27results about How to "Remove random noise" patented technology

The invention discloses a map navigation method and system based on integrated navigation, and particularly relates to the field of integrated navigation, comprising compass data acquisition, magnetic declination correction, map integration, navigation system data acquisition, navigation system data analysis and integrated navigation. According to the method, the precision and robustness of the navigation system are remarkably improved, different environments can be dynamically adapted, and the error influence is reduced; rich geographic information support is provided by presetting an electronic map and POI marks; the real-time navigation service provides visual road condition information and path planning, the user experience is enhanced, and on the whole, more accurate, reliable and convenient navigation service is provided for the user.

The invention provides an AI-based three-light image type fire detector, and relates to the technical field of data processing, and the detector comprises a flame spectrum collection module which is used for obtaining multispectral image data of a monitoring area; the preprocessing module is used for preprocessing the multispectral image data to generate multispectral data; the feature extraction module is used for generating multidimensional feature data according to the multispectral data; the environmental condition acquisition module is used for acquiring environmental condition data of a monitoring area; the AI classification module is used for carrying out layer-by-layer analysis and classification processing on the multi-dimensional feature data by a multi-layer neural network model to generate a preliminary fire recognition result; the background adaptation module is used for receiving the preliminary fire recognition result and the environmental condition data and processing the preliminary fire recognition result to obtain fire recognition data; the alarm triggering module is used for receiving the fire identification data and judging whether a preset fire alarm condition is met or not; the method has a good detection effect.

The invention relates to the technical field of RFID chip manufacturing, in particular to an RFID chip image anomaly detection method and system based on intelligent learning. The method comprises the following steps: acquiring a real-time RFID chip sampling image set, performing feature alignment based on a registration template, generating an average grey-scale map by using an alignment image set, and obtaining a chip edge mask image set by using an edge search algorithm; segmenting the edge mask image set of the RFID chip to obtain an effective edge mask variance image set of the RFID chip; integrating the variance image set to obtain a pre-generated variance image; acquiring field environment data of the RFID chip and performing sensitivity coefficient estimation to obtain an edge sensitivity coefficient; and performing convolution filtering smoothing processing on the pre-generated variance image to obtain a chip edge variance image, and constructing an RFID chip image anomaly detection model based on the edge variance image and the average grey-scale image. According to the invention, the detection accuracy and stability are improved.

The invention discloses a grinding workpiece edge shape detection method and system based on machine learning, and the method comprises the steps: obtaining an original image of a grinding workpiece, and carrying out the preliminary image preprocessing; denoising the image and enhancing edge information by combining image filtering and mathematical morphology image processing; based on a second-order differential operator, edge detection is carried out in combination with the morphological gradient, and a workpiece edge contour is obtained; performing feature extraction on the image by using a convolutional neural network, automatically learning workpiece edge features, and constructing and training an edge detection model; acquiring a workpiece image in real time, inputting the workpiece image into the edge detection model for detection, and dynamically correcting the model; and S6, outputting a final edge detection result. According to the invention, by combining image preprocessing, composite image filtering, second-order differential operator and morphological gradient edge detection and feature extraction and model construction of the convolutional neural network, high-precision and robust detection of the edge of the grinding workpiece is realized.

The invention relates to the field of industrial equipment monitoring, and discloses a multi-sensor integrated system for remote monitoring of industrial equipment, which comprises a data acquisition module, a data processing module, a fault prediction module, a self-adaptive control module and a monitoring feedback module, the data acquisition module acquires multi-dimensional data from sensor arrays of a plurality of devices; the data processing module carries out noise reduction processing on the data and extracts key features; the fault prediction module predicts a short-term fault of the equipment and predicts a long-term operation trend; the self-adaptive control module adjusts control parameters of the equipment according to the prediction result and optimizes a control strategy; and the monitoring feedback module generates real-time fault alarm information by realizing remote monitoring and feedback of the equipment. The system can improve the safety and stability of equipment operation, optimize equipment management and reduce the equipment maintenance cost, and has high real-time performance, accuracy and intelligent level.

The invention relates to the technical field of medical image recognition, and provides a medical image focus area positioning method based on deep learning, which comprises the following steps: acquiring a to-be-detected medical image; performing format unification and size unification on the to-be-measured medical image to obtain a standardized image; denoising the standardized image to obtain a denoised image; the de-noised image is converted into an RBG image; inputting the RBG image into a trained focus area positioning model to obtain a positioning result; the lesion area positioning model is constructed based on a U-Net network, and the feature of the U-Net enables the lesion area positioning model to better capture fine features, so that the problem of boundary blur is reduced. And random noise and artifacts in the image can be eliminated by denoising the standardized image, so that extraction and recognition of disease features are facilitated, and the segmentation accuracy is improved.

The invention discloses a low-illumination image enhancement algorithm based on deep convolution, which is suitable for the field of image processing, and the core of the algorithm lies in an efficient noise reduction process and accurate decomposition of an illumination image and a reflection image. Comprising the following steps that firstly, a channel noise reduction module works cooperatively through multiple noise reduction strategies, noise in an image is effectively removed, and a foundation is laid for follow-up processing; then, the illumination image decomposition network and the reflection image decomposition network accurately separate the brightness information, texture and color information of the image based on a deep learning structure. And finally, by ingeniously fusing the illumination image and the reflection image, reconstructing a high-quality enhanced image, remarkably improving the brightness, contrast and details of the image, and meanwhile, retaining the natural color of the image. The algorithm has the advantages of being high in noise reduction capacity, accurate in illumination image decomposition, efficient in reflection image decomposition, good in enhancement effect and the like, and an effective and practical solution is provided for image enhancement under the low-illumination condition.

The invention discloses a data analysis system for early diagnosis of digestive tract diseases, relates to the field of medical health, and constructs a multi-level digestive tract disease risk assessment model through machine learning and deep learning technologies. In the preliminary evaluation stage, oral cavity gas data in the standard data set C is trained by using a support vector machine (SVM), a preliminary digestive tract risk score D1 is generated, and the preliminary digestive tract risk score D1 is compared with a preset preliminary risk threshold value T1 to judge whether a patient has a digestive tract disease risk or not. And when the score D1 exceeds the initial threshold value T1, the system automatically starts a deep learning model such as an LSTM network to carry out more accurate secondary evaluation, and calculates and outputs a deep digestive tract risk score D2. By extracting big data and analyzing medical record data, the model is further optimized, a secondary risk threshold T2 is set, and finally, real-time examination prompts such as gastroscope or blood examination are sent to a doctor or an automatic system through linkage of the system and a diagnosis and treatment server.

The invention relates to the field of video processing, and aims to provide a video noise AI (Artificial Intelligence) generation processing method which can be used for improving the denoising performance, reserving details, reducing the processing cost and improving the processing efficiency. According to the technical scheme, the method comprises the steps that S1, data preprocessing is conducted, a video file is decoded into image frames, the format of the image frames is converted, a data set is obtained after standardization processing and data enhancement are conducted, and the data set is divided into a training set, a verification set and a test set; s2, establishing a model: establishing a noise detection model, a generative adversarial network and a detail enhancement model; s3, model fusion: combining the noise detection model, the generative adversarial network, the detail enhancement model and the self-attention module into a joint model; training, verifying and testing the joint model; and S4, video noisy point generation processing is carried out, the collected video is subjected to data preprocessing, the joint model is input to obtain a noiseless image frame sequence, and a noiseless video is output after merging.

The invention discloses an underwater distorted image restoration method based on nonlinear reconstruction and random image block optimization. The method comprises the following steps: eliminating periodic distortion in a video sequence according to nonlinear reconstruction; random image block optimization is carried out on the video sequence after periodic distortion elimination; fusing and splicing the selected image blocks to form a complete image; and repeating the steps until a restored image sequence is obtained. According to the method, nonlinear reconstruction and random image block optimization are creatively fused into a whole, the method is extremely sensitive to obvious periodic distortion, and the details and definition of the image are improved by selecting and fusing the optimal block in the image, so that large-amplitude distortion in the image is restored, and the restoration effect of the underwater distorted image is greatly improved.

The invention discloses a method for predicting the remaining service life of an energy storage battery and a related device. The method comprises the following steps of: establishing maximum discharge energy-circulation circle data of a full life cycle of the battery; sG filtering smoothing is carried out, the maximum discharge energy value of the first circle is used as an initial energy value to convert the maximum discharge energy of the single circle of the battery into SOH, standardization processing is carried out, and a time window sliding method is adopted to divide an obtained standardized data set; constructing a time sequence prediction model based on the LSTM, and training the time sequence prediction model based on the LSTM by using the standardized data set X; the method comprises the following steps: collecting discharge energy data of a to-be-tested battery in an actual operation process in an energy storage system in the (1-n) th circle, inputting the discharge energy data into a pre-trained LSTM time sequence prediction model after standardization processing, outputting predicted SOH values of the (n + 1) th circle to the Nth circle, and establishing a complete battery SOH attenuation curve in combination with the data after standardization processing to obtain the remaining service life of the battery.

The invention discloses a display screen flat field correction method, device and system and a storage medium, which are used for improving the uniformity of a display screen image after flat field correction. Acquiring a plurality of frames of initial bright field images through acquisition equipment; performing gray average processing on the plurality of frames of initial bright field images to generate an average bright field image; selecting an ROI region on the average bright field image according to the gray level uniformity of the average bright field image; generating an ROI region mean value according to the pixel point information of the ROI region; generating a flat field correction coefficient graph according to the average bright field image and the ROI mean value; and collecting a to-be-corrected display screen through the collection device, and carrying out flat field correction processing on the collected to-be-corrected image according to the flat field correction coefficient graph.

The invention provides a remote management system for a rotary running casing torque monitoring short section, and relates to the technical field of petroleum engineering, a multi-modal sensor is arranged and configured through a multi-modal sensing module, and sensing connection is established; the data acquisition module acquires original data based on a preset acquisition constraint, and obtains initial monitoring data through AD conversion; a data correction module performs environment correction on the strain data and the rotation data based on the environment state data to obtain standard state monitoring data; the data processing module performs self-division filtering noise reduction on the standard state monitoring data to obtain standard monitoring data; and the data transmission module transmits the standard monitoring data to an upper computer, and performs remote management decision of the lower casing torque monitoring short section. The technical problem that monitoring efficiency and decision-making quality are affected due to the fact that monitoring data obtained in the prior art are not comprehensive and accurate is solved, comprehensive and accurate monitoring data are obtained in real time, and accuracy and reliability of operation remote management are improved.

The invention belongs to the technical field of temperature measurement and analysis of power generation enterprises, and particularly relates to an automatic temperature measurement and fault analysis method and a related system, which can effectively filter environmental noise and error influence of equipment by preprocessing temperature data so as to obtain more accurate temperature data. The data preprocessing can eliminate random noise in temperature measurement, and the measurement precision is improved. Through temperature trend prediction and abnormal data extraction, temperature abnormity can be identified as early as possible, and potential faults can be predicted. The trend prediction and comparison process helps to find problems before faults occur, so that advanced intervention is realized, and the fault diagnosis efficiency is improved. According to the invention, the optimization strategy generated based on the prediction result can be correspondingly adjusted before the fault occurs, and the probability of equipment shutdown or maintenance is reduced, so that the overall operation efficiency of the system is optimized, and the maintenance cost is reduced.

The invention relates to a reflection acoustic logging method and device for adjacent well collision prevention. The method comprises the following steps that an original waveform obtained through orthogonal dipole acoustic logging is preprocessed; carrying out compression vibration processing on the original waveform; suppressing residual shaft direct waves in the original waveform; separating the waveform of the reflected wave to obtain an uplink reflected wave and a downlink reflected wave of a time-depth domain, and performing dynamic correction and common midpoint superposition on the uplink reflected wave and the downlink reflected wave; the uplink reflected waves and the downlink reflected waves are subjected to migration imaging, and the real position of the reflector is obtained; and calculating the form and azimuth information of the crack according to the real position of the reflector. According to the technical scheme, the reflection body which is tens of meters away from the well or even farther away from the well can be imaged through the adjacent well anti-collision reflection acoustic logging method, the resolution ratio is high, the adjacent well can be better detected, and well collision accidents are prevented.

The invention provides a circuit board life prediction method and system based on mass failure welding spot data and a risk matrix, electronic equipment and a storage medium, and relates to the field of welding spot failure. The method comprises the following steps: acquiring massive invalid welding spot data, and extracting welding spot failure mechanism characteristics from the invalid welding spot data based on expert knowledge to obtain welding spot failure mechanism characteristics; based on the welding spot failure mechanism characteristics, welding spot failure characteristic factors are determined; acquiring welding spot failure characteristic factor data and circuit board data of a to-be-predicted circuit board, and determining welding spot failure data of the circuit board based on the welding spot failure characteristic factor data; and determining a welding spot failure type based on the welding spot failure data, and inputting the circuit board data and the welding spot failure type into a preset life prediction model to obtain a life prediction value. By means of the method, mass data are deeply utilized, essential reasons influencing welding spot faults are found out, and accurate evaluation and prediction of the residual life of the circuit board are achieved.

The invention discloses a remote detection method for soot blowing system pipe structure defects, and relates to the technical field of thermal equipment detection. The problems that in the prior art, detection needs to be stopped, real-time monitoring cannot be achieved, and defect information is not visually presented can be at least partially solved. The method comprises the following steps: deploying an electromagnetic ultrasonic guided wave sensor array around a welding seam to perform signal excitation and synchronous acquisition; performing dual noise reduction processing by adopting synchronous average and wavelet transform to extract defect signals; constructing a space model to position defects based on the signal arrival time difference; and generating a defect image by utilizing C-scanning imaging and performing quantitative evaluation. According to the invention, on-line remote detection without shutdown is realized, the signal-to-noise ratio is obviously improved through dual noise reduction, accurate positioning and visual display of defects are realized by combining time difference analysis and C scanning imaging, and safe operation of a unit is guaranteed.

The present disclosure specifically relates to the field of image processing technology, and specifically to an image denoising method and device, a computer-readable medium and an electronic device. The method comprises: obtaining a current image and a corresponding reference image under preset exposure conditions, determining a first image parameter of the current image; querying a noise parameter lookup table according to the preset exposure conditions to obtain a second image parameter; comparing the first image parameter with the second image parameter to perform a first denoising process on the current image according to the image parameter comparison result to obtain a preliminary denoising image under the preset exposure conditions, and generating a denoising degree parameter according to the denoising result of the current image; performing a second fusion denoising process on the preliminary denoising image and the corresponding reference image based on the denoising degree parameter to obtain a fused image; performing a third fusion denoising process on the fused images under different preset exposure conditions to obtain a denoised image corresponding to the current image. The present disclosure can effectively eliminate the noise of the exposure image and ensure the image denoising effect.

The invention discloses an SVD (Singular Value Decomposition) seismic fracture-cavity body enhancement method based on structural constraint, which comprises the following steps of: taking a seismic data sub-body in a set surface element from single-channel post-stack seismic data, and performing singular value decomposition on the seismic data sub-body to obtain a singular value and a singular vector; performing data reconstruction according to the size sequence of the singular values, performing similarity structure evaluation on the reconstructed data and the original data, gradually determining a fracture-cavity response threshold value based on a similarity coefficient, and determining a noise response threshold value according to an energy retention principle; designing a soft threshold window function according to the fracture-cavity response threshold and the noise response threshold to construct a fracture-cavity enhancement operator; applying a fracture-cavity enhancement operator to the singular value matrix to obtain reconstructed fracture-cavity enhanced seismic data of the single-channel data; and repeating the process on all single-channel seismic data to obtain a final fracture-cavity enhanced seismic data volume. According to the method, the seismic response characteristics of the fracture-cavity body can be effectively enhanced, the weak signals of the reservoir are highlighted, and the interpretation precision is further improved.

The invention provides a positioning and leveling method, system and equipment of an underwater leveling machine and the underwater leveling machine, and relates to the technical field of ocean engineering. The method comprises the following steps: acquiring an original ranging data sequence of a fixed reference node; performing deviation compensation and correction on the original ranging data sequence to generate a target ranging value; performing time sequence filtering processing on the target distance measurement value of the distance measurement link between the underwater leveling machine and each fixed reference node to obtain a smooth distance measurement sequence; selecting data at multiple moments from the smooth ranging sequence to construct a time window input matrix, taking the time window input matrix as the input of the target deep learning model, and outputting the predicted position of the underwater leveling machine; and generating a control instruction based on the predicted position so as to control the underwater leveling machine to carry out positioning and leveling. According to the method, by combining multi-stage data processing and deep learning prediction, distance measurement errors and environmental interference are eliminated, so that the positioning and leveling accuracy of the underwater leveling machine can be improved.

The present invention provides an AI-based three-light image type fire detector, which relates to the technical field of data processing and includes: a flame spectrum acquisition module, which is used to obtain multi-spectral image data of a monitoring area; a preprocessing module, which is used to preprocess the multi-spectral image data and generate multi-spectral data; a feature extraction module, which is used to generate multi-dimensional feature data according to the multi-spectral data; an environmental condition acquisition module, which is used to collect environmental condition data of the monitoring area; an AI classification module, which is used to perform layer-by-layer analysis and classification processing on the multi-dimensional feature data by a multi-layer neural network model to generate a preliminary fire identification result; a background adaptation module, which is used to receive the preliminary fire identification result and the environmental condition data, and process the preliminary fire identification result to obtain fire identification data; an alarm triggering module, which is used to receive the fire identification data and determine whether the preset fire alarm conditions are met; the present invention has a good detection effect.

This invention discloses a spectral analysis method, apparatus, device, and storage medium. It includes: acquiring raw spectral data to be analyzed; preprocessing the raw spectral data to obtain preprocessed spectral data; performing multi-peak fitting on the preprocessed spectral data using an iterative residual peak-finding fitting algorithm to obtain a final fitted spectrum and a final residual spectrum; and generating spectral analysis results based on the final fitted spectrum and the final residual spectrum. Preprocessing the raw spectral data effectively eliminates background interference, cosmic ray pseudo-signals, and random noise, improving the signal-to-noise ratio of the spectral data. Using the iterative residual peak-finding fitting algorithm for multi-peak fitting accurately identifies and fits all significant spectral peaks, avoiding underfitting and overfitting problems, and improving the accuracy of multi-peak fitting. The generated spectral analysis results can accurately extract core feature indicators, and the fitting quality can be verified through residual spectroscopy, ensuring that the analysis results are accurate and reliable, truly reflecting the structural characteristics of the material.

The invention relates to the technical field of image display, in particular to a metal metallographic structure image classification display method and system. The method comprises the following steps: preprocessing a metal metallographic structure image, calculating and evaluating an image feature index Tzls, quantifying the overall quality of the metal metallographic structure image, automatically screening low-quality and repeated images, and reducing stored and processed invalid data; classifying the metal metallographic structure images by using two characteristics of structural factors and defect factors of the metal metallographic structure; the metal metallographic structure image is classified and graded by calculating and evaluating a structure characteristic index Jtzs and a defect characteristic index Qtzs; and all grade division results are transmitted to the image classification display module, so that timely and accurate data transmission is ensured, a user searches and filters according to a structure grade or a defect grade, a required image is quickly found, and the working efficiency is improved.

The invention relates to a UWB radar target motion prediction method and system based on a multi-convolutional neural network, and the method comprises the steps: obtaining echo signal features collected by a UWB radar in real time and corresponding target motion parameters, and obtaining an initial data set through time series data division and normalization processing; extracting and splicing the initial data set based on a local mean decomposition algorithm to obtain standard input feature data; constructing a motion parameter prediction model based on the echo signal features, optimizing the motion parameter prediction model by adopting a genetic algorithm, and performing training through standard input feature data; and eliminating multi-path and noise interference from echo signal characteristics corresponding to a to-be-detected target through local mean decomposition, and inputting the echo signal characteristics into the trained motion parameter prediction model to obtain predicted motion parameters of the to-be-detected target. The UWB radar motion parameter prediction precision and generalization ability can be improved through LMD interference removal and CNN combined with attention and gating and genetic algorithm optimization.

The application provides a college major talent outflow trend prediction method based on multi-channel feature fusion and adaptive weighting random forest, relates to the talent flow prediction and machine learning application technical field, and comprises the following steps: collecting college major enrollment data and corresponding macroeconomic data of each year, and preprocessing missing data; adopting a multi-channel feature decoupling and fusion mechanism; the multi-channel feature decoupling and fusion mechanism adopted by the application effectively solves the defects of weak feature expression ability and neglecting external economic driving factors in the prior art, decouples internal trends and external driving factors into three channels, comprehensively extracts enrollment sequence characteristics, regional macroeconomic characteristics and regional economic difference characteristics, realizes deep fusion of internal and external information, enhances the model's recognition ability to complex patterns, can fully capture the influence of macroeconomic variables on professional talent flow, and greatly improves the prediction accuracy.

The present invention relates to the field of spectrum detection technology, and in particular to a method, device and Raman spectrometer for denoising a Raman spectrum signal. The method for denoising a Raman spectrum signal comprises step S100, sampling a reset signal N times, and performing an addition operation on the reset signals obtained by sampling to obtain a first signal; step S200, sampling a valid signal N times, and performing an addition operation on the valid signals obtained by sampling to obtain a second signal; step S300, performing a subtraction operation on the first signal and the second signal and dividing by N to obtain an actual pixel signal. By collecting the reset signal and the valid signal multiple times and adding and then subtracting them, the reset noise can be eliminated while eliminating random noise, thereby reducing the interference of noise on the spectrum signal, improving the signal-to-noise ratio, and then effectively improving the accuracy of Raman spectrum data for material analysis.

The invention provides a pixel parasitic light sensitivity effect suppression method and an image sensor. The method comprises the following steps: acquiring a plurality of frames of basic images; calculating the signal intensity of each row of pixels by using the multiple frames of basic images to obtain a signal intensity data set; obtaining a parasitic light correction coefficient of each row of pixels according to the signal intensity data set; and performing parasitic light correction on the input original image by using the parasitic light correction coefficient. The parasitic light correction coefficient corresponding to each line is calculated according to the signal intensity of each line of the multi-frame basic image, so that the parasitic light correction coefficient of each line can be utilized to perform line-by-line parasitic light correction on the input original image, and the parasitic light sensitivity effect is effectively inhibited in a dynamic correction mode. Meanwhile, due to the fact that parasitic light correction is carried out line by line, random noise in pixel signals of the current line can be eliminated, image consistency is guaranteed, image quality is improved, and the problem of how to effectively suppress the parasitic light sensitivity effect to improve the image quality is solved.