Intelligent plankton analysis equipment

Abstract

The invention discloses intelligent plankton analysis equipment which is characterized in that automatic loading of plankton samples is automatically completed through an automatic sampling system, then image scanning is carried out on the plankton samples through an automatic scanning system, and finally intelligent identification analysis is carried out based on scanned images through an intelligent analysis system. The whole process is automatically carried out, the phytoplankton and the zooplankton can be analyzed at the same time only through one sample, the detection time is greatly shortened, the efficiency of plankton population analysis is improved, human intervention is not needed, the recognition and analysis process is not affected by subjective factors, and the method is suitable for popularization and application. And the consistency and the accuracy of analysis results are ensured.

Description

Technical Field

[0001] The present invention relates to the technical field of water ecological environment monitoring, and in particular to a plankton intelligent analysis device. Background Art

[0002] Plankton refers to organisms that cannot actively migrate horizontally over long distances. Most are tiny, often invisible to the naked eye, and lack or have very weak swimming abilities, relying solely on currents, waves, or water circulation for locomotion. They specifically include phytoplankton and zooplankton. Phytoplankton, the primary producers in lakes, are characterized by their diversity, abundance, and rapid reproduction, playing a crucial role in aquatic ecosystems. The ecological characteristics of phytoplankton, such as their species composition and abundance distribution, can sensitively reflect changes in complex environmental factors. Therefore, their community structure and changing trends are crucial research topics for water environment monitoring, water quality assessment, water pollution research, and aquatic ecological restoration and management. Zooplankton, on the other hand, are a group of animals that swim in the water and lack or have weak swimming abilities. They serve as a vital food source for fish and other commercial animals in the pelagic layer and are crucial for the development of fisheries. The distribution of many zooplankton species is linked to climate, making them useful indicators of warm and cold currents. Many species also serve as indicators of water pollution. Current plankton population analysis relies primarily on professional observation and counting of cells or individuals under a microscope. The entire process includes on-site sample collection, sample pretreatment, sample loading, microscopic observation, and counting analysis. This process is tedious and time-consuming, and given the vast variety of plankton species found in the natural environment, personnel capable of performing this analysis must undergo extensive professional training and hands-on training. These limitations result in low efficiency in current plankton population analysis, as well as subjectivity in the analysis process, making it difficult to guarantee the consistency and accuracy of the results. Summary of the Invention

[0003] The present invention provides an intelligent plankton analysis device that can realize automated detection and identification of plankton. It can simultaneously analyze phytoplankton and zooplankton using only one sample, greatly shortening the detection time and thus improving the efficiency of plankton population analysis. In addition, it does not require human intervention, and the identification and analysis process is not affected by human subjective factors, ensuring the consistency and accuracy of the analysis results.

[0004] According to one aspect of the present invention, there is provided an intelligent plankton analysis device, comprising an automatic sampling system, an automatic scanning system and an intelligent analysis system, wherein the automatic sampling system and the automatic scanning system are electrically connected to the intelligent analysis system, and the intelligent analysis system is used to control the automatic sampling system to automatically complete the automatic loading of plankton samples, and is also used to control the automatic scanning system to scan and capture images of the loaded plankton samples, and is also used to obtain biological information of the plankton samples based on the analysis of the captured plankton sample images.

[0005] Furthermore, the automatic sampling system includes a sample storage unit, a sample circulation power unit and a circulation detection pool. The sample storage unit is used to store plankton samples, and the circulation detection pool is used to provide a resting place for scanning and photographing samples. The sample storage unit is connected to the circulation detection pool through a pipeline, and the sample circulation power unit is arranged on the pipeline connecting the sample storage unit and the circulation detection pool to provide power for sample circulation.

[0006] Furthermore, the automatic sampling system also includes a pure water storage unit, a waste liquid storage unit and a sample switching unit. The pure water storage unit is used to provide pure water for pipeline cleaning, the waste liquid storage unit is used to collect waste liquid, and the sample switching unit is used to switch sampling between the pure water storage unit and the sample storage unit.

[0007] Furthermore, the automatic sampling system also includes a pipeline liquid detection unit arranged on the pipeline connecting the sample storage unit and the circulation detection pool and electrically connected to the intelligent analysis system, for detecting whether there is liquid circulating in the connecting pipeline. During sample injection, if the pipeline liquid detection unit detects that there is no circulating liquid in the connecting pipeline, the intelligent analysis system issues an alarm reminder.

[0008] Furthermore, the automatic sampling system also includes a sample concentration quality control unit arranged on the pipeline connecting the sample storage unit and the circulation detection pool and electrically connected to the intelligent analysis system, which is used to preliminarily determine whether the sample concentration meets the analysis requirements. When the sample concentration quality control unit detects that the sample concentration does not meet the analysis requirements, the intelligent analysis system issues an alarm reminder.

[0009] Furthermore, the sample storage unit includes a sample pool, a sample temperature control module and a sample mixing module. The sample pool is used to store plankton samples, the sample temperature control module is used to provide a constant low-temperature environment for plankton samples, and the sample mixing module is used to mix the samples before sample injection.

[0010] Furthermore, the automatic scanning system includes a microscope, an objective lens switching unit, a three-axis electronically controlled platform and an automatic shooting module. The microscope is used to visually magnify the plankton sample, the objective lens switching unit is used to switch the magnification of the microscope, the automatic shooting module is used to capture the image of the plankton sample after magnification by the microscope, and the three-axis electronically controlled platform is used to drive the circulation detection pool to move in the XYZ direction to realize image scanning and automatic focal length adjustment under a preset scanning path.

[0011] Furthermore, during the automatic adjustment of the focal length, the three-axis electronically controlled platform first moves upward in the Z direction with a large step distance, and records the clarity value of the current focal plane at each focal plane. If the clarity difference between the subsequent focal plane and the previous focal plane is less than a threshold value, the next step of movement is maintained with a large step distance. If the clarity difference between the subsequent focal plane and the previous focal plane is not less than the threshold value, the next step of movement is reduced. The above process is repeated until the clarity value of the subsequent focal plane is less than the clarity value of the previous focal plane, and then the platform moves in the opposite direction with the minimum step distance until the clarity value of the focal plane decreases. The focal plane is then used as the optimal focal plane, and the three-axis electronically controlled platform stops moving in the Z direction.

[0012] Furthermore, the process of generating the preset scanning path is as follows: after presetting the number of fields of view to be captured, the front of the channel of the flow detection pool is divided into an n×m rectangular topology structure according to the preset number of fields of view, and n is ensured to be less than the channel length divided by the length of a single field of view and m is less than the channel width divided by the width of a single field of view, and any one of the four vertex grids of the rectangular topology structure is selected as the origin to generate the preset scanning path according to the Z-shaped path planning.

[0013] Furthermore, the intelligent analysis system includes a plankton recognition unit, a human-computer interaction unit and a control unit. The plankton recognition unit and the human-computer interaction unit are both electrically connected to the control unit. The plankton recognition unit is equipped with a trained plankton recognition model for performing intelligent recognition and analysis based on plankton sample images to obtain biological information of plankton samples. The human-computer interaction unit is used to realize interaction between users and devices. The control unit is used to control the automatic sampling system to automatically complete the automatic loading of plankton samples, and is also used to control the automatic scanning system to scan and photograph images of the loaded plankton samples, and is also used to control the plankton recognition unit to analyze the photographed plankton sample images to obtain biological information of plankton samples.

[0014] The present invention has the following beneficial effects:

[0015] The intelligent plankton analysis equipment of the present invention first automatically completes the automatic loading of plankton samples through the automatic sampling system, then performs image scanning on the plankton samples through the automatic scanning system, and finally, performs intelligent identification and analysis based on the scanned images through the intelligent analysis system to obtain the biological information of the plankton samples. The entire process is automated, and phytoplankton and zooplankton analysis can be achieved simultaneously with only one sample, greatly shortening the detection time, thereby improving the efficiency of plankton population analysis, and without the need for human intervention. The identification and analysis process is not affected by human subjective factors, ensuring the consistency and accuracy of the analysis results.

[0016] In addition to the above-described objects, features and advantages, the present invention has other objects, features and advantages. The present invention will be further described in detail below with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] The accompanying drawings, which constitute part of this application, are intended to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are intended to explain the present invention and do not constitute an undue limitation of the present invention. In the accompanying drawings:

[0018] Figure 1 This is a schematic diagram of the module structure of the plankton intelligent analysis device according to the preferred embodiment of the present application;

[0019] Figure 2 This is a schematic diagram of the connection structure between the automatic sampling system and the automatic scanning system of the preferred embodiment of the present application;

[0020] Figure 3 This is a schematic diagram of the scanning path of the preferred embodiment of the present application.

[0021] Description of Reference Numerals

[0022] 1. Automatic sampling system; 2. Automatic scanning system; 3. Intelligent analysis system; 11. Sample storage unit; 12. Sample circulation power unit; 13. Circulation detection pool; 14. Pure water storage unit; 15. Waste liquid storage unit; 16. Sample switching unit; 17. Pipeline liquid detection unit; 18. Sample concentration quality control unit; 19. Pathway flow direction switching unit. DETAILED DESCRIPTION

[0023] It should be noted that, in the absence of conflict, the embodiments and features of the embodiments in this application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0024] Reference Figure 1A preferred embodiment of the present application provides a plankton intelligent analysis device, including an automatic sampling system 1, an automatic scanning system 2 and an intelligent analysis system 3. The automatic sampling system 1 and the automatic scanning system 2 are electrically connected to the intelligent analysis system 3. The intelligent analysis system 3 is used to control the automatic sampling system 1 to automatically complete the automatic loading of plankton samples, and is also used to control the automatic scanning system 2 to scan and capture images of the loaded plankton samples, and is also used to obtain biological information of the plankton samples based on the captured plankton sample images.

[0025] It can be understood that the plankton intelligent analysis equipment of this embodiment first automatically completes the automatic loading of plankton samples through the automatic sampling system 1, then performs image scanning on the plankton samples through the automatic scanning system 2, and finally, performs intelligent identification and analysis based on the scanned image through the intelligent analysis system 3 to obtain the biological information of the plankton sample. The entire process is automated, and phytoplankton and zooplankton analysis can be achieved simultaneously with only one sample, which greatly shortens the detection time, thereby improving the efficiency of plankton population analysis, and does not require human intervention. The identification and analysis process is not affected by human subjective factors, ensuring the consistency and accuracy of the analysis results.

[0026] Among them, such as Figure 2 As shown, the automatic sampling system 1 includes a sample storage unit 11, a sample circulation power unit 12, and a circulation detection cell 13. The sample storage unit 11 is used to store plankton samples, and the circulation detection cell 13 is used to provide a resting place for scanning and photographing the samples. The sample storage unit 11 is connected to the circulation detection cell 13 via a pipeline. The sample circulation power unit 12 is disposed on the pipeline connecting the sample storage unit 11 and the circulation detection cell 13 to provide power for sample circulation. It can be understood that the sampling process of the automatic sampling system 1 is as follows: the sample circulation power unit 12 is started, and the plankton sample is transported from the sample storage unit 11 to the circulation detection cell 13. When the sample is transported to the circulation detection cell 13 and fills the circulation detection cell 13, the sample circulation power unit 12 stops working and waits for the completion of subsequent image acquisition and intelligent analysis.

[0027] Preferably, the automatic sampling system 1 further includes a pure water storage unit 14, a waste liquid storage unit 15, and a sample switching unit 16. The pure water storage unit 14 is used to provide pure water for pipeline cleaning, the waste liquid storage unit 15 is used to collect waste liquid, and the sample switching unit 16 is used to switch sampling between the pure water storage unit 14 and the sample storage unit 11. The sample switching unit 16 is connected to the circulation detection cell 13 via a pipeline, and the circulation detection cell 13 is also connected to the waste liquid storage unit 15 via a pipeline. The sample circulation power unit 12 is provided on the pipeline connecting the sample switching unit 16 and the circulation detection cell 13, and is used to provide power for sample circulation and pure water circulation. It can be understood that the automatic sampling process of the automatic sampling system 1 is as follows: the sample switching unit 16 first switches to the sample storage unit 11 for sampling, and at the same time starts the sample circulation power unit 12, and the plankton sample is transported to the circulation detection pool 13 through the sample switching unit 16. When the sample is transported to the circulation detection pool 13 and fills the circulation detection pool 13, the sample circulation power unit 12 stops working; after the image acquisition and intelligent analysis are completed, the sample switching unit 16 switches to the pure water storage unit 14 for sampling, and at the same time starts the sample circulation power unit 12 again, and uses pure water to clean the pipeline and the circulation detection pool 13, and the waste liquid after cleaning flows back to the waste liquid storage unit 15; after cleaning is completed, the sample circulation power unit 12 stops working and waits for the next sampling.

[0028] The sample storage unit 11 specifically includes a sample pool, a sample temperature control module and a sample mixing module. The sample pool is used to store plankton samples, the sample temperature control module is used to provide a constant low-temperature environment for plankton samples, and the sample mixing module is used to mix the samples before sample injection. The sample pool provides multiple sample storage areas, which can store multiple plankton samples at the same time. The sample temperature control unit provides a constant low-temperature environment for the storage of plankton samples, generally 4°C-10°C, to prevent the plankton samples from changing while waiting for the test, ensuring the representativeness of the samples. The sample mixing module can mix the samples horizontally before sample injection to ensure the uniformity of the samples and prevent the samples from being uneven due to stratification after standing, which is conducive to improving the accuracy of the test results.

[0029] In addition, the sample switching unit 16 specifically includes a drive module and a sampling needle. The sampling needle is installed on the drive module and connected to the sample circulation detection pool 13 through a pipeline. The drive module is located above the pure water storage unit 14 and the sample storage unit 11. When sampling is required, the drive module drives the sampling needle to move in the XY direction. After moving to the top of the pure water storage unit 14 or the sample storage unit 11, the drive module drives the sampling needle downward in the Z direction, so that the sampling needle extends into the sample pool of the pure water storage unit 14 or the sample storage unit 11. After the sample circulation power unit 12 is started, the sampling of pure water or sample can be achieved. In addition, the sample circulation power unit 12 preferably adopts a peristaltic pump. Among them, the drive module adopts a three-way sliding platform or a manipulator.

[0030] Optionally, the automatic sampling system 1 also includes a pipeline liquid detection unit 17 provided on the pipeline connecting the sample storage unit 11 and the circulation detection pool 13 and electrically connected to the intelligent analysis system 3, for detecting whether there is liquid circulating in the connecting pipeline. When the plankton sample is injected, if the pipeline liquid detection unit 17 detects that there is no circulating liquid in the connecting pipeline, the intelligent analysis system 3 issues an alarm reminder. Among them, the pipeline liquid detection unit 17 can adopt a liquid level sensor. In addition, the pipeline liquid detection unit 17 is preferably provided on the pipeline connecting the sample switching unit 16 and the circulation detection pool 13. When the automatic sampling system 1 switches to pure water injection through the sample switching unit 16, if the pipeline liquid detection unit 17 detects that there is no circulating liquid in the connecting pipeline, the intelligent analysis system 3 will also issue an alarm reminder, thereby realizing the detection of water shortage of pure water.

[0031] Optionally, the automatic sampling system 1 further includes a sample concentration quality control unit 18 disposed on a pipeline connecting the sample storage unit 11 and the circulation detection cell 13 and electrically connected to the intelligent analysis system 3, for preliminarily determining whether the sample concentration meets the analysis requirements. When the sample concentration quality control unit 18 detects that the sample concentration does not meet the analysis requirements, the intelligent analysis system 3 issues an alarm. For example, when the sample concentration quality control unit 18 detects that the sample concentration is too low or too high, the intelligent analysis system 3 issues an alarm to remind staff to intervene manually to avoid errors in subsequent analysis results due to excessively low or high sample concentrations.

[0032] The sample concentration quality control unit 18 includes a micro-fluorescence detection cell, a xenon lamp, and a photomultiplier tube. Specifically, the micro-fluorescence detection cell is installed on the pipeline connecting the sample storage unit 11 and the circulation detection cell 13. Optionally, the micro-fluorescence detection cell is a quartz long cube detection cell, fixed to the detection cell seat to ensure that the detection cell is stable and light-transmitting on all four sides. A xenon lamp is provided in a direction perpendicular to one of the surfaces of the micro-fluorescence detection cell, which can emit light of a broad spectrum wavelength. The light is concentrated by a focusing convex mirror and then filtered by an excitation filter into excitation light of a preset wavelength. Optionally, the preset wavelength is 400-625nm. The excitation light is incident on the sample perpendicular to one surface of the micro-fluorescence detection cell. Chlorophyll a in the sample is excited by the excitation light to produce fluorescence of a specific wavelength band, which is emitted from a direction perpendicular to the incident surface. After being filtered by an emission filter, it is focused by an emission lens into a concentrated beam of light and irradiated to the photomultiplier tube. The photomultiplier tube converts the optical signal into an electrical signal and transmits it to the intelligent analysis system 3. The intensity of the emitted fluorescence is proportional to the concentration of chlorophyll a in the sample. The intelligent analysis system 3 can infer the concentration of chlorophyll a in the sample based on the detected fluorescence intensity, and then infer the algae density of the sample. When the algae density of the plankton sample is too high or too low, timely feedback is given or the process is terminated and the sample is re-prepared to avoid missed detections and false detections due to overlapping algae cells caused by excessively high algae concentration in the sample, and to avoid subsequent large amounts of invalid shooting and analysis due to excessively low algae concentration in the sample.

[0033] Optionally, the circulation detection pool 13 includes at least two circulation channels, which can be a circulation pool including at least two circulation channels, or at least two circulation pools, each of which includes only one circulation channel. The automatic sampling system 1 also includes a passage flow direction switching unit 19 arranged on a pipeline connecting the sample switching unit 16 and at least two circulation channels, for controlling the switching and circulation of samples or pure water between at least two circulation channels, and the passage flow direction switching unit 19 is electrically connected to the intelligent analysis system 3. For example, the number of circulation channels is two, and the passage flow direction switching unit 19 is a three-way solenoid valve, which is respectively connected to the peristaltic pump and the two circulation channels. By controlling the three-way solenoid valve, the sample or pure water transported by the peristaltic pump can be switched and circulated between the two circulation channels. In addition, the two circulation channels are also connected to the waste liquid storage unit 15 through another three-way solenoid valve. When injecting a sample into the first channel, the two three-way solenoid valves are controlled to switch to the first state, that is, switch to connect with the first channel. When injecting a sample into the second channel, the two three-way solenoid valves are controlled to switch to the second state, that is, switch to connect with the second channel. Switching between the two channels for injection is achieved by controlling the valve core switching of the two three-way solenoid valves. In addition, a sample circulation power unit 12 is also provided between the waste liquid storage unit 15 and the circulation detection cell 13 to facilitate the rapid collection of waste liquid.

[0034] In addition, the flow detection cell 13 is a three-piece glass structure, with the bottom layer being a glass substrate, the middle layer being a channel interlayer, and the top layer being a glass cover. The channel interlayer is a glass plate of a specific thickness (the thickness can be adjusted according to the required channel height) with a long strip opening in the middle along its length. The glass cover is made of optical quartz glass with a thickness of 0.16mm to 0.2mm, which can provide the working focal length required for high-power microscopic observation while ensuring optical clarity for observation and photography. The three pieces of glass are bonded together in sequence, with the long strip opening in the middle forming a long channel of a certain height. The middle section of the long channel is parallel, with both ends chamfered to prevent the deposition of plankton and other particulate impurities in the long channel. Holes are opened on both sides of the glass substrate from the middle of both ends and are opened toward the top of the glass substrate. The two openings on the top are located at the two ends of the long channel respectively. The openings on both sides of the glass substrate are connected to two metal tubes respectively. The channel inside the glass substrate connects the metal tubes to the long channel, so that the plankton sample or pure water can flow from the metal tubes into the long channel of the circulation detection pool 13. Optionally, the circulation detection pool 13 includes three channels of different sizes. Optionally, the channel heights are a, b, and c, respectively, and a<b<c. ​​Among them, the channel with a height of a is suitable for the analysis of phytoplankton and protozoa, the channel with a height of b is suitable for the analysis of rotifers, and the channel with a height of c is suitable for the analysis of cladocerans and copepods. The three channels are arranged in parallel and interconnected. When the sample is injected, the three channels will be filled in turn. The microscope can scan over the three channels to obtain images of different types of plankton. Preferably, a is 0.25 mm, b is 1 mm, and c is 5 mm.

[0035] In addition, the automatic scanning system 2 includes a microscope, an objective lens switching unit, a three-axis electronically controlled platform and a camera. The microscope is used to visually magnify the plankton sample, the objective lens switching unit is used to switch the magnification of the microscope, and the camera is used to capture the image of the plankton sample after magnification by the microscope. The circulation detection pool 13 is installed on the three-axis electronically controlled platform, and the three-axis electronically controlled platform is used to drive the circulation detection pool 13 to move in the XYZ direction. It can perform image scanning according to a preset scanning path and move the preset field of view area to below the microscope objective lens, so that the bottom of the channel in the circulation detection pool 13 moves to the optimal working distance of the microscope, so as to realize image scanning and automatic adjustment of the focal length under the preset scanning path.

[0036] In addition, during the automatic scanning process, the number of fields of view to be photographed needs to be preset, and the intelligent analysis system 3 will plan a reasonable shooting route based on the preset number of fields of view. The generation process of the preset scanning path is as follows: after the number of fields of view to be photographed is preset, the channel front of the flow detection pool 13 is divided into an n×m rectangular topological structure according to the preset number of fields of view, and n is ensured to be less than the channel length divided by the length of a single field of view and m is less than the channel width divided by the width of a single field of view, and any one of the four vertex grids of the rectangular topological structure is selected as the origin to generate the preset scanning path according to the Z-shaped path planning. For example, Figure 3 As shown, according to the preset field of view number S, the front of the strip channel of the flow detection pool 13 is disassembled into a rectangular topological structure of S=n×m. When S is not a multiple of 5, it is filled in with the multiple of 5 closest to it. During the disassembly process, a disassembly combination with n and m as close as possible is selected, n≥m, n represents the disassembly in the length direction (i.e., X direction) of the strip channel, and m represents the disassembly in the width direction (i.e., Y direction) of the strip channel, and ensure that n is less than the strip channel length L divided by the length of a single field of view, and m is less than the strip channel width W divided by the width of a single field of view, and L / n is used as the single-step displacement distance in the length direction, and W / m is used as the single-step displacement distance in the width direction, and the upper right corner grid of the rectangular topological structure is used as the origin. When the three-axis electric control platform moves, it first moves to the origin through the XY direction, and in the first The field of view is focused and photographed, and then the three-axis electronically controlled platform moves L / n in the X direction away from the origin to capture the image of the second field of view, and so on, until n fields of view are captured in the X direction of the channel; the three-axis electronically controlled platform moves W / m in the Y direction away from the origin to capture the image of the first field of view in the second row, and the three-axis electronically controlled platform moves L / n in the X direction towards the origin to capture the second field of view in the second row, and so on, until the capture of n fields of view in the second row is completed; the three-axis electronically controlled platform continues to move W / m in the Y direction away from the origin to capture the first field of view in the third row, and then repeats the above process to continue capturing in the X direction; and so on, according to the Z-shaped path planning, the capture of S fields of view is completed. This path planning ensures the uniformity of the field of view.

[0037] After planning the scanning path, the intelligent analysis system 3 first controls the three-axis electronically controlled platform to move in the XY direction so that the preset field of view moves to the bottom of the microscope objective lens, and then controls the three-axis electronically controlled platform to move in the Z direction to achieve automatic focus adjustment. In the process of automatic focus adjustment, the three-axis electronically controlled platform first moves upward in the Z direction with a large step distance, and records the clarity value of the current focal plane at each focal plane. If the clarity difference between the subsequent focal plane and the previous focal plane is less than a threshold, the next step is maintained at a large step distance. If the clarity difference between the subsequent focal plane and the previous focal plane is not less than the threshold, the next step distance is reduced. The above process is repeated until the clarity value of the subsequent focal plane is less than the clarity value of the previous focal plane. Then, the platform moves in the opposite direction with the minimum step distance until the clarity value of the focal plane decreases, that is, the clarity value of the N+1th focal plane is less than the clarity value of the Nth focal plane. Then, the Nth focal plane is regarded as the optimal focal plane, and the intelligent analysis system 3 controls the three-axis electronically controlled platform to stop moving in the Z direction. It can be understood that the present invention can ensure that the acquired image is the clearest focal plane through the above-mentioned automatic focus adjustment strategy, thereby realizing automatic focusing of the microscope.

[0038] In addition, the intelligent analysis system 3 includes a plankton recognition unit, a human-computer interaction unit, and a control unit. The plankton recognition unit and the human-computer interaction unit are both electrically connected to the control unit. The control system is also electrically connected to the automatic sampling system 1 and the automatic scanning system 2. The plankton recognition unit is equipped with a trained plankton recognition model and is used to perform intelligent recognition and analysis based on plankton sample images to obtain biological information such as genus and species information, biological density, and biomass of the plankton samples. The human-computer interaction unit is used to facilitate user interaction with the device. The control unit is used to control the automatic sampling system 1 to automatically complete the automatic loading of plankton samples, control the automatic scanning system 2 to capture images of the loaded plankton samples, and control the plankton recognition unit to analyze the captured plankton sample images to obtain biological information of the plankton samples. The plankton recognition model is trained using a standard plankton database, which includes a standard phytoplankton database and a standard zooplankton database. Specifically, the training process of the plankton recognition model is as follows: first, water samples are collected and photographed through a microscope. Professionals classify and label each phytoplankton cell and zooplankton individual in the photographed plankton pictures. The labeled photos are used as the standard plankton standard database to train the plankton recognition model; when using the plankton standard database for training, it is divided into a training set and a test set. The training set is used to train the model, and the test set is used to test the model. The model parameters are optimized according to the test results. As the number of training rounds increases, the accuracy and recall rate of the model increase accordingly, until the accuracy and recall rate values ​​increase with the training. The number of training rounds does not increase significantly; then, the trained model is deployed in the intelligent analysis system 3. In the process of plankton intelligent analysis application, the plankton pictures taken and analyzed in various lakes and reservoirs are used as the main source of the plankton standard database. For pictures with higher confidence in the analysis process, they are directly included in the plankton standard database to train the plankton recognition model, while for pictures with lower confidence, they are included in the plankton standard database for training after being reviewed and corrected by experts; after many iterative trainings, the accuracy and recall rate of the plankton recognition model can reach more than 90%, and the model is applicable to multiple water bodies and has strong generalization ability.

[0039] In addition, described plankton identification model adopts unicellular marking method when carrying out marking and result statistics, can number every kind of plankton, mark and use square frame that plankton cell is completely framed, and square frame will be as small as possible, avoid other plankton or impurity framed as far as possible.It is understandable that the present invention adopts unicellular marking method, presses unicellular identification cumulative counting, avoids the error that colony cell image segmentation counting causes, improves the accuracy of analysis result.In addition, for special colony phytoplankton, need each phytoplankton cell be marked separately, in order to avoid when colony phytoplankton cell is marked, the phytoplankton cell frame adjacent to it is entered, preferentially selects rotating frame to mark, is conducive to improving the accuracy of cell identification and positioning.In addition, in the identification process of phytoplankton cell, each phytoplankton cell that model identifies is surrounded by oblique frame, for the convenience of picture verification, algorithm is in the upper interface display process, adjacent or distance is less than the same recognition frame of certain value and is merged into a large positive frame, ensures that unicellular phytoplankton is marked with the positive frame that surrounds it in the identification result, and colony phytoplankton is marked with the positive frame that surrounds all cells in its colony with one. When the intelligent analysis system 3 counts the phytoplankton cells, each small frame identified by the model is accumulated to obtain the number of detected phytoplankton cells.

[0040] Although the preferred embodiments of the present application have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications that fall within the scope of the present application.

[0041] Obviously, those skilled in the art may make various changes and modifications to this application without departing from the spirit and scope of this application. Thus, if these modifications and variations of this application fall within the scope of the claims of this application and their equivalents, this application is intended to include these modifications and variations.

[0042] The foregoing description is merely a preferred embodiment of the present invention and is not intended to limit the present invention. Those skilled in the art will readily appreciate that various modifications and variations of the present invention are possible. Any modifications, equivalent substitutions, or improvements made within the spirit and principles of the present invention are intended to be within the scope of protection of the present invention.

Claims

1. A plankton intelligent analysis device, characterized in that: The invention comprises an automatic sampling system (1), an automatic scanning system (2) and an intelligent analysis system (3), wherein the automatic sampling system (1) and the automatic scanning system (2) are electrically connected to the intelligent analysis system (3), and the intelligent analysis system (3) is used to control the automatic sampling system (1) to automatically complete the automatic loading of plankton samples, and is also used to control the automatic scanning system (2) to scan and photograph images of the loaded plankton samples, and is also used to obtain biological information of the plankton samples based on analysis of the photographed plankton sample images.

2. The plankton intelligent analysis device according to claim 1, characterized in that: The automatic sampling system (1) comprises a sample storage unit (11), a sample circulation power unit (12) and a circulation detection pool (13); the sample storage unit (11) is used to store plankton samples; the circulation detection pool (13) is used to provide a resting place for scanning and photographing the samples; the sample storage unit (11) is connected to the circulation detection pool (13) via a pipeline; the sample circulation power unit (12) is arranged on the pipeline connecting the sample storage unit (11) and the circulation detection pool (13) and is used to provide power for sample circulation.

3. The plankton intelligent analysis device according to claim 2, characterized in that: The automatic sampling system (1) further comprises a pure water storage unit (14), a waste liquid storage unit (15) and a sample switching unit (16), wherein the pure water storage unit (14) is used to provide pure water for pipeline cleaning, the waste liquid storage unit (15) is used to collect waste liquid, and the sample switching unit (16) is used to switch sampling between the pure water storage unit (14) and the sample storage unit (11).

4. The plankton intelligent analysis device according to claim 2 or 3, characterized in that: The automatic sampling system (1) further comprises a pipeline liquid detection unit (17) provided on the pipeline connecting the sample storage unit (11) and the circulation detection pool (13) and electrically connected to the intelligent analysis system (3), for detecting whether liquid is circulating in the connecting pipeline. During sample injection, if the pipeline liquid detection unit (17) detects that no liquid is circulating in the connecting pipeline, the intelligent analysis system (3) issues an alarm.

5. The plankton intelligent analysis device according to claim 2 or 3, characterized in that: The automatic sampling system (1) further comprises a sample concentration quality control unit (18) arranged on a pipeline connecting the sample storage unit (11) and the circulation detection pool (13) and electrically connected to the intelligent analysis system (3), for preliminarily judging whether the sample concentration meets the analysis requirements. When the sample concentration quality control unit (18) detects that the sample concentration does not meet the analysis requirements, the intelligent analysis system (3) issues an alarm.

6. The plankton intelligent analysis device according to claim 2 or 3, characterized in that: The sample storage unit (11) comprises a sample pool, a sample temperature control module and a sample mixing module. The sample pool is used to store plankton samples. The sample temperature control module is used to provide a constant low-temperature environment for the plankton samples. The sample mixing module is used to mix the samples before sample injection.

7. The plankton intelligent analysis device according to claim 2 or 3, characterized in that: The automatic scanning system (2) comprises a microscope, an objective lens switching unit, a three-axis electric control platform and an automatic shooting module, wherein the microscope is used to visually magnify the plankton sample, the objective lens switching unit is used to switch the magnification of the microscope, the automatic shooting module is used to shoot the image of the plankton sample after being magnified by the microscope, and the three-axis electric control platform is used to drive the circulation detection pool (13) to move in the XYZ direction to achieve image scanning and automatic focus adjustment under a preset scanning path.

8. The plankton intelligent analysis device according to claim 7, characterized in that: During the automatic adjustment of the focal length, the three-axis electronically controlled platform first moves upward in the Z direction with a large step distance, and records the clarity value of the current focal plane at each focal plane. If the clarity difference between the subsequent focal plane and the previous focal plane is less than a threshold value, the large step distance is maintained in the next step. If the clarity difference between the subsequent focal plane and the previous focal plane is not less than the threshold value, the step distance of the next step is reduced. The above process is repeated until the clarity value of the subsequent focal plane is less than the clarity value of the previous focal plane, and then the platform moves in the opposite direction with the minimum step distance until the clarity value of the focal plane decreases. The focal plane is then used as the optimal focal plane, and the three-axis electronically controlled platform stops moving in the Z direction.

9. The plankton intelligent analysis device according to claim 7, characterized in that: The generation process of the preset scanning path is as follows: after the number of fields of view to be photographed is preset, the front of the channel of the flow detection pool (13) is divided into an n×m rectangular topological structure according to the preset number of fields of view, and it is ensured that n is less than the channel length divided by the length of a single field of view and m is less than the channel width divided by the width of a single field of view, and any one of the four vertex grids of the rectangular topological structure is selected as the origin to generate the preset scanning path according to the Z-shaped path planning.

10. The plankton intelligent analysis device according to claim 1, characterized in that: The intelligent analysis system (3) includes a plankton recognition unit, a human-computer interaction unit and a control unit. The plankton recognition unit and the human-computer interaction unit are both electrically connected to the control unit. The plankton recognition unit is equipped with a trained plankton recognition model and is used to perform intelligent recognition and analysis based on a plankton sample image to obtain biological information of the plankton sample. The human-computer interaction unit is used to realize interaction between a user and the device. The control unit is used to control the automatic sampling system (1) to automatically complete the automatic loading of the plankton sample, and is also used to control the automatic scanning system (2) to scan and photograph the image of the loaded plankton sample, and is also used to control the plankton recognition unit to analyze the photographed plankton sample image to obtain biological information of the plankton sample.

Images