Local fault rapid identification device for municipal outdoor facade equipment in urban updating

Abstract

The invention discloses a municipal outdoor facade equipment local fault rapid identification device in city updating, and belongs to the technical field of fault diagnosis, and the device comprises a data collection unit which is used for carrying out the real-time detection of a building group facade lighting region, and collecting the image data of the lighting region; the data preprocessing unit is used for preprocessing the collected image data; the feature extraction unit is used for performing feature extraction on the preprocessed image; and the fault identification unit is used for inputting the extracted feature data into the classifier and identifying the image fault according to the data features. According to the invention, an automatic detection method is adopted, so that the time and labor consumption of manual detection is reduced, and the detection efficiency is improved; fault identification is carried out based on objective data, subjective factor influence caused by manual visual inspection is avoided, and identification accuracy is improved; the image quality of the display screen can be monitored in real time, faults can be found and repaired in time, and normal operation of the display screen is guaranteed.

Description

Technical Field

[0001] The invention belongs to the technical field of fault diagnosis, and in particular relates to a device for quickly identifying local faults of municipal outdoor facade equipment in urban renewal. Background Art

[0002] The prior art discloses some invention patents in the field of fault diagnosis technology, among which the invention patent with application number CN110536505A discloses a system fault self-diagnosis device for remote LED landscape lighting, including a power supply module, a signal control module and a landscape lighting LED module, and also includes an electrical measurement monitoring module and a monitoring center; the power supply module is respectively connected to the signal control module and the electrical measurement communication module to supply power to the entire device; the signal control module is respectively connected to the power supply module, the electrical measurement communication module and the monitoring center to control the diagnostic signal and complete the signal output with clock cycle and lamp color and brightness control; the landscape lighting LED module is the lighting equipment in the device; the electrical measurement monitoring module is used to measure the voltage, current and power data source of the load LED lamp, and further transmit the measurement data; the monitoring center receives the current, voltage and power data of the electrical measurement monitoring module, and determines the number of faulty lamp tubes and the location of the faulty tubes according to the program preset values. The invention can solve the problems of high fault diagnosis cost of LED lighting devices and the need for manual inspection in the prior art; at the same time, it reduces the danger of high-altitude operation and lighting inspection at night, completes automatic remote detection, and improves the convenience of lamp maintenance; it meets the requirements of completely turning off the lights and stable deep dimming, while maintaining the conversion efficiency of the lamp to the greatest extent; in addition, adding this active consumption circuit to the output side winding can reduce the voltage resistance requirement of the switching transistor, making the solution cost-effective.

[0003] The shortcomings of the existing technology. In urban renewal, the further innovation of municipal outdoor comprehensive management equipment lies in its dedicated lighting, monitoring and display functions. These devices realize local encryption to prevent potential risks such as remote network intrusion. When a lighting failure occurs on the facade of a building complex, it depends on the direct observation and judgment of the staff, which may result in omissions or failure to discover in time. This situation may cause the faulty lighting area to be unable to be repaired in time, affecting the normal use of the lighting on the facade of the building complex, and may prevent major festivals or activities from being promoted normally.

[0004] Based on this, the present invention designs a local fault rapid identification device for municipal outdoor facade equipment in urban renewal to solve the above problems. Summary of the invention

[0005] The purpose of the present invention is to solve the problem that when a lighting failure occurs on the facade of a building complex, it depends on direct observation by staff for judgment, which may result in omissions or failure to discover the problem in time. This may result in the lighting area with the failure being unable to be repaired in time, affecting the normal use of the lighting on the facade of the building complex, and may cause major festivals or activities to be unable to be promoted normally. Therefore, a local fault rapid identification device for municipal outdoor facade equipment in urban renewal is proposed.

[0006] In order to achieve the above object, the present invention adopts the following technical solutions:

[0007] The local fault rapid identification device of municipal outdoor facade equipment in urban renewal includes:

[0008] A data acquisition unit is used to perform real-time detection of the illuminated areas of the building facades and collect image data of the illuminated areas;

[0009] A data preprocessing unit, used for preprocessing the collected image data;

[0010] A feature extraction unit, used for extracting features from the preprocessed image;

[0011] A fault identification unit, used to input the extracted feature data into a classifier and identify image faults according to the data features;

[0012] The transmission warning unit is used to wirelessly upload the identified fault results to the server to warn the on-duty personnel.

[0013] As a further description of the above technical solution:

[0014] The image data includes pixel values, brightness and color information of the image.

[0015] As a further description of the above technical solution:

[0016] The data preprocessing unit includes a noise reduction processing module and a segmentation processing module. The noise reduction processing module is used to reduce the noise of the image and improve the quality of the data;

[0017] The segmentation processing module is used to perform segmentation processing on the collected images.

[0018] As a further description of the above technical solution:

[0019] The data preprocessing unit also includes an image enhancement module, which is used to perform enhancement processing on the collected image.

[0020] As a further description of the above technical solution:

[0021] The specific steps of the fault identification are:

[0022] Obtain the faults that occur in the corresponding brightened area and the image data when the faults occur, perform feature extraction, and divide the obtained data and the data obtained by feature extraction into training data and verification data;

[0023] Select the corresponding classifier as the fault identification classifier;

[0024] Use training data to train the classifier to learn the characteristics and classification rules of the data;

[0025] Use validation data to verify and evaluate the trained classifier to determine the performance of the classifier;

[0026] The real-time data obtained is input into a qualified classifier for fault identification to determine the fault location in the illuminated area.

[0027] As a further description of the above technical solution:

[0028] The data obtained by the feature extraction includes the mean, variance, skewness and kurtosis of the data.

[0029] As a further description of the above technical solution:

[0030] The image noise reduction process is specifically as follows:

[0031] Assume the noisy model: f(k) = s(k) + n(k), k = 0, 1, 2, ... N-1

[0032] Among them, s(k) is the original image signal, and n(k) is Gaussian white noise;

[0033] Perform discrete wavelet transform on f(k): q i,k =Q s (j,k)+Q n (j,k),j=0,1,2,…,J;k=0,1,2,…,N

[0034] where q i,k , Q s (j,k) and Q n (j, k) are the noisy image signal, the original image signal and the wavelet coefficients of the noise at the nth layer, and the total number of pixels of the image with the maximum decomposition layer number J and N, respectively;

[0035] The hard threshold function expression is:

[0036]

[0037] The soft threshold function expression is:

[0038]

[0039] in To estimate the wavelet coefficients, q j,k is the decomposition wavelet coefficient, λ is the threshold;

[0040] The improved threshold function,

[0041] α is the adjustment coefficient, and 0≤α≤1, when |q j,k When |→λ, That is, the improved threshold function is continuous at the threshold point; when |q j,k When |→∞, That is, with |q j,k |Gradually increasing, With q j,k The deviation between them becomes smaller and smaller, solving the constant deviation shortcoming of the soft threshold function.

[0042] As a further description of the above technical solution:

[0043] The image segmentation process is specifically as follows:

[0044] First, the image is expressed as an undirected graph G = (V, E), where V is the set of nodes, E is the set of edges connecting nodes, and the basis of V is N = |V|; the edge connecting each two nodes is given a weight w = (u, n), which measures the similarity between nodes u and v; the image G = (V, E) can be divided into two non-intersecting sets A and B by simply removing the edges connecting the two parts, and satisfying A∪B = V, A∩B = Φ. The dissimilarity of the two parts is calculated by the sum of the weights of the removed edges, defined as,

[0045] Determine the normalized segmentation criterion: Normalized segmentation adds the ratio of the nodes in each region to the sum of the weights of all nodes on the basis of the minimum cut to balance the segmentation. The formula for normalized segmentation is expressed as:

[0046]

[0047] where assoc(A,V)=∑ u∈A,t∈V w(u,t) is the sum of the weights of the edges from all nodes in A to all nodes in the image, and assoc(B,V) is the sum of the weights of the edges from all nodes in B to all nodes in the image. In the case of segmenting isolated points, the value in the normalized segmentation is relatively large:

[0048]

[0049] Among them, assoc(A,A) and assoc(B,B) represent the sum of the weights of the edges connecting the nodes in A and B respectively. The above formula reflects the balance of the definition and represents the association degree of the nodes in each group;

[0050] The Ncut criterion can overcome the problem of dividing isolated points. The division corresponding to the minimum Ncut value is the optimal division of graph G. Minimizing Ncut can be transformed into the following standard feature system:

[0051]

[0052] Where D is a diagonal matrix of NXN, and its diagonal elements are (i, j), W is a symmetric matrix whose elements are w(i, j), λ and z are the corresponding eigenvalues ​​and eigenvectors;

[0053] The eigenvector corresponding to the second minimum eigenvalue of the eigensystem is used to complete the optimal division of the entire image and obtain a segmentation result of the corresponding image.

[0054] As a further description of the above technical solution:

[0055] The identification method of the identification device comprises the following steps:

[0056] Conduct real-time detection of the illuminated areas on the facades of buildings and collect image data of the illuminated areas;

[0057] Preprocessing the collected image data;

[0058] Perform feature extraction on the preprocessed image;

[0059] The extracted feature data is input into the classifier to identify image faults according to the data features;

[0060] The identified fault results are wirelessly uploaded to the server to alert the on-duty personnel.

[0061] As a further description of the above technical solution:

[0062] The output value of the classifier model is:

[0063]

[0064] Among them, X i and Y i is the sample observation value; σ is the smoothing factor; n is the number of samples.

[0065] In summary, due to the adoption of the above technical solution, the beneficial effects of the present invention are:

[0066] In the present invention, an automated detection method is adopted to reduce the time and labor consumption of manual detection and improve the detection efficiency; fault identification is performed based on objective data, avoiding the influence of subjective factors caused by manual visual inspection and improving the recognition accuracy; the image quality of the display screen can be monitored in real time, faults can be discovered and repaired in time, and the normal operation of the display screen can be ensured;

[0067] Adopt denoising processing on image data. Through denoising processing, noise interference can be reduced, making the image clearer, which is conducive to subsequent fault identification. Noise reduction processing can eliminate random noise in the image, making the detailed information in the image more obvious, which is very helpful for identifying tiny fault defects and can improve the accuracy of fault detection. Noise reduction processing can smooth the edges of the image and reduce the loss of edge information. This helps to maintain the integrity of the fault area and facilitates subsequent segmentation and identification processing. The presence of noise will increase the error in the identification process. Noise reduction processing can reduce this error and improve the accuracy of fault identification;

[0068] Segmentation of the image collected in the brightened area can separate the target area different from the brightened point from the background, realize data processing of the key position in the image, accurately locate the fault area, and help improve the speed and accuracy of fault identification. Segmentation can simplify complex image data and reduce the amount of calculation for subsequent analysis. This is very important for real-time monitoring systems and can improve the operating efficiency of the system. The target area after segmentation is clearer, which is conducive to extracting fault features. Features can be used for fault classification and identification, improving the accuracy of fault detection. Segmentation can use different segmentation algorithms according to different scenarios and needs to achieve accurate extraction of the fault area, which helps to improve the universality and practicality of fault identification methods.

[0069] The use of classifiers can adapt to different types of display screens and different types of faults. The fault identification system combined with the classifier can realize real-time monitoring of the screen status of the display screen. Once a fault is detected, it can respond immediately and notify maintenance personnel in time to handle it. The image can be automatically processed and analyzed without human intervention, which greatly improves work efficiency and reduces the possibility of human error. By learning a large amount of normal and faulty image data, faults in the image can be effectively identified. Especially with the support of deep learning technology, the accuracy of the classifier can be significantly improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 This is a schematic diagram of the composition of the local fault rapid identification device for municipal outdoor facade equipment in urban renewal proposed by the present invention;

[0071] Figure 2This is an internal schematic diagram of a data preprocessing unit of a device for rapid identification of local faults of municipal outdoor facade equipment in urban renewal proposed by the present invention;

[0072] Figure 3 A schematic diagram of a fault identification process of a local fault rapid identification device for municipal outdoor facade equipment in urban renewal proposed by the present invention;

[0073] Figure 4 This is a schematic diagram of the noise reduction process of the local fault rapid identification device for municipal outdoor facade equipment in urban renewal proposed by the present invention. DETAILED DESCRIPTION

[0074] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0075] Please see attached Figure 1 -Attached Figure 4 The present invention provides a technical solution: a device for quickly identifying local faults of municipal outdoor facade equipment in urban renewal, comprising:

[0076] A data acquisition unit is used to perform real-time detection of the illuminated areas of the building facades and collect image data of the illuminated areas;

[0077] A data preprocessing unit, used for preprocessing the collected image data;

[0078] A feature extraction unit, used for extracting features from the preprocessed image;

[0079] A fault identification unit, used to input the extracted feature data into a classifier and identify image faults according to the data features;

[0080] The transmission warning unit is used to wirelessly upload the identified fault results to the server to warn the on-duty personnel.

[0081] Image data includes the pixel value, brightness and color information of the image.

[0082] The data preprocessing unit includes a noise reduction processing module and a segmentation processing module. The noise reduction processing module is used to reduce the noise of the image and improve the quality of the data;

[0083] The segmentation processing module is used to perform segmentation processing on the collected images.

[0084] The data preprocessing unit also includes an image enhancement module, which is used to perform enhancement processing on the collected image.

[0085] The specific steps for fault identification are:

[0086] Obtain the faults that occur in the corresponding brightened area and the image data when the faults occur, perform feature extraction, and divide the obtained data and the data obtained by feature extraction into training data and verification data;

[0087] Select the corresponding classifier as the fault identification classifier;

[0088] Use training data to train the classifier to learn the characteristics and classification rules of the data;

[0089] Use validation data to verify and evaluate the trained classifier to determine the performance of the classifier;

[0090] The real-time data obtained is input into a qualified classifier for fault identification to determine the fault location in the illuminated area.

[0091] The data obtained by the feature extraction includes the mean, variance, skewness and kurtosis of the data.

[0092] The specific image noise reduction process is as follows:

[0093] Assume the noisy model: f(k) = s(k) + n(k), k = 0, 1, 2, ... N-1

[0094] Among them, s(k) is the original image signal, and n(k) is Gaussian white noise;

[0095] Perform discrete wavelet transform on f(k): q i,k =Q s (j,k)+Q n (j,k),j=0,1,2,…,J;k=0,1,2,…,N

[0096] where q i,k , Q s (j,k) and Q n (j, k) are the noisy image signal, the original image signal and the wavelet coefficients of the noise at the nth layer, and the total number of pixels of the image with the maximum decomposition layer number J and N, respectively;

[0097] The hard threshold function expression is:

[0098]

[0099] The soft threshold function expression is:

[0100]

[0101] in To estimate the wavelet coefficients, q j,k is the decomposition wavelet coefficient, λ is the threshold;

[0102] The improved threshold function,

[0103] α is the adjustment coefficient, and 0≤α≤1, when |q j,k When |→λ, That is, the improved threshold function is continuous at the threshold point; when |q j,k When |→∞, That is, with |q j,k |Gradually increasing, With q j,k The deviation between them becomes smaller and smaller, solving the constant deviation shortcoming of the soft threshold function.

[0104] The image segmentation process is as follows:

[0105] First, the image is expressed as an undirected graph G = (V, E), where V is the set of nodes, E is the set of edges connecting nodes, and the basis of V is N = |V|; the edge connecting each two nodes is given a weight w = (u, n), which measures the similarity between nodes u and v; the image G = (V, E) can be divided into two non-intersecting sets A and B by simply removing the edges connecting the two parts, and satisfying A∪B = V, A∩B = Φ. The dissimilarity of the two parts is calculated by the sum of the weights of the removed edges, defined as,

[0106] Determine the normalized segmentation criterion: Normalized segmentation adds the ratio of the nodes in each region to the sum of the weights of all nodes on the basis of the minimum cut to balance the segmentation. The formula for normalized segmentation is expressed as:

[0107]

[0108] where assoc(A,V)=∑ u∈A,t∈V w(u,t) is the sum of the weights of the edges from all nodes in A to all nodes in the image, and assoc(B,V) is the sum of the weights of the edges from all nodes in B to all nodes in the image. In the case of segmenting isolated points, the value in the normalized segmentation is relatively large:

[0109]

[0110] Among them, assoc(A,A) and assoc(B,B) represent the sum of the weights of the edges connecting the nodes in A and B respectively. The above formula reflects the balance of the definition and represents the association degree of the nodes in each group;

[0111] The Ncut criterion can overcome the problem of dividing isolated points. The division corresponding to the minimum Ncut value is the optimal division of graph G. Minimizing Ncut can be transformed into the following standard feature system:

[0112]

[0113] Where D is a diagonal matrix of NXN, and its diagonal elements are (i, j), W is a symmetric matrix whose elements are w(i, j), λ and z are the corresponding eigenvalues ​​and eigenvectors;

[0114] The eigenvector corresponding to the second minimum eigenvalue of the eigensystem is used to complete the optimal division of the entire image and obtain a segmentation result of the corresponding image.

[0115] The authentication method of the authentication device comprises the following steps:

[0116] Conduct real-time detection of the illuminated areas on the facades of buildings and collect image data of the illuminated areas;

[0117] Preprocessing the collected image data;

[0118] Perform feature extraction on the preprocessed image;

[0119] The extracted feature data is input into the classifier to identify image faults according to the data features;

[0120] The identified fault results are wirelessly uploaded to the server to alert the on-duty personnel.

[0121] The output value of the classifier model is:

[0122]

[0123] Among them, X i and Y i is the sample observation value; σ is the smoothing factor; n is the number of samples.

[0124] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A local fault rapid identification device for municipal outdoor facade equipment in urban renewal, characterized in that: include: A data acquisition unit is used to perform real-time detection of the illuminated areas of the building facades and collect image data of the illuminated areas; A data preprocessing unit, used for preprocessing the collected image data; A feature extraction unit, used for extracting features from the preprocessed image; A fault identification unit, used to input the extracted feature data into a classifier and identify image faults according to the data features; The transmission warning unit is used to wirelessly upload the identified fault results to the server to warn the on-duty personnel.

2. The local fault rapid identification device for municipal outdoor facade equipment in urban renewal according to claim 1 is characterized in that: The image data includes pixel values, brightness and color information of the image.

3. The local fault rapid identification device for municipal outdoor facade equipment in urban renewal according to claim 1 is characterized in that: The data preprocessing unit includes a noise reduction processing module and a segmentation processing module. The noise reduction processing module is used to reduce the noise of the image and improve the quality of the data; The segmentation processing module is used to perform segmentation processing on the collected images.

4. The local fault rapid identification device for municipal outdoor facade equipment in urban renewal according to claim 3 is characterized in that: The data preprocessing unit also includes an image enhancement module, which is used to perform enhancement processing on the collected image.

5. The local fault rapid identification device for municipal outdoor facade equipment in urban renewal according to claim 1 is characterized in that: The specific steps of the fault identification are: Obtain the faults that occur in the corresponding brightened area and the image data when the faults occur, perform feature extraction, and divide the obtained data and the data obtained by feature extraction into training data and verification data; Select the corresponding classifier as the fault identification classifier; Use training data to train the classifier to learn the characteristics and classification rules of the data; Use validation data to verify and evaluate the trained classifier to determine the performance of the classifier; The real-time data obtained is input into a qualified classifier for fault identification to determine the fault location in the illuminated area.

6. The local fault rapid identification device for municipal outdoor facade equipment in urban renewal according to claim 1 is characterized in that: The data obtained by the feature extraction includes the mean, variance, skewness and kurtosis of the data.

7. The device for rapid identification of local faults of municipal outdoor facade equipment in urban renewal according to claim 4, characterized in that: The image noise reduction process is specifically as follows: Assume the noisy model: f(k) = s(k) + n(k), k = 0, 1, 2, ... N-1 Among them, s(k) is the original image signal, and n(k) is Gaussian white noise; Perform discrete wavelet transform on f(k): q i,k =Q s (j,k)+Q n (j,k),j=0,1,2,…,J;k=0,1,2,…,N where q i,k , Q s (j,k) and Q n (j, k) are the noisy image signal, the original image signal and the wavelet coefficients of the noise at the nth layer, and the total number of pixels of the image with the maximum decomposition layer number J and N, respectively; The hard threshold function expression is: The soft threshold function expression is: in To estimate the wavelet coefficients, q j,k is the decomposition wavelet coefficient, λ is the threshold; The improved threshold function, α is the adjustment coefficient, and 0≤α≤1, when |q j,k When |→λ, That is, the improved threshold function is continuous at the threshold point; when |q j,k When |→∞, That is, with |q j,k |Gradually increasing, With q j,k The deviation between them becomes smaller and smaller, solving the constant deviation shortcoming of the soft threshold function.

8. The device for rapid identification of local faults of municipal outdoor facade equipment in urban renewal according to claim 4, characterized in that: The image segmentation process is specifically as follows: First, the image is expressed as an undirected graph G = (V, E), where V is the set of nodes, E is the set of edges connecting nodes, and the basis of V is N = |V|; the edge connecting each two nodes is given a weight w = (u, n), which measures the similarity between nodes u and v; the image G = (V, E) can be divided into two non-intersecting sets A and B by simply removing the edges connecting the two parts, and satisfying A∪B = V, A∩B = Φ. The dissimilarity of the two parts is calculated by the sum of the weights of the removed edges, defined as, Determine the normalized segmentation criterion: Normalized segmentation adds the ratio of the nodes in each region to the sum of the weights of all nodes on the basis of the minimum cut to balance the segmentation. The formula for normalized segmentation is expressed as: where assoc(A,V)=∑ u∈A,t∈V w(u,t) is the sum of the weights of the edges from all nodes in A to all nodes in the image, and assoc(B,V) is the sum of the weights of the edges from all nodes in B to all nodes in the image. In the case of segmenting isolated points, the value in the normalized segmentation is relatively large: Among them, assoc(A,A) and assoc(B,B) represent the sum of the weights of the edges connecting the nodes in A and B respectively. The above formula reflects the balance of the definition and represents the association degree of the nodes in each group; The Ncut criterion can overcome the problem of dividing isolated points. The division corresponding to the minimum Ncut value is the optimal division of graph G. Minimizing Ncut can be transformed into the following standard feature system: Where D is a diagonal matrix of NXN, and its diagonal elements are (i, j), W is a symmetric matrix whose elements are w(i, j), λ and z are the corresponding eigenvalues ​​and eigenvectors; The eigenvector corresponding to the second minimum eigenvalue of the eigensystem is used to complete the optimal division of the entire image and obtain a segmentation result of the corresponding image.

9. The local fault rapid identification device for municipal outdoor facade equipment in urban renewal according to claim 1 is characterized in that: The identification method of the identification device comprises the following steps: Conduct real-time detection of the illuminated areas on the facades of buildings and collect image data of the illuminated areas; Preprocessing the collected image data; Perform feature extraction on the preprocessed image; The extracted feature data is input into the classifier to identify image faults according to the data features; The identified fault results are wirelessly uploaded to the server to alert the on-duty personnel.

10. The local fault rapid identification device for municipal outdoor facade equipment in urban renewal according to claim 9, characterized in that: The output value of the classifier model is: Among them, X i and Y i is the sample observation value; σ is the smoothing factor; n is the number of samples.

Images