CN112308841A - Photovoltaic cell panel glass breaking risk analysis system based on vision - Google Patents
Photovoltaic cell panel glass breaking risk analysis system based on vision Download PDFInfo
- Publication number
- CN112308841A CN112308841A CN202011204270.6A CN202011204270A CN112308841A CN 112308841 A CN112308841 A CN 112308841A CN 202011204270 A CN202011204270 A CN 202011204270A CN 112308841 A CN112308841 A CN 112308841A
- Authority
- CN
- China
- Prior art keywords
- crack
- area
- panel
- crushing
- image
- 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.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20048—Transform domain processing
- G06T2207/20056—Discrete and fast Fourier transform, [DFT, FFT]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20084—Artificial neural networks [ANN]
Landscapes
- Engineering & Computer Science (AREA)
- Quality & Reliability (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Investigating Materials By The Use Of Optical Means Adapted For Particular Applications (AREA)
Abstract
The invention relates to the technical field of computer vision, in particular to a photovoltaic cell panel glass crushing risk analysis system based on vision, which comprises a network sensing module, a crack analysis module and a crushing risk analysis module which are sequentially connected, and solves the problem that the photovoltaic cell panel glass crushing not only reduces the light transmittance and the power generation efficiency of a cell panel, but also can prevent water, so that a cleaning robot can damage the cell panel glass to a greater extent when cleaning the cell panel glass; the invention utilizes the semantic perception network to automatically perceive the glass cracks on the surface of the battery panel, the complexity of a network model is low, the training is easy, and the precision is higher.
Description
Technical Field
The invention relates to the technical field of computer vision, in particular to a photovoltaic cell panel glass breaking risk analysis system based on vision.
Background
The photovoltaic cell panel, namely the solar cell panel is widely applied to photovoltaic power generation, the service life of the photovoltaic cell panel is influenced by components such as a cell piece and toughened glass, and the impact force of the toughened glass used for packaging needs to reach the international standard so as to protect a power generation main body (such as the cell piece).
However, the tempered glass may be broken due to external force during use, the broken glass may cause the light transmittance of the solar cell panel to be reduced, the output power to be insufficient, and the like, and the battery piece is oxidized due to long-time breakage, so that the power generation efficiency is greatly reduced; in addition, glass is cracked to lead to it not to possess waterproof function, and when photovoltaic cleaning machines people was clean for there being water on clean panel surface and clean mode, can make the panel intake to reduce panel charge efficiency, cleaning machines people self also can cause pressure to the panel surface simultaneously, lead to it to appear the further broken scheduling problem of panel when cleaning panel crackle area.
Disclosure of Invention
The invention provides a photovoltaic cell panel glass breakage risk analysis system based on vision, and solves the technical problems that the photovoltaic cell panel glass breakage can not only reduce the light transmittance and the power generation efficiency of a cell panel, but also prevent water, so that a cleaning robot can damage the cell panel to a greater extent when cleaning the cell panel.
In order to solve the technical problems, the invention provides a photovoltaic cell panel glass breakage risk analysis system based on vision, which comprises a network sensing module, a crack analysis module and a breakage risk analysis module which are sequentially connected;
the network perception module is used for preprocessing the battery panel image acquired in real time, inputting the preprocessed battery panel image into a semantic perception network for training and outputting a semantic perception effect graph;
the crack analysis module is used for screening a crack perception effect image from the semantic perception effect image, carrying out post-processing on the crack perception effect image to obtain a minimum external rectangle of a crack area, and taking the length-width product of the minimum external rectangle as the crack area;
the crack analysis module is also used for comparing the crack area with a preset area threshold value, and sending a prompt for replacing a battery plate to a worker if the crack area is larger than or equal to the area threshold value; if the crack area is smaller than the area threshold, inputting the crack area into the crushing risk analysis module;
the crushing risk analysis module is used for establishing a crack analysis model in advance according to historical data and inputting the received crack area into the crack analysis model to obtain the crushing area of the cleaning robot after passing through a crack area; the fracture risk analysis module is also used for inputting the crack area and the fracture area into an established fracture risk analysis model to obtain a fracture risk value;
the crushing risk analysis module is further used for comparing the crushing risk value with a preset risk threshold, and if the crushing risk value is greater than or equal to the risk threshold, a return instruction is sent to the cleaning robot; and if the crushing risk value is smaller than the risk threshold value, sending an adjusting instruction to the cleaning robot to enable the cleaning robot to be in a water-free mode.
Wherein the historical data comprises historical crack area and pressure of the cleaning robot on a panel directly below the cleaning robot; the pressure is measured in real time by a pressure sensor connected to the panel.
According to the technical scheme, the cracks of the glass on the surface of the photovoltaic cell panel are automatically sensed by computer vision, and the detection result is sent to the cleaning robot, so that the cleaning robot is adjusted, and the cell panel is prevented from being damaged to a greater extent by the robot in the cleaning process; according to the technical scheme, the crack area is detected based on the neural network, high-precision detection is guaranteed, and meanwhile, the crack analysis model established based on historical data is simple in principle, convenient to operate and high in practicability.
In a further embodiment, the crack analysis model is specifically:
A＝(S2+c)ln(F+1)
in the formula, S is the received crack area, F is the pressure of the cleaning robot on the battery plate, c is an adjustable parameter, and A is the crushing area of the cleaning robot after passing through the crack area.
This technical scheme measures the pressure of record cleaning robot to the panel under it through pressure sensor, combine cleaning robot to the pressure of panel and different panel surface glass's crack area, according to the mathematical function that passes through historical data in advance and establish, crack analysis model acquires the final broken area of panel after cleaning robot passes through panel surface crack region promptly, this scheme not only calculates the crack area of panel, still predict the broken area of cleaning robot through the crack region according to pressure, thereby avoid the panel cracked back to appear, because cleaning robot causes the emergence of the broken scheduling problem of panel bigger degree to the pressure of panel.
In a further embodiment, the crushing risk analysis model is specifically:
wherein alpha is a weight coefficient and epsilon is a crushing risk value.
This technical scheme combines the cracked area of panel and cleaning machines people through the broken area of panel behind the panel crackle area, and the size of the broken risk of large tracts of land appears in the panel around analysis cleaning machines people through the panel crackle area, sends different instructions according to the size of broken risk to prevent the broken of the clean in-process of robot increase panel, avoid it to spout water into the panel crackle area simultaneously, lead to the panel to intake, work efficiency drops etc..
In a further embodiment, the specific training process of the semantic aware network comprises:
making a label image: marking the pixels of the crack regions of the battery plate images as 1, and marking the pixels of other regions as 0;
training a model: and inputting the battery panel image and the label image into the semantic perception network, training a model by adopting cross entropy as a loss function, and outputting a semantic perception effect image with the size equal to that of the original image.
Wherein the semantic awareness network adopts an encoder-decoder network structure.
The technical scheme is that the preprocessed battery panel image is sent to a semantic perception network for training, and a glass crack area on the surface of the battery panel is perceived so as to calculate and obtain crack areas of different battery panels; according to the scheme, the crack area of the battery panel can be distinguished more accurately through the semantic perception network, and the detection quality and effect are greatly improved.
In a further embodiment, the preprocessing the panel image collected in real time specifically includes:
carrying out logarithm operation on the battery panel image, and then carrying out Fourier transform;
carrying out high-pass filtering on the battery panel image in the frequency domain, and carrying out inverse Fourier transform;
and performing exponential transformation on the panel image.
In the technical scheme, when the glass on the surface of the battery panel is cracked, white cracks can appear, namely, the brightness of the crack area on the surface of the battery panel is higher, therefore, the technical scheme firstly carries out preprocessing operation on the image, namely, the low-frequency or high-frequency components of the image are mainly estimated through a filter function, the local contrast of the image is enhanced, and the problem of uneven brightness of the image is corrected, so that the network can sense and analyze the crack area of the battery panel, and further the detection efficiency is effectively improved.
Drawings
FIG. 1 is a schematic structural diagram of a system for analyzing a risk of breaking glass of a photovoltaic cell panel based on vision according to an embodiment of the present invention;
FIG. 2 is an exemplary schematic diagram of a remotely sensed image provided by an embodiment of the present invention;
fig. 3 is a simple schematic diagram of a deep neural network model provided by an embodiment of the present invention.
Detailed Description
The embodiments of the present invention will be described in detail below with reference to the accompanying drawings, which are given solely for the purpose of illustration and are not to be construed as limitations of the invention, including the drawings which are incorporated herein by reference and for illustration only and are not to be construed as limitations of the invention, since many variations thereof are possible without departing from the spirit and scope of the invention.
Aiming at the problem that the light transmittance and the power generation efficiency of a solar panel are reduced and the solar panel cannot be prevented from being waterproofed when the solar panel is broken, so that a cleaning robot can damage the solar panel to a greater extent when the solar panel is cleaned, the embodiment of the invention provides a photovoltaic solar panel glass breaking risk analysis system based on vision, which is shown in fig. 1 and comprises a network sensing module 1, a crack analysis module 2 and a breaking risk analysis module 3 which are sequentially connected;
the embodiment of the invention collects the image of the cell panel in real time through the track camera, one side of the photovoltaic cell panel is provided with the track, the track camera is a slidable camera arranged on the track, and the image collection is carried out on the surface of the cell panel;
in the embodiment of the present invention, when the surface glass of the photovoltaic cell panel is cracked, white cracks may occur, that is, the brightness of the crack area on the surface of the cell panel is high, so that the network sensing module 1 first preprocesses the cell panel image collected in real time, mainly estimates the low frequency or high frequency component of the image through a filter function, enhances the local contrast of the image, and corrects the problem of uneven brightness of the image, so that the network detects the crack area of the cell panel, where as shown in fig. 2, the preprocessing step specifically includes:
1) carrying out logarithm taking operation on the battery panel image;
g(x,y)＝exp{n(x,y)} 1-1
f(x,y)＝i(x,y)*r(x,y) 1-2
wherein lnf represents the logarithm of f (x, y); f (x, y) are panel images, each image consisting of two components, i (x, y) being the illumination component of the image, which refers to the intensity of light incident on the scene; r (x, y) is the reflected component of the image and refers to the intensity of light reflected from the scene.
2) Carrying out Fourier transform on the logarithmic battery plate image;
F(lnf)＝F{ln(i(x,y))}+F{ln(r(x,y))}＝F(u,v) 1-3
wherein F (u, v) is the battery plate image after Fourier transformation.
3) Carrying out high-pass filtering on the battery panel image after Fourier transformation;
N(u,v)＝H(u,v)·F(u,v) 1-4
h (u, v) is a high-pass filtering function, and N (u, v) is a panel image after high-pass filtering.
4) Performing inverse Fourier transform on the high-pass filtered panel image to convert the panel image from a frequency domain back to a time domain;
n(x,y)＝F-1{N(u,v)} 1-5
5) performing exponential transformation on the battery panel image subjected to Fourier inverse transformation to obtain a preprocessed battery panel image g (x, y);
g(x,y)＝exp{n(x,y)} 1-6
in the embodiment of the present invention, the network sensing module 1 takes the preprocessed battery panel image as an input of a semantic sensing network, senses a crack area on the surface of the battery panel, and outputs a semantic sensing effect map, where as shown in fig. 3, a specific training process of the semantic sensing network includes:
making a label image: when the glass on the surface of the battery panel has cracks, the brightness of the crack area on the surface of the battery panel is higher, so that the pixel of the crack area of the image of the battery panel is marked as 1, and the pixels of other areas are marked as 0;
training a model: inputting the battery panel image and the label image into the semantic perception network, namely a crack perception encoder-decoder network structure, and performing iterative training on the network by adopting a cross entropy loss function, wherein the crack perception encoder performs feature extraction on a crack area of the battery panel and outputs the crack area as a feature map, the crack perception decoder performs up-sampling on the feature map, and finally the network outputs a semantic perception effect map with the size equal to that of an original image; the semantic perception effect graph output by the embodiment includes a crack perception effect graph and a crack-free perception effect graph, wherein the pixel values of the image are all 0 because of the crack-free area in the crack-free perception effect graph;
according to the embodiment, the crack region is detected based on the semantic perception network, and the detection accuracy is greatly improved.
In the embodiment of the invention, the crack analysis module 2 screens a crack perception effect graph from the received semantic perception effect graph, performs post-processing on the crack perception effect graph to obtain a minimum external rectangle of a crack area and length and width values of the minimum external rectangle, and takes the product of the length and the width of the minimum external rectangle as a crack area S;
in the embodiment, the minimum external rectangle of the crack area is obtained through an OpenCV interface;
the crack analysis module 2 compares the crack area S with a preset area threshold S ', and if the crack area S is larger than or equal to the area threshold S ', namely S is larger than or equal to S ', the crack degree of the battery panel is considered to be too large, and a prompt for replacing the battery panel is sent to a worker; if the crack area S is smaller than the area threshold S ', namely S is smaller than S', the current output efficiency of the panel is hardly influenced by the crack area on the surface of the panel, and at the moment, the crack area S is input into the crushing risk analysis module 3;
it should be noted that, because the glass on the surface of the photovoltaic cell panel is made of tempered glass with high transmittance, the cell and the glass are adhered together during lamination, and if the glass is taken out separately, the fragile cell is easily damaged, so that if the glass is seriously damaged, a worker usually directly replaces the whole cell.
Because when cleaning robot cleans the crackle area of panel, its pressure to the crackle area can increase panel surface crushing degree, consequently, broken risk analysis module 3 establishes the crackle analysis model according to historical data in advance to with the crackle area that receives and cleaning robot to the pressure input of panel the crackle analysis model, obtain the broken area of cleaning robot behind the crackle area, promptly the crackle analysis model specifically is:
A＝(S2+c)ln(B+1) 1-7
in the formula, S is the received crack area, B is the pressure of the cleaning robot to the battery panel, a is the crushing area of the cleaning robot after passing through the crack area, c is an adjustable parameter, and in this embodiment, c is preferably set to 5, and a person skilled in the art can adjust the crushing area according to actual conditions.
In the embodiment, the historical data comprises historical crack areas and the pressure of the cleaning robot on the battery plate, wherein the pressure of the cleaning robot on the battery plate refers to the pressure of the cleaning robot on the battery plate directly below the cleaning robot; the pressure is measured in real time by a pressure sensor connected to the panel.
The crack analysis model established based on the historical data is simple in principle, convenient to operate and high in practicability; it should be noted that the crack analysis model provided above is a mathematical model established by data obtained through experiments in this embodiment, and since actual historical data is affected by various factors such as a battery board, the crack analysis model provided in the embodiment of the present invention is not unique, and a person skilled in the art needs to establish a model according to historical data obtained through a specific implementation process.
In the embodiment of the invention, the breaking risk analysis module 3 further inputs the crack area S and the breaking area a of the cleaning robot after passing through the crack area into an established breaking risk analysis model to obtain a breaking risk value epsilon so as to analyze the large-area breaking risk of the battery panel before and after the cleaning robot passes through the crack area of the battery panel, wherein the larger the change value of the crack area of the battery panel after the cleaning robot passes through the crack area of the battery panel is, the larger the breaking risk of the battery panel is;
the crushing risk analysis model specifically comprises the following steps:
where ∈ is a fracture risk value, a is a fracture area of the cleaning robot after passing through the fracture area, S is the received fracture area, and α is a weight coefficient, where α is set to 0.8 in this embodiment, and a person skilled in the art can adjust the fracture area according to actual conditions.
The crushing risk analysis module 3 compares the crushing risk value epsilon with a preset risk threshold epsilon ', and if the crushing risk value epsilon is greater than or equal to the risk threshold epsilon ', namely epsilon is greater than or equal to epsilon ', the crushing risk of the battery panel is judged to be high, namely the battery panel is crushed in a larger area after the cleaning robot passes through a battery panel crack area, the power generation power of the battery panel is reduced, and at the moment, the system sends a detection result to the cleaning robot and sends a return instruction to the cleaning robot; if the breakage risk value epsilon is smaller than the risk threshold epsilon ', namely epsilon is smaller than epsilon', the breakage risk of the battery panel is judged to be smaller, namely when the cleaning robot passes through a crack area, the crack area of the battery panel is not affected, therefore, the cleaning robot can clean the crack area of the battery panel, at the moment, an adjusting instruction is sent to the cleaning robot to adjust the cleaning mode of the cleaning robot to be a waterless mode, the rotating speed of a brush is reduced, and the like, so that the problems that water is sprayed into the crack area of the battery panel in the cleaning process of the robot, the water enters the interior of the battery panel, the charging efficiency is reduced, and the like are avoided.
The photovoltaic cell panel glass breakage risk analysis system based on vision provided by the embodiment of the invention comprises a network sensing module 1, a crack analysis module 2 and a breakage risk analysis module 3 which are sequentially connected, and solves the problems that the light transmittance and the power generation efficiency of a cell panel are reduced and the cell panel cannot be prevented from being waterproofed due to the breakage of the photovoltaic cell panel glass, so that a cleaning robot can damage the cell panel glass to a greater extent when the cell panel glass is cleaned; according to the embodiment of the invention, the glass cracks on the surface of the photovoltaic cell panel are automatically sensed by computer vision, and the large-area crushing risk of the cell panel is analyzed by combining the cell panel crack area and the cell panel crushing area of the robot after passing through the cell panel crack area, so that the cleaning robot is adjusted, the training is easy, the damage of the cleaning robot, such as the increase of the cell panel crushing degree due to the pressure of the cleaning robot on the cell panel in the cleaning process, is avoided, the operation is convenient, the practicability is strong, and the detection efficiency can be effectively improved.
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 (7)
1. The utility model provides a broken risk analysis system of photovoltaic cell board glass based on vision which characterized in that: the system comprises a network sensing module, a crack analysis module and a crushing risk analysis module which are connected in sequence;
the network perception module is used for preprocessing the battery panel image acquired in real time, inputting the preprocessed battery panel image into a semantic perception network for training and outputting a semantic perception effect graph;
the crack analysis module is used for screening a crack perception effect image from the semantic perception effect image, carrying out post-processing on the crack perception effect image to obtain a minimum external rectangle of a crack area, and taking the length-width product of the minimum external rectangle as the crack area;
the crack analysis module is also used for comparing the crack area with a preset area threshold value, and sending a prompt for replacing a battery plate to a worker if the crack area is larger than or equal to the area threshold value; if the crack area is smaller than the area threshold, inputting the crack area into the crushing risk analysis module;
the crushing risk analysis module is used for establishing a crack analysis model in advance according to historical data and inputting the received crack area into the crack analysis model to obtain the crushing area of the cleaning robot after passing through a crack area; the fracture risk analysis module is also used for inputting the crack area and the fracture area into an established fracture risk analysis model to obtain a fracture risk value;
the crushing risk analysis module is further used for comparing the crushing risk value with a preset risk threshold, and if the crushing risk value is greater than or equal to the risk threshold, a return instruction is sent to the cleaning robot; and if the crushing risk value is smaller than the risk threshold value, sending an adjusting instruction to the cleaning robot to enable the cleaning robot to be in a water-free mode.
2. The vision-based photovoltaic panel glass breakage risk analysis system of claim 1, wherein: the historical data comprises historical crack areas and pressure of the cleaning robot on a panel directly below the cleaning robot;
the pressure is measured in real time by a pressure sensor connected to the panel.
3. The vision-based photovoltaic panel glass breakage risk analysis system of claim 1, wherein: the crack analysis model specifically comprises the following steps:
A＝(S2+c)ln(F+1)
in the formula, S is the received crack area, F is the pressure of the cleaning robot on the battery plate, c is an adjustable parameter, and A is the crushing area of the cleaning robot after passing through the crack area.
4. The vision-based photovoltaic panel glass breakage risk analysis system of claim 3, wherein: the crushing risk analysis model specifically comprises the following steps:
wherein alpha is a weight coefficient and epsilon is a crushing risk value.
5. The vision-based photovoltaic panel glass breakage risk analysis system of claim 4, wherein the specific training process of the semantic awareness network comprises:
making a label image: marking the pixels of the crack regions of the battery plate images as 1, and marking the pixels of other regions as 0;
training a model: and inputting the battery panel image and the label image into the semantic perception network, training a model by adopting cross entropy as a loss function, and outputting a semantic perception effect image with the size equal to that of the original image.
6. The vision-based photovoltaic panel glass breakage risk analysis system of claim 5, wherein: the semantic perception network adopts an encoder-decoder network structure.
7. The vision-based photovoltaic panel glass breakage risk analysis system of claim 6, wherein: the panel image that will gather in real time carries out the preliminary treatment, specifically includes:
carrying out logarithm operation on the battery panel image, and then carrying out Fourier transform;
carrying out high-pass filtering on the battery panel image in the frequency domain, and carrying out inverse Fourier transform;
and performing exponential transformation on the panel image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011204270.6A CN112308841A (en) | 2020-11-02 | 2020-11-02 | Photovoltaic cell panel glass breaking risk analysis system based on vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011204270.6A CN112308841A (en) | 2020-11-02 | 2020-11-02 | Photovoltaic cell panel glass breaking risk analysis system based on vision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112308841A true CN112308841A (en) | 2021-02-02 |
Family
ID=74333622
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011204270.6A Withdrawn CN112308841A (en) | 2020-11-02 | 2020-11-02 | Photovoltaic cell panel glass breaking risk analysis system based on vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112308841A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113469444A (en) * | 2021-07-09 | 2021-10-01 | 北京中超伟业信息安全技术股份有限公司 | Confidential storage medium crushing and screening method |
| CN113860757A (en) * | 2021-09-30 | 2021-12-31 | 长沙韶光铬版有限公司 | Manufacturing method of glass code disc, glass code disc and device |
-
2020
- 2020-11-02 CN CN202011204270.6A patent/CN112308841A/en not_active Withdrawn
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113469444A (en) * | 2021-07-09 | 2021-10-01 | 北京中超伟业信息安全技术股份有限公司 | Confidential storage medium crushing and screening method |
| CN113469444B (en) * | 2021-07-09 | 2022-02-11 | 北京中超伟业信息安全技术股份有限公司 | Confidential storage medium crushing and screening method |
| CN113860757A (en) * | 2021-09-30 | 2021-12-31 | 长沙韶光铬版有限公司 | Manufacturing method of glass code disc, glass code disc and device |
| CN113860757B (en) * | 2021-09-30 | 2023-07-14 | 长沙韶光铬版有限公司 | Manufacturing method of glass code disc, glass code disc and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| WO2020199538A1 (en) | Bridge key component disease early-warning system and method based on image monitoring data | |
| CN109215042B (en) | Photovoltaic cell panel hot spot effect detection system based on computer vision and calculation method thereof | |
| CN112308841A (en) | Photovoltaic cell panel glass breaking risk analysis system based on vision | |
| CN109916910B (en) | Photovoltaic glass edge defect detection system and corresponding method | |
| CN207908384U (en) | A kind of double light spectrum image-forming rails and fastener abnormal detector | |
| WO2015055060A1 (en) | Online detecting method for continuous casting slab surface quality | |
| CN110954968B (en) | Airport runway foreign matter detection device and method | |
| CN106584209A (en) | Real-time online monitoring method for tool wear based on cloud manufacturing | |
| CN103413150A (en) | Power line defect diagnosis method based on visible light image | |
| CN102997853A (en) | Device and method for detecting ice and snow thickness | |
| CN104134211A (en) | Band steel defect detection method based on multi-resolution filtering of Gaussian filter | |
| CN113655002B (en) | Recycled aggregate quality detection system of surface mortar based on hyperspectral technology | |
| CN104713526A (en) | Method for detecting types of foreign matters on power transmission line | |
| CN104729483A (en) | Foreign matter variety detection platform on power transmission line | |
| CN204039886U (en) | A kind of pavement damage crack detection system based on multiple stage camera stereoscopic shooting | |
| CN113706461A (en) | Micro-defect identification system of insulating blanket based on image identification | |
| CN202947691U (en) | Device for detecting ice and snow thickness | |
| WO2019076004A1 (en) | Straight-on determination apparatus for solar panel cleaning robot, and determination method thereof | |
| CN112529880A (en) | Dam image crack detection device and method based on binocular vision | |
| CN108180871B (en) | method for quantitatively evaluating composite insulator surface pulverization roughness | |
| CN103091332B (en) | Detection method and detection system of U-shaped powder pipe based on machine vision | |
| CN105140145A (en) | Photovoltaic cell grid line measuring method | |
| CN113554618B (en) | Railway track state monitoring method and device based on infrared image | |
| CN111807003A (en) | Non-metal foreign matter detection system for belt conveyor | |
| CN113408396B (en) | Bridge intelligent sensing system based on cloud computing |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20210202 |
|
| WW01 | Invention patent application withdrawn after publication |