CN117892059A - Electric energy quality disturbance identification method based on multi-mode image fusion and ResNetXt-50 - Google Patents
Electric energy quality disturbance identification method based on multi-mode image fusion and ResNetXt-50 Download PDFInfo
- Publication number
- CN117892059A CN117892059A CN202311759114.XA CN202311759114A CN117892059A CN 117892059 A CN117892059 A CN 117892059A CN 202311759114 A CN202311759114 A CN 202311759114A CN 117892059 A CN117892059 A CN 117892059A
- Authority
- CN
- China
- Prior art keywords
- image
- resnetxt
- quality disturbance
- power quality
- images
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 53
- 230000004927 fusion Effects 0.000 title claims abstract description 25
- 230000007704 transition Effects 0.000 claims abstract description 11
- 230000003044 adaptive effect Effects 0.000 claims abstract description 7
- 238000001914 filtration Methods 0.000 claims abstract description 6
- 238000012546 transfer Methods 0.000 claims abstract description 4
- 239000011159 matrix material Substances 0.000 claims description 26
- 238000012545 processing Methods 0.000 claims description 9
- 238000013507 mapping Methods 0.000 claims description 6
- 238000000354 decomposition reaction Methods 0.000 claims description 5
- 238000004590 computer program Methods 0.000 claims 1
- 238000000605 extraction Methods 0.000 claims 1
- 238000007781 pre-processing Methods 0.000 claims 1
- 230000004044 response Effects 0.000 claims 1
- 238000000638 solvent extraction Methods 0.000 claims 1
- 239000002131 composite material Substances 0.000 abstract description 2
- YLOCGHYTXIINAI-XKUOMLDTSA-N (2s)-2-amino-3-(4-hydroxyphenyl)propanoic acid;(2s)-2-aminopentanedioic acid;(2s)-2-aminopropanoic acid;(2s)-2,6-diaminohexanoic acid Chemical compound C[C@H](N)C(O)=O.NCCCC[C@H](N)C(O)=O.OC(=O)[C@@H](N)CCC(O)=O.OC(=O)[C@@H](N)CC1=CC=C(O)C=C1 YLOCGHYTXIINAI-XKUOMLDTSA-N 0.000 abstract 1
- 108010072051 Glatiramer Acetate Proteins 0.000 abstract 1
- 229940042385 glatiramer Drugs 0.000 abstract 1
- 238000004364 calculation method Methods 0.000 description 2
- 230000008569 process Effects 0.000 description 2
- 238000005070 sampling Methods 0.000 description 2
- NAWXUBYGYWOOIX-SFHVURJKSA-N (2s)-2-[[4-[2-(2,4-diaminoquinazolin-6-yl)ethyl]benzoyl]amino]-4-methylidenepentanedioic acid Chemical compound C1=CC2=NC(N)=NC(N)=C2C=C1CCC1=CC=C(C(=O)N[C@@H](CC(=C)C(O)=O)C(O)=O)C=C1 NAWXUBYGYWOOIX-SFHVURJKSA-N 0.000 description 1
- 238000012300 Sequence Analysis Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013136 deep learning model Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000011176 pooling Methods 0.000 description 1
- 238000003672 processing method Methods 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/10—Pre-processing; Data cleansing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
- G06V10/761—Proximity, similarity or dissimilarity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/02—Preprocessing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/12—Classification; Matching
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Data Mining & Analysis (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Medical Informatics (AREA)
- Multimedia (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computational Linguistics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Algebra (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a power quality disturbance identification method based on multi-mode image fusion and ResNetXt-50. First, the one-dimensional time series of power quality disturbances are processed into three different images using Markov Transition Field (MTF), recurrence Plot (RP), and Gramian Angular Field (GAF) methods, where MTF represents a markov transfer field, RP represents a reproduction plot, and GAF represents a glatiramer angle field. Then, three different color images are fused into one composite image by an adaptive filter based image fusion (ADF) method. And finally, classifying the fused color images by using a ResNetXt-50 model, and judging the type of the electric energy quality disturbance. According to the invention, the MTF, RP, GAF method is adopted to convert the one-dimensional time sequence of the power quality disturbance into the color image, the image fusion (ADF) method based on the self-adaptive filtering is utilized to fuse the images with different characteristics together, and the ResNetXt-50 model is utilized to classify the images, so that the accuracy and the reliability of the power quality disturbance identification are improved.
Description
Technical Field
The invention relates to a power quality disturbance identification method based on multi-mode image fusion and ResNetXt-50, and belongs to the field of power quality disturbance identification.
Background
The electric energy quality disturbance refers to instantaneous fluctuation of parameters such as voltage, current and the like occurring in the electric power system, and influences normal operation of the electric power system and normal use of equipment, so that the electric energy quality disturbance identification method has important practical significance for identification and classification of the electric energy quality disturbance. Most of the traditional power quality disturbance recognition methods are based on one-dimensional time sequence analysis, lack of intuitiveness and visibility, and are easily interfered by noise and artifacts, so that recognition accuracy is low.
In recent years, with the continuous development of image processing and deep learning technologies, the conversion of one-dimensional time series into color images and the identification by using image features have become a new idea. The image-based power quality disturbance recognition method can effectively improve the accuracy and reliability of recognition, but still has some problems. Typical signal processing methods often employ a single method to convert a time series into images, and there may be cases where certain data features are not adequately represented or captured. This may lead to loss or confusion of certain information, affecting the accuracy of the classification.
In addition, popular deep learning models such as LeNet, alexNet, googLeNet have the problems of low model depth, overfitting and the like, and the accuracy of power quality disturbance recognition is difficult to improve.
Disclosure of Invention
In order to overcome the problems in the prior art, the invention designs a power quality disturbance recognition method based on multi-mode image fusion and ResNetXt-50, which converts signals into images with different characteristics, performs adaptive filtering image fusion, and then utilizes a ResNetXt-50 network to realize rapid and high-precision complex power quality disturbance recognition.
In order to achieve the above purpose, the present invention adopts the following technical scheme:
a power quality disturbance identification method based on multi-mode image fusion and ResNetXt-50 comprises the following steps:
obtaining a power quality disturbance signal;
processing the power quality disturbance one-dimensional time sequence into color images with different characteristics;
fusing color images with different characteristics by using an ADF method;
power quality disturbance recognition of fused images using ResNetXt-50
Further, the one-dimensional time sequence of the power quality disturbance is processed into color images with different characteristics, specifically:
dividing the one-dimensional time sequence into p different intervals, and calculating transition probabilities between adjacent intervals to obtain a transition probability matrix; adding a time axis to expand the whole transition probability matrix to obtain an MTF matrix, and mapping each element of the MTF matrix to a color space to obtain a two-dimensional MTF image.
Performing polar coordinate coding on the one-dimensional time sequence of the power quality disturbance signal to obtain a coding result; and calculating a Grami angle field matrix of the coding result to obtain a two-dimensional GAF image.
Constructing a phase space reconstruction matrix for the one-dimensional time sequence to obtain a similarity matrix; and defining a threshold value and constructing a two-dimensional image to obtain a two-dimensional RP image.
An adaptive filtering-based image fusion (ADF) method is used to fuse three different color images into a single composite image.
Further, the method for fusing the images with different features by using the ADF method specifically comprises the following steps:
and carrying out multi-scale decomposition on the color images of the three different features.
An adaptive filter is applied at each decomposition level to fuse the images, preserving the detail and texture information of the input image, while reducing artifacts and noise.
Reconstructing the fused image into a comprehensive image through multi-scale inverse transformation.
Further, inputting the fused image into a ResNetXt-50 network, carrying out power quality disturbance identification, and determining the disturbance type, wherein the specific steps are as follows:
the input color image is preprocessed by fixed size and the like, then is sent into a network, features are extracted through a plurality of convolution layers and pooling layers, and the electric energy quality disturbance type result represented by the image is output through a full connection layer.
By means of the technical scheme, the invention provides a power quality disturbance identification method based on multi-mode image fusion and ResNetXt-50, which has at least the following beneficial effects:
1. the invention adopts Markov Transfer Field (MTF), reproducibility map (RP) and Grami Angle Field (GAF) method to process the one-dimensional time sequence of power quality disturbance into three different color images. These methods can transform the time series into a series of image representations that better reveal the data features.
2. The invention adopts an image fusion (ADF) method based on self-adaptive filtering to fuse three different color images into a comprehensive image. The method fuses the images by multi-scale decomposition of the input image, and then applying an adaptive filter on each decomposition level. The ADF method can keep the detail and texture information of the input image, reduce the artifacts and noise, has the advantages of good fusion effect, high calculation speed and the like, can effectively fuse the images with different characteristics together, and improves the accuracy of disturbance identification. Compared with direct classification of single-mode images, the multi-mode image fusion can obtain more information, and accuracy and robustness of recognition are improved.
3. And classifying the fused color images by using a ResNetXt-50 model, and judging the type of the electric energy quality disturbance. The ResNetXt-50 model is a deep convolutional neural network model, can effectively avoid model overfitting, has strong image classification capability, and can quickly and accurately identify power quality disturbance.
Drawings
The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute an undue limitation to the application. In the drawings:
FIG. 1 is a flow chart of an identification method of the present invention;
FIG. 2 is a flow chart of the present invention for processing a one-dimensional time series into a two-dimensional MTF image using MTF;
FIG. 3 is a flow chart of the present invention for processing a one-dimensional time series into a two-dimensional GAF image using GAF;
FIG. 4 is a flow chart of the present invention for processing a one-dimensional time series into a two-dimensional RP image using RP;
FIG. 5 is a flow chart of fusing two-dimensional MTF, two-dimensional GAF, two-dimensional RP images using the ADF method;
FIG. 6 is a schematic diagram of a process for type recognition of fused images using ResNetXt-50.
Detailed Description
Referring to fig. 1-6, a method for identifying power quality disturbances based on multi-modal image fusion and ResNetXt-50, the method comprising the steps of:
s1, obtaining a one-dimensional time sequence X= { X of the power quality disturbance signal 1 ,x 2 ,x 3 ,x 4 ,…x N }。
S2, referring to FIG. 2, the one-dimensional time sequence is processed into a two-dimensional MTF image by using an MTF method.
S21, dividing the one-dimensional time sequence data into Q areas Q= { Q according to the value range 1 ,q 2 ,q 3 ,q 4 ,…q n }。
S22, obtaining the state transition probability w of the Q areas according to the formula (1) ij Obtaining a Markov transfer matrix W with a Q X Q scale, wherein the matrix is shown as a formula (2);
w ij ＝P(x t ∈q j |x t-1 ∈q i ) (1)
w in ij (i, j e {1,2,3, …, Q }) is represented in region Q j One data point in (1) is q in the region i Is a probability of a data point following.
S23, adding a time dimension into W, arranging each state transition probability according to a time sequence to expand the whole Markov transition matrix, and finally generating an N multiplied by N MTF matrix M. The matrix M is shown in formula (3).
In the formula (4): m is m kh Represented in region q i X of (2) k The data points are located in region q j X of (2) h Probability of data point following.
And S24, mapping the MTF matrix M onto the color image to obtain a two-dimensional MTF image.
S3, referring to FIG. 3, processing the one-dimensional time sequence into a two-dimensional GAF image by using a GAF method
S31, scaling the one-dimensional sequence by using the formula (5) to obtain a scaled one-dimensional time sequence
S32, polar coordinate encoding
For one-dimensional time sequence after scalingAnd (3) performing polar coordinate coding, and mapping the sampling value and the time stamp of the time sequence into an angle and a radius respectively, wherein the coding result is shown in a formula (6).
Wherein:for time series +.>The i-th sampling value of (a); t is t i Representing the time sequence +.>Ith sample value +.>The corresponding timestamp, K, is the total length of the timestamp.
S33, grami Angle field
According to equation (7), a one-dimensional time series is calculatedIs a Gray Mi Jiao field G ASF And utilize the glasThe meter angle field generates a two-dimensional GAF image.
And S34, mapping the GAF matrix M onto the color image to obtain a two-dimensional GAF image.
S4, referring to FIG. 4, processing the one-dimensional time sequence into a two-dimensional RP image by using an RP method
S41, reconstructing a phase space of the one-dimensional time sequence of the power quality disturbance according to the formula (8).
X t ＝{x t ,x t+τ ,…x t+(m-1)τ } (8)
t-1,2,…,N-(m-1)τ
Wherein: x is X t Reconstructing the reconstructed point for the original signal; m is the dimension of the new artifact space; τ is the delay time of the new phase space.
S42, calculating norms of distances between each point in the new phase space according to the formula (9).
L ij ＝||X i -X j || (9)
Wherein: x is X i And X j Respectively reconstructing the original signals; and (3) taking the I as a norm calculation.
S43, obtaining the value of each point on the RP matrix according to the formula (10) to obtain the RP matrix R, wherein the RP matrix is shown as the formula (11)
R ij ＝Heaviside(ε-L ij ),i,j＝1,2,…,N (10)
Wherein: epsilon is a predetermined threshold and the Heaviside function is
S44, mapping the RP matrix R onto the color image to obtain a two-dimensional RP image
S5, referring to FIG. 5, the color images obtained by the three different features are fused by utilizing an adaptive filter ADF method
S51, three color images I 1 (x,y)、I 2 (x,y)、I 3 (x, y) having dimensions M x N, where x and y represent rows and columns, respectively, in the image.
S52, calculating a gradient image G of each color image by calculating convolution of gradient operators through the step (13) 1 (x,y)、G 2 (x,y)、G 3 (x,y)。
Wherein: g x,i (x, y) and g y,i (x, y) each represents I i (x, y) gradient in x and y directions.
S53, calculating each color image I by the formula (14) i Weighting coefficient W of (x, y) i (x,y)
Wherein: where σ is a parameter controlling the distribution range of the weighting coefficient.
S54, calculating the fused image I according to the formula (15) f (x, y) to obtain a fused color image
S6, referring to FIG. 6, inputting the color image obtained by fusion into a ResNetXt-50 network for identifying the power quality disturbance type, and outputting the corresponding power quality disturbance type, wherein the ResNetXt-50 structure is shown in a table 1.
TABLE 1 ResNetXt-50 network architecture
S7, obtaining the specific category names of the power quality disturbance signals.
Claims (6)
1. The electric energy quality disturbance recognition method based on the multi-mode image fusion and ResNetXt-50 is characterized by comprising the following steps of:
processing the one-dimensional time sequence of the power quality disturbance signal into color image representations of three different features by using Markov Transition Field (MTF), recurrence Plot (RP) and Gramian Angular Field (GAF) methods, wherein MTF represents a Markov transfer field, RP represents a reproducibility map, and GAF represents a Grami angle field;
adopting an image fusion (ADF) method based on self-adaptive filtering to fuse three color images with different characteristics into a comprehensive image, and reducing artifacts and noise while retaining the detail and texture information of an input image;
and inputting the fused image into a ResNetXt-50 model to identify the type of the power quality disturbance, and judging the type of the power quality disturbance.
2. The method for identifying the power quality disturbance based on the multi-mode image fusion and ResNetXt-50 according to claim 1, wherein the method comprises the following steps of: the one-dimensional time sequence of the power quality disturbance is converted into color image representations with different characteristics, and the method specifically comprises the following steps:
partitioning the time sequence according to the element values to obtain Q regional blocks; calculating transition probabilities among all the blocks to obtain a Markov probability transition matrix; adding a time axis into the Markov probability transition matrix to obtain an MTF matrix, and mapping to obtain a corresponding two-dimensional MTF image;
representing the continuous observation value in the time window as a point in a phase space to obtain a multidimensional phase space of an original one-dimensional time sequence; in the phase space, calculating the similarity between each pair of points to obtain a similarity matrix; setting the element at the corresponding position in the similarity matrix to be 1 or 0 according to the value of each element in the similarity matrix to obtain an RP matrix and a corresponding two-dimensional RP image;
performing polar coordinate coding on the one-dimensional time sequence signal to obtain a coding result; and calculating a Grami angle field matrix of the coding result to obtain a two-dimensional GAF image.
3. The method for identifying the power quality disturbance based on the multi-mode image fusion and ResNetXt-50 according to claim 1, wherein the method comprises the following steps of: the method for fusing three color images by using the ADF method specifically comprises the following steps:
one-dimensional reading an input image and converting it into a gray scale image;
preprocessing each input image, adjusting the image size, increasing the contrast and denoising;
respectively carrying out Gaussian pyramid decomposition on the preprocessed images to obtain an image pyramid composed of multiple layers of images;
carrying out self-adaptive filtering processing on each layer of image, and enabling the filtered image to have optimal spatial frequency response by adjusting parameters of the self-adaptive filter; parameters of the adaptive filter include filter size, standard deviation and the like;
fusing the filtered images by adopting a weighted average method to obtain a new image which contains information from all input images;
performing inverse pyramid reconstruction on the fused image to obtain a final fused image;
outputting the fused image, and storing or displaying.
4. The method for identifying the power quality disturbance based on the multi-mode image fusion and ResNetXt-50 according to claim 1, wherein the method comprises the following steps of: the identification of the power quality disturbance type is carried out by utilizing a ResNetXt-50 model, and specifically comprises the following steps:
inputting the fused image into a ResNetXt-50 network, and performing autonomous feature extraction and identification to obtain a corresponding power quality disturbance signal class; the image size and the output parameters of the input layer of the ResNetXt-50 network are respectively consistent with the size and the identification type of the fused image.
5. The method for identifying the power quality disturbance based on the multi-mode image fusion and ResNetXt-50 according to claim 1, wherein the method comprises the following steps of: comprising a computer readable storage medium, embodied as:
the storage medium stores a computer program for executing a multi-modal image fusion and ResNetXt-50 based power quality disturbance recognition method according to any one of claims 1-4.
6. The method for identifying the power quality disturbance based on the multi-mode image fusion and ResNetXt-50 according to claim 1, wherein the method comprises the following steps of: the electronic equipment comprises the following components:
a processor; a memory for storing the processor-executable instructions; the processor is configured to read the executable instructions from the memory and execute the instructions to implement a multi-modal image fusion and ResNetXt-50-based power quality disturbance recognition method as recited in any one of claims 1-4.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311759114.XA CN117892059A (en) | 2023-12-20 | 2023-12-20 | Electric energy quality disturbance identification method based on multi-mode image fusion and ResNetXt-50 |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311759114.XA CN117892059A (en) | 2023-12-20 | 2023-12-20 | Electric energy quality disturbance identification method based on multi-mode image fusion and ResNetXt-50 |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN117892059A true CN117892059A (en) | 2024-04-16 |
Family
ID=90651577
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311759114.XA Pending CN117892059A (en) | 2023-12-20 | 2023-12-20 | Electric energy quality disturbance identification method based on multi-mode image fusion and ResNetXt-50 |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117892059A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118410406A (en) * | 2024-07-02 | 2024-07-30 | 国网山东省电力公司曲阜市供电公司 | Power quality disturbance recognition method and system for power distribution network based on deep learning |
-
2023
- 2023-12-20 CN CN202311759114.XA patent/CN117892059A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN118410406A (en) * | 2024-07-02 | 2024-07-30 | 国网山东省电力公司曲阜市供电公司 | Power quality disturbance recognition method and system for power distribution network based on deep learning |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110738697B (en) | Monocular depth estimation method based on deep learning | |
| CN109191382B (en) | Image processing method, device, electronic equipment and computer readable storage medium | |
| CN110232341B (en) | Semi-supervised learning image identification method based on convolution-stacking noise reduction coding network | |
| CN112132959B (en) | Digital rock core image processing method and device, computer equipment and storage medium | |
| CN111310666B (en) | High-resolution image ground feature identification and segmentation method based on texture features | |
| CN109636722B (en) | Method for reconstructing super-resolution of online dictionary learning based on sparse representation | |
| CN117892059A (en) | Electric energy quality disturbance identification method based on multi-mode image fusion and ResNetXt-50 | |
| CN112883227B (en) | Video abstract generation method and device based on multi-scale time sequence characteristics | |
| CN115511767B (en) | Self-supervised learning multi-modal image fusion method and application thereof | |
| CN115496928A (en) | Multi-modal image feature matching method based on multi-feature matching | |
| CN114723583A (en) | Unstructured electric power big data analysis method based on deep learning | |
| CN112686830B (en) | Super-resolution method of single depth map based on image decomposition | |
| CN109002771B (en) | Remote sensing image classification method based on recurrent neural network | |
| CN117974693B (en) | Image segmentation method, device, computer equipment and storage medium | |
| CN115984110A (en) | Swin-transform-based second-order spectral attention hyperspectral image super-resolution method | |
| CN115983370A (en) | Scattered data interpolation model training method, interpolation method and device | |
| CN113723294B (en) | Data processing method and device and object recognition method and device | |
| CN111368733A (en) | Three-dimensional hand posture estimation method based on label distribution learning, storage medium and terminal | |
| CN117593666A (en) | Geomagnetic station data prediction method and system for aurora image | |
| CN116310851B (en) | Remote sensing image change detection method | |
| CN117651144A (en) | Deep learning-based building point cloud compression method and system | |
| CN117079005A (en) | Optical cable fault monitoring method, system, device and readable storage medium | |
| CN111724309B (en) | Image processing method and device, training method of neural network and storage medium | |
| CN109584194B (en) | Hyperspectral image fusion method based on convolution variation probability model | |
| CN116704378A (en) | Homeland mapping data classification method based on self-growing convolution neural network |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication |