CN114965472B - On-line automatic identification system for multi-dimensional imaging of plankton - Google Patents
On-line automatic identification system for multi-dimensional imaging of plankton Download PDFInfo
- Publication number
- CN114965472B CN114965472B CN202210608361.9A CN202210608361A CN114965472B CN 114965472 B CN114965472 B CN 114965472B CN 202210608361 A CN202210608361 A CN 202210608361A CN 114965472 B CN114965472 B CN 114965472B
- Authority
- CN
- China
- Prior art keywords
- plankton
- sample
- capillary flow
- imaging
- objective
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- 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
- G01N21/01—Arrangements or apparatus for facilitating the optical investigation
-
- 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
- G01N2021/8466—Investigation of vegetal material, e.g. leaves, plants, fruits
-
- 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
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A40/00—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production
- Y02A40/80—Adaptation technologies in agriculture, forestry, livestock or agroalimentary production in fisheries management
- Y02A40/81—Aquaculture, e.g. of fish
-
- 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
- Y02A—TECHNOLOGIES FOR ADAPTATION TO CLIMATE CHANGE
- Y02A90/00—Technologies having an indirect contribution to adaptation to climate change
- Y02A90/30—Assessment of water resources
Abstract
The invention discloses an online automatic identification system for multi-dimensional imaging of plankton, which comprises: a base; a sampling assembly comprising a sample cell, a peristaltic pump, and a filter; a sample preparation assembly comprising a plurality of capillary flow-through tubes; a mechanical transmission assembly comprising an XY axis moving platform and a mechanical arm; the optical imaging assembly comprises an object stage, a blind hole, an object lens disc, at least three multiple compound object lenses and a microscopic camera; the driving assembly is used for respectively driving the peristaltic pump, the XY-axis moving platform, the mechanical arm, the microscope camera and the objective disc; and the control center comprises an image library and an analysis module. The invention utilizes a plurality of capillary flow tubes and combines at least three multi-time composite objective lenses to carry out three-dimensional multi-dimensional panoramic imaging, thereby realizing multi-dimensional high-resolution imaging of plankton, obviously improving imaging and recognition effects and realizing high-precision automatic classification of plankton.
Description
Technical Field
The invention relates to the technical field of online detection of plankton, in particular to an online automatic identification system for multidimensional imaging of plankton.
Background
Phytoplankton refers to tiny organisms living in water bodies such as oceans, lakes and rivers and is a very important biological group in aquatic ecosystems. Parameters such as the species, biomass and biological diversity of phytoplankton are important indexes reflecting aquatic ecological health, and have clear requirements in relevant guidelines and specifications of river, lake and ocean monitoring and ecological health evaluation in China.
The existing phytoplankton online identification technology can not meet the requirements of quick, fine and accurate full-automatic monitoring, the grade of identification species is not high, the ecological effect between the phytoplankton and the water environment can not be reflected in time, and the scientific support of water ecological early warning and environmental supervision in China is restricted. The first research method of phytoplankton is to use an optical microscope to directly observe to obtain required data, but the observation by the microscope needs complicated manual operation and professional biological classification knowledge, is time-consuming and labor-consuming, is easy to cause errors due to subjective judgment of an observer, cannot accurately reflect the health condition of a water body in time, and particularly when a research area is large and involves space and time, a microscopic examination method is greatly limited, so that the analysis of a sample is slow, and further, the evaluation of an ecological environment system is delayed.
The existing developed technology for rapidly detecting phytoplankton is a Flow imager (Flow CAM), but the instrument also faces a great problem in application, firstly, the phytoplankton has a wide particle size range and is easy to aggregate into a dough or a chain, and the characteristic of Flow sample measurement of the phytoplankton easily causes the blockage of a Flow cell and the cross contamination among different samples; secondly, the phytoplankton is various in types and shapes, the accuracy of digital image recognition needs to be improved, and the classification result basically needs manual auxiliary recognition. With the progress of science and technology, the artificial intelligence image recognition technology has a qualitative breakthrough. The method is to find out a group of algae with the most similar morphology from tens of thousands of image libraries by utilizing the contour of unknown algae cells, extracting characteristic information and big data matching, and preferentially showing common algae. However, the diversity of algae in an actual sample is often very rich, and under the condition of no complex algorithm support, algae images of the same kind and different forms are difficult to be accurately identified and analyzed by the algorithm. The image recognition technology based on artificial intelligence needs to give a large number of samples to a model for deep learning and training, and finally, a species recognition task is completed. The database cannot be updated and upgraded in real time as the composition of species in the environment changes. Therefore, these instruments have not been able to perform automatic counting and identification for a while, and require a lot of manual operations, which are only assisted in image resolution and algae identification.
In summary, the following disadvantages exist in the current research of phytoplankton identification equipment:
(1) The existing imaging method can not realize three-dimensional multi-dimensional identification of phytoplankton, has low resolution, is difficult to meet the objective requirement of automatic classification, and seriously restricts the scientific research progress of the phytoplankton species composition characteristic analysis in the water ecosystem of China;
(2) The phytoplankton detection process and the instrument framework need to be optimized, are easily influenced by sundries or phytoplankton groups, and are difficult to realize full-automatic phytoplankton identification. At present, sampling, flaking, identification, data analysis and the like of phytoplankton are carried out by manpower, and the data result has serious hysteresis relative to the water environment change;
(3) The method is lack of a real-time online full-automatic phytoplankton monitoring instrument, cannot meet the requirement of field investigation real-time monitoring, and urgently needs a convenient, quick and visual phytoplankton online monitoring instrument for autonomous development, thereby providing powerful data support for improving the early warning capability of ecological disasters such as red tide, algal bloom and the like in water areas of China.
Disclosure of Invention
An object of the present invention is to solve at least the above problems and to provide at least the advantages described later.
The invention also aims to provide an on-line automatic identification system for the multi-dimensional imaging of plankton, which utilizes a plurality of capillary flow tubes and combines at least three multi-time compound objective lenses to carry out three-dimensional multi-dimensional panoramic imaging to realize the multi-dimensional high-resolution imaging of plankton, thereby obviously improving the imaging and identification effects and realizing the high-precision automatic classification of plankton.
To achieve these objects and other advantages and in accordance with the purpose of the invention, as embodied and broadly described, there is provided an online automatic identification system for multi-dimensional imaging of plankton, comprising: a base;
the sampling assembly is arranged on the base and comprises a sample pool, and the sample pool is communicated to a water body to be detected through a sample inlet pipe; the peristaltic pump and the filter are arranged on the sampling pipe;
the sample preparation assembly is arranged on the base and comprises a plurality of capillary flow tubes which are uniformly inserted on a sample table at intervals, and the sample table is close to the sample pool; the recovery platform is detachably inserted with a plurality of capillary flow-through tubes with detected samples; the mechanical transmission assembly comprises an XY axis moving platform which is arranged on the base; the mechanical arm is arranged on the XY-axis moving platform, the front end of the mechanical arm is provided with a clamp, and the front end of the clamp is detachably clamped with any capillary flow tube;
an optical imaging assembly including a stage disposed on a base; the blind hole is formed on the objective table, and the lower end of any capillary flow-through pipe is detachably inserted into the blind hole; the objective lens disk is arranged on the base in a lifting manner; the objective lens disc is uniformly distributed around the objective table in a divergent manner by taking any point on an extension line where the blind hole is located as a circle center, the plane where the at least three multi-fold composite objective lenses are located is perpendicular to the extension line where the blind hole is located, and the optical axes of the at least three multi-fold composite objective lenses are intersected at any point on the extension line where the blind hole is located; the microscopic camera is in communication connection with the at least three multiple compound objective lenses, the at least three multiple compound objective lenses and the objective lens disc synchronously lift, and a plankton multidimensional image set inserted in any capillary flow tube on the blind hole is synchronously acquired through the microscopic camera;
the driving assembly is used for respectively driving the peristaltic pump, the XY-axis moving platform, the mechanical arm, the microscope camera and the objective disc;
the control center is in wireless communication connection with the microscopic camera and the driving assembly, and further comprises an image library, wherein a classification atlas of plankton samples is prestored in the image library; and the analysis module is used for acquiring the plankton multidimensional image set shot by the microscopic camera in real time, then comparing and analyzing the plankton multidimensional image set with the classification image set of the image library, and returning an analysis result.
Preferably, the method further comprises the following steps: one end of the waste liquid outlet pipe is communicated to the bottom of the sample pool, and the other end of the waste liquid outlet pipe extends outwards; the sampling pipes comprise a first sampling pipe arranged on the water inlet side of the peristaltic pump and a second sampling pipe arranged on the outlet side of the peristaltic pump;
the three-way valve is arranged between the first sample inlet pipe and the peristaltic pump, a first water inlet of the three-way valve is communicated with a sample outlet of the first sample inlet pipe, and a water outlet of the three-way valve is communicated with a water inlet of the peristaltic pump;
the cleaner comprises a storage cavity, purified water is prestored in the storage cavity, a water outlet of the storage cavity is communicated with a second water inlet of the tee joint through a water outlet pipe, and the driving portion drives the cleaner to clean at regular time.
Preferably, the method further comprises the following steps: the sample pool and the recovery platform are respectively arranged in two accommodating cavities in the thermostat; the detection ends of the two temperature sensors respectively extend into the sample cell and above the recovery platform; the thermostat is in communication connection with the control center.
Preferably, the objective table and the objective disc are made of transparent materials, the capillary flow tubes are made of quartz materials, and numbers are respectively marked at two ends of any capillary flow tube; and when the microscopic camera acquires the plankton multidimensional image set in any capillary flow tube, synchronously acquiring and storing the serial numbers on the corresponding capillary flow tubes.
Preferably, the XY-axis moving stage includes:
the mechanical arm is arranged on the cross beam in a vertically sliding mode through the driving sliding block;
the two ends of a cross beam are arranged on the pair of longitudinal beams in a sliding manner through driving sliding blocks;
the lower ends of the at least two support columns are fixed on the base, and the upper ends of the at least two support columns are fixed on the pair of longitudinal beams.
Preferably, the objective lens disk includes two annular bodies, and at least three multiple compound lenses thereof are sandwiched between the two annular bodies; the upright post is vertically fixed on the base; and one end of the sliding block is fixed on the two annular bodies, and the other end of the sliding block is arranged on the stand column in a sliding manner.
The sample table and the recovery table are both plate bodies with a plurality of jacks uniformly distributed above, and the capillary flow tubes are correspondingly inserted into the jacks on the sample table one by one;
the capillary flow tubes with the detected samples are inserted into the jack holes on the recovery platform in a one-to-one correspondence mode.
Preferably, the depth of the blind hole is less than or equal to 1/10 of the length of any capillary flow-through tube; the volume of any capillary flow tube is more than or equal to 0.1ml.
Preferably, the depth of the blind hole is equal to 1/10 of the length of any capillary flow-through tube; the volume of any capillary flow-through is 0.1ml, 0.5ml or 1ml.
Preferably, the plankton multidimensional image set comprises a plurality of panoramic pictures obtained by shooting the same plankton at multiple angles, a plurality of panoramic pictures obtained by shooting different targets at multiple angles and the numbers of the corresponding capillary flow tubes; the analysis result is the classification result of the plankton corresponding to the number, and the number or the density of each plankton.
The invention at least comprises the following beneficial effects:
1. the invention utilizes a plurality of capillary flow tubes and combines at least three multifold compound objective lenses to carry out three-dimensional multi-dimensional panoramic imaging, thus realizing multi-dimensional high-resolution imaging of plankton, obviously improving imaging and identifying effects and realizing high-precision automatic classification of plankton;
2. by optimizing the design of instrument composition units and structural pathways, units for automatically managing sample collection, sample preparation, optical imaging, analysis and treatment and the like are realized, and the automation level of the conventional phytoplankton monitoring and identification is obviously improved;
3. an online monitoring mode is established by utilizing the technology of the Internet of things, instrument data and sample images are quickly and conveniently acquired through mobile networks such as 4G/5G, and the feedback control of the operation of the instrument is realized.
Additional advantages, objects, and features of the invention will be set forth in part in the description which follows and in part will become apparent to those having ordinary skill in the art upon examination of the following or may be learned from practice of the invention.
Drawings
FIG. 1 is a schematic structural diagram of an online plankton multi-dimensional imaging identification system according to an embodiment of the present invention;
FIG. 2 is a flow chart of an on-line automatic identification process of the plankton multi-dimensional imaging on-line identification system according to an embodiment of the present invention;
FIG. 3 is a schematic diagram of the tee and its connections in accordance with yet another embodiment of the present invention;
fig. 4 is a flow chart of an online automatic identification process of the plankton multidimensional imaging online identification system according to still another embodiment of the invention.
Detailed Description
The present invention is further described in detail below with reference to the attached drawings so that those skilled in the art can implement the invention by referring to the description text.
It will be understood that terms such as "having," "including," and "comprising," as used herein, do not preclude the presence or addition of one or more other elements or groups thereof.
As shown in fig. 1 and 2, the present invention provides an online automatic identification system for multi-dimensional imaging of plankton, comprising: a base 1; the sampling assembly 10 is arranged on the base and comprises a sample pool 101, and the sample pool is communicated to a water body to be detected through a sample inlet pipe 102; a peristaltic pump 103 and a filter, both disposed on the sample introduction tube; a sample preparation assembly 20 disposed on the base, the sample preparation assembly comprising a plurality of capillary flow tubes 201, which are uniformly spaced apart and inserted on a sample stage 202, the sample stage being disposed adjacent to the sample cell; a recovery stage 203 on which a plurality of capillary flow-through tubes 201' with the detected sample are detachably inserted; a mechanical transfer assembly 30 including an XY axis moving stage 301 disposed on the base; a mechanical arm 302 which is arranged on the XY axis moving platform, the front end of the mechanical arm is provided with a clamp 3021, and the front end of the clamp can detachably clamp any capillary flow tube; the clamp comprises two clamping pieces which are hinged; an optical imaging assembly 40 comprising a stage 401 disposed on a base; the blind hole is formed on the objective table, and the lower end of any capillary flow-through pipe is detachably inserted into the blind hole; an objective lens disk 403 which is arranged on the base in a lifting way; at least three multiple compound objective lenses 402, which are divergently and uniformly distributed around the objective table through the objective lens disc by taking any point on the extension line where the blind hole is located as the center of a circle, the plane where the at least three multiple compound objective lenses are located is perpendicular to the extension line where the blind hole is located, and the optical axes of the at least three multiple compound objective lenses are intersected at any point on the extension line where the blind hole is located; the microscopic camera is in communication connection with the at least three multi-time composite objective lenses, the at least three multi-time composite objective lenses and the objective lens disc synchronously ascend and descend, and plankton multi-dimensional image sets inserted in any capillary flow tube on the blind hole are synchronously acquired through the microscopic camera; in practical application, the devices can be integrated in a shell, so that sample pollution and the like are avoided; the driving assembly is used for respectively driving the peristaltic pump, the XY axis moving platform, the mechanical arm, the microscope camera and the objective disc; the control center is in wireless communication connection with the microscopic camera and the driving assembly, and further comprises an image library, wherein a classification atlas of the plankton sample is prestored in the image library; and the analysis module is used for acquiring the plankton multidimensional image set shot by the microscopic camera in real time, then comparing and analyzing the plankton multidimensional image set with the classification image set of the image library, and returning an analysis result.
In the scheme, the whole online automatic identification system for the plankton multidimensional imaging comprises a sample collection unit (a sampling assembly), a sample preparation unit (a sample preparation assembly), an optical imaging unit (at least three multiple compound objective lenses and a microscope camera), an analysis processing unit (an analysis module) and a core control unit (a control center and a driving assembly). A plurality of capillary flow tubes which can meet the use requirement of a maintenance period are pre-arranged in a sample preparation unit (for example, one maintenance period is a quarter, or a month, sampling is carried out every quarter or a month, after each sampling, each sample detects two tubes (two capillary flow tubes are adopted), the results of the two tubes are averaged, if the error is more than 15%, the sample detects one tube more, and if the number of sampling products is 10, at least 30 capillary flow tubes are arranged in each maintenance period); the driving component drives the XY axis moving platform to operate, after the mechanical arm is driven to move to a proper position, the front end of the mechanical arm extends to the upper end of a capillary circulation tube and is clamped by a clamp to lift, the lower end of the capillary circulation tube extends into a sample in the sample pool, the time is suspended for a certain time, after the capillary circulation tube automatically absorbs the sample, the mechanical arm is lifted, the XY axis moving platform drives the mechanical arm to move to a position right opposite to an objective table, the mechanical arm inserts the lower end of the capillary circulation tube into a blind hole to be fixed, then the driving component drives an animal lens disc to slowly lift or descend, 360-degree multi-dimensional observation is carried out on the sample in the capillary circulation tube through at least three multiple times of composite objective lenses, and a panoramic picture of each biological particle is obtained through a microscope camera; the microscopic camera receives and stores a plankton multi-dimensional image set (panoramic picture) for analysis, comparison or classification, density calculation and the like by an analysis module; and a multiple compound objective lens in the optical imaging unit transmits the sample in the capillary flow tube to an analysis processing unit, and identification and analysis software automatically classifies each picture and counts the number or density of each phytoplankton. All units are integrally controlled by a control center, and data results are transmitted to a data platform appointed by a user through the Internet of things. And the identified capillary flow-through tube is moved to a recovery platform by the mechanical arm and the XY-axis moving platform for check.
As shown in fig. 3, in a preferred embodiment, the method further includes: a waste liquid outlet pipe 1011, one end of which is communicated to the bottom of the sample cell and the other end of which extends outwards; the sampling pipes comprise a first sampling pipe 1012 arranged on the water inlet side of the peristaltic pump and a second sampling pipe 1013 arranged on the outlet side of the peristaltic pump; the tee joint 104 is arranged between the first sample inlet pipe and the peristaltic pump, a first water inlet of the tee joint is communicated to a sample outlet of the first sample inlet pipe, and a water outlet of the tee joint is communicated to a water inlet of the peristaltic pump; the cleaner comprises a storage cavity 105, purified water is prestored in the storage cavity, a water outlet of the storage cavity is communicated with a second water inlet of the tee joint through a water outlet pipe, and the driving part drives the cleaner to clean at regular time. In the scheme, a tee joint is arranged to connect a sample pool, a peristaltic pump and a cleaner, when a sample needs to be fed, a water outlet pipe of the cleaner is closed, only the sample (a sample bottle 106) is fed, and after detection of one sample is finished, waste liquid is discharged through a waste liquid outlet pipe; and then, the cleaner is started again, pure water in the storage cavity is pumped into the sample cell through the peristaltic pump, the sample tube and the sample cell are cleaned, and pollution among samples can be effectively avoided.
In a preferred embodiment, the method further comprises: the sample pool and the recovery platform are respectively arranged in two accommodating cavities in the thermostat; the detection ends of the two temperature sensors respectively extend into the sample cell and above the recovery platform; the thermostat is in communication connection with the control center. In the scheme, the thermostat is used for keeping the constant temperature of the sample pool and the recovery platform, such as the conventional room temperature of 20 ℃ and the like, so that the loss of plankton in the sample is avoided.
In a preferred scheme, the objective table and the objective disc are both made of transparent materials, the capillary flow tubes are made of quartz materials, and numbers are respectively marked at two ends of any capillary flow tube; and when the microscopic camera collects the plankton multidimensional image set in any capillary flow tube, synchronously collecting and storing the serial numbers on the corresponding capillary flow tubes. In this scheme, objective table and objective disc are transparent material, even if insert the sample in the capillary flow tube lower extreme of blind hole and also can carry out microscopic observation and shoot, guarantee to detect accurately.
As shown in fig. 1, in a preferred embodiment, the XY axis moving stage includes: the cross beam 3011 is arranged on the mechanical arm in a vertically sliding mode through the driving sliding block; a pair of longitudinal beams 3012, two ends of a cross beam are arranged on the pair of longitudinal beams in a sliding mode through driving sliding blocks; and the lower ends of the at least two support columns 3013 are fixed on the base, and the upper ends of the at least two support columns are fixed on the pair of longitudinal beams. In the scheme, a pair of longitudinal beams supports a cross beam to realize the forward and backward movement (the movement on the Y axis in the plane) of the cross beam on the horizontal plane, and the upper end of the mechanical arm is clamped on the cross beam through a sliding block to realize the sliding on the cross beam (the movement on the X axis in the plane) so as to meet the positioning and moving requirements of the mechanical arm.
In a preferred embodiment, as shown in fig. 1, the objective disc comprises two annular bodies, and at least three multiple compound lenses are sandwiched between the two annular bodies; a column 404 vertically fixed on the base; the slider 405, one end is fixed to two rings and the other end is slidably disposed on the post. In the scheme, the slide block is driven by the driving assembly to synchronously drive at least three times of composite objective lenses clamped between the two annular bodies to move upwards or downwards slowly and stably.
As shown in fig. 1, in a preferred embodiment, the sample stage and the recovery stage are both plate bodies with a plurality of insertion holes uniformly distributed above, and a plurality of capillary flow tubes are inserted into the plurality of insertion holes on the sample stage in a one-to-one correspondence manner; the capillary flow tubes with the detected samples are inserted into the jack holes on the recovery platform in a one-to-one correspondence mode. In the scheme, the sample stage and the recovery stage can effectively separate and fix the capillary flow-through tubes, so that cross infection of samples is avoided, and the accuracy of detection results is ensured; the depth of the plurality of jacks is smaller than the length of the capillary flow-through tube so as to facilitate the clamping of a clamp at the front end of the mechanical arm.
In a preferred scheme, the depth of the blind hole is less than or equal to 1/10 of the length of any capillary flow-through tube; the volume of any capillary flow tube is more than or equal to 0.1ml.
In a preferred scheme, the depth of the blind hole is equal to 1/10 of the length of any capillary flow-through pipe; the volume of any capillary flow-through is 0.1ml, 0.5ml or 1ml. In the scheme, the capillary flow tubes with different volumes can meet the detection requirements of different samples, and are suitable for various practical detection application scenes.
In a preferred scheme, the plankton multidimensional image set comprises a plurality of panoramic pictures obtained by shooting the same plankton at multiple angles, a plurality of panoramic pictures obtained by shooting different targets at multiple angles and numbers of corresponding capillary flow tubes; the analysis result is the classification result of the plankton corresponding to the number, and the number or the density of each plankton. In the scheme, samples of capillary flow tubes of different tubes can be numbered and automatically classified and analyzed according to the numbers, the number or the density can be accurately and effectively counted and calculated, and the detection efficiency can be effectively guaranteed while the detection precision is improved.
Example 1
As shown in fig. 4, an on-line automatic identification system for plankton multidimensional imaging comprises:
(1) The sample collection unit (sampling component) is used for collecting and preprocessing a water sample and is a basic link for realizing the automatic identification of phytoplankton. The peristaltic pump technology is adopted, the minimum water inflow can be as low as 0.01mL, and the automatic collection function of a trace water sample is realized through the core control unit.
The sample collection unit comprises a peristaltic pump, a pipeline, a filter, a sample pool, a cleaner and a sensor. The peristaltic pump collects a water sample to the filter through the pipeline, and the filtered water sample is collected to the sample pool for the sample preparation unit to prepare. The system can set a certain time period to clean the sample cell, the automatic cleaning of the sample cell can be realized through the cleaning device, meanwhile, the sensor can detect one or more parameters such as the water temperature of the sample cell, and the parameters can be stored together with a final test result, so that the system provides help for analysis work. In order to better ensure the living environment of plankton, the sample collection unit can also be equipped with a thermostat to balance the water temperature of the sample pool.
(2) The sample preparation unit (sample preparation component) is a main unit for embodying the automatic quantitative preparation of the sample, the sample is made into a state which can be used for visual observation, and the core element of the sample preparation unit is a cylindrical capillary flow tube with fixed volume and is made of transparent quartz material. The two ends of each flow tube have been marked with numbers to correspond to the sample numbers. The sample preparation unit is pre-loaded with a fixed volume (e.g., 0.1ml, 0.5ml, 1ml, etc.) of capillary flow-through tubes for a maintenance cycle, one capillary flow-through tube for each monitoring. And the mechanical conveying device grabs the capillary flow-through tube and sucks the sample into the capillary flow-through tube, and after the suction is finished, the capillary flow-through tube is placed on the sample table for visual observation. And after the observation is finished, the capillary circulation tube enters the recovery platform to be sealed and is recovered into the recovery box.
The application innovatively adopts the capillary flow tube to load the sample, is the best scheme for cooperating with the annular objective lens disk to observe, is also suitable for the optical angle of 360-degree multi-dimensional observation, and can show the three-dimensional posture of the particles without damage. During observation, the objective disc drives at least three times of composite objectives to uniformly translate around the capillary flow tube, and a water sample is relatively static in the capillary flow tube, so that all samples can be observed completely, and the problem of blockage of a flow cell is avoided.
(3) The optical imaging unit (at least three multiple compound objective lenses and a micro camera), the sample preparation unit is a main unit for embodying the automatic quantitative preparation of the sample, the sample is made into a state capable of being used for visual observation, and the core element of the optical imaging unit is a columnar capillary flow tube with fixed volume and made of transparent quartz material. The two ends of each flow tube have been marked with numbers to correspond to the sample numbers. The sample preparation unit is pre-loaded with a fixed volume (e.g., 0.1ml, 0.5ml, 1ml, etc.) of capillary flow-through tubes for a maintenance cycle, one capillary flow-through tube for each monitoring. The mechanical conveying device grabs the capillary flow tube and sucks the sample into the capillary flow tube, and after the suction is finished, the capillary flow tube is placed on a sample table (an object table) for visual observation. And after the observation is finished, the capillary circulation tube enters the recovery platform to be sealed and is recovered into the recovery box.
(4) An analysis processing unit (analysis module) refers to identification software and a biological species library. The establishment of the species library is that plankton samples of different water bodies are collected and purchased, and determined species images are input into identification software through dual identification of an identification instrument and a traditional microscope, so that automatic classification of observation samples is realized. The establishment of plankton libraries can not only realize the identification of samples rapidly, but also have profound significance for the classification work of aquatic organisms in different water bodies in China.
The identification software comprises an instrument control function, an image acquisition function, an image analysis function, a data storage function, a result reporting function and a network transmission function. The control function of the instrument realizes the control of each module of the instrument, and the Ethernet, RS-232, RS-485 and other buses are adopted for communication and transmission. The communication protocol is established according to each module. The image acquisition function is to acquire images in real time through an acquisition interface provided by the high-definition microscopic camera, and the acquired images are displayed on a software main interface. The image analysis function is to perform real-time or selective analysis on the acquired images through various algorithms, and can perform operations such as screening, database comparison and the like according to conditions set by a user. The data storage function is to store the acquired images and analysis results into a database. The result reporting function can generate professional report templates, can freely set report contents and formats, and can export Excel or Word document formats. The network transmission function provides an interface for the cloud platform, and the detection result 'cloud on key' can be butted with various cloud platforms.
(5) Core control unit (control center and drive assembly). The core control unit carries out systematic control on the units, and the guarantee for realizing logic operation and automation is realized. The unit comprises an embedded system, a motor driver, a signal collector and a switching value controller. The embedded system runs the core control software to control the whole instrument. The motor driver is responsible for controlling the action of mechanical transmission device (XY axle moving platform and arm), and the signal collector is responsible for gathering various signals, contains: photoelectric switch, hall element, temperature, etc., and the switching value controller is responsible for controlling various on-off. These modules constitute the core control unit.
While embodiments of the invention have been described above, it is not intended to be limited to the details shown, described and illustrated herein, but is to be accorded the widest scope consistent with the principles and novel features herein disclosed, and to such extent that such modifications are readily available to those skilled in the art, and it is not intended to be limited to the details shown and described herein without departing from the general concept as defined by the appended claims and their equivalents.
Claims (9)
1. An on-line automatic identification system for multi-dimensional imaging of plankton, characterized by comprising:
a base;
the sampling assembly is arranged on the base and comprises a sample pool, and the sample pool is communicated to a water body to be detected through a sample inlet pipe; the peristaltic pump and the filter are arranged on the sampling pipe;
the sample preparation assembly is arranged on the base and comprises a plurality of capillary flow tubes, the capillary flow tubes are uniformly inserted on the sample table at intervals, and the sample table is close to the sample pool; the recovery platform is detachably inserted with a plurality of capillary flow-through tubes with detected samples; the mechanical transmission assembly comprises an XY axis moving platform which is arranged on the base; the mechanical arm is arranged on the XY-axis moving platform, the front end of the mechanical arm is provided with a clamp, and the front end of the clamp is detachably clamped with any capillary flow tube;
an optical imaging assembly including a stage disposed on a base; the blind hole is formed on the objective table, and the lower end of any capillary flow-through pipe is detachably inserted into the blind hole; the objective lens disc is arranged on the base in a lifting way; the objective lens disc is uniformly distributed around the objective table in a divergent manner by taking any point on an extension line where the blind hole is located as a circle center, the plane where the at least three multi-fold composite objective lenses are located is perpendicular to the extension line where the blind hole is located, and the optical axes of the at least three multi-fold composite objective lenses are intersected at any point on the extension line where the blind hole is located; the objective disc comprises two annular bodies, and at least three multiple compound mirrors are clamped between the two annular bodies; the upright post is vertically fixed on the base; one end of the sliding block is fixed on the two annular bodies, and the other end of the sliding block is arranged on the stand column in a sliding manner; the microscopic camera is in communication connection with the at least three multiple compound objective lenses, the at least three multiple compound objective lenses and the objective lens disc synchronously lift, and a plankton multidimensional image set inserted in any capillary flow tube on the blind hole is synchronously acquired through the microscopic camera;
the driving assembly is used for respectively driving the peristaltic pump, the XY-axis moving platform, the mechanical arm, the microscope camera and the objective disc;
the control center is in wireless communication connection with the microscopic camera and the driving assembly, and further comprises an image library, wherein a classification atlas of the plankton sample is prestored in the image library; and the analysis module is used for acquiring the plankton multidimensional image set shot by the microscopic camera in real time, then comparing and analyzing the plankton multidimensional image set with the classification image set of the image library, and returning an analysis result.
2. The on-line automatic plankton multidimensional imaging identification system of claim 1, further comprising:
one end of the waste liquid outlet pipe is communicated to the bottom of the sample pool, and the other end of the waste liquid outlet pipe extends outwards; the sampling pipes comprise a first sampling pipe arranged on the water inlet side of the peristaltic pump and a second sampling pipe arranged on the outlet side of the peristaltic pump;
the three-way valve is arranged between the first sample inlet pipe and the peristaltic pump, a first water inlet of the three-way valve is communicated to a sample outlet of the first sample inlet pipe, and a water outlet of the three-way valve is communicated to a water inlet of the peristaltic pump;
the cleaner comprises a storage cavity, purified water is prestored in the storage cavity, a water outlet of the storage cavity is communicated with a second water inlet of the tee joint through a water outlet pipe, and the driving assembly drives the cleaner to clean at regular time.
3. The on-line automatic plankton multidimensional imaging identification system of claim 1, further comprising: the sample pool and the recovery platform are respectively arranged in two accommodating cavities in the thermostat; the detection ends of the two temperature sensors respectively extend into the sample cell and above the recovery platform; the thermostat is in communication connection with the control center.
4. The on-line automatic plankton multidimensional imaging identification system of claim 1, wherein the objective table and the objective disk are made of transparent materials, the capillary flow tubes are made of quartz materials, and numbers are respectively marked on two ends of any capillary flow tube; and when the microscopic camera acquires the plankton multidimensional image set in any capillary flow tube, synchronously acquiring and storing the serial numbers on the corresponding capillary flow tubes.
5. The on-line automatic plankton multidimensional imaging identification system according to claim 1, wherein the XY axis moving platform comprises:
the mechanical arm is arranged on the cross beam in a vertically sliding mode through the driving sliding block;
the two ends of a cross beam are arranged on the longitudinal beams in a sliding mode through driving sliding blocks;
the lower ends of the at least two support columns are fixed on the base, and the upper ends of the at least two support columns are fixed on the pair of longitudinal beams.
6. The on-line automatic plankton multidimensional imaging identification system of claim 1, wherein the sample stage and the recovery stage are both plate bodies with a plurality of jacks uniformly distributed above, and a plurality of capillary flow tubes are inserted into the jacks on the sample stage in a one-to-one correspondence manner;
the capillary flow tubes with the detected samples are inserted into the jack holes on the recovery platform in a one-to-one correspondence mode.
7. The on-line automatic plankton imaging system according to claim 1, wherein the depth of the blind hole is less than or equal to 1/10 of the length of any capillary flow tube; the volume of any capillary flow-through tube is more than or equal to 0.1ml.
8. The on-line automatic plankton multidimensional imaging identification system as recited in claim 1, wherein the depth of the blind hole is equal to 1/10 of the length of any capillary flow tube; the volume of any capillary flow-through is 0.1ml, 0.5ml or 1ml.
9. The system for the on-line automatic identification of the plankton multidimensional imaging as claimed in claim 1, wherein the plankton multidimensional image set comprises a plurality of panoramic pictures obtained by shooting the same plankton at multiple angles, a plurality of panoramic pictures obtained by shooting different targets at multiple angles, and the numbers of the corresponding capillary flow tubes; the analysis result is the plankton classification result corresponding to the number, and the number or the density of each plankton.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210608361.9A CN114965472B (en) | 2022-05-31 | 2022-05-31 | On-line automatic identification system for multi-dimensional imaging of plankton |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210608361.9A CN114965472B (en) | 2022-05-31 | 2022-05-31 | On-line automatic identification system for multi-dimensional imaging of plankton |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114965472A CN114965472A (en) | 2022-08-30 |
| CN114965472B true CN114965472B (en) | 2023-03-14 |
Family
ID=82956931
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210608361.9A Active CN114965472B (en) | 2022-05-31 | 2022-05-31 | On-line automatic identification system for multi-dimensional imaging of plankton |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114965472B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115656209B (en) * | 2022-12-12 | 2023-05-30 | 四川吉利学院 | PLC-based visual detection system for automobile manufacturing electrified equipment |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108507625A (en) * | 2018-06-21 | 2018-09-07 | 水利部中国科学院水工程生态研究所 | A kind of water body planktonic organism intellectual monitoring platform and method for integrating Internet of Things |
| CN110333236A (en) * | 2019-07-19 | 2019-10-15 | 中国海洋大学 | Planktonic organism multi-angle of view in situ detection device |
| CN111443028A (en) * | 2020-04-26 | 2020-07-24 | 长江流域水环境监测中心 | Automatic monitoring equipment and method for floating algae based on AI technology |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| AU6580700A (en) * | 1999-08-13 | 2001-03-13 | Cartesian Technologies, Inc. | Apparatus for liquid sample handling |
-
2022
- 2022-05-31 CN CN202210608361.9A patent/CN114965472B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN108507625A (en) * | 2018-06-21 | 2018-09-07 | 水利部中国科学院水工程生态研究所 | A kind of water body planktonic organism intellectual monitoring platform and method for integrating Internet of Things |
| CN110333236A (en) * | 2019-07-19 | 2019-10-15 | 中国海洋大学 | Planktonic organism multi-angle of view in situ detection device |
| CN111443028A (en) * | 2020-04-26 | 2020-07-24 | 长江流域水环境监测中心 | Automatic monitoring equipment and method for floating algae based on AI technology |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114965472A (en) | 2022-08-30 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111443028B (en) | Automatic monitoring equipment and method for floating algae based on AI technology | |
| CN108507625B (en) | Intelligent monitoring platform and method for water plankton integrating Internet of things | |
| CN107287119B (en) | Cell culture counting assembly | |
| CN114965472B (en) | On-line automatic identification system for multi-dimensional imaging of plankton | |
| CN110132989A (en) | A kind of distress in concrete detection device, method and terminal system | |
| CN102719357A (en) | Hybridization system for real-time monitoring parsing process of micro array chip | |
| CN114689384B (en) | River sewage outlet inspection device | |
| CN209656721U (en) | A kind of intelligent sample injection unit of extracorporeal diagnostic instrument | |
| CN208588728U (en) | A kind of water conservancy water quality monitoring and remote supervisory equipment | |
| CN202421056U (en) | Rapid plankton monitoring device | |
| Pollina et al. | PlanktoScope: affordable modular quantitative imaging platform for citizen oceanography | |
| CN217586937U (en) | Multi-dimensional imaging identifier for plankton | |
| CN217521029U (en) | Multi-dimensional imaging device for identifying plankton | |
| CN114112534B (en) | Intelligent sampling and detecting device suitable for microorganism suspension and solid culture medium | |
| CN114778419A (en) | High-magnification optical magnification imaging flow cytometer | |
| US20200158717A1 (en) | High-throughput imaging platform | |
| CN113092346B (en) | Algae cell counting detection system and detection method thereof | |
| Dai et al. | Cell image reconstruction for a lens-free imaging system based on linear array sensor | |
| JPH0246281A (en) | Automatically microorganism observing device | |
| CN206863042U (en) | A kind of Portable seawater nutritive salt automatical analysis device | |
| Stelzer | Automated system for sampling, counting, and biological analysis of rotifer populations | |
| CN111521753A (en) | Rice field is synthesized and is planted technology ecological remediation soil heavy metal pollution and administer detection device | |
| Smith | Image analysis for plant cell culture and micropropagation | |
| CN116296678A (en) | Automatic monitoring device and method for floating algae in water | |
| CN218976740U (en) | Biological growth image acquisition 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 |