CN116862843A - Construction site safety risk noninductive inspection method and system - Google Patents
Construction site safety risk noninductive inspection method and system Download PDFInfo
- Publication number
- CN116862843A CN116862843A CN202310697918.5A CN202310697918A CN116862843A CN 116862843 A CN116862843 A CN 116862843A CN 202310697918 A CN202310697918 A CN 202310697918A CN 116862843 A CN116862843 A CN 116862843A
- Authority
- CN
- China
- Prior art keywords
- inspection
- risk
- image data
- real
- inspection image
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000007689 inspection Methods 0.000 title claims abstract description 331
- 238000000034 method Methods 0.000 title claims abstract description 77
- 238000010276 construction Methods 0.000 title claims abstract description 66
- 230000002159 abnormal effect Effects 0.000 claims description 24
- 238000002372 labelling Methods 0.000 claims description 10
- 230000003068 static effect Effects 0.000 claims description 7
- 238000007405 data analysis Methods 0.000 claims description 6
- 238000013500 data storage Methods 0.000 claims description 5
- 238000012544 monitoring process Methods 0.000 claims description 5
- 238000004364 calculation method Methods 0.000 claims description 3
- 230000008569 process Effects 0.000 abstract description 16
- 230000002829 reductive effect Effects 0.000 abstract description 13
- 238000001514 detection method Methods 0.000 abstract description 4
- 238000004519 manufacturing process Methods 0.000 description 12
- 238000007726 management method Methods 0.000 description 6
- 230000008901 benefit Effects 0.000 description 5
- 238000005516 engineering process Methods 0.000 description 5
- 230000000694 effects Effects 0.000 description 3
- 230000006870 function Effects 0.000 description 3
- 238000010801 machine learning Methods 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 238000013473 artificial intelligence Methods 0.000 description 2
- 238000012937 correction Methods 0.000 description 2
- 238000003062 neural network model Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000012502 risk assessment Methods 0.000 description 2
- 239000002699 waste material Substances 0.000 description 2
- 206010063385 Intellectualisation Diseases 0.000 description 1
- 230000002411 adverse Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000009435 building construction Methods 0.000 description 1
- 239000004035 construction material Substances 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 239000000428 dust Substances 0.000 description 1
- 230000002427 irreversible effect Effects 0.000 description 1
- 230000000670 limiting effect Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 239000003550 marker Substances 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000036961 partial effect Effects 0.000 description 1
- 230000003449 preventive effect Effects 0.000 description 1
- 238000012545 processing Methods 0.000 description 1
- 238000007789 sealing Methods 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
- G06T7/001—Industrial image inspection using an image reference approach
-
- 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
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0635—Risk analysis of enterprise or organisation activities
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/74—Image or video pattern matching; Proximity measures in feature spaces
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Economics (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Quality & Reliability (AREA)
- Strategic Management (AREA)
- Computing Systems (AREA)
- Entrepreneurship & Innovation (AREA)
- Educational Administration (AREA)
- Game Theory and Decision Science (AREA)
- Development Economics (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application relates to the technical field of construction safety detection, in particular to a construction site safety risk noninductive inspection method and system. The method comprises the following steps: reading historical inspection image data of a preset inspection route, and performing risk marking on the historical inspection image data; establishing a risk identification template according to historical inspection image data with risk labels; receiving real-time inspection image data of a preset inspection route; comparing the real-time inspection image data with the risk identification template to obtain a risk identification result, so that the inspection operation process of constructors is effectively reduced, and the complexity of the inspection operation is reduced; and the risk identification result is sent to a responsible person, the real-time inspection image data is converted into a stored certificate hash value, and the risk identification result and the stored certificate hash value are stored, so that the inspection result is prevented from being tampered by others, and the supervision effectiveness of the safety inspection of the construction site is improved.
Description
Technical Field
The application relates to the technical field of construction safety detection, in particular to a construction site safety risk noninductive inspection method and system.
Background
In the field of building engineering and the field of part of labor intensive production and manufacturing, the coverage rate of mechanical automation is still at a low level, and large manpower is still required to be input in the construction or production process, so that the safety problem of a construction site is required to be seriously considered, and once an casualty event occurs, the enterprise has a great negative influence on society. Therefore, a large amount of manpower and material resources are input into the building construction unit and the production and manufacturing unit to reduce and avoid the occurrence of safety accidents. Nevertheless, the cost of an intelligent inspection robot without human involvement is still too high, and not all projects of the construction site have the ability to equip the intelligent inspection robot. Therefore, the effective method commonly adopted in the construction site at present is to use intelligent equipment to carry out safety inspection by constructors, to carry out integral inspection on the construction site regularly, to discover possible potential safety hazards as soon as possible, to simultaneously fill the potential safety hazards discovered in the inspection process in an inspection information system, and to inform responsible persons of timely correction.
However, in the field of building engineering and the field of part of labor intensive production and manufacturing, the inspection information system has high operation complexity and higher learning cost, and invalid work of inspection new people in the process of safety inspection is possibly caused, so that resource waste is caused, and the inspection effect is not achieved. Moreover, because the inspection information system records only some key data, but not the complete inspection records, the inspection result information is incomplete, and the accuracy of the inspection result is affected. Furthermore, in order to obtain certain benefits or avoid punishment, the patrol record stored in the patrol information system may be tampered with, so that the supervision effectiveness is insufficient.
Disclosure of Invention
The application aims to solve the technical problems that: the technical problems of complex inspection operation of constructors and poor supervision effectiveness of safety inspection exist in the prior construction site safety risk inspection, and the technical problems are solved by providing a construction site safety risk noninductive inspection method and system.
The application adopts the following technical scheme: a construction site security risk noninductive inspection method comprises the following steps:
reading historical inspection image data of a preset inspection route, and performing risk marking on the historical inspection image data;
establishing a risk identification template according to historical inspection image data with risk labels;
receiving real-time inspection image data of the preset inspection route;
comparing the real-time inspection image data with the risk identification template to obtain a risk identification result;
and sending the risk identification result to a responsible person, converting the real-time inspection image data into a evidence-storing hash value, and storing the risk identification result and the evidence-storing hash value.
The risk identification template is established, and the risk identification result is obtained according to the comparison of the risk identification template and new image data obtained in the inspection process, so that the safety risk identification is carried out on the construction site by the auxiliary constructors, the accuracy and the efficiency of the safety inspection are improved, the inspection operation process of the constructors is greatly reduced, and the complexity of the inspection operation is reduced. Meanwhile, the risk identification result and the evidence-storing hash value are stored by converting the real-time inspection image data into the evidence-storing hash value, so that the inspection result is prevented from being tampered by others, and the supervision effectiveness of the safety inspection of the construction site is improved.
Preferably, the preset routing inspection route is obtained by the following method:
receiving static image data of a construction site monitoring area;
performing inspection area division operation and inspection point position marking operation according to the static image data;
and connecting marked inspection point positions by combining the divided inspection areas to obtain a preset inspection route.
Preferably, the method for establishing the risk identification template comprises the following steps:
classifying the historical inspection image data containing the risk labels, enumerating risk items of different categories, and recording the historical inspection image data and the inspection area position information corresponding to the risk items of all the categories;
calculating the pixel mean value of the image data of each category of risk items aiming at the risk items of different categories as an abnormal template;
all classes of abnormal templates constitute risk recognition templates.
Preferably, the method for comparing the real-time inspection image data with the risk identification template to obtain a risk identification result comprises the following steps:
sequentially corresponding the real-time inspection image data to risk items of different categories;
calculating the pixel mean value of the real-time inspection image data of each category;
the pixel mean value of the real-time inspection image data of each category and the value obtained by quotient calculation of the abnormal templates of the corresponding category in the risk identification template are recorded as abnormal similarity;
if the abnormal similarity is larger than or equal to a first threshold, the risk identification result is that the area where the real-time inspection image data of the current class is located has risk, and if the abnormal similarity is smaller than the first threshold, the risk identification result is that the area where the real-time inspection image data of the current class is located is safe.
Preferably, the real-time inspection image data includes inspection position coordinates, and the method for sequentially corresponding the real-time inspection image data to risk items of different categories includes:
converting the position information of the inspection area corresponding to all the types of risk items into geographic coordinates;
and matching the patrol position coordinates with the geographical coordinates of the patrol areas corresponding to the risk items of different categories to obtain the risk item category corresponding to the real-time patrol image data.
Preferably, the method for converting the real-time inspection image data into the certificate storing hash value comprises the following steps:
terminal shooting n of patrolling and examining i After frame real-time inspection of the image, n is extracted i The hash value of the frame real-time inspection image is recorded as Hi;
at said n i Inserting a mark frame after the frame is inspected in real time, wherein the mark frame is used for recording the extracted Hi;
transmitting the last m bits of Hi to a specified server, m being an integer constant, said server feeding back n according to the last m bits of Hi i+1 The method comprises the steps of (1) sending to a patrol terminal;
and repeatedly executing the steps until the patrol terminal finishes all shooting, and extracting hash values of all frames to be recorded as the stored hash values.
Preferably, n is fed back according to the last m bits of Hi i+1 The method of (1) is as follows:
n i ＝min[max(floor(10 k ·sin(x))，N 1 )，N 2 ]
wherein n is i Initial value n of 1 For a preset value, x is the value of the last m bits of Hi, and k, N1 and N2 are constant values.
Preferably, the risk identification result is sent to a responsible person, the real-time inspection image data is converted into a forensic hash value, and the method for storing the risk identification result and the forensic hash value further comprises:
after the risk identification result is obtained, real-time inspection image data of a preset inspection route, all types of risk items and risk identification results corresponding to all risk items are formed into an inspection result report, the inspection result report is sent to a responsible person, the real-time inspection image data is converted into a certificate hash value, and the inspection result report and the certificate hash value are uploaded to a designated server for storage.
A construction site security risk noninductive inspection system, comprising:
the inspection terminal is used for reading historical inspection image data of a preset inspection route and receiving real-time inspection image data of the preset inspection route;
the data analysis module is used for carrying out risk labeling on the historical inspection image data and establishing a risk identification template according to the historical inspection image data with the risk labeling;
the risk identification module is used for comparing the real-time inspection image data with the risk identification template to obtain a risk identification result;
the data storage module is used for converting the real-time inspection image data into a storage hash value;
and the server is used for sending the risk identification result to a responsible person and storing the risk identification result and the evidence-storing hash value.
The beneficial technical effects of the application include: by means of the construction site security risk noninductive inspection method and system, a risk recognition template is established, a risk recognition result is obtained according to comparison of the risk recognition template and new image data obtained in the inspection process, so that constructors are assisted in performing security risk recognition on the construction site, accuracy and efficiency of security inspection are improved, inspection operation processes of the constructors are greatly reduced, complexity of inspection operation is reduced, and by converting real-time inspection image data into a evidence-storing hash value, the risk recognition result and the evidence-storing hash value are stored, tampering of inspection results by others is avoided, and supervision effectiveness of the construction site security inspection is improved; the risk identification result is obtained by taking the pixel mean value of the image data as the contrast characteristic, so that the operation amount of the system can be obviously reduced, the efficiency of image contrast identification is improved, and the efficiency of safety inspection on a construction site is improved; the technical scheme of geographic coordinate matching is adopted to carry out classification and identification of risk items on the real-time inspection image data, so that the prior art is replaced by adopting a neural network model or a machine learning method to carry out classification and identification, the operation complexity of the system is greatly reduced, and the efficiency of obtaining a risk identification result is improved to a certain extent; by automatically generating the inspection result report from the obtained risk identification result, the method replaces constructors to fill the safety inspection records, avoids the influence of artificial subjective factors, improves the accuracy of the inspection result report filling, and reduces the complexity of inspection operation of constructors.
Other features and advantages of the present application will be disclosed in the following detailed description of the application and the accompanying drawings.
Drawings
The application is further described with reference to the accompanying drawings:
fig. 1 is a flowchart of a construction site security risk noninductive inspection method according to an embodiment of the present application.
Fig. 2 is a flowchart of a method for obtaining a preset routing inspection route according to an embodiment of the application.
Fig. 3 is a flowchart of a method for creating a risk identification template according to an embodiment of the present application.
Fig. 4 is a flowchart of a method for obtaining a risk identification result according to an embodiment of the present application.
Fig. 5 is a flowchart of a method for converting real-time inspection image data into a certificate hash value according to an embodiment of the present application.
Fig. 6 is a schematic structural diagram of a construction site security risk-free inspection system according to an embodiment of the application.
Wherein: 1. the system comprises a patrol terminal, a data analysis module, a risk identification module, a data certification module, a server and a data storage module.
Detailed Description
The technical solutions of the embodiments of the present application will be explained and illustrated below with reference to the drawings of the embodiments of the present application, but the following embodiments are only preferred embodiments of the present application, and not all embodiments. Based on the examples in the implementation manner, other examples obtained by a person skilled in the art without making creative efforts fall within the protection scope of the present application.
In the following description, directional or positional relationships such as the terms "inner", "outer", "upper", "lower", "left", "right", etc., are presented for convenience in describing the embodiments and simplifying the description, and do not indicate or imply that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore should not be construed as limiting the application.
Before explaining the technical scheme of the present embodiment in detail, first, a description is given of a background situation to which the present embodiment is applied.
While the construction industry has become one of the fundamental industries and important props for economic construction, it also presents a number of socioeconomic problems, the first of which is the safety problem for construction. As is well known, the operators in the construction site have poor safety consciousness, high mobility and low comprehensive quality; the operation environment is complex, the labor intensity is high, and the danger projects are large; the lack of the safety management concept and the lack of safety management personnel, in particular the lag of the safety management technology, method and means, etc., and the building industry has already developed into one of three high-risk industries in China due to the influences of the characteristics and adverse factors.
At present, in the field of building engineering and the field of partial labor intensive production and manufacturing, the coverage rate of mechanical automation is still at a lower level, and larger manpower is still required to be input in the construction or production process, so that the safety problem of a construction site is more required to be seriously considered, and once an casualty event occurs, heavy loss is brought to constructors and construction enterprises, and meanwhile, negative influence is also easy to occur. Therefore, it is necessary to deeply study the mechanism of the safety accident of the construction site, carefully check the reasons of hidden danger, and apply modern scientific technology to purposefully take preventive measures of various management methods and technical means to reduce the probability of the safety accident of the construction site.
Although the artificial intelligence technology promotes the building industry to continuously develop towards automation, digitalization and intellectualization, the hidden danger of the construction site is difficult to be examined by using the new technology because the actual condition of the construction site is complex and the hidden danger of each engineering scene of the construction site is lack of systematic knowledge in the current research. Moreover, the cost of the intelligent inspection robot without manual participation is too high, and not all projects of the construction site can be equipped with the intelligent inspection robot. Therefore, the effective method generally adopted at present is to use intelligent equipment to carry out safety inspection by constructors, to perform integral inspection on construction sites regularly, to find possible potential safety hazards as soon as possible, to fill the potential safety hazards found in the inspection process in an inspection information system, and to inform responsible persons of timely correction.
However, in the field of building engineering and the field of part of labor intensive production and manufacturing, the inspection information system has high operation complexity and higher learning cost, and invalid work of inspection new people in the process of safety inspection is possibly caused, so that resource waste is caused, and the inspection effect is not achieved. Moreover, because the inspection information system records only some key data, but not the complete inspection records, the inspection result information is incomplete, and the accuracy of the inspection result is affected. Furthermore, in order to obtain certain benefits or avoid punishment, the patrol record stored in the patrol information system may be tampered with, so that the supervision effectiveness is insufficient. Therefore, it is necessary to research a construction site security risk inspection method and system capable of reducing the complexity of the inspection operation of constructors, improving the security inspection efficiency and monitoring the effectiveness.
Therefore, the embodiment of the application provides a construction site security risk noninductive inspection method, referring to fig. 1, comprising the following steps:
and A01) reading historical inspection image data of a preset inspection route, and performing risk marking on the historical inspection image data.
The operation of risk labeling on the historical inspection image data is similar to the operation of risk labeling on the data in the prior art, and the embodiment of the application is not limited to the operation. Optionally, the implementation of risk labeling on the historical inspection image data may be that the inspection personnel with abundant experience manually performs risk assessment and labeling on the historical inspection image data, or may use techniques such as machine learning and artificial intelligence to implement risk assessment and labeling.
Step A02) building a risk identification template according to the historical inspection image data with the risk labels.
Step A03) receiving real-time inspection image data of a preset inspection route.
Optionally, the method for acquiring real-time inspection image data of the preset inspection route may be: (1) The unmanned aerial vehicle is used for aerial photography, so that comprehensive field image data including a patrol route and surrounding environments can be obtained, the method is simple to operate, professional personnel are required to operate, and the cost is high; (2) The constructor carries out inspection by holding the inspection terminal, the image acquisition module of the inspection terminal is used for acquiring the image data of the inspection route of the construction site and carrying out mobile detection, and the image acquisition module of the inspection terminal comprises a camera. (3) The camera is arranged on the construction site to monitor the inspection route in real time, and meanwhile, the image data of the inspection route can be recorded. (4) The method has the advantages that the position information and the image data of the inspection route are recorded in real time by utilizing the GPS and map technology, meanwhile, the data analysis and the visualization can be performed, and professional GPS equipment and map software are needed to be used, so that the cost is high.
Preferably, in this embodiment, the real-time inspection image data of the preset inspection route is acquired by the image acquisition module of the inspection terminal, where the image acquisition module of the inspection terminal includes a camera, and in the process that the constructor holds the inspection terminal to inspect, the image acquisition module of the inspection terminal can acquire the image data of the inspection route of the construction site in real time and perform mobile detection, so as to provide data for comparing with the risk identification template.
And A04) comparing the real-time inspection image data with a risk identification template to obtain a risk identification result.
Step A05) sending the risk identification result to a responsible person, converting the real-time inspection image data into a stored hash value, and storing the risk identification result and the stored hash value.
According to the method, the risk identification template is established, the risk identification result is obtained according to the comparison of the risk identification template and the new image data obtained in the inspection process, so that constructors are assisted in carrying out safety risk identification on a construction site, the accuracy and efficiency of safety inspection are improved, the inspection operation process of the constructors is greatly reduced, the complexity of inspection operation is reduced, the real-time inspection image data are converted into the evidence-storing hash value, the risk identification result and the evidence-storing hash value are stored, the inspection result is prevented from being tampered by others, and the supervision effectiveness of safety inspection on the construction site is improved.
On the other hand, in this embodiment, referring to fig. 2, the preset routing is obtained by the following method:
step B01) receives static image data of a construction site monitoring area.
And B02) carrying out inspection area division operation and inspection point position labeling operation according to the static image data.
Further, the inspection area dividing operation and the inspection point position marking operation specifically refer to performing inspection area and inspection point position specification on static image data of a construction site monitoring area obtained through the image pickup device, and acquiring coordinate data sets corresponding to the inspection area and the inspection point position.
And B03) connecting marked inspection point positions by combining the divided inspection areas to obtain a preset inspection route.
The inspection area refers to an area in which inspection personnel must perform inspection work in the inspection process, and the inspection point location refers to equipment or a device in which the inspection personnel must perform inspection work in the inspection process.
In another aspect, referring to fig. 3, a method for establishing a risk identification template includes:
and C01) classifying the historical inspection image data containing the risk labels, listing risk items of different categories, and recording the historical inspection image data and the inspection area position information corresponding to the risk items of all the categories.
Illustratively, the categories of risk items are: falling at high places, electrical accidents, slipping of construction materials, dumping of tower cranes, improper sealing of construction areas, dust pollution, building structure problems and the like.
Step C02) calculating the pixel mean value of the image data of each category of risk items aiming at the risk items of different categories as an abnormal template.
Specifically, the implementation manner of calculating the pixel mean value of the image data of each category risk item is as follows:
(1) Reading image data; reading image data using a corresponding function in an image processing library or programming language, typically a matrix of pixels of an image;
(2) Calculating a pixel mean value: for RGB images, the pixel mean needs to be calculated separately for each channel (red, green, blue). The formula for calculating the pixel mean value is: pixel mean = sum of pixel values/total number of pixels, wherein the sum of pixel values refers to the result of adding up all pixel values, and the total number of pixels refers to the number of pixels of the image;
(3) Calculating the average value of three RGB channels; and accumulating the pixel mean values of all the channels, wherein the obtained mean value of the RGB three channels is the pixel mean value of the image data.
Step C03) forming a risk identification template by using all the abnormal templates.
In another aspect, referring to fig. 4, a method for obtaining a risk identification result includes:
step D01), the real-time inspection image data sequentially correspond to risk items of different categories.
Step D02) calculating the pixel mean value of the real-time inspection image data of each category.
The method for calculating the pixel mean value of the real-time inspection image data of each category is consistent with the implementation manner of calculating the pixel mean value of the image data of each category risk item, and this embodiment is not described herein.
Step D03) calculating the quotient of the pixel mean value of the new image data of each category and the abnormal template mean value of the corresponding category in the risk identification template, and marking the quotient as abnormal similarity.
Step D04) if the abnormal similarity is larger than or equal to a first threshold value, the risk identification result is that the region where the new image data of the current category is located has risks, and if the abnormal similarity is smaller than the first threshold value, the risk identification result is that the region where the new image data of the current category is located is safe.
According to the method for establishing the risk identification template and obtaining the risk identification result, the pixel mean value of the image data is used as the contrast characteristic, so that the operation amount of the system can be obviously reduced, the efficiency of image contrast identification is improved, and the efficiency of safety inspection on a construction site is improved.
On the other hand, in this embodiment, the method for sequentially mapping real-time patrol image data to risk items of different categories includes:
converting the position information of the inspection area corresponding to all the types of risk items into geographic coordinates;
and matching the patrol position coordinates with the geographical coordinates of the patrol areas corresponding to the risk items of different categories to obtain the risk item category corresponding to the real-time patrol image data.
The specific implementation mode of matching the patrol position coordinates with the patrol area geographic coordinates corresponding to the risk items of different categories is as follows: and calculating the distance between the patrol position coordinates and the geographical coordinates of the patrol areas corresponding to the risk items of each category by using a distance formula, such as Euclidean distance, manhattan distance and the like, and then determining which category of the patrol area corresponding to the risk items the patrol position coordinates belong to by comparing the distances.
According to the embodiment, the technical scheme of geographic coordinate matching is adopted to conduct classified identification of risk items on real-time inspection image data, a neural network model or a machine learning method is replaced in the prior art to conduct classified identification, the operation complexity of a system is greatly reduced, and therefore the efficiency of obtaining risk identification results is improved to a certain extent.
On the other hand, in this embodiment, referring to fig. 5, the method for converting the real-time inspection image data into the certificate hash value includes:
step E01) shooting n by inspection terminal i After frame real-time inspection of the image, n is extracted i Hash value of frame real-time inspection image is marked as H i 。
Step E02) at n i Frame real-time inspection image post-interpolationEntering a mark frame for recording the extracted H i 。
Step E03) H i M bits at the end of (a) are sent to a designated server, m is integer constant, and the server is based on H i Last m bits of feedback n i+1 And (5) to the inspection terminal.
The hash value is a number with a fixed length, for example, the hash value used by the SHA256 hash algorithm has a length of 256 bits.
Step E04) repeatedly executing the steps until the inspection terminal finishes all shooting, and extracting hash values of all frames to be recorded as a stored hash value.
Wherein all frames include all image frames and all marker frames.
The implementation process of converting the real-time inspection image data into the certificate hash value is as follows:
after the inspection terminal shoots 10 frames of real-time inspection images, the hash value of the 10 frames of real-time inspection images is extracted and recorded as H i ；
Inserting a frame of mark frame after the 10 frames of real-time inspection images, wherein the mark frame is used for recording the hash value H extracted from the 10 frames of real-time inspection images i ；
Will H i The last 3 digits of (a) are sent to a designated server, which is based on H i Calculating the number n of real-time inspection image frames about to be shot and hash value extracted next time by the inspection terminal according to the 3-bit numerical value at the tail of the inspection terminal i+1 And n is as follows i+1 Feeding back to the inspection terminal;
and repeatedly executing the steps until the inspection terminal finishes all shooting, namely, the inspection personnel holds all real-time inspection images shot by the inspection route by the inspection terminal, extracting hash values again for all real-time inspection image frames and mark frames, and recording the hash values extracted at the moment as the stored hash values.
Unlike encryption algorithms, the hash algorithm is an irreversible one-way function, and when a high-security hash algorithm, such as MD5 and SHA, is adopted, two different data are almost impossible to extract the same hash value result, so that once the data are modified, the data can be detected.
The security inspection is used as a management action most commonly related to security management, and is specifically divided into three parts of group basic level project department routine inspection, group headquarter inspection and government inspection, wherein the government inspection belongs to government actions and is irrelevant to construction enterprises, and the security inspection method does not consider the application; for group headquarters inspection, because headquarters lack deep knowledge of projects, inspection work is difficult to accomplish with high efficiency, and risk omission is likely to be caused easily. Particularly, for the inspection of the group headquarters, as the inspection frequency of the headquarters is lower, the basic project department can report routine inspection results through some formal display work, thereby reducing the supervision effectiveness of the group headquarters. Moreover, because the inspection information system records only some key data, but not the complete inspection records, the inspection result information is incomplete, and the accuracy of the inspection result is affected. Furthermore, in order to obtain certain benefits or avoid punishment, the patrol record stored in the patrol information system may be tampered with, so that the supervision effectiveness is insufficient. Therefore, the embodiment of the application stores the real-time inspection image data by a method for extracting the stored certificate hash value, such as uploading the extracted stored certificate hash value to a headquarter server, thereby avoiding the inspection result from being tampered by a basic project department and improving the supervision effectiveness of the safety inspection of the construction site.
On the other hand, in the present embodiment, according to H i Last m bits of feedback n i+1 The method of (1) is as follows:
n i ＝min[max(floor(10 k ·sin(x))，N 1 )，N 2 ]
wherein x is H i M bits at the end of (k) is a constant exponent coefficient, N 1 Is a minimum constant threshold, N 2 Is a maximum constant threshold.
Illustratively according to H i Last m bits of feedback n i+1 The implementation process of (1) is as follows: let k have a value of 2, N 1 Has a value of 10, N 2 Has a value of 200, H i The last 3-bit value of (a) is 136, and x=136 is substituted into the function
n i ＝min[max(floor(10 2 ·sin(136))，10)，200]＝69，
The inspection terminal will shoot and extract the real-time inspection image frame number n of the hash value next time i+1 69 frames.
On the other hand, in this embodiment, the risk identification result is sent to the responsible person, and the real-time inspection image data is converted into the stored hash value, and the method for storing the risk identification result and the stored hash value further includes:
after the risk identification result is obtained, real-time inspection image data of a preset inspection route, all types of risk items and risk identification results corresponding to all risk items are formed into an inspection result report, the inspection result report is sent to a responsible person, the real-time inspection image data is converted into a certification hash value, and the inspection result report and the certification hash value are uploaded to a designated server for storage.
Because the inspection information system in the prior art still needs manual participation for the filling of the safety inspection records, the information distortion possibly caused by artificial subjective factors can exist, and the environment and equipment hidden danger of a construction site are difficult to quickly identify and evaluate, so that serious hidden danger cannot be found in time. Therefore, according to the real-time inspection image data of the preset inspection route, all types of risk items and the risk identification results corresponding to the risk items obtained by the method, the inspection result report is automatically generated, so that the construction personnel can replace the filling of the safety inspection records, the influence of artificial subjective factors is avoided, the accuracy of the inspection result report filling is improved, and meanwhile, the complexity of the inspection operation of the construction personnel is reduced.
On the other hand, the embodiment of the application also provides a construction site security risk noninductive inspection system, referring to fig. 6, comprising:
the inspection terminal 1 is used for reading historical inspection image data of a preset inspection route and receiving real-time inspection image data of the preset inspection route;
the data analysis module 2 is used for performing risk annotation on the historical inspection image data and establishing a risk identification template according to the historical inspection image data with the risk annotation;
the risk identification module 3 is used for comparing the real-time inspection image data with a risk identification template to obtain a risk identification result;
the data storage module 4 is used for converting the real-time inspection image data into a storage hash value;
and the server 5 is used for sending the risk identification result to the responsible person and storing the risk identification result and the stored hash value.
The inspection terminal 1 includes, but is not limited to, a camera, a smart phone carrying the camera, a special smart handheld terminal device, and the like. Specialized intelligent handheld terminal devices such as EM-T695 consolidate handheld terminals. Wherein the server 5 includes, but is not limited to, a headquarter server.
On the other hand, in the present embodiment, the data analysis module 2 is configured to perform the following steps:
classifying the historical inspection image data containing the risk labels, enumerating risk items of different categories, and recording the historical inspection image data and the inspection area position information corresponding to the risk items of all the categories;
calculating the pixel mean value of the image data of each category of risk items aiming at the risk items of different categories as an abnormal template;
all classes of abnormal templates constitute risk recognition templates.
On the other hand, in the present embodiment, the risk identification module 3 is configured to perform the following steps:
sequentially corresponding real-time inspection image data to risk items of different categories;
calculating the pixel mean value of the real-time inspection image data of each category;
the pixel mean value of the real-time inspection image data of each category and the value obtained by quotient calculation of the abnormal templates of the corresponding category in the risk identification template are recorded as abnormal similarity;
if the abnormal similarity is larger than or equal to the first threshold, the risk identification result is that the area where the real-time inspection image data of the current class is located has risks, and if the abnormal similarity is smaller than the first threshold, the risk identification result is that the area where the real-time inspection image data of the current class is located is safe.
On the other hand, in the present embodiment, the data storage module 4 is configured to execute the following steps:
terminal shooting n of patrolling and examining i After frame real-time inspection of the image, n is extracted i Hash value of frame real-time inspection image is marked as H i ；
At n i After the frame real-time inspection image, a mark frame is inserted, and the mark frame is used for recording the extracted H i ；
Will H i M bits at the end of (a) are sent to a designated server, m is integer constant, and the server is based on H i Last m bits of feedback n i+1 The method comprises the steps of (1) sending to a patrol terminal;
and repeatedly executing the steps until the patrol terminal finishes all shooting, and extracting hash values of all frames to be recorded as the stored hash values.
On the other hand, in the present embodiment, the server 5 is further configured to perform the following steps:
after the risk identification result is obtained, real-time inspection image data of a preset inspection route, all types of risk items and risk identification results corresponding to all risk items are formed into an inspection result report, the inspection result report is sent to a responsible person, the real-time inspection image data is converted into a certification hash value, and the inspection result report and the certification hash value are uploaded to a designated server for storage.
The above is only a specific embodiment of the present application, but the scope of the present application is not limited thereto, and it should be understood by those skilled in the art that the present application includes but is not limited to the accompanying drawings and the description of the above specific embodiment. Any modifications which do not depart from the functional and structural principles of the present application are intended to be included within the scope of the appended claims.
Claims (9)
1. The construction site security risk noninductive inspection method is characterized by comprising the following steps of:
reading historical inspection image data of a preset inspection route, and performing risk marking on the historical inspection image data;
establishing a risk identification template according to historical inspection image data with risk labels;
receiving real-time inspection image data of the preset inspection route;
comparing the real-time inspection image data with the risk identification template to obtain a risk identification result;
and sending the risk identification result to a responsible person, converting the real-time inspection image data into a evidence-storing hash value, and storing the risk identification result and the evidence-storing hash value.
2. A construction site safety risk noninductive inspection method as set forth in claim 1, wherein,
the preset routing inspection route is obtained by the following method:
receiving static image data of a construction site monitoring area;
performing inspection area division operation and inspection point position marking operation according to the static image data;
and connecting marked inspection point positions by combining the divided inspection areas to obtain a preset inspection route.
3. A construction site safety risk noninductive inspection method as set forth in claim 1, wherein,
the method for establishing the risk identification template comprises the following steps:
classifying the historical inspection image data containing the risk labels, enumerating risk items of different categories, and recording the historical inspection image data and the inspection area position information corresponding to the risk items of all the categories;
calculating the pixel mean value of the image data of each category of risk items aiming at the risk items of different categories as an abnormal template;
all classes of abnormal templates constitute risk recognition templates.
4. A construction site safety risk noninductive inspection method as set forth in claim 3, characterized in that,
comparing the real-time inspection image data with the risk identification template to obtain a risk identification result, wherein the method comprises the following steps:
sequentially corresponding the real-time inspection image data to risk items of different categories;
calculating the pixel mean value of the real-time inspection image data of each category;
the pixel mean value of the real-time inspection image data of each category and the value obtained by quotient calculation of the abnormal templates of the corresponding category in the risk identification template are recorded as abnormal similarity;
if the abnormal similarity is larger than or equal to a first threshold, the risk identification result is that the area where the real-time inspection image data of the current class is located has risk, and if the abnormal similarity is smaller than the first threshold, the risk identification result is that the area where the real-time inspection image data of the current class is located is safe.
5. A construction site safety risk noninductive inspection method as set forth in claim 4, wherein,
the real-time inspection image data comprises inspection position coordinates, and the method for sequentially corresponding the real-time inspection image data to different types of risk items comprises the following steps:
converting the position information of the inspection area corresponding to all the types of risk items into geographic coordinates;
and matching the patrol position coordinates with the geographical coordinates of the patrol areas corresponding to the risk items of different categories to obtain the risk item category corresponding to the real-time patrol image data.
6. A construction site safety risk non-sensitive inspection method according to claim 1 to 5, wherein,
the method for converting the real-time inspection image data into the certificate hash value comprises the following steps:
terminal shooting n of patrolling and examining i After frame real-time inspection of the image, n is extracted i Hash value of frame real-time inspection image is marked as H i ；
At said n i Inserting a mark frame after the frame real-time inspection image, wherein the mark frame is used for recording the extracted H i ；
Will H i M bits at the end of (a) are sent to a specified server, m being integer constant, said server being based on H i Last m bits of feedback n i+1 The method comprises the steps of (1) sending to a patrol terminal;
and repeatedly executing the steps until the patrol terminal finishes all shooting, and extracting hash values of all frames to be recorded as the stored hash values.
7. A construction site safety risk noninductive inspection method as set forth in claim 6, wherein,
feeding back n according to the last m bits of Hi i+1 The method of (1) is as follows:
n i ＝min[max(floor(10 k ·sin(x))，N 1 )，N 2 ]
wherein n is i Initial value n of 1 Is a preset value, x is H i And (2) is a constant value.
8. A construction site safety risk noninductive inspection method as set forth in claim 4, wherein,
the risk identification result is sent to a responsible person, the real-time inspection image data are converted into a evidence-storing hash value, and the method for storing the risk identification result and the evidence-storing hash value further comprises the following steps:
after the risk identification result is obtained, real-time inspection image data of a preset inspection route, all types of risk items and risk identification results corresponding to all risk items are formed into an inspection result report, the inspection result report is sent to a responsible person, the real-time inspection image data is converted into a certificate hash value, and the inspection result report and the certificate hash value are uploaded to a designated server for storage.
9. A construction site security risk noninductive inspection system, comprising:
the inspection terminal is used for reading historical inspection image data of a preset inspection route and receiving real-time inspection image data of the preset inspection route;
the data analysis module is used for carrying out risk labeling on the historical inspection image data and establishing a risk identification template according to the historical inspection image data with the risk labeling;
the risk identification module is used for comparing the real-time inspection image data with the risk identification template to obtain a risk identification result;
the data storage module is used for converting the real-time inspection image data into a storage hash value;
and the server is used for sending the risk identification result to a responsible person and storing the risk identification result and the evidence-storing hash value.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310697918.5A CN116862843A (en) | 2023-06-13 | 2023-06-13 | Construction site safety risk noninductive inspection method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310697918.5A CN116862843A (en) | 2023-06-13 | 2023-06-13 | Construction site safety risk noninductive inspection method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116862843A true CN116862843A (en) | 2023-10-10 |
Family
ID=88229450
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310697918.5A Pending CN116862843A (en) | 2023-06-13 | 2023-06-13 | Construction site safety risk noninductive inspection method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116862843A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117549330A (en) * | 2024-01-11 | 2024-02-13 | 四川省铁路建设有限公司 | Construction safety monitoring robot system and control method |
-
2023
- 2023-06-13 CN CN202310697918.5A patent/CN116862843A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117549330A (en) * | 2024-01-11 | 2024-02-13 | 四川省铁路建设有限公司 | Construction safety monitoring robot system and control method |
| CN117549330B (en) * | 2024-01-11 | 2024-03-22 | 四川省铁路建设有限公司 | Construction safety monitoring robot system and control method |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110472858B (en) | Construction engineering supervision information management method | |
| CN116862843A (en) | Construction site safety risk noninductive inspection method and system | |
| CN112749813A (en) | Data processing system, method, electronic equipment and storage medium | |
| CN113850562B (en) | Intelligent side station supervision method and system | |
| CN113255590A (en) | Defect detection model training method, defect detection method, device and system | |
| CN112085710A (en) | Protection constant value and pressure plate inspection system and method | |
| CN111062679A (en) | Overall process intelligent engineering supervision information management and control system | |
| CN115331002A (en) | Method for realizing remote processing of heating power station fault based on AR glasses | |
| CN110543866A (en) | Safety management system and method for capital construction engineering constructors | |
| CN112528825A (en) | Station passenger recruitment service method based on image recognition | |
| CN117035419B (en) | Intelligent management system and method for enterprise project implementation | |
| CN114565845A (en) | Intelligent inspection system for underground tunnel | |
| CN113743725B (en) | Ascending qualification detection method and system for ascending operators based on edge calculation | |
| CN113297913B (en) | Identification method for dressing specification of distribution network field operators | |
| CN117294819B (en) | All-round surveillance camera head suitable for bank | |
| CN111179458A (en) | Method and device for routing inspection of rail transit equipment | |
| CN113627321A (en) | Image identification method and device based on artificial intelligence and computer equipment | |
| CN117271593A (en) | Intelligent recognition system for safety measure execution inspection based on operation type | |
| CN116128413B (en) | Intelligent warehouse material statistics system based on Bluetooth communication | |
| CN113011300A (en) | Method, system and equipment for AI visual identification of violation behavior | |
| CN116668637A (en) | Beam field video safety monitoring system and method | |
| CN115442739B (en) | Temporary communication processing method and device for subway construction, electronic equipment and medium | |
| CN116343100A (en) | Target identification method and system based on self-supervision learning | |
| CN113297914B (en) | Distribution network field operation electricity testing action recognition method | |
| CN110826473A (en) | Neural network-based automatic insulator image identification method |
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 |