CN111414500A - Steel wire rope breakage early warning system based on block chain and BIM - Google Patents
Steel wire rope breakage early warning system based on block chain and BIM Download PDFInfo
- Publication number
- CN111414500A CN111414500A CN202010382892.1A CN202010382892A CN111414500A CN 111414500 A CN111414500 A CN 111414500A CN 202010382892 A CN202010382892 A CN 202010382892A CN 111414500 A CN111414500 A CN 111414500A
- Authority
- CN
- China
- Prior art keywords
- wire rope
- steel wire
- neural network
- block
- area
- 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- Withdrawn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
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/40—Information retrieval; Database structures therefor; File system structures therefor of multimedia data, e.g. slideshows comprising image and additional audio data
- G06F16/48—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
- G06F16/483—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/27—Replication, distribution or synchronisation of data between databases or within a distributed database system; Distributed database system architectures therefor
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/08—Construction
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
Abstract
The invention discloses a steel wire rope breakage early warning system based on a block chain and a BIM. The system comprises: the steel wire rope segmentation encoder, the steel wire rope segmentation decoder, the broken wire burr region extraction unit, the edge information extraction unit and the full connection layer are main component modules of a steel wire rope fracture risk prediction depth neural network; the system further comprises a server cluster, all nodes in the server cluster load parameters required by the steel wire rope fracture risk prediction deep neural network, for the received steel wire rope color image, steel wire rope fracture risk prediction deep neural network block private links are configured in the server cluster, and network reasoning is executed to obtain a steel wire rope fracture risk grade prediction result. By using the method and the device, in the steel wire rope fracture detection, not only are the results fed back to be multivariate, but also the steel wire rope fracture risk prediction precision and the safety performance in the data processing process are improved.
Description
Technical Field
The invention belongs to the technical field of block chains, artificial intelligence, intelligent construction sites and BIM, and particularly relates to a steel wire rope breakage early warning system based on the block chains and the BIM.
Background
With the rapid development of the economy in China, real estate and industries related to buildings are playing more and more important roles. The number of construction sites is increasing at present, and with this, the management problems inside the sites are becoming more and more prominent. The tower crane is indispensable in modern buildings due to the fact that the building height of the modern buildings is higher and higher. It plays an important role in transporting construction raw materials such as reinforcing steel bars, concrete, steel pipes and the like to high-rise buildings.
The tower crane steel wire rope is woven by a plurality of thin steel wires through special equipment. The steel wire rope can produce wearing and tearing along with the increase of operation number of times, and the steel wire of its edge can fracture, and cracked steel wire can be the trend of outdiffusion under the effect of external tension, no longer hugs closely at the steel wire rope main part to produce the fracture burr. If not repair or change in time, the steel wire broken string can be more and more to the security of direct influence tower crane effect, greatly increased heavy object risk of dropping.
Because the two ends of the tower crane steel wire rope are both provided with the reel devices, the tower crane steel wire rope is finally stacked layer by layer, and slight line breakage cannot be found in time. In the operation process, the operation space is limited, the visual field is insufficient, and the detection of the wire breakage of the tower crane steel wire rope and the analysis of the wire breakage level are unrealistic only by manpower. At present, the technology for early warning of the breakage of the steel wire rope mainly adopts a mechanical proximity switch or a sensor, the two modes are used for detecting the broken steel wire rope under the condition that the steel wire rope is broken, and the risk grade of the breakage of the steel wire rope cannot be predicted. Moreover, the current supervision system usually only returns an abnormal result, and the result feedback is single. In addition, the hardware system used for calculation is easy to leak information and has low safety performance.
Therefore, the problems that the risk grade prediction cannot be realized, the result feedback is single, and the safety performance is low exist in the existing steel wire rope fracture detection field.
Disclosure of Invention
The invention aims to provide a steel wire rope breakage early warning system based on a block chain and BIM (building information modeling) aiming at the defects in the prior art, so that not only are the results fed back to multiple elements, but also the prediction precision and the safety performance in the data processing process are improved.
A wire rope fracture early warning system based on block chain and BIM, this system includes:
the steel wire rope segmentation encoder is used for encoding and extracting the characteristics of the steel wire rope color image to obtain a steel wire rope semantic characteristic diagram;
the steel wire rope segmentation decoder is used for decoding and representing the steel wire rope semantic feature map to obtain a mask image containing steel wire rope main bodies, steel wire rope broken wire burrs and other irrelevant element semantics;
the broken wire and burr area extracting unit is used for extracting a broken wire and burr area from the steel wire rope color image by using the mask image to obtain a broken wire and burr area image;
the edge information extraction unit is used for carrying out thresholding on the image of the broken wire burr area, carrying out median filtering on the image subjected to thresholding, and carrying out edge detection on the image subjected to filtering to obtain an edge image of the broken wire burr area;
the edge characteristic encoder is used for extracting encoding characteristics of the edge graph of the broken wire burr area to obtain the edge characteristic graph of the broken wire burr area;
the first full-connection network is used for carrying out weighted classification on the edge characteristic diagram of the broken wire burr area to obtain a prediction result of the breakage risk level of the steel wire rope;
the system further comprises a server cluster, for the received steel wire rope color images, steel wire rope fracture risk prediction deep neural network block private chains are configured in the server cluster, network reasoning is executed, steel wire rope fracture risk grade prediction results are obtained, and the steel wire rope fracture risk grade prediction results are sent to the building information model of the construction site area.
Further, for the received steel wire rope color image, configuring a steel wire rope fracture risk prediction deep neural network block chain private chain in the server cluster comprises:
the steel wire rope segmentation encoder, the steel wire rope segmentation decoder, the broken wire burr region extraction unit, the edge information extraction unit, the edge characteristic encoder and the full connection layer are main component modules of a steel wire rope fracture risk prediction depth neural network;
all nodes in the server cluster load parameters required by the steel wire rope fracture risk prediction deep neural network;
for the received steel wire rope color image, selecting a plurality of available nodes from a server cluster, taking the steel wire rope segmentation encoders, steel wire rope segmentation decoders, wire breakage burr region extraction units, edge information extraction units, edge feature encoders and parameters required by a steel wire rope breakage risk prediction deep neural network of a full connection layer which are respectively distributed in different available nodes as block data of corresponding nodes, connecting the node blocks according to the steel wire rope breakage risk prediction deep neural network reasoning sequence, and generating a steel wire rope breakage risk prediction deep neural network block chain private chain.
Further, the system also comprises a data acquisition unit for acquiring the color image of the steel wire rope.
Further, the block in the private chain of block chains encrypts the neural network inference intermediate result data to be transmitted to the next block, and decrypts the neural network inference intermediate result data received from the previous block.
Further, the method for encrypting and decrypting the block specifically comprises the following steps:
generating a key lookup table for the block chain private chain, wherein the key lookup table comprises the number n of invalid channels added into the block and the insertion position of each invalid channel; according to a key lookup table, tensor data C H W to be transmitted of a block respectively represent the number, height and width of tensor channels, invalid data of n channels are added as confusion data according to the insertion positions of the invalid channels, n is an integer and is 0, and the data to be transmitted of the block is encrypted and deformed into (C + n) H W; and the block eliminates n-channel invalid data from the tensor data according to the insertion position of each invalid channel in the key lookup table to finish decryption on the received tensor data.
Further, a steel wire rope segmentation encoder, a steel wire rope segmentation decoder, a broken wire burr region extraction unit, an edge information extraction unit, an edge feature encoder and a full connection layer are respectively and properly subdivided, parameters of each subdivided module which are respectively distributed in different nodes are used as block data, and a steel wire rope breakage risk grade prediction deep neural network block chain private chain is generated according to a steel wire rope breakage risk grade prediction deep neural network reasoning sequence.
Further, the construction information model of the construction site area is a three-dimensional model of the construction site area constructed based on the BIM technology, and comprises construction site scene modeling information, monitoring area information and a steel wire rope fracture risk level prediction result.
Further, a Web GIS technology is combined, a visualization unit is used for visualizing the building information model of the construction site area, and the visualization unit comprises:
the initialization module is used for acquiring building site scene modeling information from the building information model of the building site area, rendering the building information model of the building site area by combining a Web GIS technology, and displaying on a foreground Web page to obtain an initial display result of the building information model of the building site area;
the data acquisition module is used for acquiring monitoring area information and a steel wire rope fracture risk grade prediction result from the construction information model of the construction site area;
and the visualization module is used for matching the monitoring area information and the steel wire rope fracture risk grade prediction result into an initial display result according to the geographical position of the monitoring area information, and carrying out warning marking on the monitoring area according to the abnormal early warning information.
Further, the edge information extraction unit performs edge detection on the filtered image by using a Sobel operator.
Compared with the prior art, the invention has the following beneficial effects:
1. according to the invention, the steel wire rope color image is analyzed by adopting the deep neural network to obtain the steel wire rope fracture risk level, and compared with the traditional sensor-based technology, the detection efficiency is higher, and higher detection precision can be obtained.
2. The method is based on the block chain technology, reasonably divides the steel wire rope fracture risk level prediction deep neural network, dynamically generates the block chain private chain aiming at each network inference request, and compared with the traditional single-machine execution, not only improves the parallel performance of the system, but also has better fault tolerance performance.
3. The block chain private chain is generated in real time according to available nodes in the server cluster, is not easy to be tampered and attacked, and has higher confidentiality.
4. The invention encrypts the data between the private chains of the network inference block chain, prevents the leakage of the transmission data between the private chain blocks of the block chain, ensures the confidentiality of the transmission data, has convenient tensor confusion encryption operation and small calculation amount, and can not increase the system burden while improving the confidentiality.
5. According to the method, the building information model of the construction site area is designed based on the BIM technology to store the steel wire rope fracture risk level, the building information model of the construction site area is visualized, and compared with the traditional result feedback, the feedback result is more diversified, and comprises the three-dimensional display of the construction site area, the warning mark and the steel wire rope image, so that a supervisor can know the steel wire rope condition more clearly and definitely.
Drawings
FIG. 1 is a diagram of a neural network architecture of the system of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention will be described in further detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The invention provides a steel wire rope breakage early warning system based on a block chain, which adopts a non-contact scheme to judge the integral breakage risk by judging the steel wire breakage and the magnetic field induction strength of a steel wire rope. According to the method, on one hand, the color image is adopted to predict the shade information of the steel wire rope and the wire breakage burrs of the steel wire rope, on the other hand, the wire breakage burr information is combined with the magnetic induction intensity information, and finally the wire breakage risk grade information of the steel wire rope is predicted. The invention establishes a BIM model, encrypts each module of the deep neural network by adopting the thought of a block chain, and realizes distributed calculation of each module, thereby enhancing the data security, stability and disaster tolerance of the deep neural network. FIG. 1 is a diagram of a neural network architecture of the system of the present invention. The following description will be made by way of specific examples.
The first embodiment is as follows:
the steel wire rope breakage early warning prediction system based on the block chain and the BIM carries out steel wire rope breakage early warning based on an urban construction site information model. The urban construction site information model comprises construction site scene modeling information, monitoring area information and a steel wire rope fracture risk level prediction result.
Specifically, a Building Information model (Building Information Modeling) is a new datamation tool applied to design, construction and management of Building engineering, and integrates all Building data in a Building project cycle. The visualization of the building information model can be realized by using visualization software in combination with technologies such as WebGIS and the like, and convenience is provided for building design and building management.
The method and the device are combined with BIM to display the prediction result of the steel wire rope fracture risk in real time and provide early warning information for the supervisory personnel. Thus, the present invention designs a site area building information model. The building site regional building information model is based on the BIM technology and comprises building site scene modeling information, monitoring region information and steel wire rope fracture risk prediction information. The building site scene modeling information comprises building site building information, building site building material placement information, worker work area information and other various building site scene information, the type information simultaneously contains corresponding geographic position information, and the three-dimensional scene of the building site can be restored and displayed through visual software by combining a Web GIS technology. The monitoring area information comprises monitoring area geographical position information, monitoring visual sensor geographical position information, images shot by the monitoring visual sensors and a coordinate transformation matrix. The monitored area information is used to restore the monitored area image in the visualized building information model of the worksite area. The coordinate change matrix is used for converting the monitoring image into an image with better visualization effect, such as a top view and the like. The steel wire rope breakage risk prediction information comprises a steel wire rope breakage risk level prediction result, and the steel wire rope breakage risk level prediction result is integrated into a matched monitoring area to be checked by a supervisor. In the invention, the prediction result of the steel wire rope fracture risk level is transmitted to the building information model of the construction site area in real time.
The following is a detailed description of how to obtain the prediction result of the steel wire rope fracture risk level. According to the invention, a series of data collected by the sensor are processed, and the characteristics of the sensor sensing information are extracted by adopting a deep learning method, so that the purpose of predicting the fracture risk level of the steel wire rope is achieved.
As is known, steel cords are made up of a plurality of thin steel wires, which are braided by special equipment. After the steel wire rope is worn, the steel wires at the edges can be broken, namely, the phenomenon of wire breakage occurs, the broken wires tend to spread outwards under the action of external tension and are not attached to the steel wire rope body, and therefore breaking burrs are generated. This wire breakage burr increases the risk of breakage of the steel cord.
Aiming at the steel wire rope color image, firstly, a homomorphic filtering algorithm is adopted to remove the interference of ambient illumination, so that the acquired image data are in the same data distribution.
The steel wire rope fracture risk level prediction deep neural network comprises a steel wire rope semantic segmentation sub-network and a steel wire rope fracture risk level prediction sub-network. The steel wire rope semantic segmentation sub-network comprises a steel wire rope segmentation encoder and a steel wire rope segmentation decoder; the steel wire rope segmentation encoder is used for encoding and extracting the characteristics of the steel wire rope color image to obtain a steel wire rope semantic characteristic diagram; and the steel wire rope segmentation decoder is used for decoding and representing the steel wire rope semantic feature map to obtain a mask image containing steel wire rope subjects, steel wire rope broken wire burrs and other irrelevant element semantics. The steel wire rope breakage risk grade prediction sub-network comprises a wire breakage burr region extraction unit, an edge information extraction unit, an edge characteristic encoder and a first full-connection network; the broken wire and burr area extracting unit is used for extracting a broken wire and burr area from the steel wire rope color image by using the mask image to obtain a broken wire and burr area image; the edge information extraction unit is used for carrying out thresholding on the image of the broken wire burr region, carrying out median filtering on the image after thresholding, and carrying out edge detection on the image after filtering by adopting a Sobel operator to obtain an edge image of the broken wire burr region; the edge characteristic encoder is used for extracting encoding characteristics of the edge graph of the broken wire burr area to obtain the edge characteristic graph of the broken wire burr area; and the first full-connection network is used for carrying out weighted classification on the edge characteristic graph of the broken wire burr area to obtain a prediction result of the steel wire rope breakage risk level.
As shown in fig. 1, a steel wire rope semantic segmentation sub-network firstly extracts a steel wire rope semantic feature map featuremaa of a steel wire rope and burrs thereof through a steel wire rope segmentation encoder EncoderA, inputs the obtained feature map into a steel wire rope segmentation decoder DecoderA, and finally obtains mask information of the steel wire rope and the broken wire burrs thereof, and distinguishes semantics of a steel wire rope body, the broken wire burrs of the steel wire rope and other irrelevant elements. It should be noted that the predicted mask information not only includes the category of the current pixel, but also indirectly obtains the specific geographic position coordinate information of the wire rope area.
In the training process, the semantic segmentation sub-network of the steel wire rope inputs a color picture, the output is the shielding information of the steel wire rope area and the broken wire burrs of the steel wire rope area, and the monitoring information is a three-value image which is marked manually. The three-value image means that the pixel value of the steel wire rope body is marked as 1, the pixel value of the broken wire burr is marked as 2, and other background pixel information is marked as 0. And training the network by adopting cross entropy loss to obtain the mask information. And finishing the extraction of the information of the steel wire rope and the broken wire burrs.
As is known, construction site environments are complex and variable, illumination conditions are unstable, and if feature extraction is directly performed on a filament-broken and burr-broken RGB three-channel color image mapped to an original image after segmentation, due to illumination differences, the filament-broken and burr-broken image is not in the same data distribution, so that a lot of interference information is introduced, which affects convergence of a model on one hand and adversely affects classification results on the other hand. The edge information of the broken wire burrs is very rich, and the edge characteristics can truly reflect the amount of the burrs and the breaking degree of the burrs. The invention designs the method for predicting the edge image of the broken wire burr, which is in the same data distribution, can overcome the problem of model convergence caused by unstable illumination, accelerate the convergence speed and improve the generalization capability of the model.
The invention adopts image post-processing operation to obtain the burr image edge information. And the broken wire burr area extraction unit obtains an image of the broken wire burr area through segmentation. Specifically, a broken wire and burr area in the mask image is set to be 1, other areas are set to be 0, point-to-point multiplication operation is carried out on the obtained mask image and a steel wire rope color image, and then a broken wire and burr area image can be extracted.
Obtaining a binary image of the burr area by adopting an Otsu threshold value method in an edge information extraction unit, thereby obtaining foreground information of burr and broken wire; and then removing the high-frequency noise information of the binary image by adopting a median filtering method. And then, extracting the edge information of the filtered binary image by using a Sobel operator in an edge information extraction unit.
The Sobel operator detects the edge according to the gray weighting difference of the upper, lower, left and right adjacent points of the pixel point, and the phenomenon that the edge reaches an extreme value. In the present invention, the thickness of the edge is set to one pixel unit. And newly building a blank single-channel picture sobel _ image to store the edge information of the broken wire burr area. The picture size is equal to the size of the original color image, and the edge information extracted by the sobel operator is copied to the single-channel picture sobel _ image. This completes the post-processing operation.
And then, extracting the features of the sobel _ image by adopting an edge feature encoder EncoderB to obtain an edge feature map FeatureMapp B, and finally predicting the steel wire rope fracture risk level through the weighted classification of full-connection operation in the first full-connection network.
In the training process, the input of the sub-network for predicting the fracture risk level of the steel wire rope is the edge image of the wire-breaking burr area, the output is the fracture risk level of the steel wire rope, and the monitoring information is manually marked fracture risk level marking information of the wire-breaking burr information. And (3) updating parameters by adopting the commonly used cross entropy loss, and finally obtaining a fracture risk grade discrimination model of the steel wire rope. The invention divides the fracture risk of the steel wire rope into four grades: index 0 represents no risk of fracture, index 1 represents low risk of fracture, index 2 represents medium risk of fracture, and index 3 represents high risk of fracture. In the implementation scenario, the implementer may manually label the ratings as appropriate. The larger the edge change of the broken wire burrs is, the more the broken wire burrs are, and the higher the risk is.
In the neural network, an encoder is used for carrying out feature extraction on input multi-channel two-dimensional data, and a decoder is used for carrying out up-sampling decoding characterization on the features. The encoder and the decoder can be realized in various ways, and CNN Block, ResBlock and the like can be adopted. In order to take account of the size of a large target and a small target, the invention proposes to adopt an hourglass network to extract features, and an implementer can also select a proper module design in a neural network, such as Residual Block, Bottleneck Block, CNN Block and the like according to the size of an image and the occupation of a video memory. The encoder, the decoder and the full-connection network of the invention adopt which network design, an implementer can select according to the specific implementation requirements, and the modularization idea is the protection content of the invention.
In order to improve the confidentiality of the system and prevent data leakage, the invention designs the private chain of the block chain by combining the block chain technology.
The present invention is described in detail herein in connection with the block chain technique and the DNN technique. The block chain adopts block division data, a chain data structure is used, the data are used as blocks for verification and storage, the whole data structure is summarized, centralized hardware and management mechanisms do not exist, and decentralization is achieved. The block chain technology of the first generation is mainly applied to a distributed account book, the block chain technology of the second generation mainly realizes an intelligent contract, and the block chain idea and other field technologies can be combined by the block chain technology of the third generation, so that more and more presentation forms exist, and more emphasis is placed on system function service. The block chain private chain completely inherits the characteristics of the public chain, is not bound by a game mechanism, focuses more on data transmission and encryption and decryption processing of practical application, and can be better combined with technologies in other fields. For deep neural network computation in artificial intelligence, it is not necessary to store intermediate result data, and chained logic is retained to match the principle of neural network forward propagation.
The invention considers the problem that the contents of the image data in the plaintext are leaked in the uploading process and the processing process when the data are directly uploaded to be processed, so that the invention uses the form of a block chain private chain, takes different modules of a deep neural network as blocks, carries out dispersed reasoning and carries out encryption processing on the data transmitted among the blocks, thereby realizing excellent performances of parallel reasoning, fault tolerance and data leakage prevention.
Based on the thought, firstly, the steel wire rope fracture risk grade prediction depth neural network is subjected to module division, and a steel wire rope segmentation encoder, a steel wire rope segmentation decoder, a broken wire burr region extraction unit, an edge information extraction unit, an edge characteristic encoder and a full connection layer are used as different modules of the network. Thus, according to the inference sequence of the neural network shown in fig. 1, the steel wire rope fracture risk level prediction depth neural network inference chain can be obtained.
The system also comprises a server cluster, wherein all nodes in the server cluster are loaded with weights and parameters required by the steel wire rope fracture risk level prediction deep neural network; and selecting a plurality of available nodes from the server cluster according to the prediction depth neural network inference request of each steel wire rope fracture risk level, and taking the steel wire rope segmentation encoders, the steel wire rope segmentation decoders, the broken wire burr region extraction units, the edge information extraction units, the edge characteristic encoders, the weights and the parameters required by the full-connection layer which are respectively distributed at different available nodes as block data of corresponding nodes. Therefore, the wire rope segmentation encoder block, the wire rope segmentation decoder block, the wire breakage burr area extraction unit block, the edge information extraction unit block and the full connection layer block which are distributed on different available nodes can be obtained. And connecting the blocks according to the steel wire rope fracture risk grade prediction deep neural network reasoning sequence to generate a steel wire rope fracture risk grade prediction deep neural network reasoning block chain private chain, and executing steel wire rope fracture risk grade prediction deep neural network reasoning. The private chain order of the block chain is consistent with the network inference chain order, as shown in fig. 1.
When selecting available nodes and performing node sorting, preferably, the available nodes in the server cluster are randomly sorted, and the number of computing nodes is selected from the nodes, wherein the number of the computing nodes is the same as the number of the blocks. For example, 10 available nodes are selected, 6 nodes are selected from the available nodes, one node is randomly selected, and parameters such as weights required by a wire rope segmentation encoder in the nodes are used as block data; and randomly taking another node, taking parameters such as weight required by the steel wire rope segmentation decoder in the node as block data, linking the block data with the previous block, and by analogy, generating a corresponding steel wire rope fracture risk level prediction deep neural network inference block chain private chain according to the neural network inference sequence. Therefore, a plurality of steel wire rope fracture risk level prediction deep neural network inference block private chains generated according to different requests can exist in the server cluster at the same time, and the block private chains are dynamically generated, are not easy to crack by attack, and have better confidentiality. And after the block chain private chain is obtained according to the inference sequence, performing network inference calculation on the inference request according to the inference sequence.
Some modules of the neural network are difficult to be put into one node at one time, namely, the computation of some modules is large, and the computation is difficult to be completed in a short time. Therefore, the blocks, i.e., the encoder and the decoder, may be further divided in advance. The granularity of the further segmentation can be adjusted by the implementer according to the specific implementation. In this embodiment, preferably, the wire rope segmentation encoder is segmented into 3 blocks, the wire rope segmentation decoder is segmented into 3 blocks, the edge feature encoder is segmented into 3 blocks, the first fully-connected network is segmented into 4 blocks, and the rest of the units may not be segmented. Therefore, according to the inference sequence of the neural network, a more subdivided steel wire rope fracture risk grade prediction deep neural network inference chain can be obtained. Correspondingly, aiming at each steel wire rope breakage risk grade prediction deep neural network reasoning request, selecting a plurality of available nodes from a server cluster, taking the subdivided steel wire rope segmentation encoders, steel wire rope segmentation decoders, wire breakage burr region extraction units, edge information extraction units, edge feature encoders and sub-module required weights and parameters of a full connection layer which are respectively distributed in different available nodes as block data, predicting a deep neural network reasoning sequence according to the steel wire rope breakage risk grade, and generating a steel wire rope breakage risk grade prediction deep neural network reasoning block chain private chain.
Meanwhile, the data acquisition unit can be used as a block and added into a block chain private chain, so that the purpose of ensuring that the output of the image acquisition unit is confidential through a subsequent encryption strategy and is not easy to intercept and tamper is achieved. Meanwhile, the monitoring center receiving the prediction result can be added into the block chain, so that the confidentiality of data transmission between the server cluster and the monitoring center is ensured.
Further, in order to ensure the integrity and security of data received by each block and prevent data from being attacked and tampered during transmission, data transmitted between blocks needs to be encrypted. That is, the data transmission from the private link node to the node of the block chain needs to use encryption means.
Specifically, the following encryption and decryption methods are adopted: generating a key lookup table for the block chain private chain, wherein the key lookup table comprises the number n of invalid channels added into the block and the insertion position of each invalid channel; according to a key lookup table, tensor data C H W to be transmitted of a block respectively represent the number, height and width of tensor channels, invalid data of n channels are added as confusion data according to the insertion positions of the invalid channels, n is an integer and is 0, and the data to be transmitted of the block is encrypted and deformed into (C + n) H W; and the block eliminates n-channel invalid data from the tensor data according to the insertion position of each invalid channel in the key lookup table to finish decryption on the received tensor data. For the case where the intermediate result is a vector or a single data, the implementer may use a general encryption algorithm to place the encryption and decryption keys in the key lookup table.
And applying the encryption algorithm to the data to be transmitted among all the blocks to finish the encryption operation of the data. After the data is input into the next block, the data is decrypted according to the reverse reasoning of the encryption rule, so that the encrypted transmission of the data between the computing nodes is completed.
It should be noted that encryption parameters required for data transmission between the wire rope fracture risk level prediction deep neural network modules, namely between blocks, should be generated by the trusted nodes. The trusted node may be fixed, for example, a master node may be configured to broadcast encryption parameters and allocation tables periodically to prevent the encryption parameters from being broken. However, the fixed trusted node is vulnerable to attack, and therefore, preferably, the encryption parameters required for the next inference request are generated by the node of the last block of the private chain of the blockchain and broadcast to all nodes in the server cluster.
In order to visually present the BIM information state of the current construction site area and enable a construction site manager to visually acquire the camera perception information of the current area and the information of whether the network forecast has the fracture risk, the BIM information model is displayed on the Web through the Web GIS technology by combining the Web GIS visualization technology.
And the visualization unit is used for acquiring data from the building information model of the construction site area, rendering the building information model of the construction site area by combining the Web GIS, and displaying the fracture condition of the steel wire rope on a foreground Web page.
And the initialization module is used for acquiring the building site scene modeling information from the building site area building information model, rendering the building site area building information model by combining the Web GIS technology, and displaying on a foreground Web page to obtain an initial display result of the building site area building information model.
And the data acquisition module is used for acquiring monitoring area information and a steel wire rope fracture risk prediction result from the construction information model of the construction site area.
And the steel wire rope fracture condition visualization module is used for matching the monitoring area information and the steel wire rope fracture risk prediction result into an initial display result from the geographical position according to the monitoring area information, and carrying out warning marking on the monitoring area according to the abnormal early warning information.
Meanwhile, when the risk of breaking the steel wire rope is predicted to be high, an alarm can be sent out, and the corresponding position is marked in the building information model, so that the supervision personnel can take corresponding preventive and emergency measures according to the geographic position coordinates of the supervision personnel.
The above embodiments are merely preferred embodiments of the present invention, which should not be construed as limiting the present invention, and any modifications, equivalents and improvements made within the spirit and principle of the present invention should be included in the scope of the present invention.
Claims (9)
1. The utility model provides a wire rope fracture early warning system based on block chain and BIM which characterized in that, this system includes:
the steel wire rope segmentation encoder is used for encoding and extracting the characteristics of the steel wire rope color image to obtain a steel wire rope semantic characteristic diagram;
the steel wire rope segmentation decoder is used for decoding and representing the steel wire rope semantic feature map to obtain a mask image containing steel wire rope main bodies, steel wire rope broken wire burrs and other irrelevant element semantics;
the broken wire and burr area extracting unit is used for extracting a broken wire and burr area from the steel wire rope color image by using the mask image to obtain a broken wire and burr area image;
the edge information extraction unit is used for carrying out thresholding on the image of the broken wire burr area, carrying out median filtering on the image subjected to thresholding, and carrying out edge detection on the image subjected to filtering to obtain an edge image of the broken wire burr area;
the edge characteristic encoder is used for extracting encoding characteristics of the edge graph of the broken wire burr area to obtain the edge characteristic graph of the broken wire burr area;
the first full-connection network is used for carrying out weighted classification on the edge characteristic diagram of the broken wire burr area to obtain a prediction result of the breakage risk level of the steel wire rope;
the system further comprises a server cluster, for the received steel wire rope color images, steel wire rope fracture risk prediction deep neural network block private chains are configured in the server cluster, network reasoning is executed, steel wire rope fracture risk grade prediction results are obtained, and the steel wire rope fracture risk grade prediction results are sent to the building information model of the construction site area.
2. The system of claim 1, wherein configuring, for the received wire rope color image, a wire rope breakage risk prediction deep neural network block chaining private chain in the server cluster comprises:
the steel wire rope segmentation encoder, the steel wire rope segmentation decoder, the broken wire burr region extraction unit, the edge information extraction unit, the edge characteristic encoder and the full connection layer are main component modules of a steel wire rope fracture risk prediction depth neural network;
all nodes in the server cluster load parameters required by the steel wire rope fracture risk prediction deep neural network;
for the received steel wire rope color image, selecting a plurality of available nodes from a server cluster, taking steel wire rope segmentation encoders, steel wire rope segmentation decoders, wire breakage burr region extraction units, edge information extraction units and parameters required by a full-connection layer, namely steel wire rope breakage risk prediction depth neural network, distributed at different available nodes as block data of corresponding nodes, and connecting the node blocks according to the steel wire rope breakage risk prediction depth neural network reasoning sequence to generate a steel wire rope breakage risk prediction depth neural network block chain private chain.
3. The system of claim 1, further comprising a data acquisition unit for acquiring a color image of the wire rope.
4. The system of claim 1, wherein a block in the private chain of blocks encrypts neural network inference intermediate result data that it is to transmit to a next block and decrypts neural network inference intermediate result data that it receives from a previous block.
5. The system of claim 4, wherein the block encryption and decryption method comprises:
generating a key lookup table for the block chain private chain, wherein the key lookup table comprises the number n of invalid channels added into the block and the insertion position of each invalid channel; according to a key lookup table, adding invalid data of n channels into tensor data C H W to be transmitted of a block according to the inserting position of the invalid channel to serve as confusion data, wherein n is an integer and n > is 0, and encrypting and deforming the data to be transmitted of the block into (C + n) H W; and the block eliminates n-channel invalid data from the tensor data according to the insertion position of each invalid channel in the key lookup table to finish decryption on the received tensor data.
6. The system of claim 1, wherein the wire rope segmentation encoder, the wire rope segmentation decoder, the wire breakage burr region extraction unit, the edge information extraction unit, the edge feature encoder and the full connection layer are respectively and properly subdivided, parameters of each subdivided module respectively distributed in different nodes are used as block data, and a wire rope breakage risk level prediction deep neural network block chain private chain is generated according to a wire rope breakage risk level prediction deep neural network inference sequence.
7. The system of any one of claims 1-6, wherein the worksite area building information model is a three-dimensional model of a worksite area constructed based on BIM technology, and comprises worksite scene modeling information, monitoring area information and prediction results of steel wire rope breakage risk level.
8. The system of claim 7, wherein the site area building information model is visualized with a visualization unit in conjunction with Web GIS technology, the visualization unit comprising:
the initialization module is used for acquiring building site scene modeling information from the building information model of the building site area, rendering the building information model of the building site area by combining a Web GIS technology, and displaying on a foreground Web page to obtain an initial display result of the building information model of the building site area;
the data acquisition module is used for acquiring monitoring area information and a steel wire rope fracture risk grade prediction result from the construction information model of the construction site area;
and the visualization module is used for matching the monitoring area information and the steel wire rope fracture risk grade prediction result into an initial display result according to the geographical position of the monitoring area information, and carrying out warning marking on the monitoring area according to the abnormal early warning information.
9. The system according to claim 1, wherein the edge information extraction unit performs edge detection on the filtered image using a Sobel operator.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010382892.1A CN111414500A (en) | 2020-05-08 | 2020-05-08 | Steel wire rope breakage early warning system based on block chain and BIM |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010382892.1A CN111414500A (en) | 2020-05-08 | 2020-05-08 | Steel wire rope breakage early warning system based on block chain and BIM |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111414500A true CN111414500A (en) | 2020-07-14 |
Family
ID=71495045
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010382892.1A Withdrawn CN111414500A (en) | 2020-05-08 | 2020-05-08 | Steel wire rope breakage early warning system based on block chain and BIM |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111414500A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112529893A (en) * | 2020-12-22 | 2021-03-19 | 郑州金惠计算机系统工程有限公司 | Hub surface flaw online detection method and system based on deep neural network |
| CN112748116A (en) * | 2020-12-23 | 2021-05-04 | 郑州金惠计算机系统工程有限公司 | Medical gauze surface defect online detection method and device |
-
2020
- 2020-05-08 CN CN202010382892.1A patent/CN111414500A/en not_active Withdrawn
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112529893A (en) * | 2020-12-22 | 2021-03-19 | 郑州金惠计算机系统工程有限公司 | Hub surface flaw online detection method and system based on deep neural network |
| CN112748116A (en) * | 2020-12-23 | 2021-05-04 | 郑州金惠计算机系统工程有限公司 | Medical gauze surface defect online detection method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Mansouri et al. | A novel one-dimensional sine powered chaotic map and its application in a new image encryption scheme | |
| Khan et al. | DNA and plaintext dependent chaotic visual selective image encryption | |
| Liu et al. | Cryptanalyzing a RGB image encryption algorithm based on DNA encoding and chaos map | |
| CN101044761B (en) | Method for processing an input image securely | |
| Wu et al. | Genetic algorithm based methodology for breaking the steganalytic systems | |
| CN111563433A (en) | Wisdom building site is monitored system of overflowing water based on block chain | |
| Gayathri et al. | An efficient spatiotemporal chaotic image cipher with an improved scrambling algorithm driven by dynamic diffusion phase | |
| Somaraj et al. | Performance and security analysis for image encryption using key image | |
| CN111414500A (en) | Steel wire rope breakage early warning system based on block chain and BIM | |
| CN111428694A (en) | Abnormal smoke monitoring system of wisdom building site based on block chain | |
| CN111414499A (en) | Operation personnel safety belt wearing detection system based on block chain and BIM | |
| CN111461075A (en) | Guardrail crossing detection method combining deep neural network and block chain | |
| Wang et al. | An efficient image encryption algorithm based on a novel chaotic map | |
| CN115049070A (en) | Screening method and device of federal characteristic engineering data, equipment and storage medium | |
| Jiao et al. | Image encryption with the cross diffusion of two chaotic maps | |
| CN111583068A (en) | Statistical method for stay time and heat degree of personnel in construction area combined with block chain | |
| Liu et al. | Image encryption via complementary embedding algorithm and new spatiotemporal chaotic system | |
| Wang et al. | An image encryption scheme using bit-plane cross-diffusion and spatiotemporal chaos system with nonlinear perturbation | |
| CN111414501A (en) | Wood flammability risk prediction system based on block chain and CIM | |
| CN111524131A (en) | Wall crack detection system based on block chain and BIM | |
| CN111414504A (en) | Building site flood detection system based on block chain and CIM | |
| CN111414503A (en) | Wood accumulation detection system based on block chain and CIM | |
| CN111476208A (en) | Intelligent building site constructor gathering monitoring system utilizing block chains | |
| CN111414502A (en) | Steel wire rope burr detection system based on block chain and BIM | |
| Fu et al. | A watermarking scheme based on rotating vector for image content authentication |
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 | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20200714 |