CN114757948B - Deep learning-based method and device for detecting content of recycled aggregate mortar - Google Patents
Deep learning-based method and device for detecting content of recycled aggregate mortar Download PDFInfo
- Publication number
- CN114757948B CN114757948B CN202210664982.9A CN202210664982A CN114757948B CN 114757948 B CN114757948 B CN 114757948B CN 202210664982 A CN202210664982 A CN 202210664982A CN 114757948 B CN114757948 B CN 114757948B
- Authority
- CN
- China
- Prior art keywords
- mortar
- aggregate
- data set
- recycled aggregate
- content
- 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.)
- Active
Links
Images
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
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/11—Region-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/12—Edge-based segmentation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/187—Segmentation; Edge detection involving region growing; involving region merging; involving connected component labelling
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
- G06T7/62—Analysis of geometric attributes of area, perimeter, diameter or volume
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20081—Training; Learning
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/30—Subject of image; Context of image processing
- G06T2207/30108—Industrial image inspection
- G06T2207/30132—Masonry; Concrete
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02W—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO WASTEWATER TREATMENT OR WASTE MANAGEMENT
- Y02W30/00—Technologies for solid waste management
- Y02W30/50—Reuse, recycling or recovery technologies
- Y02W30/91—Use of waste materials as fillers for mortars or concrete
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Biomedical Technology (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- Quality & Reliability (AREA)
- Biophysics (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Geometry (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention discloses a method and a device for detecting the content of recycled aggregate mortar based on deep learning, belonging to the field of visual detection of deep learning, wherein a first data set mixed with pure aggregate and pure mortar is adopted to train an image segmentation model to obtain a basic model; segmenting the second data set to obtain a third data set; disordering the first data set, the third data set and the fourth data set to obtain a fifth data set, and using the fifth data set to train a semantic segmentation model based on deplab to obtain a final model; obtaining images of different surfaces of the recycled aggregate to be detected, and calculating the mortar content and the area ratio of the connected area after obtaining a second segmentation result through final model segmentation; the method comprises the steps of obtaining the relation between the water absorption of the standard aggregate and the mortar content and the ratio of the communicating area, and calculating the water absorption of the recycled aggregate to be detected according to the relation and the ratio of the mortar content and the communicating area, so that the problems that the performance of the recycled aggregate cannot be detected in real time, the detection efficiency is low and the like are solved.
Description
Technical Field
The invention relates to the field of deep learning, in particular to a method and a device for detecting the content of recycled aggregate mortar based on deep learning.
Background
The recycled aggregate is formed by crushing, cleaning and grading waste concrete. The use of the recycled aggregate effectively utilizes the waste construction waste, and avoids the accumulation of the construction waste to destroy the environment. Compared with natural aggregate, the recycled aggregate can completely replace the natural aggregate in various physical and chemical index performances, has stable yield, can completely make up the current situation of insufficient natural aggregate, and effectively protects the environment and the homeland resources.
The content of the mortar on the surface of the recycled aggregate influences the service performance of the recycled aggregate, and the content of the mortar on the surface of the recycled aggregate has the characteristics of large pores, high water absorption and low strength compared with the natural aggregate, so that the quality performance of the recycled aggregate can be evaluated by the content of the mortar to a certain extent.
The conventional method for measuring the mortar content of the recycled aggregate is generally a high-temperature ball milling method, a chemical method and the like, and the methods are complicated in process and cannot meet the requirement of real-time production.
Disclosure of Invention
The invention mainly aims to improve the detection efficiency of the mortar content on the recycled aggregate, and compared with the traditional detection method which needs to be completed in a laboratory, the invention uses the deep learning technology to segment the mortar content on the recycled aggregate, and can quickly obtain the mortar content on a batch of recycled aggregates so as to determine the quality and the performance of the recycled aggregates. The embodiment of the application aims to provide a method and a device for detecting the content of recycled aggregate mortar based on deep learning, so as to solve the technical problems mentioned in the background technology section.
In a first aspect, an embodiment of the present application provides a method for detecting a content of recycled aggregate mortar based on deep learning, including the following steps:
s1, training the image segmentation model by adopting a first data set mixed with pure aggregate and pure mortar to obtain a basic model;
s2, segmenting the second data set containing the aggregate of the mortar by adopting the basic model to obtain a first segmentation result, and obtaining a third data set based on the first segmentation result;
s3, obtaining a fourth data set containing aggregate of mortar, disordering the first data set, the third data set and the fourth data set to obtain a fifth data set, and training a semantic segmentation model based on deplab by adopting the fifth data set to obtain a final model;
s4, obtaining images of different surfaces of the recycled aggregate to be detected, segmenting the images of different surfaces by adopting a final model to obtain a second segmentation result, and calculating the mortar content and the communication domain area ratio of the recycled aggregate to be detected according to the second segmentation result;
and S5, obtaining the relation between the water absorption of the standard aggregate which has the same material as the recycled aggregate to be detected and the mortar content and the communication area ratio, and calculating the water absorption of the recycled aggregate to be detected according to the relation and the mortar content and the communication area ratio of the recycled aggregate to be detected.
Preferably, the image segmentation model comprises a semantic segmentation model and an instance segmentation model.
Preferably, the base model is a semantic segmentation model based on deplab.
Preferably, the semantic segmentation model based on deplab is deplab V3+, in the encoding stage, firstly, the DCNN network of Xceptation is used for extracting low semantic information in the image, the low semantic information is transmitted into the decoding stage, the feature map is extracted, the ASPP multi-scale convolution is carried out, the low semantic information is transmitted into the decoding stage, in the decoding stage, two data transmitted in the encoding stage are fused to obtain a result, and the semantic information is extracted.
Preferably, the first data set in step S1 is obtained according to the following steps:
acquiring a picture of pure aggregate or pure mortar, extracting a mask of the picture of pure aggregate or pure mortar by using graying, binaryzation and median filtering, determining the coordinate of the boundary of the mask by using an edge searching function in opencv, and storing the category, the coordinate information and the like of the picture of pure aggregate or pure mortar in a json format file; the step S2 of obtaining the third data set based on the segmentation result specifically includes: correcting and adjusting the first segmentation result to extract a segmentation mask and generate a third data set; the fourth data set in step S3 is obtained by manual annotation.
Preferably, in step S4, the calculating the mortar content and the connected domain area ratio in the recycled aggregate to be detected according to the second division result specifically includes: calculating the areas S1, S2, … … and Sn of the mortar and the areas T1, T2, … … and Tn of the recycled aggregate on n different surfaces according to the second division result, and respectively averaging the areas S and T to obtain the average area S of the mortar and the average area T of the recycled aggregate:
calculating the mortar content as follows: mu = S/T;
calculating the average value of the areas of the recycled aggregates larger than the threshold value alpha of the area of the connected domain to obtain the area of the connected domain
Alpha is set according to the area characteristics of the recycled aggregate;
calculating the area ratio of the connected domain as follows:
。
preferably, step S5 specifically includes:
carrying out water absorption test on the standard aggregate to obtain the water absorption of the standard aggregate, calculating by adopting a final model to obtain the mortar content and the communication domain area ratio of the standard aggregate, carrying out regression analysis by taking the water absorption of the standard aggregate as a dependent variable and the mortar content and the communication domain area ratio of the standard aggregate as independent variables to obtain the relation between the water absorption of the standard aggregate and the mortar content and the communication domain area ratio;
and substituting the mortar content and the communication area ratio of the recycled aggregate to be detected into the relation, and calculating to obtain the water absorption of the recycled aggregate to be detected.
In a second aspect, an embodiment of the present application provides a device for detecting a content of recycled aggregate mortar based on deep learning, including:
the basic model training module is configured to train the image segmentation model by adopting a first data set mixed with pure aggregate and pure mortar to obtain a basic model;
the first segmentation module is configured to segment a second data set containing aggregate of the mortar by adopting a basic model to obtain a first segmentation result, and obtain a third data set based on the first segmentation result;
the self-training module is configured to obtain a fourth data set of aggregate containing mortar, disorder the first data set, the third data set and the fourth data set to obtain a fifth data set, and train the deep semantic segmentation model based on the fifth data set to obtain a final model;
the second segmentation module is configured to acquire images of different surfaces of the to-be-detected recycled aggregate, segment the images of the different surfaces by adopting a final model to obtain a second segmentation result, and calculate the mortar content and the communication domain area ratio of the to-be-detected recycled aggregate according to the second segmentation result;
and the water absorption calculation module is configured to acquire the relation between the water absorption of the standard aggregate which has the same material as the to-be-detected recycled aggregate and the mortar content and the communication area ratio, and calculate the water absorption of the to-be-detected recycled aggregate according to the relation and the mortar content and the communication area ratio of the to-be-detected recycled aggregate.
In a third aspect, embodiments of the present application provide an electronic device comprising one or more processors; a storage device for storing one or more programs which, when executed by one or more processors, cause the one or more processors to implement a method as described in any implementation of the first aspect.
In a fourth aspect, embodiments of the present application provide a computer-readable storage medium storing a computer program, which when executed by a processor implements the method as described in any of the implementations of the first aspect.
Compared with the prior art, the invention has the following beneficial effects:
(1) according to the method, the semantic segmentation model based on deplab trained in a self-training similar manner is adopted, so that the accuracy of mortar segmentation on the recycled aggregate can be effectively improved, the quality of the recycled aggregate is rapidly determined, and the analysis efficiency is improved.
(2) The invention uses deep learning technology to detect the recycled aggregate, can quickly determine the content of the coated mortar, and can ensure that the identification precision of the model is higher by continuously adding samples.
(3) The method adopts an industrial camera fixed above a conveyor belt to shoot in real time to obtain the recycled aggregate to be detected, and carries out prediction through a trained semantic segmentation model based on deplab, so that a prediction result can be obtained in real time, and the water absorption of the recycled aggregate to be detected is calculated, thereby judging the quality of the recycled aggregate.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is an exemplary device architecture diagram in which one embodiment of the present application may be applied;
FIG. 2 is a schematic flow chart of a method for detecting the content of recycled aggregate mortar based on deep learning according to an embodiment of the invention;
FIG. 3 is an example of an image of an aggregate and mortar data set for a deep learning-based recycled aggregate mortar content detection method according to an embodiment of the present invention;
FIG. 4 is an example of an image of a manually labeled tag-containing dataset of the deep learning-based recycled aggregate mortar content detection method of an embodiment of the present invention;
FIG. 5 is a diagram showing an example of a result predicted by a final model of a deep learning-based recycled aggregate mortar content detection method according to an embodiment of the present invention;
FIG. 6 is a schematic diagram of a device for detecting the content of recycled aggregate mortar based on deep learning according to an embodiment of the present invention;
fig. 7 is a schematic structural diagram of a computer device suitable for implementing an electronic apparatus according to an embodiment of the present application.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the present invention will be described in further detail with reference to the accompanying drawings, and it is apparent that the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Fig. 1 illustrates an exemplary device architecture 100 to which the deep learning-based recycled aggregate mortar content detection method or the deep learning-based recycled aggregate mortar content detection device according to the embodiment of the present application may be applied.
As shown in fig. 1, the apparatus architecture 100 may include terminal devices 101, 102, 103, a network 104, and a server 105. Network 104 is the medium used to provide communication links between terminal devices 101, 102, 103 and server 105. Network 104 may include various connection types, such as wired, wireless communication links, or fiber optic cables, to name a few.
The user may use the terminal devices 101, 102, 103 to interact with the server 105 via the network 104 to receive or send messages or the like. Various applications, such as data processing type applications, file processing type applications, etc., may be installed on the terminal apparatuses 101, 102, 103.
The terminal apparatuses 101, 102, and 103 may be hardware or software. When the terminal devices 101, 102, 103 are hardware, they may be various electronic devices including, but not limited to, smart phones, tablet computers, laptop portable computers, desktop computers, and the like. When the terminal apparatuses 101, 102, 103 are software, they can be installed in the electronic apparatuses listed above. It may be implemented as multiple pieces of software or software modules (e.g., software or software modules used to provide distributed services) or as a single piece of software or software module. And is not particularly limited herein.
The server 105 may be a server that provides various services, such as a background data processing server that processes files or data uploaded by the terminal devices 101, 102, 103. The background data processing server can process the acquired file or data to generate a processing result.
It should be noted that the method for detecting the content of the recycled aggregate mortar based on deep learning provided in the embodiment of the present application may be executed by the server 105, or may be executed by the terminal devices 101, 102, and 103, and accordingly, the device for detecting the content of the recycled aggregate mortar based on deep learning may be provided in the server 105, or may be provided in the terminal devices 101, 102, and 103.
It should be understood that the number of terminal devices, networks, and servers in fig. 1 is merely illustrative. There may be any number of terminal devices, networks, and servers, as desired for implementation. In the case where the processed data does not need to be acquired from a remote location, the above device architecture may not include a network, but only a server or a terminal device.
Fig. 2 shows a method for detecting the content of recycled aggregate mortar based on deep learning, which includes the following steps:
and S1, training the image segmentation model by adopting the first data set mixed with the pure aggregate and the pure mortar to obtain a basic model.
In a specific embodiment, the first data set in step S1 is obtained according to the following steps:
acquiring a picture of pure aggregate or pure mortar, extracting a mask of the picture of pure aggregate or pure mortar by using graying, binaryzation and median filtering, determining the coordinate of the boundary of the mask by using an edge searching function in opencv, and storing the category, the coordinate information and the like of the picture of pure aggregate or pure mortar in a json format file;
specifically, a common industrial global shutter area-array camera is used for collecting a batch of pure aggregate pictures and collecting a batch of pure mortar pictures. The background of the image of the recycled aggregate data set is a fixed conveyor belt, the image set is captured from the top down by an industrial camera fixed above the conveyor belt, and an example of the image of the aggregate and mortar data set is shown in fig. 3. The pure aggregate data set A1 and the pure mortar data set A2 are labeled using an automatic labeling method, and the label information includes contour information of the object and category information of the object. Specifically, a mask of a picture of pure aggregate or pure mortar is extracted by using graying, binarization, median filtering and other methods, the coordinate of the boundary of the mask is determined by using an edge finding function in opencv, and the category, the coordinate information and the like of the picture of pure aggregate or pure mortar are stored in a json format file. The first data set A3 was generated by scrambling the pure aggregate data set a1 with the pure mortar data set a2 using a multiple data mixing approach in data enhancement.
In particular embodiments, the image segmentation model includes, but is not limited to, a semantic segmentation model and an instance segmentation model. Specifically, a model is selected from the image segmentation models, and the model is trained using the first data set a3 to obtain a base model. In a preferred embodiment, the base model is a deplab-based semantic segmentation model. That is, the base model is trained for the first time by the first data set a3, and is also used for the second training in the following step S3 to update the model to obtain a model more capable of being accurately segmented.
And S2, segmenting the second data set containing the aggregate of the mortar by adopting the basic model to obtain a first segmentation result, and obtaining a third data set based on the first segmentation result.
Specifically, the second data set is an unmarked data set containing aggregate of mortar, and the segmentation is performed by using the basic model to obtain a segmentation result, and the step S2 of obtaining the third data set based on the first segmentation result specifically includes: and correcting and adjusting the segmentation result to extract a segmentation mask, and generating a third data set. The segmentation cases of each class are reviewed and the first segmentation result is manually adjusted to extract the segmentation mask to generate a third data set D2. Since the basic model is the same as the semantic segmentation model based on deplab below, the self-training effect is achieved.
S3, obtaining a fourth data set containing aggregate of mortar, disordering the first data set, the third data set and the fourth data set to obtain a fifth data set, and training the semantic segmentation model based on deplab by adopting the fifth data set to obtain a final model.
Specifically, referring to fig. 4, a batch of fourth data set B1 containing aggregate from the mortar is manually annotated, and the fourth data set B1 is merged with the first data set A3 and the third data set D2 and shuffled to form a fifth data set a 4. Converting the data set file in the json format into the forms of an original image and a png image; dividing the converted fifth data set into a training set, a verification set and a test set; and training the semantic segmentation model based on deplab by adopting a training set, evaluating the effect of the trained model by adopting a verification set, evaluating the generalization capability of the final model by adopting a test set, and training to obtain the final model with the best performance.
In a specific embodiment, the semantic segmentation model based on deplab is deplab V3+, and includes an Xception backbone network, an ASPP feature extraction network and other structures. In the encoding stage, firstly, low semantic information in an image is extracted by using a DCNN (binary-coded neural network) of Xception and transmitted into a decoding stage, a feature map is extracted to perform ASPP (asynchronous transfer protocol) multi-scale convolution and transmitted into the decoding stage, and two data transmitted in the encoding stage are fused in the decoding stage to obtain a result and extract semantic information.
And S4, acquiring images of different surfaces of the recycled aggregate to be detected, segmenting the images of different surfaces by adopting the final model to obtain a second segmentation result, and calculating the mortar content and the communication domain area ratio of the recycled aggregate to be detected according to the second segmentation result.
In a specific embodiment, the segmentation result is shown in fig. 5, and the step S4 of calculating the mortar content and the connected domain area ratio in the recycled aggregate to be detected according to the second segmentation result specifically includes: calculating the areas S1, S2, … … and Sn of the mortar and the areas T1, T2, … … and Tn of the recycled aggregate on n different surfaces according to the second division result, and respectively averaging the areas S and T to obtain the average area S of the mortar and the average area T of the recycled aggregate:
calculating the mortar content as follows: mu = S/T;
calculating the average value of the areas of the recycled aggregates larger than the threshold value alpha of the area of the connected domain to obtain the area of the connected domain
Alpha is set according to the area characteristics of the recycled aggregate;
calculating the area ratio of the connected domain as follows:
。
specifically, the connected domain area threshold alpha is set according to the area characteristics of the recycled aggregate, and the average value of the areas of the partial recycled aggregates with the areas larger than the threshold alpha is obtained to obtain the area of the connected domain
。
And S5, obtaining the relation between the water absorption of the standard aggregate which has the same material as the recycled aggregate to be detected and the mortar content and the communication area ratio, and calculating the water absorption of the recycled aggregate to be detected according to the relation and the mortar content and the communication area ratio of the recycled aggregate to be detected.
In a specific embodiment, step S5 specifically includes:
carrying out water absorption test on the standard aggregate to obtain the water absorption of the standard aggregate, calculating by adopting a final model to obtain the mortar content and the communication domain area ratio of the standard aggregate, carrying out regression analysis by taking the water absorption of the standard aggregate as a dependent variable and the mortar content and the communication domain area ratio of the standard aggregate as independent variables to obtain the relation between the water absorption of the standard aggregate and the mortar content and the communication domain area ratio;
and substituting the mortar content and the communication area ratio of the recycled aggregate to be detected into the relation, and calculating to obtain the water absorption of the recycled aggregate to be detected.
Specifically, the water absorption rate of the standard aggregate which has the same material as the to-be-detected recycled aggregate, the mortar content mu of the to-be-detected recycled aggregate and the area ratio of the connected region are established
And substituting the mortar content and the communication domain area ratio of the recycled aggregate to be detected, which are calculated in real time, into the regression relationship to calculate the water absorption of the recycled aggregate to be detected. Because a planar image of the to-be-detected recycled aggregate is adopted, a real apparent density cannot be obtained, when the water absorption of the to-be-detected recycled aggregate is calculated, the apparent density of the standard aggregate is taken as the apparent density of the to-be-detected recycled aggregate, the water absorption is taken as a dependent variable, the mortar content and the communication domain area ratio are taken as independent variables, a regression relation is established, and the water absorption of the to-be-detected recycled aggregate is further calculated according to the mortar content and the communication domain area ratio of the to-be-detected recycled aggregate calculated in real time, so that the quality of the recycled aggregate is judged.
With further reference to fig. 6, as an implementation of the methods shown in the above figures, the present application provides an embodiment of a device for detecting recycled aggregate mortar content based on deep learning, which corresponds to the embodiment of the method shown in fig. 2, and which can be applied to various electronic devices.
The embodiment of the application provides a recycled aggregate mortar content detection device based on deep learning, includes:
the basic model training module 1 is configured to train an image segmentation model by adopting a first data set mixed with pure aggregate and pure mortar to obtain a basic model;
the first segmentation module 2 is configured to segment a second data set containing aggregate of the mortar by adopting a basic model to obtain a first segmentation result, and obtain a third data set based on the first segmentation result;
the self-training module 3 is configured to obtain a fourth data set of aggregate containing mortar, disorder the first data set, the third data set and the fourth data set to obtain a fifth data set, and train the deep semantic segmentation model based on the fifth data set to obtain a final model;
the second segmentation module 4 is configured to acquire images of different surfaces of the to-be-detected recycled aggregate, segment the images of the different surfaces by using a final model to obtain a second segmentation result, and calculate the mortar content and the communication domain area ratio of the to-be-detected recycled aggregate according to the second segmentation result;
and the water absorption calculation module 5 is configured to acquire a relation between the water absorption of the standard aggregate having the same material as the to-be-detected recycled aggregate and the mortar content and the communication area ratio, and calculate the water absorption of the to-be-detected recycled aggregate according to the relation and the mortar content and the communication area ratio of the to-be-detected recycled aggregate.
Referring now to fig. 7, a schematic diagram of a computer device 700 suitable for use in implementing an electronic device (e.g., the server or terminal device shown in fig. 1) according to an embodiment of the present application is shown. The electronic device shown in fig. 7 is only an example, and should not bring any limitation to the functions and the scope of use of the embodiments of the present application.
As shown in fig. 7, the computer apparatus 700 includes a Central Processing Unit (CPU) 701 and a Graphics Processing Unit (GPU) 702, which can perform various appropriate actions and processes in accordance with a program stored in a Read Only Memory (ROM) 703 or a program loaded from a storage section 709 into a Random Access Memory (RAM) 704. In the RAM 704, various programs and data necessary for the operation of the apparatus 700 are also stored. The CPU 701, GPU702, ROM 703, and RAM 704 are connected to each other via a bus 705. An input/output (I/O) interface 706 is also connected to bus 705.
The following components are connected to the I/O interface 706: an input section 707 including a keyboard, a mouse, and the like; an output section 708 including a display such as a Liquid Crystal Display (LCD) and a speaker; a storage section 709 including a hard disk and the like; and a communication section 710 including a network interface card such as a LAN card, a modem, or the like. The communication section 710 performs communication processing via a network such as the internet. The driver 711 may also be connected to the I/O interface 706 as needed. A removable medium 712 such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, or the like is mounted on the drive 711 as necessary, so that a computer program read out therefrom is mounted into the storage section 709 as necessary.
In particular, according to an embodiment of the present disclosure, the processes described above with reference to the flowcharts may be implemented as computer software programs. For example, embodiments of the present disclosure include a computer program product comprising a computer program embodied on a computer readable medium, the computer program comprising program code for performing the method illustrated in the flow chart. In such embodiments, the computer program may be downloaded and installed from a network via the communication section 710, and/or installed from the removable media 712. The computer program performs the above-described functions defined in the method of the present application when executed by a Central Processing Unit (CPU) 701 and a Graphics Processing Unit (GPU) 702.
It should be noted that the computer readable medium described herein can be a computer readable signal medium or a computer readable medium or any combination of the two. The computer readable medium can be, for example, but not limited to, an electronic, magnetic, optical, electromagnetic, infrared, or semiconductor device, apparatus, or a combination of any of the foregoing. More specific examples of the computer readable medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer diskette, a hard disk, a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the foregoing. In the present application, a computer readable medium may be any tangible medium that can contain, or store a program for use by or in connection with an instruction execution apparatus, device, or apparatus. In this application, however, a computer readable signal medium may include a propagated data signal with computer readable program code embodied therein, for example, in baseband or as part of a carrier wave. Such a propagated data signal may take many forms, including, but not limited to, electro-magnetic, optical, or any suitable combination thereof. A computer readable signal medium may be any computer readable medium that is not a computer readable storage medium and that can communicate, propagate, or transport a program for use by or in connection with an instruction execution apparatus, device, or apparatus. Program code embodied on a computer readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, fiber optic cable, RF, etc., or any suitable combination of the foregoing.
Computer program code for carrying out operations for aspects of the present application may be written in any combination of one or more programming languages, including an object oriented programming language such as Java, Smalltalk, C + + or the like and conventional procedural programming languages, such as the "C" programming language or similar programming languages. The program code may execute entirely on the user's computer, partly on the user's computer, as a stand-alone software package, partly on the user's computer and partly on a remote computer or entirely on the remote computer or server. In the case of a remote computer, the remote computer may be connected to the user's computer through any type of network, including a Local Area Network (LAN) or a Wide Area Network (WAN), or the connection may be made to an external computer (for example, through the Internet using an Internet service provider).
The flowchart and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of apparatus, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based devices that perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The modules described in the embodiments of the present application may be implemented by software or hardware. The modules described may also be provided in a processor.
As another aspect, the present application also provides a computer-readable medium, which may be contained in the electronic device described in the above embodiments; or may exist separately without being assembled into the electronic device. The computer readable medium carries one or more programs which, when executed by the electronic device, cause the electronic device to: training an image segmentation model by adopting a first data set mixed with pure aggregate and pure mortar to obtain a basic model; segmenting the second data set containing the aggregate of the mortar by adopting a basic model to obtain a segmentation result, and obtaining a third data set based on the segmentation result; obtaining a fourth data set containing aggregate of mortar, disordering the first data set, the third data set and the fourth data set to obtain a fifth data set, and training a semantic segmentation model based on deplab by adopting the fifth data set to obtain a final model; acquiring images of different surfaces of the recycled aggregate to be detected, segmenting the images of the different surfaces by adopting a final model to obtain a segmentation result, and calculating the mortar content and the communication domain area ratio of the recycled aggregate to be detected according to the segmentation result; and obtaining the relation between the water absorption of the standard aggregate which has the same material as the recycled aggregate to be detected and the mortar content and the communication area ratio, and calculating the water absorption of the recycled aggregate to be detected according to the relation and the mortar content and the communication area ratio of the recycled aggregate to be detected.
The above description is only a preferred embodiment of the application and is illustrative of the principles of the technology employed. It will be appreciated by those skilled in the art that the scope of the invention herein disclosed is not limited to the particular combination of features described above, but also encompasses other arrangements formed by any combination of the above features or their equivalents without departing from the spirit of the invention. For example, the above features may be replaced with (but not limited to) features having similar functions disclosed in the present application.
Claims (10)
1. A method for detecting the content of recycled aggregate mortar based on deep learning is characterized by comprising the following steps:
s1, training the image segmentation model by adopting a first data set mixed with pure aggregate and pure mortar to obtain a basic model;
s2, segmenting a second data set containing aggregate of the mortar by adopting the basic model to obtain a first segmentation result, and obtaining a third data set based on the first segmentation result;
s3, obtaining a fourth data set containing aggregate of mortar, disordering the first data set, the third data set and the fourth data set to obtain a fifth data set, and training a semantic segmentation model based on deplab by adopting the fifth data set to obtain a final model;
s4, obtaining images of different surfaces of the recycled aggregate to be detected, segmenting the images of different surfaces by adopting the final model to obtain a second segmentation result, and calculating the mortar content and the communication domain area ratio of the recycled aggregate to be detected according to the second segmentation result;
and S5, obtaining the relation between the water absorption of the standard aggregate which has the same material as the recycled aggregate to be detected and the mortar content and the communication area ratio, and calculating the water absorption of the recycled aggregate to be detected according to the relation and the mortar content and the communication area ratio of the recycled aggregate to be detected.
2. The deep learning-based recycled aggregate mortar content detection method according to claim 1, wherein the image segmentation model comprises a semantic segmentation model and an instance segmentation model.
3. The method for detecting the content of the recycled aggregate mortar based on deep learning of claim 1, wherein the basic model is the semantic segmentation model based on depeplab.
4. The deep learning-based method for detecting the content of the recycled aggregate mortar according to claim 1 or 3, wherein the deep learning-based semantic segmentation model is deep lab V3+, in the encoding stage, Xception is firstly used for extracting low semantic information in the image and transmitting the low semantic information into the decoding stage and extracting feature maps for ASPP multi-scale convolution and transmitting the feature maps into the decoding stage, and in the decoding stage, two data transmitted in the encoding stage are fused to obtain a result and extract semantic information.
5. The method for detecting the content of the recycled aggregate mortar based on deep learning according to claim 1, wherein the first data set in step S1 is obtained according to the following steps:
acquiring a picture of pure aggregate or pure mortar, extracting a mask of the picture of the pure aggregate or the pure mortar by using graying, binarization and median filtering, determining the coordinate of the boundary of the mask by using an edge searching function in opencv, and storing the category, the coordinate information and the like of the picture of the pure aggregate or the pure mortar in a json format file; the obtaining of the third data set based on the segmentation result in step S2 specifically includes: correcting and adjusting the first segmentation result to extract a segmentation mask, and generating a third data set; the fourth data set in step S3 is obtained by manual annotation.
6. The method for detecting mortar content in recycled aggregate based on deep learning of claim 1, wherein the step S4 includes calculating the mortar content and the area ratio of the connected domain in the recycled aggregate to be detected according to the second segmentation result, and specifically includes: calculating the areas S1, S2, … … and Sn of the mortar and the areas T1, T2, … … and Tn of the recycled aggregate on n different surfaces according to the second division result, and respectively averaging the areas S and T to obtain the average area S of the mortar and the average area T of the recycled aggregate:
calculating the mortar content as follows: mu = S/T;
calculating the average value of the areas of the recycled aggregates larger than the threshold value alpha of the area of the connected domain to obtain the area of the connected domain
Alpha is set according to the area characteristics of the recycled aggregate;
calculating the area ratio of the connected domain as follows:
。
7. the method for detecting the content of the recycled aggregate mortar based on deep learning according to claim 1, wherein the step S5 specifically comprises:
carrying out water absorption test on the standard aggregate to obtain the water absorption of the standard aggregate, calculating by adopting the final model to obtain the mortar content and the communication domain area ratio of the standard aggregate, carrying out regression analysis by taking the water absorption of the standard aggregate as a dependent variable and the mortar content and the communication domain area ratio of the standard aggregate as independent variables to obtain the relation between the water absorption of the standard aggregate and the mortar content and the communication domain area ratio;
and substituting the mortar content and the communication area ratio of the recycled aggregate to be detected into the relation, and calculating to obtain the water absorption of the recycled aggregate to be detected.
8. The utility model provides a regeneration aggregate mortar content testing device based on deep learning which characterized in that includes:
the basic model training module is configured to train the image segmentation model by adopting a first data set mixed with pure aggregate and pure mortar to obtain a basic model;
the first segmentation module is configured to segment a second data set containing aggregate of mortar by using the basic model to obtain a first segmentation result, and obtain a third data set based on the first segmentation result;
the self-training module is configured to obtain a fourth data set containing aggregate of mortar, disorder the first data set, the third data set and the fourth data set to obtain a fifth data set, and train the semantic segmentation model based on deplab by adopting the fifth data set to obtain a final model;
the second segmentation module is configured to acquire images of different surfaces of the to-be-detected recycled aggregate, segment the images of the different surfaces by adopting the final model to obtain a second segmentation result, and calculate the mortar content and the communication area ratio of the to-be-detected recycled aggregate according to the second segmentation result;
and the water absorption calculation module is configured to acquire the relation between the water absorption of the standard aggregate which has the same material as the to-be-detected recycled aggregate and the mortar content and the communication area ratio, and calculate the water absorption of the to-be-detected recycled aggregate according to the relation and the mortar content and the communication area ratio of the to-be-detected recycled aggregate.
9. An electronic device, comprising:
one or more processors;
a storage device for storing one or more programs,
the one or more programs, when executed by the one or more processors, cause the one or more processors to implement the method recited in any of claims 1-7.
10. A computer-readable storage medium, on which a computer program is stored which, when being executed by a processor, carries out the method according to any one of claims 1-7.
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210664982.9A CN114757948B (en) | 2022-06-14 | 2022-06-14 | Deep learning-based method and device for detecting content of recycled aggregate mortar |
| PCT/CN2022/110831 WO2023240776A1 (en) | 2022-06-14 | 2022-08-08 | Deep learning-based muck classification processing guidance method and apparatus, and readable medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210664982.9A CN114757948B (en) | 2022-06-14 | 2022-06-14 | Deep learning-based method and device for detecting content of recycled aggregate mortar |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114757948A CN114757948A (en) | 2022-07-15 |
| CN114757948B true CN114757948B (en) | 2022-09-06 |
Family
ID=82336789
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210664982.9A Active CN114757948B (en) | 2022-06-14 | 2022-06-14 | Deep learning-based method and device for detecting content of recycled aggregate mortar |
Country Status (2)
| Country | Link |
|---|---|
| CN (1) | CN114757948B (en) |
| WO (1) | WO2023240776A1 (en) |
Families Citing this family (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116503845B (en) * | 2023-06-25 | 2023-09-12 | 四川省交通勘察设计研究院有限公司 | Method, system and medium for detecting content of false aggregate in recycled asphalt mixture |
| CN116678885B (en) * | 2023-08-03 | 2023-12-19 | 福建南方路面机械股份有限公司 | Deep learning-based detection control method and device for mud content of water-washed coarse aggregate |
| CN116689133B (en) * | 2023-08-04 | 2023-12-15 | 福建南方路面机械股份有限公司 | Deep learning-based recycled aggregate quality control method and device and readable medium |
Family Cites Families (14)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| IT201700058505A1 (en) * | 2017-05-30 | 2018-11-30 | Volta Robots S R L | Method of control of a soil processing vehicle based on image processing and related system |
| CN108288266A (en) * | 2018-01-12 | 2018-07-17 | 河南工业大学 | A kind of regenerative bone material by concrete corner loss late and mortar peel off the method for quantitatively evaluating of area |
| JP7455660B2 (en) * | 2019-07-19 | 2024-03-26 | 太平洋セメント株式会社 | How to diagnose or predict concrete deterioration |
| CN110750819A (en) * | 2019-09-19 | 2020-02-04 | 长沙理工大学 | Discrete element and finite difference method coupled asphalt mixture simulation modeling method |
| CN112906700B (en) * | 2021-01-15 | 2022-10-21 | 重庆交通大学 | Self-compacting concrete image semantic segmentation method and device and data set generation method |
| CN112801953A (en) * | 2021-01-15 | 2021-05-14 | 清华大学 | Freshly-mixed self-compacting concrete segregation resistance analysis method and device |
| CN113177909B (en) * | 2021-04-01 | 2023-06-20 | 华侨大学 | Multi-mode visual detection method and system for recycled aggregate with mortar on surface |
| CN113655002B (en) * | 2021-08-30 | 2023-11-24 | 华侨大学 | Recycled aggregate quality detection system of surface mortar based on hyperspectral technology |
| CN113744237A (en) * | 2021-08-31 | 2021-12-03 | 华中科技大学 | Deep learning-based automatic detection method and system for muck fluidity |
| CN114266989A (en) * | 2021-11-15 | 2022-04-01 | 北京建筑材料科学研究总院有限公司 | Concrete mixture workability determination method and device |
| CN114118266A (en) * | 2021-11-24 | 2022-03-01 | 华侨大学 | Visual detection classification method and system for recycled aggregate with mortar on surface |
| CN114219993A (en) * | 2021-12-15 | 2022-03-22 | 西安建筑科技大学 | CNN-based construction waste classification method |
| CN114565838A (en) * | 2022-01-26 | 2022-05-31 | 福建南方路面机械股份有限公司 | Formula adjustment control method and device based on muck image and readable medium |
| CN114758184B (en) * | 2022-06-14 | 2022-09-06 | 福建南方路面机械股份有限公司 | Deep learning-based muck classification processing guide method and device and readable medium |
-
2022
- 2022-06-14 CN CN202210664982.9A patent/CN114757948B/en active Active
- 2022-08-08 WO PCT/CN2022/110831 patent/WO2023240776A1/en unknown
Also Published As
| Publication number | Publication date |
|---|---|
| WO2023240776A1 (en) | 2023-12-21 |
| CN114757948A (en) | 2022-07-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114757948B (en) | Deep learning-based method and device for detecting content of recycled aggregate mortar | |
| KR102002024B1 (en) | Method for processing labeling of object and object management server | |
| CN110363220B (en) | Behavior class detection method and device, electronic equipment and computer readable medium | |
| CN109214501B (en) | Method and apparatus for identifying information | |
| CN109285181B (en) | Method and apparatus for recognizing image | |
| CN111311480B (en) | Image fusion method and device | |
| KR101602591B1 (en) | Methods and apparatuses for facilitating detection of text within an image | |
| CN114359590A (en) | NFT image work infringement detection method and device and computer storage medium | |
| CN113408507B (en) | Named entity identification method and device based on resume file and electronic equipment | |
| CN111967449B (en) | Text detection method, electronic device and computer readable medium | |
| CN111914850B (en) | Picture feature extraction method, device, server and medium | |
| CN111259676A (en) | Translation model training method and device, electronic equipment and storage medium | |
| CN116092101A (en) | Training method, image recognition method apparatus, device, and readable storage medium | |
| CN115546554A (en) | Sensitive image identification method, device, equipment and computer readable storage medium | |
| CN115761698A (en) | Target detection method, device, equipment and storage medium | |
| CN111738454B (en) | Target detection method, device, storage medium and equipment | |
| CN114445751A (en) | Method and device for extracting video key frame image contour features | |
| CN117333487B (en) | Acne classification method, device, equipment and storage medium | |
| CN113657230B (en) | Method for training news video recognition model, method for detecting video and device thereof | |
| CN110826421B (en) | Method and device for filtering faces with difficult gestures | |
| CN115861255A (en) | Model training method, device, equipment, medium and product for image processing | |
| CN111914863A (en) | Target detection method and device, terminal equipment and computer readable storage medium | |
| CN115035528A (en) | Method, device, electronic equipment and medium for detecting interested fields of value-added tax invoice | |
| Zhang et al. | Improved U-net network asphalt pavement crack detection method | |
| CN116309655A (en) | Display screen edge detection method, device and equipment based on deep learning |
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 | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |