CN115410003A - Power distribution room temperature early warning method, system and medium based on deep learning and infrared detection - Google Patents
Power distribution room temperature early warning method, system and medium based on deep learning and infrared detection Download PDFInfo
- Publication number
- CN115410003A CN115410003A CN202211062359.2A CN202211062359A CN115410003A CN 115410003 A CN115410003 A CN 115410003A CN 202211062359 A CN202211062359 A CN 202211062359A CN 115410003 A CN115410003 A CN 115410003A
- Authority
- CN
- China
- Prior art keywords
- infrared
- thermal imaging
- picture
- early warning
- power distribution
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/56—Extraction of image or video features relating to colour
-
- 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
- G06N3/084—Backpropagation, e.g. using gradient descent
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Physics & Mathematics (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Multimedia (AREA)
- Software Systems (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Medical Informatics (AREA)
- Databases & Information Systems (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Molecular Biology (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Radiation Pyrometers (AREA)
Abstract
The invention discloses a power distribution room temperature early warning method, a system and a medium based on deep learning and infrared detection, wherein the method comprises the steps of obtaining a thermal imaging picture according to thermal imaging equipment; carrying out picture preprocessing on the thermal imaging picture; inputting the preprocessed thermal imaging picture into an infrared temperature measurement identification model for detection; correcting the infrared temperature measurement result; judging a temperature interval according to the color depth of the identification picture; and outputting alarm information. The infrared thermal imaging monitoring system has the advantages of real-time detection, accurate alarm, automatic pushing of alarm information and the like, 24 x 7 all-weather and non-omission real-time detection of the surface temperature of the transformer in the power room is realized through video monitoring, the real-time detection is compared with a normal standard, a visible thermal distribution image is analyzed, infrared thermal imaging identification is carried out according to the change condition of an infrared thermal image, online monitoring and safety early warning are carried out, and the safety operation emergency guarantee capability of equipment in the power distribution room is improved.
Description
Technical Field
The invention belongs to the technical field of thermal imaging temperature measurement, and particularly relates to a power distribution room temperature early warning method, system and medium based on deep learning and infrared detection.
Background
With the increasing demand of people for energy resources, the load bearing capacity of power equipment is also increasing, and transformer equipment as important equipment in a power system can be subjected to the action of power during operation, so that the temperature is increased. Although there are many types of failures of the transformer, the surface temperature of the transformer rises, and when a short circuit caused by corrosion of the metal winding occurs, the surface temperature of the transformer rises sharply. According to the research, the infrared temperature measurement technology is adopted to detect the temperature according to the temperature and the change condition of the surface of the transformer, the temperature and the change condition are compared with normal standards, the visible heat distribution image is analyzed, the infrared thermal imaging identification is carried out according to the change condition of the infrared thermal image, the transformer in the power distribution room is subjected to online monitoring and safety early warning, and the emergency guarantee capability of the safe operation of equipment in the power distribution room is promoted.
Disclosure of Invention
The invention mainly aims to overcome the defects and shortcomings of the prior art and provides a power distribution room temperature early warning method, system and medium based on deep learning and infrared detection.
In order to achieve the purpose, the invention adopts the following technical scheme:
one aspect of the invention provides a power distribution room temperature early warning method based on deep learning and infrared detection, which comprises the following steps:
acquiring a thermal imaging picture according to a thermal imaging device;
performing picture pretreatment on the thermal imaging picture;
inputting the preprocessed thermal imaging picture into an infrared temperature measurement identification model for detection;
correcting the infrared temperature measurement result;
judging a temperature interval according to the color depth of the identification picture;
and outputting alarm information.
As a preferred technical solution, the acquiring a thermal imaging picture according to a thermal imaging device specifically includes:
and setting a video frame selection interval, and converting the intercepted single-frame thermal imaging picture into a format which can be processed by the model.
As a preferred technical solution, the image preprocessing on the thermal imaging image specifically includes:
and removing partial noise interference in the picture and enabling the data preprocessing operation of the training picture and the picture to be predicted to be consistent.
As a preferred technical solution, the infrared thermometry identification model is an improved SSD detection framework, and specifically includes:
a convolutional neural network MobileNet is adopted to replace a VGG-16 network in a traditional SSD detection framework;
the MobileNet replaces conventional convolutional layers with depth separable convolutions, decomposing standard convolutions into depth convolutions and point-by-point convolutions.
As a preferred technical solution, the infrared temperature measurement recognition model adopts a Batch Normalization layer during training, and the Batch Normalization layer is incorporated into the convolutional layer at the deployment stage of the network model.
As a preferred technical solution, the Batch Normalization layer performs Normalization and scaling operations;
the normalization is specifically as follows:
wherein mu is a mean value, sigma is a variance, and epsilon is a small number;
the scaling is specifically as follows:
where γ is the scaling factor, β is the offset, W new As new convolution weights, b old Is originally biased, b new Is the new bias.
As a preferred technical scheme, the temperature correction is carried out on the infrared temperature measurement result through a BP neural network, and the method specifically comprises the following steps:
the input of the BP neural network input layer is an infrared temperature measurement result of a measured target and a measurement distance of the target, and the output is corrected temperature;
the hidden layer of the BP neural network comprises m nodes, the input layer comprises n nodes, and the output layer comprises o nodes; the transfer function of the neuron between the output layer and the hidden layer is a linear transfer function.
As a preferred technical solution, the input state x of the neuron between the output layer and the hidden layer of the BP neural network i And output y i Is a linear change, as follows:
y i ＝f(x i )
the corresponding outputs of the inputs of the BP neural network are:
E i ＝D i ，i＝1，2，…，o
wherein D is i I =1,2, \8230, o is the input of BP neural network respectively;
the input of the jth node in the hidden layer is:
F i ＝L j1 ×E 1 +L j2 ×E 2 +…+L jo ×E o +M j
the output of the jth node in the hidden layer is:
H j ＝f(x j )，i＝1，2，…，m
the input to the jth node in the output layer is:
the output of the jth node in the output layer is:
Y j ＝f(x j )＝f(k)＝k
wherein L is ji ，i＝1，2，…，o、
For the weight of the connection, M 2 、M j Is an offset value.
The invention also provides a power distribution room temperature early warning system based on deep learning and infrared detection, which is applied to the power distribution room temperature early warning method based on deep learning and infrared detection and comprises an image acquisition module, a preprocessing module, an infrared temperature measurement identification module, a temperature correction module and a result output module;
the image acquisition module is used for acquiring a thermal imaging picture according to thermal imaging equipment;
the preprocessing module is used for preprocessing the thermal imaging picture;
the infrared temperature measurement identification module is used for inputting the preprocessed thermal imaging picture into the infrared temperature measurement identification model for detection;
the temperature correction module is used for correcting the infrared temperature measurement result;
and the result output module is used for judging the temperature interval according to the color depth of the identification picture and outputting alarm information.
In another aspect of the present invention, a storage medium is further provided, which stores a program, and when the program is executed by a processor, the method for early warning of power distribution room temperature based on deep learning and infrared detection is implemented.
Compared with the prior art, the invention has the following advantages and beneficial effects:
(1) The infrared thermal imaging monitoring system has the advantages of real-time detection, accurate alarm, automatic pushing of alarm information and the like, 24 x 7 all-weather and non-omission real-time detection of the surface temperature of the transformer in the power room is realized through video monitoring, the real-time detection is compared with a normal standard, a visible thermal distribution image is analyzed, infrared thermal imaging identification is carried out according to the change condition of an infrared thermal image, online monitoring and safety early warning are carried out, and the safety operation emergency guarantee capability of equipment in the power distribution room is improved.
(2) The invention carries out model training based on a Tensorfolw deep learning framework, does not need manual monitoring and calculation processing after an algorithm model runs, and realizes all-round real-time monitoring and alarm pushing all day by accessing intelligent and automatic data of the model to model calculation and alarm result classification pushing.
Drawings
FIG. 1 is a flow chart of a power distribution room temperature early warning method based on deep learning and infrared detection according to an embodiment of the present invention;
FIG. 2 is a schematic structural diagram of an infrared thermometry identification model after improvement of an embodiment of the present invention;
FIG. 3 is a schematic structural diagram of a temperature-corrected BP neural network according to an embodiment of the present invention;
FIG. 4 is a schematic diagram of determining a temperature range according to an embodiment of the present invention;
FIG. 5 is a schematic diagram of a system according to an embodiment of the present invention;
fig. 6 is a schematic structural diagram of a storage medium according to an embodiment of the present invention.
Detailed Description
In order to make the technical solutions better understood by those skilled in the art, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. It should be apparent that the described embodiments are only a few embodiments of the present application, and not all 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 application.
Examples
The method comprises the steps of firstly obtaining a thermal imaging picture according to thermal imaging equipment, converting the intercepted single-frame picture into a picture in a JPG format which can be processed by a model, then carrying out picture preprocessing on a real-time thermal imaging picture shot by infrared thermal imaging equipment, carrying out preprocessing operation before recognition on the picture according to a preprocessing result, and then inputting the picture into an infrared imaging recognition model to detect temperature difference and infrared imaging color depth.
As shown in fig. 1, the embodiment provides a power distribution room temperature early warning method based on deep learning and infrared detection, which includes the following steps:
s1, acquiring a thermal imaging picture according to thermal imaging equipment;
and (3) comprehensively considering according to three aspects of scenes, requirements and performance, designing a reasonable video frame selection interval, and converting the intercepted single-frame thermal imaging picture into a JPG format picture which can be processed by the model.
S2, carrying out picture preprocessing on the thermal imaging picture;
and carrying out picture preprocessing on a real-time thermal imaging picture shot by the infrared thermal imaging equipment, removing partial noise interference in the picture and enabling the training picture to be consistent with the data preprocessing operation of the picture to be predicted.
And S3, infrared thermal imaging identification.
And inputting the preprocessed thermal imaging picture into an infrared temperature measurement identification model to detect the temperature difference and the infrared imaging color depth.
The infrared temperature measurement identification model is an improved SSD detection framework, and the SSD is adopted mainly because the SSD has a denser anchor frame arrangement on a feature map of a small receptive field. In infrared temperature measurement problem, this application has carried out some pertinence optimization to SSD, reduces the complexity of model and promotes the accuracy of priori hypothesis simultaneously to promote the accuracy of whole oil level discernment, specifically be:
the original SSD adopts a VGG network as a feature extraction network, and in order to adapt to the complex environment of an electric room, the application uses a more efficient and lightweight convolutional neural network MobileNet to replace a VGG-16 network (shown in figure 2) in the traditional SSD, so that compared with an SSD algorithm, the method has better environmental adaptability, robustness and higher precision.
MobileNet replaces the conventional convolutional layer with a depth separable convolution (depth separable convolutions), which decomposes the standard convolution into a depth convolution and a point-by-point convolution, and when the input Feature mAP is m × n × 16 and 32 channels are desired to be output, the convolution kernel should be 16 × 3 × 3 × 32, and can be decomposed into a depth convolution: 16 × 3 × 3, obtaining a feature map of 16 channels, point convolution: 16 × 1 × 1 × 32, if standard convolution is used, the amount calculated is: m × n × 16 × 3 × 3 × 32= m × n × 4608, and the amount of calculation after deconvolution by depth separable is: and m multiplied by n multiplied by 16 multiplied by 3+ m multiplied by n multiplied by 16 multiplied by 1 multiplied by 32= m multiplied by n multiplied by 656, the calculated amount and the parameter quantity of the convolutional neural network are reduced, and the network operation efficiency is improved.
Furthermore, a Batch Normalization layer is adopted in the process of training the infrared temperature measurement identification model, so that the training speed can be accelerated, and the network convergence speed can be increased; in the convolutional neural network, the Batch Normalization layer is generally placed behind the convolutional layer or the full-link layer, and after the Batch Normalization layer normalizes data, the problems of gradient disappearance and gradient explosion can be effectively solved, and the training fitting speed is accelerated. The Batch Normalization layer plays a certain positive role in the training stage of the deep convolutional neural network, but in the deployment stage of the network model, one more layer of calculation is added in the model prediction, the overall operation speed of the model is influenced, and the occupied space of a video memory and a memory is increased. Therefore, in the deployment phase of the network model, the Batch Normalization layer needs to be incorporated into the convolution layer to increase the speed of the network model.
Assuming that the input of each layer is represented as X, W is convolution weight, b is convolution offset, convolution operation is performed first, and the operation formula of the convolution layer is:
W×X+b
the Batch Normalization layer performs Normalization and scaling operations;
the normalization is specifically as follows:
wherein mu is a mean value, sigma is a variance, epsilon is a small number, and the prevention denominator is zero;
the second operation of the Batch Normalization layer is scaling: γ X + β;
after combining the convolutional layer and the Batch Normalization layer, we obtained:
where γ is the scaling factor, β is the offset, W new As new convolution weights, b new Is the new bias.
Further, the embodiment performs model training based on the Tensorfolw deep learning framework.
S4, correcting the infrared temperature measurement result;
further, temperature correction is carried out on the infrared temperature measurement result through a BP neural network;
generally, the intensity of infrared radiation emitted from the surface of an object is attenuated in air propagation, so that the infrared temperature measurement result of a measured target is usually lower than the actual temperature of the measured target, and the longer the distance is, the larger the actual difference is. Therefore, the infrared temperature measurement result of the measured object needs to be corrected. The BP neural network algorithm is widely applied to various scenes, and compared with a linear interpolation method and a multiple linear regression method, the BP neural network algorithm has strong nonlinear mapping capability, strong adaptability and high accuracy when the influence parameters of infrared temperature measurement are explored. The temperature correction module adopts a BP neural network to correct the temperature of the infrared temperature measurement result. As shown in the figure, the input of the input layer is the infrared temperature measurement result of the measured target and the measurement distance of the target, and the two types of data are input into the BP neural network to obtain the final corrected temperature.
The input of the BP neural network input layer is an infrared temperature measurement result of a measured target and a measurement distance of the target, and the output is corrected temperature;
as shown in fig. 3, the hidden layer of the BP neural network includes m nodes, the input layer includes n nodes, and the output layer includes o nodes; the transfer function of the neuron between the output layer and the hidden layer is a linear transfer function;
input state x of a neuron i And output y i Is a linear change, as follows:
y i ＝f(x i )
the corresponding outputs of the inputs of the BP neural network are:
E i ＝D i ，i＝1，2，…，o
wherein D is i I =1,2, \8230, o is the input of BP neural network respectively;
the input of the jth node in the hidden layer is:
F i ＝L j1 ×E 1 +L j2 ×E 2 +…+L jo ×E o +M j
the output of the jth node in the hidden layer is:
H j ＝f(x j )，i＝1，2，…，m
the input to the jth node in the output layer is:
the output of the jth node in the output layer is:
Y j ＝f(x j )＝f(k)＝k
wherein L is ji ，i＝1，2，…，o、
For the weight of the connection, M 2 、M j Is an offset value.
And S5, judging the temperature interval according to the color depth of the identification picture, as shown in figure 4.
And S6, outputting alarm information.
As shown in fig. 5, in another embodiment of the present application, a power distribution room temperature early warning system based on deep learning and infrared detection is provided, and the system includes an image acquisition module, a preprocessing module, an infrared temperature measurement identification module, a temperature correction module, and a result output module;
the image acquisition module is used for acquiring a thermal imaging picture according to thermal imaging equipment;
the preprocessing module is used for preprocessing the thermal imaging picture;
the infrared temperature measurement identification module is used for inputting the preprocessed thermal imaging picture into the infrared temperature measurement identification model for detection;
the temperature correction module is used for correcting the infrared temperature measurement result;
and the result output module is used for judging a temperature interval according to the color depth of the identification picture and outputting alarm information.
It should be noted that the system provided in the above embodiment is only illustrated by dividing the above function modules, and in practical applications, the above function allocation may be completed by different function modules according to needs, that is, the internal structure is divided into different function modules to complete all or part of the above described functions.
As shown in fig. 6, in another embodiment of the present application, there is further provided a storage medium storing a program, and when the program is executed by a processor, the method for early warning of power distribution room temperature based on deep learning and infrared detection in the foregoing embodiment is implemented, specifically:
acquiring a thermal imaging picture according to a thermal imaging device;
performing picture pretreatment on the thermal imaging picture;
inputting the preprocessed thermal imaging picture into an infrared temperature measurement identification model for detection;
correcting the infrared temperature measurement result;
judging a temperature interval according to the color depth of the identification picture;
and outputting alarm information.
It should be understood that portions of the present application may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
The above embodiments are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents thereof, and all such changes, modifications, substitutions, combinations, and simplifications are intended to be included in the scope of the present invention.
Claims (10)
1. A power distribution room temperature early warning method based on deep learning and infrared detection is characterized by comprising the following steps:
acquiring a thermal imaging picture according to a thermal imaging device;
performing picture pretreatment on the thermal imaging picture;
inputting the preprocessed thermal imaging picture into an infrared temperature measurement identification model for detection;
correcting the infrared temperature measurement result;
judging a temperature interval according to the color depth of the identification picture;
and outputting alarm information.
2. The deep learning and infrared detection-based power distribution room temperature early warning method according to claim 1, wherein the obtaining of the thermal imaging picture according to the thermal imaging device specifically comprises:
and setting a video frame selection interval, and converting the intercepted single-frame thermal imaging picture into a format which can be processed by the model.
3. The deep learning and infrared detection-based power distribution room temperature early warning method according to claim 1, wherein the image preprocessing of the thermal imaging image specifically comprises:
and removing partial noise interference in the picture and enabling the data preprocessing operation of the training picture and the picture to be predicted to be consistent.
4. The deep learning and infrared detection-based power distribution room temperature early warning method according to claim 1, wherein the infrared temperature measurement identification model is an improved SSD detection framework, and specifically comprises:
a convolutional neural network MobileNet is adopted to replace a VGG-16 network in a traditional SSD detection framework;
the MobileNet replaces the conventional convolutional layer with a deep separable convolution, decomposing the standard convolution into a deep convolution and a point-by-point convolution.
5. The deep learning and infrared detection-based power distribution room temperature early warning method according to claim 1, wherein a Batch Normalization layer is adopted in the infrared temperature measurement recognition model during training, and the Batch Normalization layer is merged into the convolutional layer in the deployment stage of the network model.
6. The deep learning and infrared detection-based power distribution room temperature early warning method according to claim 5, wherein the Batch Normalization layer performs Normalization and scaling operations;
the normalization is specifically as follows:
wherein mu is a mean value, sigma is a variance, and epsilon is a small number;
the scaling is specifically as follows:
where γ is the scaling factor, β is the offset, W new As new convolution weights, b old Is originally biased, b new Is the new bias.
7. The deep learning and infrared detection-based power distribution room temperature early warning method according to claim 1, characterized in that the infrared temperature measurement result is subjected to temperature correction through a BP neural network, specifically:
the input of the BP neural network input layer is an infrared temperature measurement result of a measured target and a measurement distance of the target, and the output is corrected temperature;
the hidden layer of the BP neural network comprises m nodes, the input layer comprises n nodes, and the output layer comprises o nodes; the transfer function of the neuron between the output layer and the hidden layer is a linear transfer function.
8. The deep learning and infrared detection-based power distribution room temperature early warning method as claimed in claim 7, wherein the input state x of the neuron between the output layer and the hidden layer of the BP neural network is the input state x of the neuron i And output y i Is a linear change, as follows:
y i ＝f(x i )
the corresponding outputs of the inputs of the BP neural network are:
E i ＝D i ，i＝1，2，…，o
wherein D is i I =1,2, \8230, o is the input of BP neural network respectively;
the input of the jth node in the hidden layer is:
F i ＝L j1 ×E 1 +L j2 ×E 2 +…+L jo ×E o +M j
the output of the jth node in the hidden layer is:
H j ＝f(x j )，i＝1，2，…，m
the input to the jth node in the output layer is:
the output of the jth node in the output layer is:
Y j ＝f(x j )＝f(k)＝k
wherein L is ji ，i＝1，2，…，o、
For the weight of the connection, M 2 、M j Is an offset value.
9. The power distribution room temperature early warning system based on deep learning and infrared detection is characterized by being applied to the power distribution room temperature early warning method based on deep learning and infrared detection in any one of claims 1-8, and comprising an image acquisition module, a preprocessing module, an infrared temperature measurement identification module, a temperature correction module and a result output module;
the image acquisition module is used for acquiring a thermal imaging picture according to thermal imaging equipment;
the preprocessing module is used for preprocessing the thermal imaging picture;
the infrared temperature measurement identification module is used for inputting the preprocessed thermal imaging picture into the infrared temperature measurement identification model for detection;
the temperature correction module is used for correcting the infrared temperature measurement result;
and the result output module is used for judging a temperature interval according to the color depth of the identification picture and outputting alarm information.
10. A storage medium storing a program, characterized in that: when being executed by a processor, the program realizes the power distribution room temperature early warning method based on deep learning and infrared detection as claimed in any one of claims 1-8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211062359.2A CN115410003A (en) | 2022-08-31 | 2022-08-31 | Power distribution room temperature early warning method, system and medium based on deep learning and infrared detection |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211062359.2A CN115410003A (en) | 2022-08-31 | 2022-08-31 | Power distribution room temperature early warning method, system and medium based on deep learning and infrared detection |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115410003A true CN115410003A (en) | 2022-11-29 |
Family
ID=84164601
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211062359.2A Pending CN115410003A (en) | 2022-08-31 | 2022-08-31 | Power distribution room temperature early warning method, system and medium based on deep learning and infrared detection |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115410003A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115876334A (en) * | 2023-02-22 | 2023-03-31 | 瑞纳智能设备股份有限公司 | Infrared thermal image indoor temperature measurement method and system and computer readable storage medium |
-
2022
- 2022-08-31 CN CN202211062359.2A patent/CN115410003A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115876334A (en) * | 2023-02-22 | 2023-03-31 | 瑞纳智能设备股份有限公司 | Infrared thermal image indoor temperature measurement method and system and computer readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Zhao et al. | A new steel defect detection algorithm based on deep learning | |
| CN111091233B (en) | Short-term wind power prediction modeling method for wind power plant | |
| CN111695620B (en) | Method and system for detecting and correcting time sequence abnormal data of power system | |
| CN113822880A (en) | Crack identification method based on deep learning | |
| CN115410003A (en) | Power distribution room temperature early warning method, system and medium based on deep learning and infrared detection | |
| CN110297218B (en) | Method for detecting unknown modulation mode of radar signal based on generation countermeasure network | |
| CN111753870B (en) | Training method, device and storage medium of target detection model | |
| CN111008726A (en) | Class image conversion method in power load prediction | |
| CN111259733A (en) | Point cloud image-based ship identification method and device | |
| CN111753666B (en) | Small target fault detection method, detection system and storage medium for power transmission line | |
| Liu et al. | Fault diagnosis method of substation equipment based on You Only Look Once algorithm and infrared imaging | |
| CN114863361A (en) | Distribution line bird nest detection method based on USRNet and YOLOv3 | |
| CN116246060A (en) | Transmission line bolt defect detection method based on context reasoning | |
| Xue et al. | Detection of Various Types of Metal Surface Defects Based on Image Processing. | |
| CN115423091A (en) | Conditional antagonistic neural network training method, scene generation method and system | |
| CN113780644A (en) | Photovoltaic output prediction method based on online learning | |
| CN113516187A (en) | Infrared weak and small target detection algorithm adopting local characteristic contrast | |
| CN114492540A (en) | Training method and device of target detection model, computer equipment and storage medium | |
| CN111929335A (en) | X-ray cable detection system | |
| CN113496159A (en) | Multi-scale convolution and dynamic weight cost function smoke target segmentation method | |
| Arosh et al. | Composite imagery-based non-uniform illumination sensing for system health monitoring of solar power plants | |
| CN117523605B (en) | Substation animal intrusion detection method based on multi-sensor information fusion | |
| CN113971755B (en) | All-weather sea surface target detection method based on improved YOLOV model | |
| CN116611267B (en) | All-constraint configuration method for priori boundary structure in geophysical prospecting data regularized inversion | |
| CN113627310B (en) | SAR ship target detection method based on background and scale sensing |
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 |