CN111563433A - Wisdom building site is monitored system of overflowing water based on block chain - Google Patents
Wisdom building site is monitored system of overflowing water based on block chain Download PDFInfo
- Publication number
- CN111563433A CN111563433A CN202010344965.8A CN202010344965A CN111563433A CN 111563433 A CN111563433 A CN 111563433A CN 202010344965 A CN202010344965 A CN 202010344965A CN 111563433 A CN111563433 A CN 111563433A
- Authority
- CN
- China
- Prior art keywords
- construction site
- monitoring
- ground
- neural network
- overflowing
- 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- Withdrawn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
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/26—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion
- G06V10/267—Segmentation of patterns in the image field; Cutting or merging of image elements to establish the pattern region, e.g. clustering-based techniques; Detection of occlusion by performing operations on regions, e.g. growing, shrinking or watersheds
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Biomedical Technology (AREA)
- Molecular Biology (AREA)
- Biophysics (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Alarm Systems (AREA)
Abstract
The invention discloses a smart building site overflowing water monitoring system based on a block chain, which comprises a terminal cluster and a computing cluster, wherein the computing cluster is loaded with parameters required by a smart building site overflowing water monitoring deep neural network, configures a smart building site overflowing water monitoring deep neural network block chain private chain, and executes smart building site overflowing water monitoring deep neural network reasoning to realize building site overflowing water monitoring; wisdom building site flood detection degree of depth neural network's input is the monitoring area ground image that the terminal was gathered, and the output includes ground element semantic segmentation picture, building site flood grade classification result, comprises a plurality of modules, cuts apart encoder, building site ground semantic segmentation decoder, building site flood grade classification encoder, first full connection network including building site ground semantic segmentation encoder, building site ground semantic segmentation. By using the method and the system, the construction site overflowing water monitoring precision and efficiency as well as the safety and confidentiality in the data processing and transmitting process are improved in the construction site environment monitoring.
Description
Technical Field
The invention relates to the technical field of artificial intelligence, block chains, CIM and intelligent construction sites, in particular to an intelligent construction site overflowing monitoring system based on the block chains.
Background
At present, the technical means in the aspect of site environment management is not perfect. The construction site is flooded to bring some potential influences to the construction, for example, the vehicle passes through fast to cause the condition such as waters sputtering, formation rivers, causes inconvenience to building site materials, workman, and if serious, can influence the use of construction materials, bring the loss of cost.
The monitoring of the construction site flood has no corresponding technical scheme, and the flood condition is mostly supervised by manpower. The traditional manual monitoring mode has low efficiency, and the monitoring precision is low due to the subjectivity of manual monitoring. Moreover, current site condition monitoring typically returns only abnormal results, with a single feedback of results. And moreover, information is easy to leak and the safety performance is low for the calculation cluster. Therefore, the existing construction site overflowing water monitoring technology has the problems of low monitoring precision and monitoring efficiency, single result feedback and low safety performance in the data processing and transmission processes.
The development of technologies such as artificial intelligence and the Internet of things is benefited, and an intelligent construction site becomes a brand-new engineering full life cycle management idea. The invention improves the prior art on the aspects of flood monitoring in an intelligent construction site, result feedback, monitoring precision efficiency, system safety and the like.
Disclosure of Invention
The invention provides an intelligent building site overflowing monitoring system based on a block chain, which not only feeds back a plurality of results, but also improves monitoring precision, monitoring efficiency and safety and confidentiality in a data processing and transmitting process.
A smart building site overflowing water monitoring system based on a block chain comprises a terminal cluster and a computing cluster, wherein the computing cluster is loaded with parameters required by a smart building site overflowing water monitoring deep neural network, configures a smart building site overflowing water monitoring deep neural network block chain private chain, and executes smart building site overflowing water monitoring deep neural network reasoning to realize smart building site overflowing water monitoring;
wherein, wisdom building site flood detection degree of depth neural network's input is the monitoring area ground image that the terminal was gathered, and the output includes ground element semantic segmentation picture, building site flood grade classification result, comprises a plurality of modules:
the construction site ground semantic segmentation encoder is used for encoding RGB data of construction site ground images acquired by the terminal and extracting features to obtain a construction site ground semantic feature map;
the construction site ground semantic segmentation decoder is used for decoding and restoring the construction site ground semantic feature map to obtain a ground element semantic segmentation map and distinguishing the semantics of a water area, the ground and other elements;
the construction site overflowing water grade classification encoder is used for extracting encoding characteristics of RGB data and depth map data of construction site ground images acquired by the terminal to obtain a construction site overflowing water grade classification characteristic map;
and the first full-connection network is used for carrying out weighted classification on the combined feature vector obtained by jointing and unfolding the construction site ground semantic feature map and the construction site overflowing level classification feature map to obtain a construction site overflowing level classification result.
The construction site ground semantic segmentation encoder and the construction site ground semantic segmentation decoder form a construction site ground semantic segmentation sub-network, when the sub-network is trained, firstly, semantic annotation is carried out on a water area and the ground in a sample image, then, other irrelevant elements are annotated to obtain annotation data of the sample image, and the sample image and the annotation data are input into the sub-network together to be trained by adopting a cross entropy loss function.
Dispose wisdom building site flood monitoring degree of depth neural network block chain private chain includes:
all computing nodes of the computing cluster are loaded with parameters required by the intelligent construction site water-overflowing monitoring deep neural network;
according to each intelligent construction site overflowing monitoring depth neural network inference request, selecting a plurality of available nodes from a computing cluster, taking camera parameters of a terminal, construction site ground semantic segmentation encoders, construction site ground semantic segmentation decoders, construction site overflowing level classification encoders and parameters required by a first full-connection network which are respectively distributed in different available nodes as block data of corresponding nodes, connecting the blocks according to an intelligent construction site overflowing monitoring depth neural network inference sequence, and generating an intelligent construction site overflowing monitoring depth neural network block chain private chain.
Selecting a plurality of available nodes comprises:
taking one channel data in RGB three channels of the image acquired by the terminal to perform down-sampling and binarization processing to obtain binarized single channel data; expanding the single-channel binary data to obtain a binary sequence, and converting the binary sequence into a decimal number serving as a random number seed; and generating a random number sequence by using the random number seed, and selecting an available node corresponding to the index according to the size index of the numerical value in the random number sequence.
And the blocks in the private chain of the block chain encrypt the neural network inference intermediate result data to be transmitted to the next block, and decrypt the neural network inference intermediate result data received from the previous block.
The encryption and decryption adopts a byte masking encryption mechanism.
The method comprises the steps of conducting appropriate subdivision on a construction site ground semantic segmentation encoder, a construction site ground semantic segmentation decoder, a construction site overflowing level classification encoder and a first full-connection network, taking parameters of each module distributed in different nodes after subdivision as block data, and generating a block chain private chain of the intelligent construction site overflowing detection depth neural network according to an intelligent construction site overflowing detection depth neural network reasoning sequence.
Based on CIM technology, constructing an urban construction site information model, wherein the urban construction site information model comprises the following steps: the method comprises the steps of urban construction site three-dimensional space modeling information, monitoring area information and intelligent construction site overflowing monitoring result information; and rendering the city construction site information model by using a visualization unit in combination with a Web GIS technology, and displaying the rendered city construction site information model on a foreground page.
The visualization unit includes:
the system comprises an initialization module, a display module and a display module, wherein the initialization module is used for acquiring urban construction site three-dimensional space modeling information from an urban construction site information model, rendering the urban construction site information model by combining a WebGIS technology, and displaying on a foreground Web page to obtain an initial visualization result of the urban construction site information model;
the data acquisition module is used for acquiring monitoring area information and intelligent construction site overflowing monitoring result information from the city construction site information model;
and the flood visualization module is used for integrating the monitoring area information and the smart site flood monitoring result into an initial visualization result of the urban site information model according to the geographic position.
The invention has the beneficial effects that:
1. the method adopts the deep neural network to analyze the construction site ground image of the monitored area, and has more accurate result response and higher monitoring efficiency compared with the traditional manual supervision method.
2. The depth data and the RGB data are analyzed, and the sub-network is segmented by using the ground semantic of the construction site to assist training, so that the convergence of the whole network and the sub-network for classifying the grades of the flood of the construction site is accelerated.
3. According to the method, based on the block chain technology, the intelligent construction site flood monitoring deep neural network is reasonably divided, the block chain private chain is dynamically generated according to each network inference request, and compared with the traditional single-machine execution, the parallel performance of the system is improved.
4. The block chain private chain is generated in real time according to available nodes in the computing cluster, and compared with the traditional fixed distribution, the block chain private chain is not easy to attack and crack, and the confidentiality of system data is improved.
5. The invention encrypts the data between the private chains of the block chain of the network inference, prevents the leakage of the transmission data between the private chain blocks of the block chain, ensures the confidentiality of the transmission data, has small calculation amount of a byte mask encryption mechanism, and can not increase the system burden while improving the confidentiality and the safety.
6. The city construction site information model is designed based on the CIM technology to store construction site overflowing monitoring results and visualize the city construction site information model, compared with traditional result feedback, the feedback result of the city construction site overflowing monitoring method is more diversified, the city construction site overflowing monitoring method comprises construction site area three-dimensional display, warning marks and monitoring area images, and supervision personnel can know construction site overflowing conditions more clearly and definitely.
Drawings
FIG. 1 is a diagram of an intelligent construction site flood monitoring deep neural network structure 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 existing deep learning method obtains the latest achievement in the field of computer vision, introduces deep learning into construction site water overflowing monitoring, can realize automatic water overflowing monitoring, provides an early warning function, and has a certain application value. In summary, the invention provides an intelligent construction site flood monitoring system based on a convolutional neural network and a block chain. Firstly, the images obtained by the RGB-D camera are subjected to channel separation, and are respectively calculated through two parallel neural networks, one is obtained a flood segmentation image, and the other is obtained flood grade information of the image, so that the monitoring of the flood of the construction site can be realized, and the segmentation image of the construction site can be provided for visualizing data. FIG. 1 is a diagram of a deep 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 utility model provides a wisdom building site flood monitoring system based on block chain, includes terminal cluster and calculation cluster, and the required parameter of wisdom building site flood monitoring depth neural network is loaded into to the calculation cluster, through configuration wisdom building site flood monitoring depth neural network block chain private chain, carries out wisdom building site flood monitoring depth neural network and infers. The terminal is arranged in the monitoring area, can collect images of the monitoring area and has certain computing capacity.
The invention mainly monitors whether the construction site overflows or not, and realizes the safety management of the construction site environment. Specifically, the flood monitoring is realized based on a deep neural network, and a construction site flood grade classification is combined with a semantic segmentation network. The reason for designing the intelligent construction site water overflowing monitoring deep neural network is as follows: the classification of the existence of the overflowing water cannot be fit for the actual situation, the condition of the overflowing water can be judged as long as the water exists, the practicability of the application is not met, and the range of the overflowing water can be effectively judged by adopting the grade of the overflowing water; the water is a non-rigid body, the texture has single characteristic on the representation, the semantic segmentation can solve the detection problem of the non-rigid body, and other networks are not used; for water area detection of a construction site, the color and the fluidity of the water area detection are special, and the water area detection has stronger generalization capability by combining a semantic segmentation network in order to deal with different flooding conditions, such as the condition that a water area is doped with silt; the encoder of the semantic segmentation network can integrate more features into the first fully-connected network, and is more beneficial to classifying the flooding grade, so that the convergence of the flooding grade is accelerated.
The input of the intelligent construction site overflowing water monitoring depth neural network is a monitoring area ground image collected by a terminal, and the output of the intelligent construction site overflowing water monitoring depth neural network comprises a ground element semantic segmentation graph and a construction site overflowing water grade classification result. The invention is used for acquiring the image of the monitoring area by adopting the RGB-D sensor. Wisdom building site flood monitoring degree of depth neural network includes: the system comprises a construction site ground semantic segmentation encoder, a construction site ground semantic segmentation decoder, a construction site overflowing level classification encoder and a first full-connection network.
The construction site ground semantic segmentation sub-network comprises a construction site ground semantic segmentation encoder and a construction site ground semantic segmentation decoder and is used for realizing semantic analysis on construction site ground elements, the output result is a ground element semantic segmentation image of an image, and the features extracted by the construction site ground semantic segmentation encoder are helpful for carrying out level classification on construction site flood conditions. Firstly, a channel separation technology is adopted to separate RGB data in a construction site ground image acquired by an RGB-D sensor. The reason for the channel separation (channel split) is because the semantic of the worksite floor is independent of depth and does not result in a difference in the worksite floor categories due to differences in height. However, the flooding water may have the depth of the water area, and the generalization capability of the flooding level classification network can be better improved by adding the depth map feature.
Specifically, the construction site ground semantic segmentation encoder is used for encoding RGB data of construction site ground images acquired by the terminal and extracting features to obtain a construction site ground semantic feature map; the construction site ground semantic segmentation decoder is used for decoding and restoring the construction site ground semantic feature map to obtain a ground element semantic segmentation map. The ground element semantic segmentation graph mainly distinguishes a water area from the ground and other elements.
The training process for the site-ground semantic segmentation sub-network is as follows: and (3) carrying out normalization processing on the acquired image data, namely changing the picture matrix into floating point numbers between [0 and 1] so as to facilitate better convergence of the model. The pixel types of the label data are divided into three types, only the water area and the ground type are marked during marking, and the other types are marked through a program. The pixel value 0 represents the water area, 1 represents the ground, and 2 represents the others. And then, sending the processed image data and the tag data (which need to be subjected to one-hot coding) to a network, and training a construction site ground semantic segmentation encoder and a construction site ground semantic segmentation decoder to obtain a ground element semantic segmentation image of the construction site ground. The construction site ground semantic segmentation encoder is used for extracting the characteristics of the RGB image, inputting the RGB image data into a normalized RGB image data and outputting the RGB image data into a construction site ground semantic characteristic diagram; the construction site ground semantic segmentation decoder performs up-sampling on the construction site ground semantic feature map and performs pixel-level classification, inputs construction site ground semantic feature map2 generated by the construction site ground semantic segmentation encoder, outputs the construction site ground semantic feature map as a segmentation probability map, and outputs a ground element semantic segmentation image through argmax operation. In consideration of the fact that the situation of the overflowing water at the construction site is not frequent, samples of the overflowing water are relatively few, and therefore the Loss function adopts the weighted cross entropy to better solve the problem of sample proportion imbalance by distributing different weights of water areas and ground categories. Therefore, the extraction of the ground semantics of the construction site can be completed.
The construction site overflowing water grade classification sub-network extracts the characteristics of the RGB-D image obtained by the camera, combines the characteristic diagram extracted by the semantic segmentation network construction site ground semantic segmentation encoder, and outputs the construction site overflowing water grade through the full-connection network. The worksite flood level classification subnetwork comprises: a worksite flood level classification encoder and a first fully-connected network. The construction site overflowing water grade classification encoder is used for extracting encoding characteristics of RGB data of construction site ground images acquired by the terminal and the depth map to obtain a construction site overflowing water grade classification characteristic map. The first full-connection network is used for carrying out weighted classification on a combined feature map (which is obtained by combining the construction site ground semantic feature map and the construction site overtopping level classification feature map and needs to be subjected to a flatten expansion operation to be converted into a combined feature vector) so as to obtain a construction site overtopping level classification result. The construction site water overflowing grades are divided into a plurality of grades according to different water overflowing ranges, and the grades are recommended to be divided by referring to the water overflowing conditions of the urban construction sites, and the grades are shown as no water overflowing, small-range water overflowing, large-range water overflowing and the like.
The training process for the worksite flood level classification subnetwork is as follows: the RGB-D image is subjected to z-score normalization, which is expressed as:
μ represents the mean of all sample data channels and σ is the standard deviation of all sample data channels. And the mu and the sigma adopt all sample data used for training to calculate the obtained mean value and standard deviation in the reasoning process, and the adoption of the standardization helps the network to converge to the optimal solution. During the training process, the data for each batch of sequences is normalized. The label data are respectively replaced by continuous Arabic numerals according to the grades, such as no water overflow (0), small-range water overflow (1), large-range water overflow (2) and large-range water overflow (3). And then, the RGB-D data after the standardization processing and the label data subjected to the one-hot coding are sent to a construction site overflowing water grade classification network for training. The construction site overflowing water grade classifier performs feature extraction on the RGB-D data, inputs the RGB-D data subjected to standardization processing, and outputs a construction site overflowing water grade classification feature map. The first fully-connected network plays a role in mapping the construction site overtopping level classification feature map1 to a sample mark space, and the first fully-connected network is input as a result of the union (concatenate) of the construction site ground semantic feature map2 generated by the construction site ground semantic segmentation encoder and the construction site overtopping level classification feature map1 generated by the construction site overtopping level classifier, and outputs the probability of each level of construction site overtopping. The loss function uses cross entropy.
The construction site ground semantic segmentation sub-network plays a role in assisting training and accelerating convergence on the construction site overtopping level classification sub-network. Because the input of the first fully-connected network is combined with the feature map output by the construction site ground semantic segmentation encoder, the training of the construction site overflow water level classification encoder is supervised by two items, and the generated construction site overflow water level classification feature map is more beneficial to the training of the first fully-connected network.
The invention proposes to adopt an hourglass network to extract the features in order to give consideration to the speed and the precision of the network, and an implementer can also select a proper module design in a neural network according to the image size and the video memory occupation, such as ResidualBlock, Bottleneck Block, CNN Block and the like. The encoder and the decoder of the invention adopt what kind of network design, and the implementer can select according to the specific implementation requirements, and the modularization idea is the protection content of the invention. For example, a building site ground semantic segmentation encoder-a building site ground semantic segmentation decoder is designed by adopting a skip structure, and a block is designed by adopting a block of a lightweight network such as ShuffleNet and MobileNet. The construction site flood level classification encoder suggests applying the EfficientNet image classification network to extract features.
The invention considers that a temporary machine room cannot be built on a construction site for calculation of construction site flood monitoring, the power supply of the construction site is unstable, and no dust-free environment exists, so that a rack server is easy to crash when being placed on the construction site (static electricity is caused by low humidity, connector aging is caused by high humidity, static electricity adsorption is caused by dust, and links of insects, mice and other animals are damaged, so that the system is not suitable for centralized calculation), and therefore, the calculation cluster calculation is adopted. In order to improve the confidentiality of the system, prevent data leakage and improve the parallelism of the system, the invention designs the private chain of the block chain by combining the block chain technology. Distributed storage may be performed, with distributed computing performed at multiple camera ends, servers. Each device is responsible for a part of work, and the workload is reduced. The distributed and block chain private chain idea is used, and the system has excellent performances of information encryption, high disaster tolerance and the like.
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 of AI, it is not necessary to store intermediate result data, and the logic of the chain is preserved 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 the computing cluster for processing, so that the invention uses the form of a block chain private chain, uses different modules of the deep neural network as blocks to carry 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 intelligent construction site water overflowing monitoring deep neural network is required to be divided into modules. The construction site ground semantic segmentation encoder, the construction site ground semantic segmentation decoder, the construction site overflowing level classification encoder and the first full-connection network are used as different modules of the network. Thus, according to the reasoning sequence of the neural network shown in fig. 1, the reasoning chain of the intelligent construction site flood monitoring deep neural network can be obtained.
The computing cluster in this embodiment is a public cloud, and one host instance of the public cloud is used as one computing node. And calculating parameters such as weights required by loading all nodes in the cluster into the intelligent construction site waterlogging monitoring deep neural network. And selecting a plurality of available nodes from the computing cluster according to each intelligent construction site overflowing monitoring depth neural network reasoning request, and taking the camera parameters and construction site ground semantic segmentation encoders, construction site ground semantic segmentation decoders, construction site overflowing level classification encoders and parameters required by the first fully-connected network which are respectively distributed in different available nodes as block data. That is, parameters required by the building site ground semantic segmentation encoder are used as blocks, parameters required by the building site ground semantic segmentation decoder are used as block data, and by analogy, 4 blocks respectively located at different nodes can be obtained. According to the intelligent construction site overflowing water monitoring deep neural network reasoning sequence, generating an intelligent construction site overflowing water monitoring deep neural network block chain private chain, and executing intelligent construction site overflowing water monitoring deep neural network reasoning. That is, the blocks are connected according to the inference sequence, and the corresponding private chain of the block chain can be obtained. When selecting available nodes and performing node sorting, preferably, the available nodes in the computing cluster are randomly sorted, and computing nodes with the same number as the number of blocks are selected from the sorted nodes. For example, there are 10 available nodes, and 4 nodes are selected from the available nodes, and first, the camera parameters of the opposite end corresponding to the inference request are used as block data, and the block is a first block; randomly taking a node from 4 nodes, taking parameters such as weight required by a ground semantic segmentation encoder in the node as block data, and linking the block data to a block where a terminal is located; randomly taking another node, taking parameters such as weight required by a ground semantic segmentation decoder in the node as block data, and linking the block data with a semantic segmentation encoder block; by parity of reasoning, according to the neural network reasoning sequence, a corresponding intelligent construction site overflowing water monitoring deep neural network block chain private chain is generated. The block chain order is consistent with the network inference chain order, which is shown in fig. 1. Therefore, a plurality of intelligent construction site overflowing monitoring deep neural network block chain private chains generated according to different requests can exist at the same time, and the block chain private chains are dynamically generated, are not easy to crack by attack, and have better confidentiality. In available node selection, node selection is randomized, wherein a Hash method is adopted, namely, an R single waveband is extracted from image data acquired by a camera, sampling is carried out again, the sampling is carried out to a fixed size, then the maximum value max and the minimum value min of a pixel are calculated, a threshold value is set through a formula (max + min)/2 to carry out binarization processing, then the binarization data is subjected to a flatten operation to be changed into one-dimensional sequence data, the one-dimensional sequence data is converted into decimal data to be used as a seed of a random number, a random number sequence is generated, and the number in the random number sequence is the same as the number of available nodes. The nodes are selected according to the numerical value index of the random number sequence, and chain reasoning is performed according to a new sequence every time, so that the decryption difficulty is further increased.
If the available node selection operation is put into operation at one node, when the node fails, the neural network reasoning operation is put into a halt. Moreover, when the computation is concentrated on one node, it is easy to attack and crack. Therefore, to achieve decentralized block chaining, the available node selection operation may be performed by the last node after the inference process of each deep neural network inference request is completed.
And adding the node where the terminal is located into the block chain private chain, and taking the camera parameters as block data. The purpose of this is to ensure that the output of the terminal is secret through the encryption strategy and is not easy to be intercepted and tampered. The node where the monitoring center receiving the monitoring result is located can be added into the block chain private chain, so that the purpose of ensuring that the system analysis result can be safely transmitted to the monitoring center is achieved.
When the module division is carried out on the deep neural network, certain modules of the neural network are difficult to be put into one node at one time, namely, the computation of certain modules is large, and the computation is difficult to be completed in a short time. Therefore, the construction site ground semantic segmentation encoder, the construction site ground semantic segmentation decoder, the construction site overflowing level classification encoder and the first full-connection network can be properly subdivided, the number of segmentation blocks of the neural network is increased, the task granularity is reduced, and the parallelism is improved. And a construction site ground semantic segmentation encoder submodule group, a construction site ground semantic segmentation decoder submodule group, a construction site overflowing level classification encoder submodule group and a first fully-connected network submodule group can be obtained. Therefore, according to the reasoning sequence of the neural network, a more subdivided intelligent construction site overflowing monitoring deep neural network reasoning chain can be obtained. Correspondingly, selecting a plurality of available nodes from the computing cluster according to each intelligent construction site overflowing monitoring depth neural network reasoning request, taking construction site ground semantic segmentation encoders, construction site ground semantic segmentation decoders, construction site overflowing level classification encoders and sub-modules of the first fully-connected network which are respectively distributed in different available nodes and required weights and parameters as block data, and generating intelligent construction site overflowing monitoring depth neural network block chain private chains according to the intelligent construction site overflowing monitoring depth neural network reasoning sequence.
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. Specifically, the invention adopts a byte masking encryption mechanism to encrypt the transmission data between the blocks. The encryption and decryption operations are performed on the data transmission among all the blocks, and the best security performance can be ensured. After the node where the next block is located receives the encrypted result, the decryption operation is performed first, and then the subsequent processing is performed.
The byte-masking encryption mechanism employed by the present invention is described in detail below. Since the monitoring center receiving the monitoring result is authentic, an 8-bit Mask may be periodically broadcast by the monitoring center as a key, as exemplified herein. Assuming that one byte of the intermediate data is 0b11001100, the 8-bit Mask is 0b10101010, performing exclusive or operation on the intermediate data to obtain an encrypted byte 0b01100110, and so on, encrypting the whole data, and finally decrypting the next node when receiving the encrypted byte: and performing exclusive OR operation on 0b01100110 and 8-bit Mask 0b10101010 to obtain 0b11001100, namely recovering the original data.
The CIM (City Information Modeling) technology is based on three-dimensional City space geographic Information, and superimposes CIM Information of City buildings, underground facilities and City internet of things Information to construct a City Information model of a three-dimensional digital space. The intelligent construction site water overflowing monitoring method based on CIM technology is used for displaying and early warning the intelligent construction site water overflowing monitoring result. The results of the construction site waterlogging monitoring system are displayed in Web through a CIM city information model in combination with a Web GIS technology, and the visualization of construction site personnel conditions and data is realized. In the system, the CIM technology realizes the datamation and informatization of buildings in cities by constructing a three-dimensional model of the buildings in urban areas, provides geographical position information for subsequent construction site flood monitoring, and realizes all-weather construction site flood monitoring by combining the DNN technology, and has convenient application and low cost.
Therefore, the invention designs an urban construction site information model. The city construction site information model mainly comprises a three-dimensional city space model and regional construction site information, and can update the model and information content in real time along with the continuous promotion of construction progress. The city construction site information model is based on a CIM technology and comprises construction site three-dimensional space modeling information, monitoring area information and construction site overflowing monitoring results. The building site three-dimensional space modeling information comprises various building site scene information such as building site building information, building site building material placement information and worker work area information of each building site in an area, the type information simultaneously contains corresponding geographic position information, and the three-dimensional scenes of each building site in the urban area can be restored and displayed through visual software by combining a Web GIS technology. The monitoring area information comprises images and coordinate transformation matrixes which are shot by the monitoring area geographical position information monitoring visual sensor. 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. And the construction site overflowing monitoring result is integrated into a matched monitoring area for a supervisor to check. In the invention, the construction site overtopping monitoring result is transmitted to the urban construction site information model in real time.
The system of the invention also comprises a visualization unit combined with the Web GIS technology to visualize the city construction site information model and display the visualization unit on the monitoring page. Specifically, the visualization unit includes: the system comprises an initialization module, a display module and a display module, wherein the initialization module is used for acquiring building site three-dimensional space modeling information from an urban building site information model, rendering the urban building site information model by combining a Web GIS technology, and displaying on a foreground Web page to obtain an initial visualization result of the urban building site information model; the data acquisition module is used for acquiring monitoring area information and a flood monitoring result from the city construction site information model; and the visualization module is used for integrating the monitoring area information and the construction site overflowing monitoring result into an initial visualization result of the urban construction site information model. The manager can check the construction site water overflowing situation in the area through the visualization result.
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. A smart building site flood monitoring system based on a block chain comprises a terminal cluster and a computing cluster, and is characterized in that the computing cluster is loaded with parameters required by a smart building site flood monitoring deep neural network, a smart building site flood monitoring deep neural network block chain private chain is configured, and smart building site flood monitoring deep neural network reasoning is executed to realize smart building site flood monitoring;
wherein, wisdom building site flood detection degree of depth neural network's input is the monitoring area ground image that the terminal was gathered, and the output includes ground element semantic segmentation picture, building site flood grade classification result, comprises a plurality of modules:
the construction site ground semantic segmentation encoder is used for encoding RGB data of construction site ground images acquired by the terminal and extracting features to obtain a construction site ground semantic feature map;
the construction site ground semantic segmentation decoder is used for decoding and restoring the construction site ground semantic feature map to obtain a ground element semantic segmentation map and distinguishing the semantics of a water area, the ground and other elements;
the construction site overflowing water grade classification encoder is used for extracting encoding characteristics of RGB data and depth map data of construction site ground images acquired by the terminal to obtain a construction site overflowing water grade classification characteristic map;
and the first full-connection network is used for carrying out weighted classification on the combined feature vector obtained by jointing and unfolding the construction site ground semantic feature map and the construction site overflowing level classification feature map to obtain a construction site overflowing level classification result.
2. The system of claim 1, wherein the site-based semantic segmentation encoder and the site-based semantic segmentation decoder form a site-based semantic segmentation sub-network, and when training the sub-network, semantic labeling is performed on a water area and a ground area in the sample image, and then other irrelevant elements are labeled to obtain labeled data of the sample image, and the sample image and the labeled data are input into the sub-network together for training by adopting a cross entropy loss function.
3. The system of claim 1, wherein configuring the intelligent worksite flood monitoring deep neural network block chaining private chain comprises:
all computing nodes of the computing cluster are loaded with parameters required by the intelligent construction site water-overflowing monitoring deep neural network;
according to each intelligent construction site overflowing monitoring depth neural network inference request, selecting a plurality of available nodes from a computing cluster, taking camera parameters of a terminal, construction site ground semantic segmentation encoders, construction site ground semantic segmentation decoders, construction site overflowing level classification encoders and parameters required by a first full-connection network which are respectively distributed in different available nodes as block data of corresponding nodes, connecting the blocks according to an intelligent construction site overflowing monitoring depth neural network inference sequence, and generating an intelligent construction site overflowing monitoring depth neural network block chain private chain.
4. The system of claim 3, wherein the selecting the plurality of available nodes comprises:
taking one channel data in RGB three channels of the image acquired by the terminal to perform down-sampling and binarization processing to obtain binarized single channel data; expanding the single-channel binary data to obtain a binary sequence, and converting the binary sequence into a decimal number serving as a random number seed; and generating a random number sequence by using the random number seed, and selecting an available node corresponding to the index according to the size index of the numerical value in the random number sequence.
5. The system of claim 3, 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.
6. The system of claim 5, wherein the encryption and decryption employs a byte masking encryption mechanism.
7. The system according to any one of claims 1 to 6, wherein the ground semantic segmentation encoder, the ground semantic segmentation decoder, the construction site flood level classification encoder and the first fully-connected network are respectively and properly subdivided, parameters of each subdivided module respectively distributed in different nodes are used as block data, and a block chain private chain of the intelligent construction site flood detection deep neural network is generated according to an inference sequence of the intelligent construction site flood detection deep neural network.
8. The system of claim 1, wherein the city site information model is constructed based on CIM techniques, the city site information model comprising: the method comprises the steps of urban construction site three-dimensional space modeling information, monitoring area information and intelligent construction site overflowing monitoring result information; and rendering the city construction site information model by using a visualization unit in combination with a Web GIS technology, and displaying the rendered city construction site information model on a foreground page.
9. The system of claim 8, wherein the visualization unit comprises:
the system comprises an initialization module, a display module and a display module, wherein the initialization module is used for acquiring urban construction site three-dimensional space modeling information from an urban construction site information model, rendering the urban construction site information model by combining a Web GIS technology, and displaying on a foreground Web page to obtain an initial visualization result of the urban construction site information model;
the data acquisition module is used for acquiring monitoring area information and intelligent construction site overflowing monitoring result information from the city construction site information model;
and the flood visualization module is used for integrating the monitoring area information and the smart site flood monitoring result into an initial visualization result of the urban site information model according to the geographic position.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010344965.8A CN111563433A (en) | 2020-04-27 | 2020-04-27 | Wisdom building site is monitored system of overflowing water based on block chain |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202010344965.8A CN111563433A (en) | 2020-04-27 | 2020-04-27 | Wisdom building site is monitored system of overflowing water based on block chain |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111563433A true CN111563433A (en) | 2020-08-21 |
Family
ID=72071811
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202010344965.8A Withdrawn CN111563433A (en) | 2020-04-27 | 2020-04-27 | Wisdom building site is monitored system of overflowing water based on block chain |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111563433A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112947360A (en) * | 2021-01-26 | 2021-06-11 | 山东省计算中心（国家超级计算济南中心） | Water distribution system abnormity detection method and system based on state transition time delay diagram |
| CN113610718A (en) * | 2021-07-19 | 2021-11-05 | 中国烟草总公司郑州烟草研究院 | Distribution calculation method and system for construction site air quality detection and working condition defogging |
| CN114710796A (en) * | 2022-03-18 | 2022-07-05 | 深圳技师学院 | Sensor abnormity detection method, device and system based on block chain |
| CN116336398A (en) * | 2023-05-24 | 2023-06-27 | 成都秦川物联网科技股份有限公司 | Intelligent gas leakage safety monitoring method, system and medium based on Internet of things |
| CN116935289A (en) * | 2023-09-13 | 2023-10-24 | 长江信达软件技术（武汉）有限责任公司 | Open channel embankment detection method based on video monitoring |
| CN117994737A (en) * | 2024-04-07 | 2024-05-07 | 辽宁云通汇智能科技有限公司 | Monitoring alarm system and method for intelligent building site management and control platform |
-
2020
- 2020-04-27 CN CN202010344965.8A patent/CN111563433A/en not_active Withdrawn
Cited By (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112947360A (en) * | 2021-01-26 | 2021-06-11 | 山东省计算中心（国家超级计算济南中心） | Water distribution system abnormity detection method and system based on state transition time delay diagram |
| CN113610718A (en) * | 2021-07-19 | 2021-11-05 | 中国烟草总公司郑州烟草研究院 | Distribution calculation method and system for construction site air quality detection and working condition defogging |
| CN114710796A (en) * | 2022-03-18 | 2022-07-05 | 深圳技师学院 | Sensor abnormity detection method, device and system based on block chain |
| CN116336398A (en) * | 2023-05-24 | 2023-06-27 | 成都秦川物联网科技股份有限公司 | Intelligent gas leakage safety monitoring method, system and medium based on Internet of things |
| CN116336398B (en) * | 2023-05-24 | 2023-07-25 | 成都秦川物联网科技股份有限公司 | Intelligent gas leakage safety monitoring method, system and medium based on Internet of things |
| US12018996B2 (en) | 2023-05-24 | 2024-06-25 | Chengdu Qinchuan Iot Technology Co., Ltd. | Methods, systems, and mediums for monitoring gas leakage safety based on internet of things |
| CN116935289A (en) * | 2023-09-13 | 2023-10-24 | 长江信达软件技术（武汉）有限责任公司 | Open channel embankment detection method based on video monitoring |
| CN116935289B (en) * | 2023-09-13 | 2023-12-19 | 长江信达软件技术（武汉）有限责任公司 | Open channel embankment detection method based on video monitoring |
| CN117994737A (en) * | 2024-04-07 | 2024-05-07 | 辽宁云通汇智能科技有限公司 | Monitoring alarm system and method for intelligent building site management and control platform |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111563433A (en) | Wisdom building site is monitored system of overflowing water based on block chain | |
| CN111428694A (en) | Abnormal smoke monitoring system of wisdom building site based on block chain | |
| CN101044761B (en) | Method for processing an input image securely | |
| CN101044759B (en) | Method for processing an input image securely | |
| Hamza et al. | An efficient cryptosystem for video surveillance in the internet of things environment | |
| CN111461075A (en) | Guardrail crossing detection method combining deep neural network and block chain | |
| CN111611949A (en) | Intelligent urban water area garbage density detection method based on CIM and block chain | |
| CN111524131A (en) | Wall crack detection system based on block chain and BIM | |
| CN116347045B (en) | Monitoring method, device, equipment and storage medium based on communication and satellite technology | |
| CN111583068A (en) | Statistical method for stay time and heat degree of personnel in construction area combined with block chain | |
| CN111476208A (en) | Intelligent building site constructor gathering monitoring system utilizing block chains | |
| CN115424059A (en) | Remote sensing land use classification method based on pixel level comparison learning | |
| CN111414499A (en) | Operation personnel safety belt wearing detection system based on block chain and BIM | |
| CN115632839A (en) | Smart campus environment network supervision method and system | |
| CN113901166B (en) | Electronic map construction method, device, equipment and storage medium | |
| CN111553269A (en) | Wisdom building site vehicle blocks up monitoring system based on block chain | |
| WO2022221122A1 (en) | Filling gaps in electric grid models | |
| CN111414501A (en) | Wood flammability risk prediction system based on block chain and CIM | |
| CN111523434A (en) | Building site personnel abnormal gathering detection system based on block chain and BIM | |
| CN111414504A (en) | Building site flood detection system based on block chain and CIM | |
| CN111414500A (en) | Steel wire rope breakage early warning system based on block chain and BIM | |
| CN111461074A (en) | Channel blockage detection method combining block chain and deep learning | |
| CN111553270A (en) | Monitoring system is piled up to wisdom building site cement bag based on block chain | |
| CN116471103B (en) | Internal and external network data security exchange method, device and equipment based on boundary network | |
| CN117351360A (en) | Remote sensing image road extraction method based on attention mechanism improvement |
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: 20200821 |
|
| WW01 | Invention patent application withdrawn after publication |