CN113655002B - Recycled aggregate quality detection system of surface mortar based on hyperspectral technology - Google Patents
Recycled aggregate quality detection system of surface mortar based on hyperspectral technology Download PDFInfo
- Publication number
- CN113655002B CN113655002B CN202111005886.5A CN202111005886A CN113655002B CN 113655002 B CN113655002 B CN 113655002B CN 202111005886 A CN202111005886 A CN 202111005886A CN 113655002 B CN113655002 B CN 113655002B
- Authority
- CN
- China
- Prior art keywords
- module
- recycled aggregate
- hyperspectral
- model
- area
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 239000004570 mortar (masonry) Substances 0.000 title claims abstract description 37
- 238000001514 detection method Methods 0.000 title claims abstract description 33
- 238000005516 engineering process Methods 0.000 title claims abstract description 11
- 238000000611 regression analysis Methods 0.000 claims abstract description 23
- 238000012549 training Methods 0.000 claims abstract description 16
- 238000013135 deep learning Methods 0.000 claims abstract description 14
- 238000012545 processing Methods 0.000 claims abstract description 14
- 238000005286 illumination Methods 0.000 claims abstract description 10
- 238000010521 absorption reaction Methods 0.000 claims abstract description 8
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 claims abstract description 8
- 238000010987 Kennard-Stone algorithm Methods 0.000 claims abstract description 6
- 230000003595 spectral effect Effects 0.000 claims abstract description 4
- 238000007781 pre-processing Methods 0.000 claims description 19
- 238000000605 extraction Methods 0.000 claims description 18
- 229910052736 halogen Inorganic materials 0.000 claims description 7
- 150000002367 halogens Chemical class 0.000 claims description 6
- 238000000034 method Methods 0.000 claims description 6
- 230000009467 reduction Effects 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 4
- 239000006185 dispersion Substances 0.000 claims description 3
- 239000002245 particle Substances 0.000 claims description 3
- 230000008569 process Effects 0.000 claims description 2
- 230000004069 differentiation Effects 0.000 description 6
- 230000000694 effects Effects 0.000 description 5
- 230000009466 transformation Effects 0.000 description 3
- 239000002699 waste material Substances 0.000 description 3
- 238000010276 construction Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000001228 spectrum Methods 0.000 description 2
- -1 50W halogen Chemical class 0.000 description 1
- 230000002146 bilateral effect Effects 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000003912 environmental pollution Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 238000004064 recycling Methods 0.000 description 1
- 238000003860 storage Methods 0.000 description 1
- 238000005728 strengthening Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N21/00—Investigating or analysing materials by the use of optical means, i.e. using sub-millimetre waves, infrared, visible or ultraviolet light
- G01N21/17—Systems in which incident light is modified in accordance with the properties of the material investigated
- G01N21/25—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands
- G01N21/27—Colour; Spectral properties, i.e. comparison of effect of material on the light at two or more different wavelengths or wavelength bands using photo-electric detection ; circuits for computing concentration
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- 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
- G06T7/13—Edge detection
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2201/00—Features of devices classified in G01N21/00
- G01N2201/06—Illumination; Optics
- G01N2201/061—Sources
- G01N2201/06146—Multisources for homogeneisation, as well sequential as simultaneous operation
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2201/00—Features of devices classified in G01N21/00
- G01N2201/12—Circuits of general importance; Signal processing
- G01N2201/129—Using chemometrical methods
-
- G—PHYSICS
- G01—MEASURING; TESTING
- G01N—INVESTIGATING OR ANALYSING MATERIALS BY DETERMINING THEIR CHEMICAL OR PHYSICAL PROPERTIES
- G01N2201/00—Features of devices classified in G01N21/00
- G01N2201/12—Circuits of general importance; Signal processing
- G01N2201/129—Using chemometrical methods
- G01N2201/1293—Using chemometrical methods resolving multicomponent spectra
-
- 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)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Evolutionary Computation (AREA)
- Health & Medical Sciences (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Biochemistry (AREA)
- General Health & Medical Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention provides a recycled aggregate quality detection system of surface mortar based on hyperspectral technology, wherein a recycled aggregate conveying module is used for conveying recycled aggregate to an acquisition area of a hyperspectral image acquisition module; the light source module provides illumination for an acquisition area of the hyperspectral image acquisition module; the hyperspectral image acquisition module is used for acquiring the original data of the spectral reflection intensity of the recycled aggregate, converting the original data into a pseudo-color image and outputting the pseudo-color image to the deep learning module; the deep learning module picks samples by using a KS algorithm, uses the extracted features for training a model, and uses the trained model for online detection; the image processing module provides parameter input for the regression analysis module; the regression analysis module builds a regression model related to the water absorption and apparent density of the recycled aggregate.
Description
Technical Field
The invention relates to the field of quality grade detection of recycled aggregate surface mortar, in particular to a recycled aggregate quality detection system of surface mortar based on hyperspectral technology.
Background
Along with continuous investment of China to the capital construction, the rapidly increased aggregate consumption causes serious shortage of the virgin aggregate in some places, on the other hand, a large amount of building rubbish can be generated in China every year, the shortage of the storage land causes environmental pollution, and the waste concrete is the main component of the building rubbish. Mortar is inevitably remained on the surface of recycled aggregate in the surface strengthening and crushing process of waste concrete, and the existence and the content of the mortar have serious influence on the quality of the recycled aggregate, so that the quality grade of the recycled aggregate is detected, and whether the quality of the recycled aggregate meets the recycling requirement is judged, so that the recycled aggregate is used for producing concrete. The development of a set of high-efficiency and high-quality recycled aggregate quality grade detection system has great significance for accelerating the utilization rate and the utilization degree of recycled aggregates in China.
Disclosure of Invention
The invention aims to solve the main technical problems of providing a recycled aggregate quality detection system of surface mortar based on hyperspectral technology, which can realize the judgment of the quality grade of recycled aggregate, thereby controlling the production process of the recycled aggregate and solving the problems of construction waste treatment and raw aggregate shortage.
In order to solve the technical problems, the invention provides a recycled aggregate quality detection system of surface mortar based on hyperspectral technology, comprising: the system comprises a recycled aggregate conveying module, a light source module, a hyperspectral image acquisition module, a deep learning module, an image processing module and a regression analysis module;
the recycled aggregate conveying module is used for conveying the recycled aggregate to the collecting area of the hyperspectral image collecting module;
the light source module provides illumination for an acquisition area of the hyperspectral image acquisition module;
the hyperspectral image acquisition module is used for acquiring original data of the spectral reflection intensity of the recycled aggregate and converting the original data into a pseudo-color image to be output;
the deep learning module selects samples by using a KS algorithm, further differentiates hyperspectral data characteristics of the samples by preprocessing data, performs dimension reduction processing on the preprocessed sample data characteristics, and finally uses the extracted characteristics for training a model and uses the trained model for online detection;
the different objects detected by the trained model on line are output in the form of pictures, the image processing module carries out statistics on the number of pixel points of the different objects on the output classified pictures, then the pixel duty ratio of mortar is counted, the area size and the convex hull ratio of the recycled aggregate are calculated, and the input of parameters is provided for the regression analysis module;
the regression analysis module builds a regression model related to the water absorption and apparent density of the recycled aggregate.
In a preferred embodiment: the recycled aggregate conveying module comprises a vibration dispersing feeding device, an encoder and a conveyor belt; the vibration dispersion feeding device is used for providing stable and dispersed recycled aggregate for the conveyor belt; the encoder is used for reading the speed of the current conveyor belt; the conveyor belt device conveys the recycled aggregate dispersed by the vibration dispersing device to a collecting area of the hyperspectral image collecting module.
In a preferred embodiment: the light source module includes two halogen lamps.
In a preferred embodiment: the hyperspectral image acquisition module is a hyperspectral camera.
In a preferred embodiment: the deep learning module comprises a light sample selection module, a data preprocessing module, a characteristic extraction module, a model training module and an online detection module;
the sample selecting module screens the collected hyperspectral data to select samples; the data preprocessing module further expands the selected sample characteristics, so that the characteristic differences of the samples are more obvious; the feature extraction module performs data dimension reduction on the hyperspectral data, and removes features with larger relativity; the model training module trains the extracted features so as to obtain a model with stronger generalization capability and obtain model parameters for online detection; the on-line detection module is used for on-line classification of the recycled aggregate by using the trained model parameters, and outputting the classification result by using pictures.
In a preferred embodiment: the image processing module comprises image preprocessing and characteristic parameter extraction;
the image preprocessing uses Gaussian filtering to remove noise points of the classified images and remove smaller particles in the images; and the feature extraction is used for counting mortar and raw aggregate pixels in the classified images, calculating the duty ratio of the mortar pixels and calculating the area and convex hull ratio of the recycled aggregate.
In a preferred embodiment: the regression analysis module comprises parameter input and regression analysis model acquisition;
the parameter input is used for inputting the area ratio of mortar, the area of recycled aggregate and the convex hull ratio of the recycled aggregate into a regression model; the regression analysis model carries out multiple regression on the area ratio of mortar, the area of recycled aggregate, the convex hull ratio of recycled aggregate and the like, the water absorption rate and the apparent density, so as to obtain regression parameters, and a regression equation is constructed for quality grade detection of the recycled aggregate.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions of the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, and it is obvious that the drawings in the following description are some embodiments of the present invention, and other drawings may be obtained according to the drawings without inventive effort to a person skilled in the art.
FIG. 1 is a schematic diagram of a device of a recycled aggregate quality detection system of surface mortar based on hyperspectral technology.
Fig. 2 is an overall flow chart of the recycled aggregate quality detection system of the surface mortar based on the hyperspectral technology.
Fig. 3 is a detailed flowchart of the recycled aggregate quality detection system of the surface mortar based on the hyperspectral technology.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention; it is apparent that the described embodiments are only some embodiments of the present invention, not all embodiments, and that all other embodiments obtained by persons of ordinary skill in the art without making creative efforts based on the embodiments in the present invention are within the protection scope of the present invention.
In the description of the present invention, it should be noted that the positional or positional relationship indicated by the terms such as "upper", "lower", "inner", "outer", "top/bottom", etc. are based on the positional or positional relationship shown in the drawings, are merely for convenience of describing the present invention and simplifying the description, and do not indicate or imply that the apparatus or elements referred to must have a specific orientation, be constructed and operated in a specific orientation, and thus should not be construed as limiting the present invention. Furthermore, the terms "first," "second," and the like, are used for descriptive purposes only and are not to be construed as indicating or implying relative importance.
In the description of the present invention, unless explicitly specified and limited otherwise, the terms "mounted," configured to, "" engaged with, "" connected to, "and the like are to be construed broadly, and may be, for example," connected to, "wall-mounted," connected to, removably connected to, or integrally connected to, mechanically connected to, electrically connected to, directly connected to, or indirectly connected to, through an intermediary, and may be in communication with each other between two elements, as will be apparent to those of ordinary skill in the art, in view of the detailed description of the terms herein.
Referring to fig. 1-3, the present embodiment provides a recycled aggregate quality detection system for surface mortar based on hyperspectral technology, comprising: the system comprises a recycled aggregate conveying module 10, a light source module 20, a hyperspectral image acquisition module 30, a deep learning module 40, an image processing module 50 and a regression analysis module 60;
the recycled aggregate conveying module 10 is used for conveying the recycled aggregate to the collecting area of the hyperspectral image collecting module 30;
the light source module 20 provides illumination for the acquisition region of the hyperspectral image acquisition module 30;
the hyperspectral image acquisition module 30 is used for acquiring the original data of the spectral reflection intensity of the recycled aggregate, and converting the original data into a pseudo-color image to be output;
the deep learning module 40 selects samples by using a KS algorithm, further differentiates hyperspectral data features of the samples by preprocessing data, performs dimension reduction processing on the preprocessed sample data features, and finally uses the extracted features for training a model and the trained model for online detection;
the different objects detected by the trained model on line are output in the form of pictures, the image processing module 50 carries out statistics on the number of pixel points of the different objects on the output classified pictures, then the pixel duty ratio of mortar is counted, the area size and the convex hull ratio of the recycled aggregate are calculated, and the input of parameters is provided for the regression analysis module 60;
the regression analysis module 60 builds a regression model that relates the water absorption and apparent density of the recycled aggregate.
The working flow of the recycled aggregate conveying module is shown in fig. 1, the speed of the conveyor belt is controlled by the controller to convey the recycled aggregate placed on the conveyor belt to the data acquisition area of the hyperspectral image acquisition module, and the encoder reads the current speed of the conveyor belt.
The working flow of the light source module is shown in fig. 1, and two 50W halogen lamps are utilized, wherein the halogen lamps are symmetrically distributed on two sides of the hyperspectral camera in a left-right mode, a stable and uniform ideal illumination environment is provided for a measurement area of the hyperspectral camera, and the influence of the light source on the collected spectrum data is reduced as much as possible.
In this embodiment, the recycled aggregate conveying module 10 includes a vibration dispersing feeder 11, an encoder 12, and a conveyor belt 13; the vibration dispersion feeding device 11 is used for providing stable and dispersed recycled aggregate for the conveyor belt 13; the encoder 12 is used for reading the current speed of the conveyor belt 13; the conveyor belt device 13 conveys the recycled aggregate dispersed by the vibration dispersing device to the collection area of the hyperspectral image collection module 30.
The light source module 20 includes two halogen lamps 21.
The hyperspectral image capturing module 30 is a hyperspectral camera 31.
The deep learning module 40 comprises a light sample selecting module 41, a data preprocessing module 42, a feature extraction module 43, a model training module 44 and an online detection module 45;
the sample selecting module 41 screens the collected hyperspectral data to select samples; the data preprocessing module 42 further expands the selected sample characteristics, so that the characteristic differences of the samples are more obvious; the feature extraction module 43 performs data dimension reduction on the hyperspectral data, and removes features with larger relativity; the model training module 44 trains the extracted features so as to obtain a model with stronger generalization capability and obtain model parameters for online detection; the online detection module 45 uses the trained model parameters for online classification of the recycled aggregate, and outputs the classification result by using pictures.
The image processing module 50 comprises image preprocessing 51 and characteristic parameter extraction 52;
the image preprocessing 51 uses Gaussian filtering to remove noise points of the classified images and remove smaller particles in the images; the feature extraction 52 counts mortar and raw aggregate pixels in the classified images, calculates the duty ratio of the mortar pixels, and calculates the area and convex hull ratio of the recycled aggregate. The feature extraction 52 finds the edge contour of the recycled aggregate after image binarization, then respectively counts the number of pixels of the mortar and the raw aggregate in the contour in the original image, and the area occupation ratio of the mortar can be obtained by comparing the total number of pixels of the mortar with the number of pixels of the whole recycled aggregate. In addition, the area, convex hull ratio and the like of the recycled aggregate are calculated.
The regression analysis module 60 comprises a parameter input 61 and a regression analysis model 62;
the parameter input 61 inputs the area ratio of mortar, the area of recycled aggregate and the convex hull ratio of the recycled aggregate into a regression model; the regression analysis model 62 carries out multiple regression on the area ratio of mortar, the area of recycled aggregate, the convex hull ratio of recycled aggregate and the like, the water absorption rate and the apparent density, so as to obtain regression parameters, and construct a regression equation for quality grade detection of the recycled aggregate.
The installation mode and the working flow chart of the hyperspectral image acquisition module are shown in figures 1 and 3, the hyperspectral camera acquires the illumination intensity reflected on the surface of the recycled aggregate through the line scanning, so that hyperspectral data are acquired, meanwhile, the conveyor belt drives the recycled aggregate to move, the illumination intensity of the surface of the whole recycled aggregate is acquired, and the corresponding relation between the stereoscopic vision systems is calculated, so that the pseudo-color image of the recycled aggregate is reconstructed.
The deep learning module workflow is shown in fig. 3, and the KS algorithm is used to screen the collected hyperspectral data to pick out representative samples. The data preprocessing is to compare the differentiation degree of the selected hyperspectral data characteristics by the methods of first-order differentiation, logarithmic transformation of derivative, envelope removal and the like, and select the preprocessing method with the best hyperspectral data characteristic differentiation effect. The feature extraction module performs feature extraction on the preprocessed hyperspectral data by using principal component analysis PCA and wavelet transformation WT so as to remove redundant data, compares the obtained features with training effects of the model, and selects a feature extraction method with the best model training effect. Model training the best training model is selected for online detection by comparing training model effects of an extreme learning machine, a Gaussian kernel extreme learning machine and a wavelet kernel extreme learning machine. The obtained model with the best generalization capability is used for online detection and classification of the recycled aggregate.
The workflow of the image processing module is shown in fig. 3, and the classified images are subjected to Gaussian filtering to remove noise and small points in the images. The extraction of the characteristic parameters is carried out by finding the edge contour of the recycled aggregate after the image binarization, then respectively counting the number of pixels of the mortar and the virgin aggregate in the contour in the original image, and comparing the total number of pixels of the mortar with the number of pixels of the whole recycled aggregate to obtain the area occupation ratio of the mortar. In addition, the area, convex hull ratio and the like of the recycled aggregate are calculated.
The work flow chart of the regression analysis module is shown in fig. 3, and the parameter input comprises the mortar area ratio in the recycled aggregate, the area of the recycled aggregate, the convex hull ratio of the recycled aggregate and the like. The regression analysis model carries out multiple regression on the mortar area ratio in the recycled aggregate, the area of the recycled aggregate, the convex hull ratio of the recycled aggregate and the like and the water absorption rate and apparent density of the recycled aggregate, so as to obtain a multiple regression equation, thereby realizing the online detection of the quality grade of the recycled aggregate.
The aggregate conveying module comprises a vibration dispersing device, a conveyor belt device and an encoder device. The controller device controls the conveying speed of the conveyor belt, the conveyor belt device sequentially sends the regenerated aggregate dispersed by the vibration dispersing device to each image acquisition area, and the encoder device reads the speed of the current transfer belt.
The light source module utilizes the halogen lamps with the power of 50W, the halogen lamps are distributed on the two sides of the hyperspectral camera in a bilateral symmetry mode, a stable and uniform ideal illumination environment is provided for a measurement area of the hyperspectral camera, and the influence of the light source on the collected spectrum data is reduced as much as possible.
The hyperspectral camera in the hyperspectral image acquisition module acquires the illumination intensity reflected on the surface of the recycled aggregate through the line scanning, so that hyperspectral data are acquired, meanwhile, the conveyor belt drives the recycled aggregate to move, so that the surface illumination intensity of the whole recycled aggregate is acquired, and the corresponding relation between the stereoscopic vision systems is calculated, so that the pseudo-color image of the recycled aggregate is reconstructed.
Based on the technical scheme, the deep learning module further comprises sample selection, data preprocessing, feature extraction, model training and online detection. The sample selection uses KS algorithm to screen the collected hyperspectral data, and representative samples are selected. The data preprocessing is to compare the degree of differentiation of the selected hyperspectral data characteristics by the methods of first-order differentiation, logarithmic transformation of derivatives, envelope removal and the like, and select the preprocessing method with the best hyperspectral data characteristic differentiation effect.
The foregoing is only a preferred embodiment of the present invention, but the design concept of the present invention is not limited thereto, and any person skilled in the art will be able to make insubstantial modifications of the present invention within the scope of the present invention disclosed herein by this concept, which falls within the actions of invading the protection scope of the present invention.
Claims (1)
1. The recycled aggregate quality detection system of the surface mortar based on hyperspectral technology is characterized by comprising the following components: the system comprises a recycled aggregate conveying module (10), a light source module (20), a hyperspectral image acquisition module (30), a deep learning module (40), an image processing module (50) and a regression analysis module (60);
the recycled aggregate conveying module (10) is used for conveying the recycled aggregate to the collecting area of the hyperspectral image collecting module (30);
the light source module (20) provides illumination for an acquisition area of the hyperspectral image acquisition module (30);
the hyperspectral image acquisition module (30) is used for acquiring original data of the spectral reflection intensity of the recycled aggregate, converting the original data into a pseudo-color image and outputting the pseudo-color image to the deep learning module (40);
the deep learning module (40) selects samples by using a KS algorithm, pre-processes the data to further differentiate hyperspectral data characteristics of the samples, then performs dimension reduction processing on the pre-processed sample data characteristics, finally uses the extracted characteristics for training a model, and uses the trained model for online detection;
the different objects detected by the trained model on line are output in the form of pictures, an image processing module (50) carries out statistics on the number of pixel points of the different objects on the output classified pictures, then the pixel duty ratio of mortar is counted, the area size and the convex hull ratio of the recycled aggregate are calculated, and the input of parameters is provided for a regression analysis module (60);
the regression analysis module (60) constructs a regression model related to the water absorption and apparent density of the recycled aggregate;
the recycled aggregate conveying module (10) comprises a vibration dispersing feeding device (11), an encoder (12) and a conveyor belt (13); the vibration dispersion feeding device (11) is used for providing stable and dispersed recycled aggregate for the conveyor belt (13); -said encoder (12) is adapted to read the speed of the current conveyor belt (13); the conveyor belt (13) conveys the recycled aggregate dispersed by the vibration dispersing device to a collecting area of the hyperspectral image collecting module (30);
the light source module (20) comprises two halogen lamps (21);
the hyperspectral image acquisition module (30) is a hyperspectral camera (31);
the deep learning module (40) comprises a light sample selection module (41), a data preprocessing module (42), a feature extraction module (43), a model training module (44) and an online detection module (45);
the sample selection module (41) screens the collected hyperspectral data to select samples; the data preprocessing module (42) further expands the selected sample characteristics so that the characteristic differences of the samples are more obvious; the characteristic extraction module (43) performs data dimension reduction on the hyperspectral data, and removes the characteristic with larger relativity; the model training module (44) trains the extracted features so as to obtain a model with stronger generalization capability and obtain model parameters for online detection; the online detection module (45) uses the trained model parameters for online classification of the recycled aggregate, and outputs the classified result by using pictures;
the image processing module (50) comprises image preprocessing (51) and characteristic parameter extraction (52);
the image preprocessing (51) uses Gaussian filtering to remove noise points of the classified images and remove smaller particles in the images; the characteristic parameter extraction (52) is used for counting mortar and raw aggregate pixels in the classified images, calculating the duty ratio of the mortar pixels and calculating the area and convex hull ratio of the recycled aggregate;
the regression analysis module (60) comprises parameter input (61) and a regression analysis model (62);
the parameter input (61) inputs the area ratio of mortar, the area of recycled aggregate and the convex hull ratio of the recycled aggregate into the acquired regression analysis model (62); and the regression analysis model (62) carries out multiple regression on the area occupation ratio of mortar, the area of the recycled aggregate, the convex hull ratio of the recycled aggregate, the water absorption and the apparent density, so as to obtain regression parameters, and a regression equation is constructed for quality grade detection of the recycled aggregate.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111005886.5A CN113655002B (en) | 2021-08-30 | 2021-08-30 | Recycled aggregate quality detection system of surface mortar based on hyperspectral technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111005886.5A CN113655002B (en) | 2021-08-30 | 2021-08-30 | Recycled aggregate quality detection system of surface mortar based on hyperspectral technology |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN113655002A CN113655002A (en) | 2021-11-16 |
| CN113655002B true CN113655002B (en) | 2023-11-24 |
Family
ID=78493250
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111005886.5A Active CN113655002B (en) | 2021-08-30 | 2021-08-30 | Recycled aggregate quality detection system of surface mortar based on hyperspectral technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113655002B (en) |
Families Citing this family (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114118266B (en) * | 2021-11-24 | 2024-08-20 | 华侨大学 | Visual detection classification method and system for recycled aggregate with mortar on surface |
| CN114166845A (en) * | 2021-11-26 | 2022-03-11 | 浙江交投矿业有限公司 | Automatic get material on-line measuring building stones quality integration and equip |
| CN114757948B (en) * | 2022-06-14 | 2022-09-06 | 福建南方路面机械股份有限公司 | Deep learning-based method and device for detecting content of recycled aggregate mortar |
| 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 |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2952551A1 (en) * | 2016-11-30 | 2018-05-30 | National Research Council Of Canada | Method and system for characterizing an aggregate sample by using laser-induced breakdown spectroscopy |
| CN109145992A (en) * | 2018-08-27 | 2019-01-04 | 西安电子科技大学 | Cooperation generates confrontation network and sky composes united hyperspectral image classification method |
| CN110282926A (en) * | 2019-06-26 | 2019-09-27 | 内蒙古巨华集团大华建筑安装有限公司 | A kind of building waste materials recycling concrete and preparation method thereof |
| CN110458202A (en) * | 2019-07-18 | 2019-11-15 | 东莞弓叶互联科技有限公司 | Image processing method, device, equipment and computer readable storage medium |
| CN111062078A (en) * | 2019-12-18 | 2020-04-24 | 南京航空航天大学 | Quality grading method suitable for mixed type recycled coarse aggregate |
| CN111368712A (en) * | 2020-03-02 | 2020-07-03 | 四川九洲电器集团有限责任公司 | Hyperspectral image disguised target detection method based on deep learning |
| CN111751253A (en) * | 2020-07-06 | 2020-10-09 | 重庆理工大学 | Forming method and quality detection method of concrete aggregate detection model |
| CN111754463A (en) * | 2020-06-02 | 2020-10-09 | 石家庄铁道大学 | Ballastless track CA mortar layer defect detection method based on convolutional neural network |
| WO2021082428A1 (en) * | 2019-10-29 | 2021-05-06 | 平安科技（深圳）有限公司 | Semi-supervised learning-based image classification method and apparatus, and computer device |
| CN113177909A (en) * | 2021-04-01 | 2021-07-27 | 华侨大学 | Multi-mode visual detection method and system for recycled aggregate with mortar on surface |
Family Cites Families (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10806334B2 (en) * | 2017-02-28 | 2020-10-20 | Verily Life Sciences Llc | System and method for multiclass classification of images using a programmable light source |
| US10902581B2 (en) * | 2017-06-19 | 2021-01-26 | Apeel Technology, Inc. | System and method for hyperspectral image processing to identify foreign object |
-
2021
- 2021-08-30 CN CN202111005886.5A patent/CN113655002B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CA2952551A1 (en) * | 2016-11-30 | 2018-05-30 | National Research Council Of Canada | Method and system for characterizing an aggregate sample by using laser-induced breakdown spectroscopy |
| CN109145992A (en) * | 2018-08-27 | 2019-01-04 | 西安电子科技大学 | Cooperation generates confrontation network and sky composes united hyperspectral image classification method |
| CN110282926A (en) * | 2019-06-26 | 2019-09-27 | 内蒙古巨华集团大华建筑安装有限公司 | A kind of building waste materials recycling concrete and preparation method thereof |
| CN110458202A (en) * | 2019-07-18 | 2019-11-15 | 东莞弓叶互联科技有限公司 | Image processing method, device, equipment and computer readable storage medium |
| WO2021082428A1 (en) * | 2019-10-29 | 2021-05-06 | 平安科技（深圳）有限公司 | Semi-supervised learning-based image classification method and apparatus, and computer device |
| CN111062078A (en) * | 2019-12-18 | 2020-04-24 | 南京航空航天大学 | Quality grading method suitable for mixed type recycled coarse aggregate |
| CN111368712A (en) * | 2020-03-02 | 2020-07-03 | 四川九洲电器集团有限责任公司 | Hyperspectral image disguised target detection method based on deep learning |
| CN111754463A (en) * | 2020-06-02 | 2020-10-09 | 石家庄铁道大学 | Ballastless track CA mortar layer defect detection method based on convolutional neural network |
| CN111751253A (en) * | 2020-07-06 | 2020-10-09 | 重庆理工大学 | Forming method and quality detection method of concrete aggregate detection model |
| CN113177909A (en) * | 2021-04-01 | 2021-07-27 | 华侨大学 | Multi-mode visual detection method and system for recycled aggregate with mortar on surface |
Non-Patent Citations (2)
| Title |
|---|
| 基于深度学习的高光谱图像目标检测及场景分类;张顺;中国优秀硕士学位论文全文数据库;第工程科技Ⅱ辑卷(第2021 年 第02期期);C028-296 * |
| 基于高光谱成像和深度学习的基于高光谱成像和深度学习的可回收垃圾分类方法研究;吴瑞;中国优秀硕士学位论文全文数据库;第工程科技Ⅰ辑卷(第2020 年 第09期期);B027-169 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN113655002A (en) | 2021-11-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113655002B (en) | Recycled aggregate quality detection system of surface mortar based on hyperspectral technology | |
| CN113177909B (en) | Multi-mode visual detection method and system for recycled aggregate with mortar on surface | |
| CN105488816B (en) | A kind of mineral floating foam flow velocity on-line measuring device and method based on 3D vision information | |
| CN109598715B (en) | Material granularity online detection method based on machine vision | |
| CN111862028B (en) | Wood defect detecting and sorting device and method based on depth camera and depth learning | |
| CN101685006A (en) | Automatic detection system for vision of contact part feature machine | |
| CN103163141A (en) | Strip steel surface online detection system and method based on embedded image processing system | |
| CN107486415A (en) | Thin bamboo strip defect on-line detecting system and detection method based on machine vision | |
| CN101059425A (en) | Method and device for identifying different variety green tea based on multiple spectrum image texture analysis | |
| CN105388162A (en) | Raw material silicon wafer surface scratch detection method based on machine vision | |
| CN101949686A (en) | Online nondestructive testing (NDT) method and device for comprehensive internal/external qualities of fruits | |
| CN103913468A (en) | Multi-vision defect detecting equipment and method for large-size LCD glass substrate in production line | |
| CN1335496A (en) | Image inspecting method and device for detecting faults | |
| CN110220908A (en) | A kind of on-line detecting system and its online test method of floatation clean coal ash content | |
| CN104048966B (en) | The detection of a kind of fabric defect based on big law and sorting technique | |
| CN212442146U (en) | Household garbage recognition system | |
| CN103752533A (en) | Method and device for sorting coal and gangue online through image method | |
| CN108120715B (en) | Corn kernel breaking rate online detection method and device based on machine vision | |
| CN113362276B (en) | Visual detection method and system for plates | |
| CN103308525A (en) | Online detection method and device for metal wire production | |
| YANG et al. | Bubble size estimation using interfacial morphological information for mineral flotation process monitoring | |
| CN106248680A (en) | A kind of engine commutator quality detecting system based on machine vision and detection method | |
| CN102855641A (en) | Fruit level classification system based on external quality | |
| CN102980892A (en) | On-line examination system and method for steel pipe | |
| CN106093051A (en) | Paper roll tangent plane burr detection method based on machine vision and device |
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 |