so that the amplitude of the vibrating plate is changed, and the electrostatic driver and the vibrating plate can be selectively made to make contact with or be separated from each other according to the detected pressure change or noise of an outlet of the turbocharger; and the influence of synchronization impulse noise can be reduced or weakened.

The invention relates to the technical field of communication, in particular to a signal detection method and system based on Gini gamma correlation coefficient matrix eigenvalues. The method comprises the following steps: acquiring signals received by multiple paths of sensors, and calculating rank statistics of each path of received signal; calculating Gini gamma correlation coefficients of any two paths of received signals to form a correlation coefficient matrix; obtaining characteristic values of the Gini gamma correlation coefficient matrix, and calculating a geometric mean value and an arithmetic mean value of the characteristic values; constructing test statistics based on the ratio of the geometric mean value to the arithmetic mean value; comparing the test statistic with the detection threshold to complete the detection process; according to the method, the influence of a large abnormal value in the pulse noise is effectively suppressed through Gini gamma correlation transformation, the test statistic constructed based on the ratio of the geometric mean value to the arithmetic mean value of the characteristic values shows excellent robustness in a pulse interference environment, and the signal detection performance is remarkably improved; signal detection can be carried out under background noise containing pulse components.

The invention discloses an aircraft fuel quantity measurement self-adaptive filtering method, system and device based on multi-dimensional statistic dynamic adaptation and a medium. The method comprises the following steps that S01, signals are collected and preprocessed; dividing a fuel quantity digital signal output by a sensor into continuous time blocks according to a fixed time window; s02, calculating a multi-dimensional statistical magnitude; comprising the mean value, variance and kurtosis of the current time block; s03, performing multi-stage filtering processing; s04, performing data fusion and conversion; and keeping the original value of the data pulse point, and combining the original value with the fine filtering result of the non-pulse point to obtain a complete filtered signal. By monitoring the multi-dimensional statistical characteristics (mean value, variance and peak value) of a fuel quantity signal in real time and combining a coarse filtering-pulse detection-fine filtering multi-stage structure, filtering parameters are dynamically adjusted, efficient suppression of mixed noise (low-frequency shaking, Gaussian noise and pulse noise) is achieved, and the precision of fuel measurement under complex working conditions is improved.

The application relates to the technical field of power line carrier communication, and discloses a PLC signal pulse noise processing method and device and communication equipment, the method comprising the following steps: acquiring a current time domain signal collected by a current sampling point; slidingly buffering the current time domain signal into a buffer window, the buffer window comprising a plurality of buffer positions arranged continuously, and each buffer position storing a time domain signal; judging whether there are continuous target time domain signals of a preset number on any side of a target buffer position, the target buffer position being one of the plurality of buffer positions and being used for outputting a time domain signal of the sampling point, and the target time domain signal being a time domain signal with a signal strength greater than or equal to a preset threshold; and processing the time domain signal output by the target buffer position according to the judgment result. According to whether the signal strength of the continuous number of time domain signals is greater than the preset threshold, the time domain signal output from the specific buffer position in the buffer window is processed, the misjudgment can be reduced, and the signal quality is improved.

The invention discloses an underwater acoustic anti-interference transmission method and system based on sparse time-frequency characteristic mapping, and the method comprises the steps: adaptively generating a fractional order linear frequency modulation waveform with a specific time-frequency shearing slope at a transmitting end according to the Doppler state of an underwater acoustic channel, and achieving the physical layer focusing of transmitted energy; at a receiving end, constructing a hybrid observation model comprising a wide block sparse channel component and a narrow block sparse burst noise component, and inferring posterior probability distribution of environmental burst noise in a fractional order domain through joint iteration by using a dual-channel variational Bayesian algorithm; and finally, recovering an original signal through soft threshold interference cancellation and fractional order channel equalization. Through active feature mapping of the waveform and heterogeneous sparse joint inference of the receiving end, the problem of communication failure caused by underwater acoustic high-dynamic Doppler diffusion and marine organism impulse noise interference is effectively solved, and the transmission reliability in a severe underwater acoustic environment is remarkably improved.

The invention provides a vehicle-mounted pantograph-catenary temperature detection method and device, and the method comprises the steps: collecting an initial infrared image set containing a target train related pantograph-catenary through an infrared imaging device disposed at the roof of a target train; screening the initial infrared images contained in the initial infrared image set based on a pre-trained definition recognition model to obtain a to-be-processed infrared image set; based on a pre-trained noisy point elimination model, processing the to-be-processed infrared images contained in the to-be-processed infrared image set to obtain a target infrared image set; and according to the target infrared image set, determining the pantograph-catenary temperature associated with the target train. Automatic pantograph-catenary temperature monitoring is achieved, and the influence on the temperature monitoring process caused by the problems of image blurring, quality reduction, isolated pulse noise points and the like due to train movement is avoided.

The present invention discloses a method and system for estimating a direction of arrival in an impulse noise scenario. The method repeatedly iterates and updates impulse noise parameters and a sparse sound source signal matrix according to an overcomplete array manifold matrix, a received data vector matrix, an initial sparse sound source signal matrix, an initial impulse noise matrix, an initial impulse noise distribution accuracy parameter matrix and an initial impulse noise distribution degree of freedom parameter until an iteration condition is satisfied, thereby obtaining a final sparse sound source signal matrix, and estimating a direction of arrival according to the final sparse sound source signal matrix, so as to better adapt to a complex impulse noise environment, thereby improving the estimation accuracy of the direction of arrival, and solving the problem that, in a complex impulse noise statistical characteristic environment, a fitting error matrix with a p value selected by experience cannot fit the impulse noise well, resulting in low impulse noise fitting accuracy, thereby resulting in low estimation accuracy of the direction of arrival.

The application provides an echo cancellation method and system based on a non-standard FIR structure, and relates to the technical field of telephone communication.The method comprises the following steps: sampling a far-end signal to obtain a discrete value of the far-end signal; filtering an input vector of an adaptive filter at a current time through the adaptive filter to obtain an output vector; wherein the adaptive filter is a non-standard FIR filter, and a tap weight coefficient of an i-th tap of the non-standard FIR filter is calculated at an n-i time; obtaining a near-end sampling signal at the current time, subtracting the near-end sampling signal from the obtained output vector to obtain a residual signal at the current time, and returning the residual signal to the far-end; updating a tap weight coefficient vector of the adaptive filter, and repeating the above steps until the call is ended.The application can reduce the error of echo cancellation in a pulse noise environment.

The application provides a high-degree-of-freedom DOA estimation method based on a mirror image co-prime array in a pulse noise environment, and relates to the technical field of array signal processing, and comprises the following steps: S1: obtaining an observation signal by relying on non-circular signal characteristics; S2: calculating a pseudo-covariance matrix based on the information of the observation signal and a square-based correlation entropy operator to suppress the pulse noise; S3: vectorizing the pseudo-covariance matrix to obtain a virtual vector, removing redundant rows in the virtual vector, and performing truncation and sorting according to positions corresponding to continuous virtual array elements to obtain a virtual array receiving signal; S4: performing spatial smoothing on the virtual array receiving signal to obtain a full-rank reconstructed covariance matrix; S5: obtaining a noise subspace by performing eigenvalue decomposition on the covariance matrix; and S6: obtaining an accurate DOA estimation value by performing spectral peak searching on the noise subspace by using a MUSIC algorithm. The method has a high degree of freedom, can accurately complete DOA estimation in a pulse noise environment, and has good performance.

The method comprises the following steps: dividing real-time ocean noise time domain data of a polar region into a plurality of observation time periods, calculating a kurtosis value of each observation time period, further dividing each observation time period into a plurality of short time periods, and calculating short-time energy of each short time period; if the kurtosis value of the observation time period is greater than a threshold value and the short-time energy of a short time period in the observation time period is greater than an average value, judging that pulse noise exists in the observation time period, extracting a pulse noise signal, detecting a pulse event in the pulse noise signal, and taking a key parameter of the pulse event as a feature extraction result; and for an observation time period without pulse noise, acoustic characteristics of background noise without pulse are extracted through a frequency domain method and characteristic parameters. Through signal separation and feature extraction, the data transmission quantity is remarkably reduced, the transmission efficiency and the effective information quantity of the data are improved, meanwhile, the noise detection accuracy is improved, the polar region severe environment can be better adapted, and the requirement for long-term monitoring is met.

The invention discloses a transmission optimization method and device and related equipment, and belongs to the technical field of power line carrier communication. The transmission optimization method is applied to a first device, and comprises the following steps: carrying out pulse noise monitoring on a channel to obtain first channel state information, and carrying out white noise monitoring on the channel to obtain second channel state information; determining diversity mode parameters according to the first channel state information, and determining modulation mode parameters according to the second channel state information; and according to the communication transmission parameters, generating a first frame, and sending the first frame to the second device, the first frame being used for indicating the second device to communicate based on the communication transmission parameters, the communication transmission parameters comprising diversity mode parameters and modulation mode parameters. According to the method, the accuracy of channel state evaluation is improved by jointly monitoring the white noise and the pulse noise, the diversity times and the modulation mode are dynamically adjusted according to the channel state, collaborative optimization of the diversity times and the modulation mode is achieved, and the method can be applied to interference complex scenes.

The application relates to the technical field of communication interference elimination, in particular to a pulse interference elimination method for a power carrier communication chip, which comprises the following steps: acquiring carrier signals of each communication node in a power carrier communication network and the length of a power line between any two communication nodes; acquiring suspected pulse noise points; calculating pulse noise weights of the suspected pulse noise points; intercepting each suspected pulse signal in the carrier signal; obtaining a pulse noise spectrum diagram of each communication node; calculating line loss ratios between any two communication nodes; obtaining a comprehensive pulse spectrum diagram of each communication node; calculating an attenuation ratio of each frequency position; obtaining a noise suppression spectrum diagram; and obtaining a carrier signal after noise suppression based on the noise suppression spectrum diagram, so as to eliminate pulse interference in the carrier signal. The application dynamically filters the frequency domain characteristics of the carrier signal, and improves the communication quality of the power carrier communication chip.

The invention discloses a method and system for suppressing non-pulse noise to accurately measure pulse arrival time, and the method comprises the steps: S1, generating pulsar simulation parameters which comprise a rotation period P, a flow density S1250 and a pulse arrival phase delay standard deviation sigma delta p; s2, constructing a pulsar standard contour Q (x), and generating based on a pulse contour type Ptype, a sub-pulse number and a sub-pulse contour shape; s3, simulating multi-epoch observation data, injecting noise, adding phase delay delta p, and generating an average contour q (x); s4, judging a pulse region: positioning a candidate pulse region through baseline identification, and screening a final pulse region according to a contour type; s5, calculating a phase difference delta x between the standard contour and the average contour by using a phase gradient method; and S6, comparing a difference value between the phase difference delta x and the analog phase delay delta p, and verifying whether higher timing accuracy is obtained by removing the noise in the non-pulse region or not. The pulsar timing precision is remarkably improved by reducing noise interference in a non-pulse area, and the pulsar timing method is suitable for a high-precision scientific target.

The present application relates to a kind of free weight training monitoring method using wearable PPG sensor, belong to mobile computing application and human training auxiliary technical field.The present application uses infrared light and green light on PPG sensor to irradiate wrist skin, and collects the optical density change data containing muscle movement information.By analyzing change data, the load weight currently used by exerciser, the number of repetitions and the type of movement are obtained, and further predict the adjusted load weight required in next step.The present application proposes a novel arterial noise elimination method, effectively reduces the influence of pulse noise in artery, designs a novel feature extraction method based on recurrence plot, maps each action segment to two-dimensional phase space to extract space-time features, and constructs a novel multi-task learning model based on convolutional neural network.The present application can effectively identify load weight and movement type, realize the monitoring of free weight training.

The application relates to the field of underwater acoustic signal processing and discloses a deep learning method and device for weak line spectrum joint detection and reconstruction under non-Gaussian pulse noise. In order to solve the problem that in the prior art, a weak line spectrum detection method mainly extracts target features by using experience knowledge, and the performance of the method depends on the cognitive degree of an operator for samples to a great extent, and the problem that underwater acoustic signal processing technology under a Gaussian noise assumption is directly applied to signal detection under non-Gaussian noise, which causes the performance to decrease or even fail, the technical scheme provided by the application is as follows: a weak line spectrum joint detection and reconstruction method comprises the following steps: collecting pulse noise, and constructing a ship line spectrum data set according to the noise; pre-processing the data set to obtain a lofar graph; extracting weak features of the lofar graph, and obtaining a line spectrum detection and reconstruction result; training a preset model according to the lofar graph and the result, and outputting a trained model. The application is suitable for weak line spectrum detection and reconstruction under Gaussian pulse noise.