CN110243768A - A kind of detection method of clay compactness - Google Patents
A kind of detection method of clay compactness Download PDFInfo
- Publication number
- CN110243768A CN110243768A CN201910512891.1A CN201910512891A CN110243768A CN 110243768 A CN110243768 A CN 110243768A CN 201910512891 A CN201910512891 A CN 201910512891A CN 110243768 A CN110243768 A CN 110243768A
- Authority
- CN
- China
- Prior art keywords
- clay
- compactness
- textural characteristics
- camera
- clay sample
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- E—FIXED CONSTRUCTIONS
- E02—HYDRAULIC ENGINEERING; FOUNDATIONS; SOIL SHIFTING
- E02D—FOUNDATIONS; EXCAVATIONS; EMBANKMENTS; UNDERGROUND OR UNDERWATER STRUCTURES
- E02D1/00—Investigation of foundation soil in situ
-
- 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
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/044—Recurrent networks, e.g. Hopfield networks
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Mathematical Physics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Health & Medical Sciences (AREA)
- Biochemistry (AREA)
- Biophysics (AREA)
- Spectroscopy & Molecular Physics (AREA)
- Civil Engineering (AREA)
- Paleontology (AREA)
- Mining & Mineral Resources (AREA)
- General Life Sciences & Earth Sciences (AREA)
- Immunology (AREA)
- Pathology (AREA)
- Artificial Intelligence (AREA)
- Biomedical Technology (AREA)
- Structural Engineering (AREA)
- Computational Linguistics (AREA)
- Data Mining & Analysis (AREA)
- Soil Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Investigating Or Analysing Materials By Optical Means (AREA)
Abstract
The invention discloses a kind of detection methods of clay compactness comprising following steps: step A: passing through the high spectrum image of camera and spectrometer collection clay sample;Step B: obtaining the textural characteristics of the high spectrum image by the Matlab software of computer, and the textural characteristics include: contrast, average brightness, consistency, uniformity, energy, the degree of correlation, third moment value, average contrast, smoothness and entropy;Step C: prediction model is established by BP neural network using the textural characteristics of the clay sample with different compactnesss;Step D: the textural characteristics for obtaining clay sample to be measured input the prediction model to obtain the compactness of clay sample to be measured.The present invention has the advantages that reducing cost of human and material resources, detection efficiency is high, and testing result is reliable, and the relationship of compactness and high spectrum image textural characteristics is used multiple times.
Description
Technical field
The present invention relates to a kind of detection methods of clay compactness.
Background technique
The compactness of roadbed is to control an important indicator of roadbed quality, it directly affects the intensity and stabilization of roadbed
Property, the service performance and service life of bedding of roadbed are influenced, but how to realize accurate, the quick inspection of Subgrade Compaction Quality
It surveys, is but always puzzlement one of unit in charge of construction and the main problem of quality testing department, the main reason is that being current normal
More or less there are certain places not fully up to expectations in rule detection method.
Conventional detection method has sand replacement method, core cutter method, douche and nucleon density hygronom method;Sand replacement method measures roadbed
Compactness is primarily adapted for use in fine grained soil and coarse-grained soil, the disadvantages of the method are as follows needing to carry more amount sand, weighing number is more,
The moisture determination time is long, and measuring speed is slow;Core cutter method measures Subgrade Compaction and is suitable for fine grained soil, is not suitable for coarse-grained soil,
The disadvantage is that destroying structure sheaf, human factor influences greatly, and the moisture determination time is long;Douche is suitable for coarse-grained soil, over coarse grained soil, original
Shape sand, calculous soil and situations such as rockfill, but douche will obtain enough samples very not in the case of packing material size is larger
It is easy, usually surveys a density value and sampling is needed to exceed thousand kilograms, spend a large amount of manpower and material resources, detect heavy workload, time-consuming, from
And limit the practical application of this method;Nucleon density hygronom measures Subgrade Compaction and is suitable for coarse-grained soil and fine grained soil, tool
Have the advantages that using easy, detection speed is fast, but the radioactive substance that this method uses is harmful to the human body, and accuracy
It is lower;Therefore, it is badly in need of studying a kind of detection that speed is fast, accuracy is high, the clay detection methods of compaction degree of safety non-pollution.
Summary of the invention
It is an object of the invention to overcome the deficiency of the prior art, provides a kind of detection method of clay compactness.
The technical solution adopted by the present invention to solve the technical problems is:
A kind of detection method of clay compactness comprising opaque cabinet, light source, camera, spectrometer and have image
The computer of capture card, the light source, camera and spectrometer are mounted in the opaque cabinet, the light source emit beam to
On clay sample, the camera connects the spectrometer to acquire the high spectrum image of the clay sample, and the computer connects
It connects the camera and obtains the high spectrum image comprising following steps:
Step A: by the high spectrum image of camera and spectrometer collection clay sample to having described image capture card
Computer;
Step B: the textural characteristics of the high spectrum image, the textural characteristics are obtained by the Matlab software of computer
Include: contrast, average brightness, consistency, uniformity, energy, the degree of correlation, third moment value, average contrast, smoothness and
Entropy;
Step C: prediction mould is established by BP neural network using the textural characteristics of the clay sample with different compactnesss
Type, the input variable after the textural characteristics are normalized as the prediction model make the compactness of clay
For output variable;
Step D: the textural characteristics for obtaining clay sample to be measured input the prediction model to obtain clay sample to be measured
Compactness.
In a further preferred embodiment, the clay sample is placed on the box house, and the camera and spectrometer are set
It sets at the cabinet upper opening.
In a further preferred embodiment, mobile platform is equipped in the cabinet, the clay sample is placed on the movement
On platform.
In a further preferred embodiment, the camera and the spectrometer are installed on the mobile apparatus, the mobile device
It is mounted on the cabinet internal upper part.
In a further preferred embodiment, the light source is fiber optic bundle.
In a further preferred embodiment, in step C, the implicit layer functions of the prediction model are tansig, output layer letter
Number is logsig;The anticipation error of the prediction model is 0.001, and trained maximum times are 1000 times, and learning rate is
0.05。
In a further preferred embodiment, in step C, the clay sample of different compactnesss is obtained by compaction test.
The beneficial effects of the present invention are:
High light spectrum image-forming measuring technique is applied in Subgrade Compaction detection by the present invention, which can both save people
Power shortens the construction period, and can guarantee construction quality, safety non-pollution, has important practical application value;Using impermeable finish
Cabinet made of matter avoids external light influence;It on a mobile platform by the setting of clay sample, can be flat by the movement movement
Platform realizes acquisition to the high spectrum image of clay sample different location by mobile camera and spectrometer;Light source passes through a light
Light is radiated on the clay sample by fibre, and optical fiber has many advantages, such as that size is small, the light-weight and service life is long;Pass through BP neural network
Prediction model is established to predict that clay compactness uses manpower and material resources sparingly, rapid and convenient, accuracy is high;By acquiring clay sample
The high spectrum image of different location reduces error;The convergence rate and precision of prediction model are promoted by normalized;Pass through
Compaction test obtains the clay sample of different compactnesss to be trained to prediction model.
Invention is further described in detail with reference to the accompanying drawings and embodiments;But a kind of clay compactness of the invention
Detection method be not limited to the embodiment.
Detailed description of the invention
Fig. 1 is the structural schematic diagram of a preferred embodiment of the present invention;
Fig. 2 is the error curve diagram of the BP neural network training of a preferred embodiment of the present invention.
Specific embodiment
Embodiment, a kind of shown in Figure 1, the detection method of clay compactness of the invention comprising such as lower component: light
Source 10, camera 20, spectrometer 30, clay sample 40, the computer 50 with image pick-up card, mobile platform 60 and wooden case 70,
The mobile platform 60 is mounted in wooden case 70, and the clay sample 40 is placed on the mobile platform 60, the light source 10
It is radiated on the clay sample 40 by the optical fiber 11 being mounted among 70 inner wall of wooden case, the camera 20 and the spectrometer
30 are mounted on the top of the wooden case, and the camera 20 connects the spectrometer 30 to acquire the EO-1 hyperion of the clay sample 40
Image, the computer 50 connect the camera 20 and obtain the high spectrum image.
The present embodiment clay compactness detection method the following steps are included:
Step A: the mobile mobile platform 60 acquires the different positions of the clay sample 40 by camera 20 and spectrometer 30
The high spectrum image set is to the computer 50 with image pick-up card;
Step B: the height is obtained by the textural characteristics that the Matlab software of computer 50 extracts the high spectrum image
The textural characteristics of spectrum picture, the textural characteristics include: contrast, average brightness, consistency, uniformity, energy, correlation
Degree, third moment value, average contrast, smoothness and entropy;
Step C: obtaining the clay sample of different compactnesss by compaction test, utilizes the clay sample of different compactnesss
Textural characteristics establish prediction model by BP neural network, and the prediction is used as after the textural characteristics are normalized
The input variable of model, using the compactness of clay as output variable, the implicit layer functions for configuring the prediction model are
Tansig, output layer functions are logsig;The anticipation error of the prediction model is 0.001, and trained maximum times are 1000
Secondary, learning rate 0.05 is input to 10 textural characteristics data and is instructed by the prediction model that BP neural network is established
Practice, the prediction model identification accuracy established by BP neural network has reached 93.2%;
Step D: the textural characteristics for obtaining clay sample to be measured input the prediction model to obtain clay sample to be measured
Compactness.
Above-described embodiment is only used to further illustrate a kind of detection method of clay compactness of the invention, but the present invention is simultaneously
It is not limited to the embodiment, according to the technical essence of the invention any simple modification to the above embodiments, equivalent change
Change and modify, falls within the scope of protection of technical solution of the present invention.
Claims (7)
1. a kind of detection method of clay compactness, it is characterised in that: it include opaque cabinet, light source, camera, spectrometer and
Computer with image pick-up card, the light source, camera and spectrometer are mounted in the opaque cabinet, the light source hair
Out on light to clay sample, the camera connects the spectrometer to acquire the high spectrum image of the clay sample, described
Computer connects the camera and obtains the high spectrum image comprising following steps:
Step A: pass through the high spectrum image of camera and spectrometer collection clay sample to the calculating with described image capture card
Machine;
Step B: the textural characteristics of the high spectrum image, the textural characteristics packet are obtained by the Matlab software of computer
Include: contrast, average brightness, consistency, uniformity, energy, the degree of correlation, third moment value, average contrast, smoothness and
Entropy;
Step C: establishing prediction model by BP neural network using the textural characteristics of the clay sample with different compactnesss, will
Input variable as the prediction model after the textural characteristics are normalized, using the compactness of clay as output
Variable;
Step D: the textural characteristics for obtaining clay sample to be measured input the prediction model to obtain the compacting of clay sample to be measured
Degree.
2. a kind of detection method of clay compactness according to claim 1, it is characterised in that: the clay sample is placed
In the box house, the camera and spectrometer are arranged at the cabinet upper opening.
3. a kind of detection method of clay compactness according to claim 1, it is characterised in that: be equipped with and move in the cabinet
Moving platform, the clay sample are placed on the mobile platform.
4. a kind of detection method of clay compactness according to claim 1, it is characterised in that: the camera and the light
Spectrometer is installed on the mobile apparatus, and the mobile device is mounted on the cabinet internal upper part.
5. a kind of detection method of clay compactness according to claim 1, it is characterised in that: the light source passes through optical fiber
Light is radiated on the clay sample.
6. a kind of detection method of clay compactness according to claim 1, it is characterised in that: in step C, the prediction
The implicit layer functions of model are tansig, and output layer functions are logsig;The anticipation error of the prediction model is 0.001, instruction
Experienced maximum times are 1000 times, learning rate 0.05.
7. a kind of detection method of clay compactness according to claim 1, it is characterised in that: in step C, by hitting reality
Test obtains the clay sample of different compactnesss.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910512891.1A CN110243768A (en) | 2019-06-13 | 2019-06-13 | A kind of detection method of clay compactness |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910512891.1A CN110243768A (en) | 2019-06-13 | 2019-06-13 | A kind of detection method of clay compactness |
Publications (1)
Publication Number | Publication Date |
---|---|
CN110243768A true CN110243768A (en) | 2019-09-17 |
Family
ID=67886981
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910512891.1A Pending CN110243768A (en) | 2019-06-13 | 2019-06-13 | A kind of detection method of clay compactness |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110243768A (en) |
Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2288242A (en) * | 1994-04-05 | 1995-10-11 | Univ Cardiff | Ground compaction apparatus and method |
CN1868610A (en) * | 2006-04-29 | 2006-11-29 | 江西农业大学 | Small-size movable fruit and birds, beasts and eggs intelligence grading plant |
CN102519965A (en) * | 2011-12-15 | 2012-06-27 | 南京工程学院 | Online roadbed compactness detection method based on machine vision |
CN102621190A (en) * | 2012-03-23 | 2012-08-01 | 山东大学 | Soil body sample compressing consolidation and resistivity real-time imaging monitoring device and soil sampler thereof |
CN107255637A (en) * | 2017-06-06 | 2017-10-17 | 河海大学 | A kind of grinding coagulation soil compactness detection method based on laser image |
CN108344670A (en) * | 2018-01-09 | 2018-07-31 | 国电大渡河流域水电开发有限公司 | A kind of subregion material circle pollution monitor system applied to dam facing filling construction scene |
-
2019
- 2019-06-13 CN CN201910512891.1A patent/CN110243768A/en active Pending
Patent Citations (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
GB2288242A (en) * | 1994-04-05 | 1995-10-11 | Univ Cardiff | Ground compaction apparatus and method |
CN1868610A (en) * | 2006-04-29 | 2006-11-29 | 江西农业大学 | Small-size movable fruit and birds, beasts and eggs intelligence grading plant |
CN102519965A (en) * | 2011-12-15 | 2012-06-27 | 南京工程学院 | Online roadbed compactness detection method based on machine vision |
CN102621190A (en) * | 2012-03-23 | 2012-08-01 | 山东大学 | Soil body sample compressing consolidation and resistivity real-time imaging monitoring device and soil sampler thereof |
CN107255637A (en) * | 2017-06-06 | 2017-10-17 | 河海大学 | A kind of grinding coagulation soil compactness detection method based on laser image |
CN108344670A (en) * | 2018-01-09 | 2018-07-31 | 国电大渡河流域水电开发有限公司 | A kind of subregion material circle pollution monitor system applied to dam facing filling construction scene |
Non-Patent Citations (2)
Title |
---|
刘木华: "《农产品质量安全光学无损检测技术及应用》", 30 June 2011 * |
李细荣: "基于激光图像土的压实度检测方法的研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Stanier et al. | Improved image-based deformation measurement in the centrifuge environment | |
CN203275401U (en) | Novel road soil engineering compaction rapid thickness measuring accommodometer | |
KR102148812B1 (en) | Apparatus for measuring air quality and system therefor | |
CN1800820A (en) | Pollutant gas concentration field measurement method in environmental wind tunnel | |
CN106855522B (en) | White light neutron imaging method and the material composition lossless detection method for using it | |
CN102803953A (en) | Method for determining weights of eggs, and apparatus | |
CN113933195A (en) | Concrete compressive strength prediction method and system based on image digital processing | |
CN115877555A (en) | Microscope illumination simulation system | |
CN207081670U (en) | It is a kind of can automatic range the cannot-harm-detection device | |
CN112149356A (en) | Method, device, equipment and medium for predicting structural crack propagation path | |
CN111830070A (en) | Automatic defect identification and judgment system and method based on edge calculation | |
CN107271466A (en) | A kind of nondestructive detection system | |
CN111784803A (en) | Automatic acquisition system and method for drill core correlation data | |
CN110243768A (en) | A kind of detection method of clay compactness | |
CN108013887A (en) | A kind of automatic exposure control method and device and auto exposure system | |
CN106934856A (en) | Three-dimension disclocation based on X-ray detection technology is rebuild and slice display method | |
CN107316131A (en) | A kind of electric power meter mounting process quality detecting system based on image recognition | |
CN106225736A (en) | A kind of interior formwork detection device | |
Owens et al. | A photographic technique for repeated mapping of rangeland plant populations in permanent plots. | |
CN203479725U (en) | Glass internal defect detection device | |
CN212569866U (en) | Automatic acquisition system for drilling core correlation data | |
CN112700423A (en) | Method and system for automatically detecting surface damage defects of fuselage based on deep learning | |
CN209277213U (en) | A kind of test device measuring Pile Soil 3-D displacement field | |
CN110873724B (en) | Shale core image acquisition method | |
CN114943693B (en) | Jetson Nano bridge crack detection method and system |
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 | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190917 |