CN111198186A - Turbidity detection method based on image recognition and optical transmission fusion - Google Patents

Turbidity detection method based on image recognition and optical transmission fusion Download PDF

Info

Publication number
CN111198186A
CN111198186A CN202010047828.8A CN202010047828A CN111198186A CN 111198186 A CN111198186 A CN 111198186A CN 202010047828 A CN202010047828 A CN 202010047828A CN 111198186 A CN111198186 A CN 111198186A
Authority
CN
China
Prior art keywords
turbidity
liquid
water tank
pictures
preset number
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.)
Granted
Application number
CN202010047828.8A
Other languages
Chinese (zh)
Other versions
CN111198186B (en
Inventor
周灿
刘天豪
唐燕伟
阳春华
李勇刚
朱红求
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Central South University
Original Assignee
Central South University
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by Central South University filed Critical Central South University
Priority to CN202010047828.8A priority Critical patent/CN111198186B/en
Publication of CN111198186A publication Critical patent/CN111198186A/en
Application granted granted Critical
Publication of CN111198186B publication Critical patent/CN111198186B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

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
    • G01N21/84—Systems specially adapted for particular applications
    • 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
    • G01N21/47—Scattering, i.e. diffuse reflection
    • G01N21/49—Scattering, i.e. diffuse reflection within a body or fluid
    • G01N21/51—Scattering, i.e. diffuse reflection within a body or fluid inside a container, e.g. in an ampoule

Landscapes

  • Physics & Mathematics (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Biochemistry (AREA)
  • General Health & Medical Sciences (AREA)
  • General Physics & Mathematics (AREA)
  • Immunology (AREA)
  • Pathology (AREA)
  • Investigating Or Analysing Materials By Optical Means (AREA)

Abstract

The invention provides a turbidity detection method based on image recognition and optical transmission fusion, which comprises the following steps: when the liquid level of the liquid to be measured entering the water tank reaches a preset value, a visible light LED of a composite light source in the water tank is started; continuously shooting the liquid to be detected in the water tank through a first high-speed camera in the water tank to obtain a first preset number of pictures; turning off the visible light LED and turning on the infrared light emitting diode; continuously shooting the liquid to be measured in the water tank through a second high-speed camera in the water tank to obtain a second preset number of pictures; inputting a first preset number of pictures into a first convolutional neural network for turbidity prediction to obtain a first turbidity; inputting a second preset number of pictures into a second convolutional neural network for turbidity prediction to obtain a second turbidity; and fusing the first turbidity and the second turbidity, and taking the fused turbidity as the turbidity of the liquid to be detected. The invention can improve the detection range of turbidity under the condition of reducing the detection cost of turbidity.

Description

Turbidity detection method based on image recognition and optical transmission fusion
Technical Field
The invention relates to the technical field of water body detection, in particular to a turbidity detection method based on image recognition and optical transmission fusion.
Background
Turbidity is an optical property of a water sample. The suspended particles in the solution obstruct or enable the light passing through the water sample to generate scattering and transmission phenomena, and the content of the suspended matters in the solution can be reflected by detecting the transmission or scattering light intensity, so that the turbidity is represented.
At present, optical turbidimeters are widely used, and can be classified into a transmission light method and a scattering light method according to different receiving angles of scattering light intensity. Traditional optics turbidity appearance equipment is inside accurate, and the price is higher, is limited by the light source, and detection range receives certain restriction, can not satisfy low turbidity and high turbidity's measurement requirement simultaneously.
Disclosure of Invention
The invention provides a turbidity detection method based on image recognition and optical transmission fusion, and aims to solve the problems of high turbidity detection cost and limited detection range.
In order to achieve the above object, an embodiment of the present invention provides a turbidity detection method based on image recognition and optical transmission fusion, including:
when the liquid level height of liquid to be detected entering a water tank reaches a preset value, a visible light LED of a composite light source arranged in the water tank is turned on, and an infrared light emitting diode of the composite light source is turned off;
continuously shooting the liquid to be measured in the water tank through a first high-speed camera arranged in the water tank to obtain a first preset number of pictures;
turning off the visible light LED and turning on the infrared light emitting diode;
continuously shooting the liquid to be measured in the water tank through a second high-speed camera arranged in the water tank to obtain a second preset number of pictures; the distance between the second high-speed camera and the bottom of the water tank is smaller than the distance between the first high-speed camera and the bottom of the water tank;
inputting the first preset number of pictures into a first convolutional neural network for turbidity prediction to obtain a first turbidity;
inputting the second preset number of pictures into a second convolutional neural network for turbidity prediction to obtain a second turbidity;
and fusing the first turbidity and the second turbidity according to DS evidence rules to obtain fused turbidity, and taking the fused turbidity as the turbidity of the liquid to be detected.
Wherein, the turbidity detection method further comprises:
the liquid level height of the liquid to be detected entering the water tank is detected through a liquid level sensor arranged in the water tank.
The water tank is communicated with a liquid inlet pipeline, and the liquid to be measured enters the water tank through the liquid inlet pipeline.
Wherein, a first pipeline valve is arranged on the liquid inlet pipeline;
when the liquid level height of the liquid to be detected entering the water tank reaches a preset value, the turbidity detection method further comprises the following steps:
closing the first pipeline valve.
Wherein, be equipped with first flow sensor on the liquid inlet pipeline.
The water tank is communicated with a liquid outlet pipeline, and the liquid to be detected flows out of the water tank through the liquid outlet pipeline.
Wherein, a second pipeline valve is arranged on the liquid outlet pipeline;
after the step of taking the turbidity after the fusion as the turbidity of the liquid to be detected, the turbidity detecting method further includes:
opening the second pipeline valve.
Wherein, a second flow sensor is arranged on the liquid outlet pipeline.
The scheme of the invention has at least the following beneficial effects:
in the embodiment of the invention, when the liquid level height of the liquid to be detected entering the water tank reaches a preset value, the visible light LED of the composite light source in the water tank is started, and the liquid to be detected is shot by the first high-speed camera in the water tank to obtain a first preset number of pictures; then turning off the visible light LED, turning on an infrared light emitting diode of the combined light source, shooting the liquid to be detected through a second high-speed camera in the water tank to obtain a second preset number of pictures, inputting the first preset number of pictures into a pre-trained first convolutional neural network for turbidity prediction to obtain a first turbidity under visible light, and inputting the second preset number of pictures into a pre-trained second convolutional neural network for turbidity prediction to obtain a second turbidity under scattered infrared light; and finally, according to the DS evidence rule, the turbidities under the two light sources are fused to obtain the turbidity of the liquid to be detected, so that the detection range of the turbidity is enlarged. Meanwhile, the turbidity detection can be realized by adopting two high-speed cameras, a combined light source and a convolutional neural network operation platform, so that the turbidity detection cost is greatly reduced.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only some embodiments of the present invention, and for those skilled in the art, other drawings can be obtained according to the structures shown in the drawings without creative efforts.
FIG. 1 is a flow chart of a turbidity detection method based on image recognition and optical transmission fusion according to an embodiment of the present invention;
FIG. 2 is a schematic structural view of a detection apparatus for detecting turbidity according to an embodiment of the present invention;
fig. 3 is a schematic structural diagram of a composite light source according to an embodiment of the invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, not all, embodiments of the present invention. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention. 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.
It should be noted that the terms "first" and "second" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of the present invention, "a plurality" means two or more unless specifically defined otherwise.
As shown in fig. 1, an embodiment of the present invention provides a turbidity detection method based on image recognition and optical transmission fusion, which includes the following steps:
and step 11, when the liquid level height of the liquid to be detected entering the water tank reaches a preset value, starting a visible light LED of a composite light source arranged in the water tank, and closing an infrared light emitting diode of the composite light source.
In the embodiment of the present invention, as shown in fig. 2, a liquid level sensor 202, a first high-speed camera 203, a second high-speed camera 204 and a composite light source 205 are disposed in the water tank 201; the water tank 201 is communicated with a liquid inlet pipeline 206, the liquid to be measured enters the water tank 201 through the liquid inlet pipeline 206, and a first pipeline valve 207 and a first flow sensor 208 are arranged on the liquid inlet pipeline 206; the water tank 201 is further communicated with a liquid outlet pipe 209, the liquid to be measured flows out of the water tank 201 through the liquid outlet pipe 209, and a second pipe valve 210 and a second flow sensor 211 are arranged on the liquid outlet pipe 209.
It should be noted that the liquid level sensor 202, the first high-speed camera 203, the second high-speed camera 204, the compound light source 205, the first pipeline valve 207, the first flow sensor 208, the second pipeline valve 210 and the second flow sensor 211 are all connected to an embedded control platform 212. The liquid level sensor 202 is configured to collect a liquid level of a liquid to be measured in the water tank 201, and output the collected liquid level to the embedded control platform 212 (that is, the embedded control platform 212 may detect the liquid level of the liquid to be measured entering the water tank 201 through the liquid level sensor 202 disposed in the water tank 201), so that when the embedded control platform 212 detects that the liquid level of the liquid to be measured entering the water tank 201 reaches a preset value (a specific value of the preset value may be set according to an actual situation), the first pipeline valve 207 is closed (wherein, the first flow sensor 208 on the liquid inlet pipeline 206 may collect a flow rate of the liquid to be measured on the liquid inlet pipeline 206, and output collected flow rate information to the embedded control platform 212, so that a user may conveniently measure a flow rate of the liquid to be measured on the liquid inlet pipeline 206), and the first high-speed camera 203, The second high speed camera 204 and the compound light source 205.
Specifically, in the embodiment of the present invention, the embedded control platform 212 may control the composite light source 205 through the pulse width modulation circuit 213, such as turning on the visible light LED (or the infrared light emitting diode), turning off the visible light LED (or the infrared light emitting diode), and the like. The structure of the composite light source is shown in fig. 3, and the composite light source includes a visible light LED31 and an infrared light emitting diode 32. As a preferred example, the infrared light emitting diode may be an infrared light emitting diode having a wavelength of 850 nm.
And step 12, continuously shooting the liquid to be detected in the water tank through a first high-speed camera arranged in the water tank to obtain a first preset number of pictures.
In an embodiment of the invention, the first predetermined number may be set by the embedded control platform, and a specific value of the first predetermined number may be set according to actual needs. As a preferred example, the first high-speed camera may be a high-speed continuous shooting camera.
And step 13, turning off the visible light LED and turning on the infrared light emitting diode.
In the embodiment of the invention, after the first high-speed camera shoots the first preset number of pictures, the visible light LED of the composite light source can be turned off, and the infrared light emitting diode is turned on, so that the second high-speed camera can shoot the liquid to be detected.
And step 14, continuously shooting the liquid to be detected in the water tank through a second high-speed camera arranged in the water tank to obtain a second preset number of pictures.
And the distance between the second high-speed camera and the bottom of the water tank is smaller than the distance between the first high-speed camera and the bottom of the water tank. It should be noted that the second preset number can be set by the embedded control platform, and the specific numerical value of the second preset number can be set according to actual needs. As a preferred example, the second high-speed camera may be a high-speed continuous shooting camera.
And step 15, inputting the first preset number of pictures into a first convolutional neural network for turbidity prediction to obtain a first turbidity.
In an embodiment of the present invention, the first convolutional neural network is trained in advance, specifically, a plurality of turbidity pictures of different liquids can be taken by the first high-speed camera, turbidity values of the turbidity pictures are calibrated, and the pictures and the turbidity value corresponding to each picture are input into the first convolutional neural network, so that training of the first convolutional neural network can be completed. Therefore, after the first preset number of pictures are input into the first convolution neural network, the first convolution neural network can predict the turbidity corresponding to the first preset number of pictures, wherein the turbidity is the turbidity of the liquid to be measured under the visible light.
And step 16, inputting the second preset number of pictures into a second convolutional neural network for turbidity prediction to obtain a second turbidity degree.
In an embodiment of the present invention, the second convolutional neural network is trained in advance, specifically, a second high-speed camera may be used to take several turbidity pictures of different liquids, calibrate turbidity values of the turbidity pictures, and input the pictures and the turbidity value corresponding to each picture into the second convolutional neural network, so as to complete training of the second convolutional neural network. Therefore, after the second preset number of pictures are input into the second convolutional neural network, the second convolutional neural network can predict the turbidity corresponding to the second preset number of pictures, wherein the turbidity is the turbidity of the liquid to be detected under the scattered infrared light.
And step 17, fusing the first turbidity and the second turbidity according to a DS evidence rule to obtain fused turbidity, and taking the fused turbidity as the turbidity of the liquid to be detected.
In the embodiment of the invention, after the turbidity of the liquid to be measured is obtained, the embedded control platform opens the second pipeline valve, so that the liquid to be measured in the water tank flows out of the water tank. The second flow sensor on the liquid outlet pipeline can collect the flow of the liquid to be detected on the liquid outlet pipeline and output the collected flow information to the embedded control platform, so that a user can know the flow of the liquid to be detected on the liquid outlet pipeline.
It is worth mentioning that in the embodiment of the invention, when the liquid level height of the liquid to be measured entering the water tank reaches a preset value, the visible light LED of the composite light source in the water tank is turned on, and the liquid to be measured is shot by the first high-speed camera in the water tank to obtain a first preset number of pictures; then turning off the visible light LED, turning on an infrared light emitting diode of the combined light source, shooting the liquid to be detected through a second high-speed camera in the water tank to obtain a second preset number of pictures, inputting the first preset number of pictures into a pre-trained first convolutional neural network for turbidity prediction to obtain a first turbidity under visible light, and inputting the second preset number of pictures into a pre-trained second convolutional neural network for turbidity prediction to obtain a second turbidity under scattered infrared light; and finally, according to DS evidence rules, the turbidities under the two light sources are fused to obtain the turbidity of the liquid to be measured, and the advantages of transmission, scattering and image identification are fully combined, so that the detection range of the turbidity is enlarged, and the measurement precision is improved. Meanwhile, the turbidity detection can be realized by adopting two high-speed cameras, a combined light source and a convolutional neural network operation platform, so that the turbidity detection cost is greatly reduced.
In addition, it should be noted that the turbidity detection method can detect turbidity of liquid in a flowing state, the flowing liquid enters the water tank through the liquid inlet pipeline, the turbidity detection process is completed in the liquid flowing process, and the turbidity detection can be completed on line without manual sample preparation. And the turbidity detection method has universal adaptability to different types of turbidity liquid and high mobility.
Next, the implementation of step 17 will be further described.
The basic flow in the fusion algorithm of the DS evidence rules is explained first. First, a mass function is introduced, in a decision problem, a set of all possible answers to the problem is represented by an identification frame Θ, the answers to the problem at any time can only take a certain subset of Θ, and the identification frame can be represented as Θ ═ Θ1，Θ2，…ΘnIn which, ΘiTo identify an element or event of a frame, n represents the number of elements or events in the identified frame, and i is an integer greater than or equal to 1 and less than or equal to n.
The mass function is 2ΘTo [0,1 ]]A is 2ΘAny subset of (1), denoted as
Figure BDA0002370053090000071
And satisfies the following conditions:
Figure BDA0002370053090000072
the mass fusion rule is as follows:
for the
Figure BDA0002370053090000073
Identifying a finite number of mass functions m on a framework Θ1，m2，...，mnThe fusion rule is:
Figure BDA0002370053090000074
wherein the content of the first and second substances,
Figure BDA0002370053090000075
the fusion algorithm flow is described next:
the detection range of the turbidity is 1 to 1000NTU, and the resolution is 10NTU, namely the turbidity which can be output by the system (the system for executing the turbidity detection method) is 10NTU, 20NTU, 30NTU, … and 1000 NTU. A first convolutional neural network of d1The second convolutional neural network is d2The identification frame of each convolutional neural network is omega ═ { omega ═ omega1，ω2，ω3，...ω100And defining the similarity of x and y of two gray level images as Rxy，
Figure BDA0002370053090000076
Where m and n represent the length and width of the input image, and I and K represent the corresponding pixels of the two test pictures, respectively. It can be seen that RxyIn the range of [0,1]R isxyWhen 1 indicates that the two images are completely similar, RxyLess than 1 is less similar. Hypothesis convolutional neural network d1Prediction sample y1Turbidity of u1＝ωsWhen y is1And a training set (i.e. for training the convolutional neural network d)1Data of (d) turbidity of ωsWhen the picture similarity is high, x at this time is indicatedsThe prediction is relatively reliable.
Defining a training distribution matrix T as:
Figure BDA0002370053090000081
it can be seen that the training distribution matrix T is a 100 x 100 matrix, defining T1、T2Respectively, a convolutional neural network of d1、d2Each row i (1 ≦ i ≦ 100) in the training distribution matrix represents the true turbidity ωiEach column j (1. ltoreq. j. ltoreq.100) represents the turbidity ω predicted by the neural networkjNumber of test samples. With convolutional neural network as d1For example, when d1Predicted turbidity of uk＝ωsAccording to the training distribution matrix T1The conditional probability vector p under the current prediction can be obtainedk：
Figure BDA0002370053090000082
pk＝{pk(uk|ω1)，pk(uk|ω2)，...，pk(uk|ω100)}
Then, the similarity R obtained in the previous step is utilizedkAnd correcting each conditional probability to obtain 100 conditional mass functions:
mk，i(uk|ωi)＝Rpk(uk|ωi)，1≤i≤100
mk，i(uk|Ω)＝1-Rpk(uk|ωi)，1≤i≤100
to this end, d for convolutional neural networks1Prediction of turbidity as uk＝ωsIn this case, the conditional probability vector p is passedkThe sum similarity obtains 100 conditional mass functions { massk，1，massk，2，...massk，100}. The 100 conditional mass functions are fused through a DS evidence fusion rule to obtain d1Mass function mas ofsd1. By the same token, d can be obtained2Mass function ofd2. And finally, fusing the two functions into a final mass function by utilizing a DS evidence rule, so as to obtain the fused predicted turbidity.
It should be noted that, because the turbidity detection method of the embodiment of the present invention adopts a fusion algorithm, the detection accuracy is improved by the decision-level fusion of turbidity information under the transmission of visible light and the scattering of infrared light. Meanwhile, concepts such as a training distribution matrix, reliability and the like are established under a fusion algorithm, and the method can be used for correcting the predicted values (namely the first turbidity and the second turbidity) of the convolutional neural network.
While the foregoing is directed to the preferred embodiment of the present invention, it will be understood by those skilled in the art that various changes and modifications may be made without departing from the spirit and scope of the invention as defined in the appended claims.

Claims (8)

1. A turbidity detection method based on image recognition and optical transmission fusion is characterized by comprising the following steps:
when the liquid level height of liquid to be detected entering a water tank reaches a preset value, a visible light LED of a composite light source arranged in the water tank is turned on, and an infrared light emitting diode of the composite light source is turned off;
continuously shooting the liquid to be measured in the water tank through a first high-speed camera arranged in the water tank to obtain a first preset number of pictures;
turning off the visible light LED and turning on the infrared light emitting diode;
continuously shooting the liquid to be measured in the water tank through a second high-speed camera arranged in the water tank to obtain a second preset number of pictures; the distance between the second high-speed camera and the bottom of the water tank is smaller than the distance between the first high-speed camera and the bottom of the water tank;
inputting the first preset number of pictures into a first convolutional neural network for turbidity prediction to obtain a first turbidity;
inputting the second preset number of pictures into a second convolutional neural network for turbidity prediction to obtain a second turbidity;
and fusing the first turbidity and the second turbidity according to DS evidence rules to obtain fused turbidity, and taking the fused turbidity as the turbidity of the liquid to be detected.
2. A method of turbidity detection according to claim 1, further comprising:
the liquid level height of the liquid to be detected entering the water tank is detected through a liquid level sensor arranged in the water tank.
3. A method of turbidity measurements according to claim 1, wherein said water tank is in communication with a liquid inlet conduit through which said liquid to be measured enters said water tank.
4. A method according to claim 3, wherein the liquid inlet conduit is provided with a first conduit valve;
when the liquid level height of the liquid to be detected entering the water tank reaches a preset value, the turbidity detection method further comprises the following steps:
closing the first pipeline valve.
5. A method of turbidity according to claim 3, wherein a first flow sensor is provided on said liquid inlet conduit.
6. A method of turbidity measurements according to claim 1, wherein said tank is in communication with a liquid outlet conduit through which said liquid to be measured flows out of said tank.
7. A method according to claim 1, wherein the liquid outlet conduit is provided with a second conduit valve;
after the step of taking the turbidity after the fusion as the turbidity of the liquid to be detected, the turbidity detecting method further includes:
opening the second pipeline valve.
8. A method of turbidity according to claim 6, wherein a second flow sensor is provided on said liquid outlet conduit.
CN202010047828.8A 2020-01-16 2020-01-16 Turbidity detection method based on image recognition and optical transmission fusion Active CN111198186B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202010047828.8A CN111198186B (en) 2020-01-16 2020-01-16 Turbidity detection method based on image recognition and optical transmission fusion

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202010047828.8A CN111198186B (en) 2020-01-16 2020-01-16 Turbidity detection method based on image recognition and optical transmission fusion

Publications (2)

Publication Number Publication Date
CN111198186A true CN111198186A (en) 2020-05-26
CN111198186B CN111198186B (en) 2021-08-31

Family

ID=70744840

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202010047828.8A Active CN111198186B (en) 2020-01-16 2020-01-16 Turbidity detection method based on image recognition and optical transmission fusion

Country Status (1)

Country Link
CN (1) CN111198186B (en)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112782097A (en) * 2020-12-21 2021-05-11 中国科学院合肥物质科学研究院 Liquid turbidity measuring device and method based on convolutional neural network

Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0864855A1 (en) * 1997-03-10 1998-09-16 Fuji Electric Co., Ltd. Method and apparatus for measuring turbidity
EP1096248A2 (en) * 1999-10-28 2001-05-02 Matsushita Electric Industrial Co., Ltd. Method and apparatus for measuring concentration of a solution
EP1926983A1 (en) * 2005-09-22 2008-06-04 ETR-Unidata Limited Scattering centre detector assembly and method
CN104748791A (en) * 2013-03-18 2015-07-01 吴昊 Operating method of water environmental monitoring device adopting image vision processing technology
CN109142296A (en) * 2018-08-16 2019-01-04 中国科学院合肥物质科学研究院 The black smelly quick identification measuring method of urban water-body based on multi-source optical spectrum feature
CN109187534A (en) * 2018-08-01 2019-01-11 江苏凯纳水处理技术有限公司 Water quality detection method and its water sample pattern recognition device
CN109459399A (en) * 2018-12-26 2019-03-12 南京波思途智能科技股份有限公司 A kind of spectral water quality COD, turbidity detection method
CN109635249A (en) * 2019-01-09 2019-04-16 中国科学院遥感与数字地球研究所 Water turbidity inverse model method for building up, water turbidity detection method and device
CN110672524A (en) * 2019-10-22 2020-01-10 浙江卓锦环保科技股份有限公司 Water body turbidity detection method suitable for intelligent water environment
CN110672523A (en) * 2019-11-14 2020-01-10 厦门华联电子股份有限公司 Turbidity sensor

Patent Citations (10)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
EP0864855A1 (en) * 1997-03-10 1998-09-16 Fuji Electric Co., Ltd. Method and apparatus for measuring turbidity
EP1096248A2 (en) * 1999-10-28 2001-05-02 Matsushita Electric Industrial Co., Ltd. Method and apparatus for measuring concentration of a solution
EP1926983A1 (en) * 2005-09-22 2008-06-04 ETR-Unidata Limited Scattering centre detector assembly and method
CN104748791A (en) * 2013-03-18 2015-07-01 吴昊 Operating method of water environmental monitoring device adopting image vision processing technology
CN109187534A (en) * 2018-08-01 2019-01-11 江苏凯纳水处理技术有限公司 Water quality detection method and its water sample pattern recognition device
CN109142296A (en) * 2018-08-16 2019-01-04 中国科学院合肥物质科学研究院 The black smelly quick identification measuring method of urban water-body based on multi-source optical spectrum feature
CN109459399A (en) * 2018-12-26 2019-03-12 南京波思途智能科技股份有限公司 A kind of spectral water quality COD, turbidity detection method
CN109635249A (en) * 2019-01-09 2019-04-16 中国科学院遥感与数字地球研究所 Water turbidity inverse model method for building up, water turbidity detection method and device
CN110672524A (en) * 2019-10-22 2020-01-10 浙江卓锦环保科技股份有限公司 Water body turbidity detection method suitable for intelligent water environment
CN110672523A (en) * 2019-11-14 2020-01-10 厦门华联电子股份有限公司 Turbidity sensor

Non-Patent Citations (6)

* Cited by examiner, † Cited by third party
Title
PINGPING CAO: "Using a Digital Camera Combined With Fitting", 《IEEE》 *
WEN-CHIN LEE: "An Integrated Water Turbidity Detection Method based on multi-spectral Analysis", 《2011 INTERNATIONAL CONFERENCE ON CONSUMER ELECTRONICS, COMMUNICATIONS AND NETWORKS (CECNET)》 *
张远: "用于自来水浊度检测的多传感器信息融合技术", 《自动化与仪表》 *
李昂: "基于图像识别的水质浊度检测系统的设计与实现", 《中国优秀硕士学位论文全文数据库 工程科技Ⅰ辑》 *
王清泉: "基于CCD的多浊度快速检测技术", 《仪表技术与传感器》 *
郑铁生: "《临床生物化学检验》", 31 August 2015 *

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112782097A (en) * 2020-12-21 2021-05-11 中国科学院合肥物质科学研究院 Liquid turbidity measuring device and method based on convolutional neural network

Also Published As

Publication number Publication date
CN111198186B (en) 2021-08-31

Similar Documents

Publication Publication Date Title
CN106650913B (en) A kind of vehicle density method of estimation based on depth convolutional neural networks
CN105975941B (en) A kind of multi-direction vehicle detection identifying system based on deep learning
CN108921206A (en) A kind of image classification method, device, electronic equipment and storage medium
CN111028939B (en) Multigroup intelligent diagnosis system based on deep learning
Duan et al. A machine vision inspector for beer bottle
JP2012514751A (en) Method and apparatus for measuring liquid level in a container using imaging
CN108956484A (en) A kind of method and apparatus of integration tracking pollution sources
CN111198186B (en) Turbidity detection method based on image recognition and optical transmission fusion
KR100423115B1 (en) Water Quality Measuring Apparatus and Method Using Image Data
CN109711322A (en) A kind of people's vehicle separation method based on RFCN
CN115808324B (en) Light safety management monitoring method and system for small and medium span bridges
US11047807B2 (en) Defect detection
CN105426830B (en) Passage aisle flow pattern of gas-liquid two-phase flow identifying system and method based on multi-visual information integration technology
CN113688751A (en) Method and device for analyzing alum blossom characteristics by using image recognition technology
CN110298410A (en) Weak target detection method and device in soft image based on deep learning
Zhang et al. Automatic scratch detector for optical surface
WO2022229413A1 (en) Method for detecting and quantifying viruses in fluid samples by means of digital processing of hyperspectral images of diffuse optical reflectance obtained in the visible and near infrared ranges
CN106845051A (en) A kind of near infrared no-wound blood sugar test wavelength Variable Selection method based on Combinatorial Optimization
TWI700489B (en) Device for instantaneously inspecting waste quality and recovery device and method using the same
CN113851001A (en) Automatic auditing method, system, device and storage medium for multilane merging violation
WO2004038403A1 (en) Water quality measuring apparatus and method using image
CN116952654B (en) Environment monitoring and early warning system for administrative supervision
CN116403284B (en) Wisdom running examination training system based on bluetooth transmission technology
Zhang et al. Automatic segmentation of airport pavement damage by AM‐Mask R‐CNN algorithm
CN203405417U (en) Device for measuring gas-liquid two-phase flow in pipeline through image method

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant