CN115761211A - Petrochemical enterprise pump room equipment temperature monitoring method and device based on RGB image and thermal imaging coupling - Google Patents
Petrochemical enterprise pump room equipment temperature monitoring method and device based on RGB image and thermal imaging coupling Download PDFInfo
- Publication number
- CN115761211A CN115761211A CN202111020351.5A CN202111020351A CN115761211A CN 115761211 A CN115761211 A CN 115761211A CN 202111020351 A CN202111020351 A CN 202111020351A CN 115761211 A CN115761211 A CN 115761211A
- Authority
- CN
- China
- Prior art keywords
- pixel
- image
- temperature
- target
- equipment
- 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
Images
Landscapes
- Radiation Pyrometers (AREA)
Abstract
The invention discloses a petrochemical enterprise pump room equipment temperature monitoring method and device based on RGB image and thermal imaging coupling. Firstly, constructing training data, pre-training a segmentation algorithm, and extracting pixel characteristics of a target equipment region based on a Gaussian mixture model; then, judging whether all pixels in the RGB image to be segmented belong to a target area or not according to the pre-training model, and further updating parameters of the Gaussian mixture model; then, constructing a topological graph by taking the pixels as nodes, taking the similarity between the pixels and the target area and the similarity between the pixel nodes as edge weights, and dividing the topological graph by using a minimum segmentation algorithm to finish image segmentation; and finally, positioning a target equipment area according to the image segmentation result, matching the temperature image to obtain the surface temperature data of the equipment, identifying the abnormal area of the surface temperature of the equipment according to the improved OSTU algorithm, and judging whether the abnormal area exceeds a preset threshold value.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to an image segmentation method based on multivariate information coupling.
Background
With the development of machine vision technology, image processing, image analysis, and image segmentation have been applied to various industries. In a complex context, accurate positioning of the detection object remains an important research problem in machine vision. In the pump house inspection task of the petrochemical enterprise, temperature information of the pump house is acquired through thermal imaging, but the whole image is not interested by people, but some areas for representing equipment are interested by people, and whether the surface temperature of the equipment is in a proper range can be confirmed only by segmenting the equipment and a background. The device and the background usually have features with different properties, and in order to identify and analyze the target device, the target device region needs to be segmented in an RGB space by using the feature difference between the background and the target region.
As a basis for image analysis, image segmentation is one of the most basic and difficult problems in the field of machine vision. The purpose of image segmentation is to separate a target device from a background, noise often exists in an image, and meanwhile, a target area is affected by factors such as angle, lighting conditions and shielding, so that the image segmentation problem becomes more complicated. With the development of statistical learning technology, image segmentation is converted into a pixel classification problem, and the degree of correlation between pixels is further utilized to further improve the segmentation precision, which has become a trend of image segmentation. Therefore, how to embody the characteristic difference between the background and the target area and obtain the relation information between the pixels has important research significance and application value.
Disclosure of Invention
The invention aims to provide a method and a device for monitoring the temperature of pump room equipment of a petrochemical enterprise based on RGB image and thermal imaging coupling, which aim to solve the technical problems of converting image segmentation into pixel classification, improving segmentation precision by utilizing the association degree between pixels, further reflecting the characteristic difference between a background and a target area, acquiring the relationship information between pixels and the like.
In order to achieve the purpose of the invention, the technical scheme adopted by the invention is as follows:
a petrochemical enterprise pump room equipment temperature monitoring method based on RGB image and thermal imaging coupling is suitable for automatic monitoring of computer equipment, and comprises the following steps:
(1) Pre-training images, namely extracting pixel distribution characteristics of a target area and a background area by using a Gaussian Mixture Model (GMM) based on labeling RGB image data;
(2) Distributing Gaussian components in GMM (Gaussian mixture model) for all pixels in the image to be segmented according to the pre-training result, and judging whether each pixel point of the image to be segmented belongs to a target area;
(3) Establishing a topological graph by taking each pixel as a node, adding two terminal nodes at the same time, wherein one terminal node represents a background region, the other terminal node represents a target region, the terminal nodes establish undirected connection with all the nodes, the edge weight is calculated according to the entropy of probability distribution of the nodes belonging to the background or the target region, the neighborhood pixel nodes perform undirected connection, and the edge weight is calculated according to the similarity of the pixel nodes;
(4) Segmenting the topological graph generated in the step (3) by a graph minimal cut algorithm, and dividing all pixels into a background or a target area;
(5) Carrying out digital processing on the original temperature image, and positioning the surface temperature of the target equipment according to a 0-1 matrix of an image segmentation result;
(6) And dividing the surface temperature data of the target equipment, and identifying the temperature abnormal area.
In the step (4), characteristic information of a target area image and a background image based on GMM representation is obtained in a pre-training mode, gaussian components with the maximum probability are automatically distributed to each pixel of an image to be segmented, and whether the pixel belongs to a target area is judged;
the pixel x is the target pixel, and the RGB value of the pixel x is substituted into each gaussian component in the target region GMM, where the pixel x is most likely to be generated, i.e. belongs to the kth pixel x The individual gaussian components:
θ＝{π(a,k),u(a,k),∑(,K),a＝0,1,k＝1......K}
wherein, L represents a likelihood function, x is a pixel RGB vector, a takes a value of 0 or 1 to represent whether the pixel belongs to a target region (a takes 1 to represent that the pixel belongs to the target region, and a takes 0 to represent that the pixel does not belong to the target region). In the temperature monitoring method for the pump house equipment of the petrochemical enterprise based on the coupling of the RGB image and the thermal imaging, in the step (6), the coupling of the temperature image and the RGB image can be realized by the image segmentation and the target equipment identification, the temperature image is subjected to digital processing, and the surface temperature of the target equipment is positioned according to a 01 matrix of an image segmentation result; the OSTU algorithm is improved, the multi-classification OSTU algorithm is provided for segmenting the temperature map, and the method aims at monitoring different temperature abnormal areas of equipment at the same time:
wherein M is the number of the division categories, p is each grouping frequency, N is the grouping number of the division histograms, and T is the inter-category segmentation boundary vector.
According to the petrochemical enterprise pump room equipment temperature monitoring method based on RGB image and thermal imaging coupling, temperature abnormal region classification points are converted into stagnation points based on an outer envelope curve of a gray distribution histogram, different equipment temperatures are different, areas in the image are different, the number of pixels of different gray values in the gray distribution histogram is different, each temperature similar region is a peak on the outer envelope curve of the gray distribution histogram, and therefore the image is segmented according to the stagnation points between the peaks, and different temperature abnormal regions can be obtained.
The utility model provides a petrochemical industry pump house equipment temperature monitoring devices based on RGB image and thermal imaging coupling, includes image acquisition equipment, treater and storage device, and wherein the storage device is applicable to the storage many instructions, and its instruction is applicable to the treater and loads and carry out.
Collecting RGB images of the internal environment of a pump room of a petrochemical enterprise, using the RGB images as training data and marking the RGB images;
step 1: collecting a certain number of RGB images and corresponding temperature images of the internal environment of the pump room of the petrochemical enterprise as training data;
step 2: labeling the RGB image, and manually separating a target equipment area from a background area;
and step 3: respectively carrying out unsupervised learning on a background area and a target area in a training set, representing pixels in different areas by using a Gaussian mixture model to carry out cluster analysis, and training to obtain GMM model parameters;
calculating and marking pixels as a target area or a background, and updating parameters and pixel weights;
step 1: calculating the probability that all pixels belong to a background or a target area according to the GMM model parameters obtained by training, further establishing an initial judgment matrix, and marking that the pixel belongs to the target area or the background;
and 2, step: for given image data Z, learning parameters that optimize the GMM; calculating the probability of each pixel belonging to each Gaussian component according to the pixel segmentation result of the target area and the background in the judgment matrix, judging which Gaussian component the pixel belongs to, further updating GMM model parameters, and determining the weight of the pixel according to the number of the pixels in different Gaussian components;
and step 3: establishing a topological graph by taking each pixel as a node, simultaneously adding two terminal nodes, wherein one terminal node represents a background region, the other terminal node represents a target region, the terminal nodes establish undirected connection with all the nodes, the edge weight is calculated according to the entropy of probability distribution of the nodes belonging to the background or the target region, the neighboring pixel nodes are connected undirectly, and the edge weight is calculated according to the similarity of the pixel nodes;
and 4, step 4: segmenting the topological graph generated in the step 3 by a graph minimal cut algorithm, and dividing all pixels into a background or a target area;
and 5: returning to the step 2, updating the model parameters, and performing iterative optimization until convergence;
thirdly, positioning the surface temperature of the equipment and detecting whether the temperature is abnormal or not;
step 1: carrying out digital processing on the temperature image, and positioning the surface temperature of the target equipment according to the 01 matrix of the image segmentation result;
step 2: and carrying out OSTU segmentation on the surface temperature data of the target equipment, identifying an abnormal temperature area, positioning the average temperature of the abnormal area, judging whether the average temperature is within a preset threshold range, and further determining whether to send alarm information.
The invention has the beneficial effects that: compared with the prior art, the characteristic information of the target area image and the background image based on GMM representation is obtained in a pre-training mode, and further the image to be segmented can be preliminarily estimated to divide the background and the target area; in the image segmentation process, artificial information interaction is not needed, meanwhile, target identification can be carried out on the segmented target area, and whether pre-trained target equipment is contained in the image to be segmented or not is detected. The segmentation of the image and the identification of the target equipment can realize the coupling of the temperature image and the RGB image, locate the abnormal area of the surface temperature of the equipment, eliminate the influence of the background temperature on the temperature monitoring of the target equipment in the inspection process and improve the detection precision.
Drawings
FIG. 1 is a flow chart of the present invention.
FIG. 2 is a schematic diagram of the present invention.
Detailed Description
The invention relates to a temperature monitoring method of petrochemical enterprise pump room equipment coupled by image and thermal imaging, which is suitable for computer equipment to carry out target detection and comprises the following steps:
(1) Pre-training images, namely extracting pixel distribution characteristics of a target region and a background region by using a Gaussian Mixture Model (GMM) based on labeled RGB image data;
(2) Distributing Gaussian components in GMM (Gaussian mixture model) for all pixels in the image to be segmented according to the pre-training result, and judging whether each pixel point of the image to be segmented belongs to a target area;
(3) Establishing a topological graph by taking each pixel as a node, simultaneously adding two terminal nodes, wherein one terminal node represents a background region, the other terminal node represents a target region, the terminal nodes establish undirected connection with all the nodes, the edge weight is calculated according to the entropy of probability distribution of the nodes belonging to the background or the target region, the neighboring pixel nodes are connected undirectly, and the edge weight is calculated according to the similarity of the pixel nodes;
(4) Segmenting the topological graph generated in the step (3) by a graph minimal cut algorithm, and dividing all pixels into a background or a target area;
(5) Carrying out digital processing on the original temperature image, and positioning the surface temperature of the target equipment according to a 0-1 matrix of an image segmentation result;
(6) And dividing the surface temperature data of the target equipment, and identifying the temperature abnormal area.
In the temperature monitoring method for the petrochemical enterprise pump room equipment based on RGB image and thermal imaging coupling, in the step (4), characteristic information of a target area image and a background image based on GMM representation is obtained in a pre-training mode, gaussian components with the maximum probability are automatically distributed to each pixel of an image to be segmented, and whether the pixel belongs to a target area is further judged;
the pixel x is the target pixel, and the RGB value of the pixel x is substituted into each gaussian component in the target region GMM, where the pixel x is most likely to be generated, i.e. belongs to the kth pixel x The individual gaussian components:
θ＝{π(a,k),u(a,k),∑(,K),a＝0,1,k＝1......K}
wherein, L represents a likelihood function, x is a pixel RGB vector, a takes a value of 0 or 1 to represent whether the pixel belongs to a target region (a takes 1 to represent that the pixel belongs to the target region, and a takes 0 to represent that the pixel does not belong to the target region). According to the temperature monitoring method of the petrochemical enterprise pump house equipment based on RGB image and thermal imaging coupling, in the step (6), the segmentation of the image and the identification of target equipment can realize the coupling of the temperature image and the RGB image, the temperature image is subjected to digital processing, and the surface temperature of the target equipment is positioned according to a 01 matrix of an image segmentation result; the OSTU algorithm is improved, a multi-classification OSTU algorithm is provided for segmenting a temperature map, and the method aims at monitoring different temperature abnormal areas of equipment at the same time:
wherein M is the number of the division categories, p is each grouping frequency, N is the grouping number of the division histograms, and T is the inter-category segmentation boundary vector.
According to the petrochemical enterprise pump room equipment temperature monitoring method based on RGB image and thermal imaging coupling, temperature abnormal region classification points are converted into stagnation points based on an outer envelope curve of a gray distribution histogram, different equipment temperatures are different, areas in the image are different, the number of pixels of different gray values in the gray distribution histogram is different, each temperature similar region is a peak on the outer envelope curve of the gray distribution histogram, and therefore the image is segmented according to the stagnation points between the peaks, and different temperature abnormal regions can be obtained. The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
As shown in fig. 1, an embodiment of the present invention provides a temperature monitoring apparatus for a pump room device of a petrochemical enterprise based on RGB image and thermal imaging coupling, including an image capturing device, a processor, and a storage device, where the storage device is adapted to store a plurality of instructions, and the instructions are adapted to be loaded and executed by the processor.
1. Pre-training
Step 1: collecting a certain number of RGB images and corresponding temperature images of the internal environment of the pump room of the petrochemical enterprise as training data, and marking a target equipment area in the RGB images;
step 2: in an RGB color space, taking Euclidean distance between RGB values of each pixel as evaluation of similarity between the pixels, respectively carrying out K-means clustering on all pixels of a target equipment area and a background area, and dividing all pixels of the target area and the background area into K classes;
and step 3: in the RGB color space, the target and background are modeled with a full covariance GMM of K gaussian components, respectively. Where GMM model D (x) may be expressed as:
where x is the pixel component, g i Is the ith Gaussian component, π i For the ith gaussian component weight, u and Σ represent the mean vector and covariance matrix of the gaussian components. For each pixel, either a gaussian component from the target region GMM or a gaussian component from the background GMM. And (3) according to the clustering result in the step (2), regarding each type of pixel as a Gaussian component and calculating a mean vector and a covariance matrix of the pixel.
2. Image segmentation
Step 1: calculating the probability that all pixels in the image to be segmented belong to the background and the target area according to the GMM model parameters obtained by training, and judging whether the pixels belong to the target area or the background area according to the probability value; step 2: assigning Gaussian components in GMM to all pixels in the image to be segmented, for example, pixel x is a target pixel, substituting the RGB value of pixel x into each Gaussian component in the target region GMM, and generating pixel x most probably with the highest probability, that is, pixel x belongs to kth x Individual gaussian components:
wherein, L represents the likelihood function, x is the pixel RGB vector, and a value 0,1 represents whether the pixel belongs to the target area.
And step 3: according to the result of the step 2, all pixel points belonging to each Gaussian component can be obtained, a mean value and a covariance matrix are calculated for all the pixel points belonging to a certain Gaussian component, and meanwhile, the weight of each Gaussian component is determined according to the number of pixels in different Gaussian components, so that GMM model parameters are updated;
and 4, step 4: and taking each pixel as a node to establish a topological graph, simultaneously adding two terminal nodes, wherein the S node represents a background area, the T node represents a target area, the terminal nodes establish undirected connection with all the nodes, and the neighborhood pixel nodes perform undirected connection.
Calculating the edge weight W between the terminal node and the pixel node according to the entropy of the probability distribution of the node belonging to the background or the target area 1 ：
Wherein, w 1 i Representing the edge weight of the ith pixel and the terminal node; calculating edge weight W according to the similarity of pixel nodes and neighborhood nodes in RGB space 2 ：
Wherein m and n represent two pixel nodes in the neighborhood, [ 2 ]]Representing a symbolic function, a n And (3) indicating whether the nth pixel belongs to the target area, wherein gamma is a hyper-parameter and is generally 50, beta is determined by the contrast of the image to be segmented, and the image with small contrast is a larger value.
And 4, step 4: segmenting the topological graph generated in the step 3 by utilizing a min-cut algorithm, and dividing all pixels into a background or a target area;
and 5: and returning to the step 2, updating the model parameters, and performing iterative optimization until convergence.
3. Temperature monitoring
Step 1: carrying out digital processing on the temperature image, and positioning the surface temperature of the target equipment according to a 01 matrix of an image segmentation result;
step 2: and dividing the surface temperature data of the target equipment, and identifying the temperature abnormal area. The invention improves the OSTU algorithm in order to more accurately describe the abnormal temperature condition, provides the multi-classification OSTU algorithm to segment the temperature map, and aims to simultaneously monitor different abnormal temperature areas of the equipment:
wherein M is the number of the division categories, p is each grouping frequency, N is the grouping number of the division histograms, and T is the inter-category segmentation boundary vector. In order to solve the model, the classification points are converted into stagnation points based on an outer envelope curve of a gray distribution histogram, different equipment temperatures are different, areas in an image are different, the number of pixels of different gray values in the gray distribution histogram is different, each temperature similar area is a peak on the outer envelope curve of the gray distribution histogram, and therefore the image is segmented according to the stagnation points between the peaks, and different temperature abnormal areas can be obtained.
The mapping structure of the present invention is shown in fig. 2, in which the original image is on the left side.
Claims (5)
1. A petrochemical enterprise pump room equipment temperature monitoring method based on RGB image and thermal imaging coupling is suitable for automatic monitoring on computer equipment, and comprises the following steps:
(1) Pre-training images, namely extracting pixel distribution characteristics of a target region and a background region by using a Gaussian Mixture Model (GMM) based on labeled RGB image data;
(2) Distributing Gaussian components in GMM (Gaussian mixture model) for all pixels in the image to be segmented according to the pre-training result, and judging whether each pixel point of the image to be segmented belongs to a target area;
(3) Establishing a topological graph by taking each pixel as a node, simultaneously adding two terminal nodes, wherein one terminal node represents a background region, the other terminal node represents a target region, the terminal nodes establish undirected connection with all the nodes, the edge weight is calculated according to the entropy of probability distribution of the nodes belonging to the background or the target region, the neighboring pixel nodes are connected undirectly, and the edge weight is calculated according to the similarity of the pixel nodes;
(4) Segmenting the topological graph generated in the step (3) by a graph minimal cut algorithm, and dividing all pixels into a background or a target area;
(5) Carrying out digital processing on the original temperature image, and positioning the surface temperature of the target equipment according to a 0-1 matrix of an image segmentation result;
(6) And dividing the surface temperature data of the target equipment, and identifying the temperature abnormal area.
2. The method for monitoring the temperature of the pumping room equipment of the petrochemical enterprise based on the coupling of the RGB image and the thermal imaging as claimed in claim 1, wherein in the step (4), the characteristic information of the target area image and the background image based on the GMM representation is obtained in a pre-training mode, and a Gaussian component with the maximum probability is automatically allocated to each pixel of the image to be segmented, so as to judge whether the pixel belongs to the target area;
pixel x is the target pixel, substituting the RGB value of pixel x into each of the target regions GMMOf the Gaussian components, the probability is the highest, i.e., the pixel x is most likely to be generated, i.e., the pixel x belongs to the kth x Individual gaussian components:
θ＝{π(a,k),u(a,k),∑(,K),a＝0,1,k＝1......K}
wherein, L represents a likelihood function, x is a pixel RGB vector, a takes a value of 0 or 1 to represent whether the pixel belongs to a target area, a takes 1 to represent that the pixel belongs to the target area, and a takes 0 to represent that the pixel does not belong to the target area.
3. The method for monitoring the temperature of the pumping room equipment of the petrochemical enterprise based on the coupling of the RGB image and the thermal imaging as claimed in claim 1, wherein in the step (6), the segmentation of the image and the identification of the target equipment can realize the coupling of the temperature image and the RGB image, the temperature image is processed digitally, and the surface temperature of the target equipment is positioned according to a 0-1 matrix of the image segmentation result; the OSTU algorithm is improved, the multi-classification OSTU algorithm is provided for segmenting the temperature map, and the method aims at monitoring different temperature abnormal areas of equipment at the same time:
wherein M is the number of the division categories, p is each grouping frequency, N is the grouping number of the division histograms, and T is the inter-category segmentation boundary vector.
4. The method as claimed in claim 3, wherein the temperature monitoring method for the equipment in the pump room of the petrochemical enterprise based on the coupling of the RGB image and the thermal imaging converts the classification points of the abnormal temperature areas into stagnation points based on an outer envelope curve of a gray distribution histogram, different equipment temperatures are different, areas in the image are different, the number of pixels with different gray values in the gray distribution histogram is different, each temperature similar area on the outer envelope curve of the gray distribution histogram is a peak, and therefore the image is segmented according to the stagnation points between the peaks, and different abnormal temperature areas can be obtained.
5. The utility model provides a petrochemical industry pump house equipment temperature monitoring devices based on RGB image and thermal imaging coupling, includes image acquisition equipment, treater and storage device, and wherein the storage device is applicable to the storage many instructions, and its instruction is applicable to the treater and loads and carry out.
Collecting RGB images of the internal environment of a pump room of a petrochemical enterprise, using the RGB images as training data and marking the RGB images;
step 1: collecting a certain number of RGB images and corresponding temperature images of the internal environment of the pump room of the petrochemical enterprise as training data;
step 2: labeling the RGB image, and manually separating a target equipment area from a background area;
and step 3: respectively carrying out unsupervised learning on a background area and a target area in a training set, representing pixels in different areas by using a Gaussian mixture model to carry out cluster analysis, and training to obtain GMM model parameters;
calculating and marking pixels as a target area or a background, and updating parameters and pixel weights;
step 1: calculating the probability that all pixels belong to a background or a target area according to the GMM model parameters obtained by training, further establishing an initial judgment matrix, and marking that the pixel belongs to the target area or the background;
step 2: for given image data Z, learning parameters that optimize the GMM; calculating the probability of each pixel belonging to each Gaussian component according to the pixel segmentation result of the target area and the background in the judgment matrix, judging which Gaussian component the pixel belongs to, further updating GMM model parameters, and determining the weight of the pixel according to the number of the pixels in different Gaussian components;
and step 3: establishing a topological graph by taking each pixel as a node, simultaneously adding two terminal nodes, wherein one terminal node represents a background region, the other terminal node represents a target region, the terminal nodes establish undirected connection with all the nodes, the edge weight is calculated according to the entropy of probability distribution of the nodes belonging to the background or the target region, the neighboring pixel nodes are connected undirectly, and the edge weight is calculated according to the similarity of the pixel nodes;
and 4, step 4: segmenting the topological graph generated in the step 3 by a graph minimal cut algorithm, and dividing all pixels into a background or a target area;
and 5: returning to the step 2, updating the model parameters, and performing iterative optimization until convergence;
thirdly, positioning the surface temperature of the equipment and detecting whether the temperature is abnormal or not;
step 1: carrying out digital processing on the temperature image, and positioning the surface temperature of the target equipment according to a 01 matrix of an image segmentation result;
and 2, step: and carrying out OSTU segmentation on the surface temperature data of the target equipment, identifying an abnormal temperature area, positioning the average temperature of the abnormal area, judging whether the average temperature is within a preset threshold range, and further determining whether to send out alarm information.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111020351.5A CN115761211A (en) | 2021-09-01 | 2021-09-01 | Petrochemical enterprise pump room equipment temperature monitoring method and device based on RGB image and thermal imaging coupling |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111020351.5A CN115761211A (en) | 2021-09-01 | 2021-09-01 | Petrochemical enterprise pump room equipment temperature monitoring method and device based on RGB image and thermal imaging coupling |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115761211A true CN115761211A (en) | 2023-03-07 |
Family
ID=85332174
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111020351.5A Pending CN115761211A (en) | 2021-09-01 | 2021-09-01 | Petrochemical enterprise pump room equipment temperature monitoring method and device based on RGB image and thermal imaging coupling |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115761211A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117951350A (en) * | 2024-03-26 | 2024-04-30 | 西安航天动力试验技术研究所 | Factory building temperature data visualization method, system, equipment and storage medium |
-
2021
- 2021-09-01 CN CN202111020351.5A patent/CN115761211A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117951350A (en) * | 2024-03-26 | 2024-04-30 | 西安航天动力试验技术研究所 | Factory building temperature data visualization method, system, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109977808B (en) | Wafer surface defect mode detection and analysis method | |
| CN111860670B (en) | Domain adaptive model training method, image detection method, device, equipment and medium | |
| CN115082683B (en) | Injection molding defect detection method based on image processing | |
| CN110148130B (en) | Method and device for detecting part defects | |
| CN107437245B (en) | High-speed railway contact net fault diagnosis method based on deep convolutional neural network | |
| CN114419025A (en) | Fiberboard quality evaluation method based on image processing | |
| CN115351598A (en) | Numerical control machine tool bearing detection method | |
| CN113724231B (en) | Industrial defect detection method based on semantic segmentation and target detection fusion model | |
| CN115294140A (en) | Hardware part defect detection method and system | |
| CN109801305B (en) | SAR image change detection method based on deep capsule network | |
| CN112200121A (en) | Hyperspectral unknown target detection method based on EVM and deep learning | |
| CN116152242B (en) | Visual detection system of natural leather defect for basketball | |
| CN111369526A (en) | Multi-type old bridge crack identification method based on semi-supervised deep learning | |
| CN117705815B (en) | Printing defect detection method based on machine vision | |
| CN116309575A (en) | Electric plug production quality detection method based on image processing | |
| CN115294109A (en) | Real wood board production defect identification system based on artificial intelligence, and electronic equipment | |
| CN115082781A (en) | Ship image detection method and device and storage medium | |
| CN116206208B (en) | Forestry plant diseases and insect pests rapid analysis system based on artificial intelligence | |
| CN113313179A (en) | Noise image classification method based on l2p norm robust least square method | |
| CN113344047A (en) | Platen state identification method based on improved K-means algorithm | |
| CN111199228B (en) | License plate positioning method and device | |
| CN115082449B (en) | Electronic component defect detection method | |
| CN116523922A (en) | Bearing surface defect identification method | |
| CN116664540A (en) | Rubber sealing ring surface defect detection method based on Gaussian line detection | |
| CN116188445A (en) | Product surface defect detection and positioning method and device and terminal equipment |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |