CN114092724A - Project supervision method and system based on image recognition technology - Google Patents
Project supervision method and system based on image recognition technology Download PDFInfo
- Publication number
- CN114092724A CN114092724A CN202111393026.3A CN202111393026A CN114092724A CN 114092724 A CN114092724 A CN 114092724A CN 202111393026 A CN202111393026 A CN 202111393026A CN 114092724 A CN114092724 A CN 114092724A
- Authority
- CN
- China
- Prior art keywords
- construction
- image
- project
- construction stage
- supervision
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- 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
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computational Linguistics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The project supervision method based on the image recognition technology comprises the steps of setting an image sample; training a neural network with the image sample; collecting project images on site at each construction stage; identifying the collected images of the items by using a neural network; sending the collected images in the construction stage and the result of image identification to a server; judging whether the construction stage meets the requirements of project supervision or not according to the image recognition result; if the construction stage does not meet the requirements of project supervision, searching a corresponding solution in a construction supervision database according to the problems existing in the construction stage; when the construction stage is completed according to the solution, the construction stage is checked again; finishing the acceptance of the current construction stage and sending an acceptance result to the server; the project supervision method based on the image recognition technology can efficiently realize project supervision.
Description
Technical Field
The invention relates to the technical field of project supervision, in particular to a project supervision method and system based on an image recognition technology.
Background
After the construction is finished, the completion condition of the project needs to be checked and accepted, and the construction quality is ensured. The general construction supervision is carried out in a manual mode, and workers collect the completion conditions of projects in a manual mode on site. Taking house building supervision as an example, after a building project is finished, a plurality of aspects of the house building need to be detected, and whether each aspect of the house meets the requirements of construction supervision is judged. When the information of the construction project needing to be detected is more, the manual supervision mode is low in efficiency, and the supervision requirement cannot be met.
In view of the above-mentioned related art, the inventor considers that the manual supervision method is inefficient.
Disclosure of Invention
The invention aims to provide a project supervision method and a project supervision system based on an image recognition technology, which can efficiently realize project supervision.
In order to realize the purpose, the technical scheme of the invention is as follows:
in a first aspect, the present application provides an engineering supervision method based on an image recognition technology, which adopts the following scheme:
an engineering supervision method based on an image recognition technology comprises the following steps:
setting an image sample;
training a neural network with the image sample;
collecting project images on site at each construction stage;
identifying the collected images of the items by using a neural network;
sending the collected images in the construction stage and the result of image identification to a server;
judging whether the construction stage meets the requirements of project supervision or not according to the image recognition result;
if the construction stage does not meet the requirements of project supervision, searching a corresponding solution in a construction supervision database according to the problems existing in the construction stage;
when the construction stage is completed according to the solution, the construction stage is checked again;
and finishing the acceptance of the current construction stage and sending an acceptance result to the server.
By adopting the technical scheme, the neural network is trained by the image sample, so that the neural network can efficiently and accurately identify the construction project similar to the image sample. The construction stage is identified through the neural network, and when the construction stage does not meet the requirements, a solution can be quickly searched through the construction supervision database, so that reference is provided for modification of the construction stage.
Preferably, the setting the image sample includes:
dividing a project into a plurality of construction stages according to actual construction requirements;
setting corresponding image samples for each construction stage according to detection requirements;
wherein the image sample is a photograph of a construction project or a three-dimensional modeled model.
By adopting the technical scheme, the project is divided into a plurality of different construction stages, different construction stages can be respectively checked and accepted, the pertinence checking and accepting of construction projects is realized, problems are found in a targeted mode, and the quality of construction supervision is improved.
Preferably, the training of the neural network with the image sample specifically includes: determining a loss function of the neural network based on the image samples:
y-true value, which is a true label for some set of inputs x; f (x) -predictor, which is a prediction label for input x; m-number of samples.
Preferably, the identifying the acquired image of the item by using the neural network specifically includes: and the neural network identifies the image in the construction stage and judges the similarity between the image in the construction stage and the image sample according to the identification result.
Preferably, the step of judging whether the construction stage meets the requirements of project supervision according to the image recognition result specifically comprises: and judging whether the construction stage meets the requirement of project supervision according to the similarity of the image and the image sample in the construction stage.
By adopting the technical scheme, whether the engineering quality in the construction stage meets the construction requirement can be determined by judging the similarity between the image collected in the construction stage and the image sample. And if the similarity is low, the construction quality in the construction stage does not meet the construction requirement. On the contrary, the construction quality in the construction stage meets the construction requirements.
In a second aspect, the present application provides an engineering supervision method based on an image recognition technology, which adopts the following scheme:
an engineering supervision method based on an image recognition technology comprises the following steps:
establishing a construction supervision database at a server side, and storing construction modification suggestions in the construction supervision database;
and storing the received image of the construction stage and the result of image identification in a server, and establishing a project life cycle management database.
By adopting the technical scheme, the informationized supervision of each construction stage is realized by image acquisition of each construction stage and image identification through the neural network. By informationized supervision of all construction stages of the construction project, the efficiency of supervision of the construction project is improved, targeted supervision of each construction stage of the construction project is realized, the problems existing in the construction project are solved accurately, and the quality of acceptance check of the project is improved.
Preferably, the establishing a construction supervision database at the service end, and the storing the construction modification suggestion in the construction supervision database includes:
setting a detection standard for each construction stage;
and when the construction stage does not meet the detection standard, giving a corresponding modification suggestion.
By adopting the technical scheme, the inspection and acceptance of the construction stage are realized by detecting the construction stage and giving the modification suggestion, and the modification suggestion is given in time, so that the improvement of the construction stage is facilitated.
Preferably, the step of storing the received image of the construction stage and the result of image recognition in a server, and establishing a project lifecycle management database comprises:
identifying the collected project image by using a neural network to judge whether the construction stage meets the construction detection standard or not, and obtaining the result of image identification;
the collected images of the construction stage, the identification result of the images of the construction stage, the information of the responsible person of the construction stage and the detection time of the construction stage form a detection record and the detection record is stored in a server;
the detection records of all construction stages of a construction project form the detection records of the construction project so as to be convenient for maintaining the construction project.
By adopting the technical scheme, the identification is carried out on each construction stage of the construction project, and the identification results of each construction stage form a detection record, so that the omnibearing detection of the whole construction project is realized.
In a third aspect, the present application provides an engineering supervision system based on an image recognition technology, which adopts the following scheme:
an engineering supervision system based on an image recognition technology is characterized by comprising a client and a server which are in communication connection. The client comprises a first memory and a first processor; the first processor executes a program stored in the first memory to implement the image recognition technology-based project supervision method of the first aspect. The server comprises a second memory and a second processor; the second processor executes a program stored in the second memory to implement the project supervision method based on the image recognition technology described in the second aspect.
The invention has the following advantages:
in the whole construction supervision process, the client side is used for collecting the image information of the supervised project, then the neural network is used for identifying the collected image information so as to judge whether the supervised project meets the construction requirement, the detection can be efficiently completed for the same type of project needing to be supervised, and the construction supervision efficiency is improved.
And sending and storing the collected images in the construction stage and the identification results of the images in a server, thereby realizing the whole-course recording of the construction supervision process. And after the project is judged to have problems, searching a corresponding solution in the construction supervision database according to the problems, and timely providing a modification suggestion for the construction scheme.
The whole process of construction supervision realizes high-efficiency supervision of the project through interaction of the client and the server, efficiently finds problems and provides a solution, and can meet the requirements of a large number of project supervision.
Drawings
Fig. 1 is a flowchart of an engineering supervision method based on an image recognition technology according to embodiment 1 of the present invention.
Fig. 2 is a flowchart of an engineering supervision method based on an image recognition technology according to embodiment 1 of the present invention.
Fig. 3 is a flowchart of an engineering supervision method based on an image recognition technology according to embodiment 2 of the present invention.
Fig. 4 is a flowchart of an engineering supervision method based on an image recognition technology according to embodiment 2 of the present invention.
Fig. 5 is a flowchart of an engineering supervision method based on an image recognition technology according to embodiment 2 of the present invention.
Detailed Description
The following examples are intended to illustrate the invention but are not intended to limit the scope of the invention.
In order to make the technical means, the original characteristics, the achieved purposes and the effects of the invention easy to understand, the invention is further described with reference to the figures and the specific embodiments.
The project supervision method and system based on the image recognition technology of the invention are explained in detail with reference to fig. 1 to 5.
Example 1
Referring to fig. 1 and 2, the project supervision method based on the image recognition technology of the present embodiment includes the steps of:
in step S1, an image sample is set at the client, which is an industrial camera in the present application.
And step S11, dividing the project into a plurality of construction stages according to the actual construction requirements.
And step S12, setting corresponding image samples according to the detection requirements for each construction stage.
Wherein the image sample is a photograph of a construction project or a three-dimensional modeled model.
For example, in a house construction project, it is necessary to sequentially construct a foundation of a house, a house frame construction, a wall construction, a door and window installation, and a finishing. According to the sequence of building construction, the building construction is divided into a foundation construction stage, a building frame construction stage, a wall construction stage, a door and window installation stage and a house decoration stage in sequence.
An image sample is set corresponding to each stage. For example, in the door and window installation stage, the pictures of the door and window installation of the same kind of buildings or the images of the door and window installation model of the three-dimensional modeling design can be collected, and the pictures of the same kind of buildings or the three-dimensional modeling model can be used as image samples. Depending on the actual situation, multiple photographs may be selected as the image sample.
Step S2, training the neural network by using the image sample, specifically, determining a loss function of the neural network based on the image sample:
y-true value, which is a true label for some set of inputs x; f (x) -predicted values, which are prediction labels for input x; m-number of samples.
And step S3, acquiring images of the project on site at each stage of construction.
For example, after the end of each stage of a house building, image information of the current house building is captured with an industrial camera. In the wall construction stage, after the building wall construction is finished, the industrial camera is used for collecting the image information of the building wall construction.
Step S4, identifying the collected project image by using a neural network, specifically:
and step S41, the neural network identifies the image in the construction stage and judges the similarity between the image in the construction stage and the image sample according to the identification result. The method adopts an average Hash algorithm (aHash) to judge the similarity of the image and the image sample in the construction stage.
And step S5, transmitting the collected image of the construction stage and the result of the image recognition to the server.
Step S6, judging whether the construction stage meets the requirements of project supervision according to the image recognition result specifically comprises the following steps: and judging whether the construction stage meets the requirement of project supervision according to the similarity of the image and the image sample in the construction stage.
For example, for the detection of the shape of a building, the similarity between the shape of the building and the shape of the image sample is determined. The angle between the two walls in the image sample is a right angle. When the wall construction is finished, the included angle of the part of the wall of the building collected by the industrial camera is not a right angle, the similarity is low, and the building wall is judged not to meet the construction requirements.
And step S7, if the construction stage does not meet the requirements of project supervision, searching a corresponding solution in a construction supervision database according to the problems existing in the construction stage, and after the construction stage is completed according to the solution, executing step S3 and re-checking and accepting the construction stage.
For example, the ratio of the size of the window to the size of the wall is large, and the construction supervision requirement is not met. The construction supervision database gives the recommendation to reduce the size of the window so that the ratio of the size of the window to the size of the wall meets the construction supervision requirements.
After the window size is modified according to the recommendations from the construction administration database, the process goes to step S3 to re-check the construction phase.
And step S8, finishing the acceptance of the current construction stage and sending the acceptance result to the server. And after the acceptance of a certain construction stage is finished, transmitting the relevant information of the construction stage, including the name of the construction stage and the acceptance result of the construction stage to the non-service state and storing the information.
In the whole construction supervision process, the image information of the supervised project is collected through the client, then the neural network identifies the collected image information to judge whether the supervised project meets the construction requirements, and for the same type of project needing to be supervised, the detection can be efficiently completed, and the efficiency of the project supervision is improved.
And sending and storing the collected images in the construction stage and the identification results of the images in a server, thereby realizing the whole-course recording of the construction supervision process. And after the project is judged to have problems, searching a corresponding solution in the construction supervision database according to the problems, and timely proposing a modification suggestion.
The whole construction supervision process realizes high-efficiency supervision of the project through interaction of the client and the server, efficiently finds problems and provides a solution, and can meet the requirements of supervision of a large number of projects.
Example 2
The application provides an engineering supervision method based on an image recognition technology, which comprises the following steps:
referring to fig. 3 and 4, in step S9, a construction supervision database is built at the service end, and construction modification suggestions are stored in the construction supervision database.
Step S91, setting detection standard for each construction stage.
And step S92, when the construction stage does not meet the detection standard, giving a corresponding modification suggestion.
For example, for the wall of a house building, the corresponding detection standards are set for the included angle between two adjacent walls, the ratio of the height of a door to the height of a room, and the like.
Two adjacent walls require the included angle to be a right angle, and when the included angle of the wall does not meet the requirements, the corresponding modification suggestion is to remove one or two walls and rebuild the wall so that the included angle of the wall meets the detection standard.
And setting an upper limit on the ratio of the height of the room door to the height of the room, and when the ratio of the height of the room door to the height of the room exceeds the upper limit, correspondingly modifying and proposing that the height of the room door is reduced so that the ratio of the height of the room door to the height of the room meets the requirement.
Referring to fig. 5, step S10 is to store the received image of the construction stage and the result of the image recognition in the server, and to establish a project lifecycle management database.
And S101, identifying the acquired project image by the neural network to judge whether the construction stage meets the construction detection standard or not and obtaining the result of image identification.
Taking a house building as an example, an included angle between two walls is required to be a right angle. The industrial camera captures an image of a wall of a house, and recognizes the captured image through a neural network. And when the included angle of the house wall is not a right angle, identifying and judging that the included angle of the wall does not meet the construction detection standard through a neural network.
And S102, forming a detection record by the collected construction stage image, the identification result of the construction stage image, the information of a responsible person in the construction stage and the detection time in the construction stage, and storing the detection record in a server.
For example, it is determined through neural network judgment that one of the included angles of the wall is not a right angle and does not meet the construction detection standard. The method comprises the steps that when images are collected through an industrial camera, information of a responsible person in a construction stage and detection time of the construction stage are sent to a server through an intelligent terminal. And sending and storing the acquired images of the included angles of the walls of the house, the recognition results of the images by the neural network, the information of the responsible persons in the construction stage and the detection time in the construction stage to a server to form house detection records. The intelligent terminal is a mobile phone and is communicated with the server.
Step S103: the detection records of all construction stages of a construction project form the detection records of the construction project so as to be convenient for maintaining the construction project. In the house use and maintenance process in later stage, the problem that exists in the house is found in time through house detection record, and the house maintenance in later stage of being convenient for.
Example 3
The project supervision system based on the image recognition technology comprises a client and a server which are in communication connection.
The client comprises a first memory and a first processor, and the first processor executes a program stored on the first memory to realize the project supervision method based on the image recognition technology in embodiment 1.
The server includes a second memory and a second processor. The second processor executes the program stored in the second memory to implement the project supervision method based on the image recognition technology described in embodiment 2. The client is an industrial camera.
Although the invention has been described in detail above with reference to a general description and specific examples, it will be apparent to one skilled in the art that modifications or improvements may be made thereto based on the invention. Accordingly, such modifications and improvements are intended to be within the scope of the invention as claimed.
Claims (9)
1. An engineering supervision method based on an image recognition technology is characterized by comprising the following steps:
setting an image sample;
training a neural network with the image sample;
collecting project images on site at each construction stage;
identifying the collected images of the items by using a neural network;
sending the collected images in the construction stage and the result of image identification to a server;
judging whether the construction stage meets the requirements of project supervision or not according to the image recognition result;
if the construction stage does not meet the requirements of project supervision, searching a corresponding solution in a construction supervision database according to the problems in the construction stage;
when the construction stage is completed according to the solution, the construction stage is checked again;
and finishing the acceptance of the current construction stage and sending an acceptance result to the server.
2. The project supervision method based on the image recognition technology as claimed in claim 1, wherein the setting of the image sample comprises:
dividing a project into a plurality of construction stages according to actual construction requirements;
setting corresponding image samples for each construction stage according to the detection requirements;
wherein the image sample is a photograph of a construction project or a three-dimensional modeled model.
3. The project supervision method based on the image recognition technology as claimed in claim 1, wherein the training of the neural network by the image sample is specifically: determining a loss function of the neural network based on the image samples:
y-true value, which is a true label for some set of inputs x; f (x) -predicted values, which are prediction labels for input x; m-number of samples.
4. The project supervision method based on the image recognition technology as claimed in claim 1, wherein the recognition of the collected project image by the neural network is specifically:
and the neural network identifies the image in the construction stage, and the similarity between the image in the construction stage and the image sample is judged according to the identification result.
5. The project supervision method based on the image recognition technology as claimed in claim 4, wherein the step of judging whether the construction stage meets the project supervision requirement according to the image recognition result is specifically as follows: and judging whether the construction stage meets the requirement of project supervision according to the similarity of the image and the image sample in the construction stage.
6. An engineering supervision method based on an image recognition technology is characterized by comprising the following steps:
establishing a construction supervision database at a server side, and storing construction modification suggestions in the construction supervision database;
and storing the received image of the construction stage and the result of image identification in a server, and establishing a project life cycle management database.
7. The project supervision method based on the image recognition technology as claimed in claim 6, wherein the building of the construction supervision database at the server end, and the storing of the construction modification suggestions in the construction supervision database comprises:
setting a detection standard for each construction stage;
and when the construction stage does not meet the detection standard, giving a corresponding modification suggestion.
8. The project supervision method based on the image recognition technology as claimed in claim 7, wherein the step of storing the received images of the construction stage and the result of the image recognition in a server, and establishing a project lifecycle management database comprises:
identifying the collected project image by using a neural network to judge whether the construction stage meets the construction detection standard or not, and obtaining the result of image identification;
the collected construction stage image, the identification result of the construction stage image, the information of the responsible person in the construction stage and the construction stage detection time form a detection record and are stored in a server;
the detection records of all construction stages of a construction project form the detection records of the construction project so as to be convenient for maintaining the construction project.
9. An engineering supervision system based on an image recognition technology is characterized by comprising a client and a server which are in communication connection;
the client comprises a first memory and a first processor; the first processor executes a program stored in the first memory to implement the project supervision method based on the image recognition technology according to any one of claims 1 to 5;
the server comprises a second memory and a second processor; the second processor executes a program stored in the second memory to implement the project supervision method based on the image recognition technology according to any one of claims 6 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111393026.3A CN114092724A (en) | 2021-11-23 | 2021-11-23 | Project supervision method and system based on image recognition technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111393026.3A CN114092724A (en) | 2021-11-23 | 2021-11-23 | Project supervision method and system based on image recognition technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114092724A true CN114092724A (en) | 2022-02-25 |
Family
ID=80303082
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111393026.3A Pending CN114092724A (en) | 2021-11-23 | 2021-11-23 | Project supervision method and system based on image recognition technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114092724A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114677123A (en) * | 2022-04-19 | 2022-06-28 | 赵德群 | Management system for low-carbon green building supervision engineering project |
-
2021
- 2021-11-23 CN CN202111393026.3A patent/CN114092724A/en active Pending
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114677123A (en) * | 2022-04-19 | 2022-06-28 | 赵德群 | Management system for low-carbon green building supervision engineering project |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110544293B (en) | Building scene recognition method through visual cooperation of multiple unmanned aerial vehicles | |
| US20130243249A1 (en) | Electronic device and method for recognizing image and searching for concerning information | |
| CN108416707A (en) | House house type appraisal procedure and device | |
| CN108921038A (en) | A kind of classroom based on deep learning face recognition technology is quickly called the roll method of registering | |
| CN112820071B (en) | Behavior recognition method and device | |
| CN110765134A (en) | File establishing method, equipment and storage medium | |
| CN105005852A (en) | Image analysis based intelligent monitoring system for dormitory environment | |
| CN112349057A (en) | Deep learning-based indoor smoke and fire detection method | |
| CN109446946A (en) | A kind of multi-cam real-time detection method based on multithreading | |
| CN114092724A (en) | Project supervision method and system based on image recognition technology | |
| CN115862113A (en) | Stranger abnormity identification method, device, equipment and storage medium | |
| CN110827263A (en) | Magnetic shoe surface defect detection system and detection method based on visual identification technology | |
| CN113962660A (en) | AI technology-based construction worker behavior safety management correction method and system | |
| CN113626917A (en) | Component engineering quantity calculation method and system based on panoramic technology and BIM model | |
| CN111567060B (en) | Quality inspection device, quality inspection method, and program | |
| CN114663834B (en) | On-site monitoring method for express storage | |
| JP2020204708A (en) | Map information update system | |
| Sun et al. | Automatic building age prediction from street view images | |
| CN115344761A (en) | Memory occupation optimization method and system for nodes of Internet of things | |
| CN114584585A (en) | Industrial equipment self-diagnosis system and method based on Internet of things | |
| CN113051981B (en) | Face recognition method and device | |
| CN112883213B (en) | Picture archiving method and device and electronic equipment | |
| CN111709346A (en) | Historical building identification and detection method based on deep learning and high-resolution images | |
| CN113032629A (en) | Hotel peer detection and determination method and device | |
| KR102520218B1 (en) | System and Method for improving hardware usage in control server using artificial intelligence image processing |
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 |