205 results about "Turbid water" patented technology

The invention provides an ICESat-2 turbid water area sounding correction and water depth inversion method based on optical characteristics of a water body. The method comprises the following steps: S1, obtaining photon data and an optical remote sensing image of a near-shore turbid water research area, and carrying out preprocessing operation on the data; s2, performing spectral sensitivity analysis by acquiring an ICESat-2 photon error value and combining optical characteristics of a water body; s3, constructing an ICESat-2 turbid water area sounding compensation model based on water optical characteristics, and selecting the model preferentially according to evaluation indexes; s4, synthesizing a multi-temporal image, and realizing model adaptive adjustment of different space weight distribution based on the spectral sensitivity difference of the turbid water area; s5, constructing an ICESat-2 and remote sensing image joint inversion water depth model based on turbidity partitioning; s6, spatial interpolation is carried out on a water depth inversion result, residual noise and adjacent pixel differences possibly occurring in the image are suppressed, and continuous water depth is obtained; according to the invention, large-range water depth inversion without field actual measurement work can be realized.

The invention discloses an automatic mixing water tank for a sediment density flow circulation test, comprising a water tank which is divided into a test pool and a muddy water pool by a lifting gate; the water circulation system comprises a clear water recovery pool and a muddy water recovery pool, the clear water recovery pool is used for injecting water into the test pool, the muddy water recovery pool is used for receiving muddy water discharged from the test pool, the clear water recovery pool is communicated with the muddy water recovery pool, and a muddy water filter is arranged at the communication part; the silt mixing unit comprises a silt storage tank, a mixer and a layered pressure measurer, the silt storage tank is used for releasing silt to the muddy water pool, the mixer is used for stirring in the muddy water pool, and the layered pressure measurer is used for monitoring the sand density of different water layers in the muddy water pool in real time; and the control system is used for controlling the water circulation system and the silt mixing unit to carry out automatic mixing.

The utility model discloses an underwater detection robot. The underwater detection robot comprises a robot body and a multi-beam assembly, the multi-beam assembly is arranged on a frame of the robot body, the multi-beam assembly comprises a plurality of mutually connected multi-beam depth sounder, and the multi-beam depth sounder is respectively connected with an energy system and a control system of the robot body; the multi-beam depth sounder is provided with beam emergent surfaces, the directions of the beam emergent surfaces are different, the multiple multi-beam depth sounder are matched, and the emitted beams can perform 360-degree annular acoustic detection on the diversion tunnel to be detected. When the robot is applied, the defect detection precision of the diversion tunnel can be improved, and tunnel defects can be easily distinguished in a turbid water environment.

The invention provides a turbid water area mud flat shoreline fine extraction method based on a hyperspectral satellite, and relates to the technical field of image processing, and the method comprises the following steps: S1, obtaining a plurality of Landsat series satellite original hyperspectral satellite remote sensing images according to a research target area and research time; s2, obtaining a preprocessed image; s3, performing multi-scale segmentation on the preprocessed image and calculating a normalized differential water body index of image data; and S4, performing threshold classification on the image after multi-scale segmentation in combination with a normalized differential water body index, adjusting and setting a proper normalized differential water body index, distinguishing a water body from a non-water body, and extracting a shoreline. According to the method, the long-time change rule of the coastline can be better researched and analyzed, the change trend of the coastline can be predicted, a basis is provided for development of coastal areas, and the method has a good application prospect in the fields of coastline research, coastline engineering construction and the like.

A water body suspended sediment concentration monitoring method based on FUI water body classification comprises the steps that firstly, a target watershed Sentinel-2 L1C image from a certain year to a certain year is obtained from GEE, and after resampling is conducted till the resolution is 10 m, preprocessing such as 6S model atmospheric correction, proximity effect correction and quality control is conducted; 5 wave bands are used for calculating spectral angles to obtain FUI, and 14 is used as a threshold value to divide the water body into clear water and muddy water. And then selecting an optimal wave band combination according to the FUI classification to construct a semi-empirical inversion model, and finally calculating the suspended sediment concentration distribution by using the model and verifying the suspended sediment concentration distribution. The method solves the problem that the model generalization ability is low in the monitoring of the suspended sediment concentration of the target drainage basin, and the inversion precision is high.

The invention discloses a multi-mode dam crack detection method based on acoustics-optics, and belongs to the technical field of underwater target recognition. According to the method, an underwater camera and an imaging sonar sensor are utilized to perform joint detection on cracks possibly existing on the surface of a dam: firstly, an optical image and a sonar image are respectively acquired, and crack feature extraction and preliminary recognition are performed in respective modes; and then, mapping detection results of the two modes to the same coordinate system through coordinate transformation, fusing optical detection information and acoustic detection information based on spatial consistency, and confirming the position and range of a crack target. According to the method, the complementary advantages that optical imaging definition is high and sonar penetrates through a turbid water body are brought into full play, the high crack detection rate and precision can still be kept in the underwater environment with extremely low visibility, compared with a single sensor, missing detection and false detection can be remarkably reduced, and reliable and automatic recognition of underwater dam structure cracks is achieved.

The invention relates to the technical field of image enhancement, in particular to a turbid water fish image enhancement system based on image enhancement, which comprises an attenuation feature layering module used for acquiring a turbid water fish image, extracting a color attenuation value and a local contrast value of each pixel point, and outputting the color attenuation value and the local contrast value; and calculating the pixel attenuation degree influenced by the suspended matters in the water body and the background scattered light to obtain a pixel attenuation distribution diagram, and setting a plurality of attenuation thresholds. According to the method, a turbid image is divided into a plurality of layers such as a foreground, a middle scene and a background by calculating a color attenuation value and a local contrast ratio of each pixel influenced by suspended matters in a water body and scattered light, and then differential gain coefficients are applied to extracted initial details such as fish scale and fin textures according to the layers to carry out injection; high-fidelity enhancement of a key foreground target is realized, and excessive sharpening and noise amplification of a background area are avoided.

The invention discloses an underwater image restoration method based on polarization imaging and low-frequency filtering, and the method comprises the steps: S1, collecting polarization images of different turbid water bodies at three angles through an underwater polarization imaging experiment platform, and calculating a Stokes vector of an underwater target scene; s2, calculating a polarization degree and a polarization angle image of an underwater target scene according to the Stokes vector, and establishing a target light model according to a relation of polarization information; s3, based on the low-frequency characteristics of the background light, extracting low-frequency background light components of the turbid polarization image at each angle, and calculating a corresponding background light Stokes vector according to the low-frequency background light components; and S4, the processed parameters are substituted into the imaging model, and the optimal cut-off frequency of the filter is adaptively determined in combination with enhanced measurement evaluation indexes, so that high-quality underwater image restoration is realized. According to the method, texture features and edge details of more target areas can be effectively restored.

The invention provides an underwater structure crack detection method for underwater image enhancement of a turbid water area, and the method comprises the steps: collecting water quality parameters in real time through a plurality of sensors integrated at a front end, constructing an environment feature encoder in combination with a neural network, and mapping an environment state to a physical prior generation network to predict transmissivity and a channel attenuation coefficient. The prior information guides an image defogging and color recovery module, combines enhanced image definition feedback, dynamically optimizes network weight, and adopts synthesis and real data mixed training and loss function adaptive adjustment to improve the adaptability of the model to a complex underwater environment and the image enhancement effect. According to the invention, high-quality recovery and color correction of underwater images under different water body conditions are realized.

The invention discloses an image processing system and method suitable for a turbid water area, and belongs to the technical field of image processing. The system comprises an underwater camera component for collecting original image data of a turbid water area in real time and an embedded image processing unit. The embedded image processing unit comprises a preprocessing module used for carrying out noise suppression and image calibration processing on original image data, and a physical parameter estimation module used for estimating environment background light intensity of a current water area, a water body attenuation coefficient and a distance map of an object in a scene and calculating initial transmissivity. The scattering correction module is used for removing back scattering and preliminarily compensating a color channel based on a physical imaging model in combination with physical parameters, and the deep learning enhancement module is used for improving image local contrast, adjusting brightness and finely correcting colors by adopting a lightweight deep convolutional neural network. According to the image processing system and method suitable for the turbid water area, the image definition and color of the turbid water area can be improved, the real-time performance is high, and the system and method are suitable for different water area environments.

The invention provides an intelligent optical sensing system and method for imaging through an underwater turbid environment, and belongs to the field of underwater optical imaging. The problem of low underwater object recognition efficiency caused by difficulty in obtaining clear images in turbid water by existing equipment is solved; the system comprises an optical imaging module, an image acquisition module and computing equipment which are carried on diving equipment, computer programs of an image preprocessing module, a depth sensing module and a neural network denoising recognition module run on the computing equipment, and the optical imaging module generates a structured light field carrying an orbital angular momentum mode based on a spiral phase plate; the image acquisition module is used for acquiring reflection information of the target object irradiated by the vortex light; the image preprocessing module is used for enhancing the acquired image; the depth sensing module obtains the current water pressure through a pressure sensor, and the current water depth is inversely solved through the computing device; the neural network denoising recognition module carries out real-time recognition and denoising on the image; the method is applied to turbid underwater imaging.

The invention belongs to the field of ocean remote sensing information, and particularly relates to a laver culture area rapid extraction method based on sentinel No.1 radar images. Comprising the following steps: 1) acquiring a sentry No.1 radar image, and analyzing and processing pixels in the image to obtain a nori raft frame initial product; 2) pretreating the initial product of the laver raft frame to obtain a continuous raft frame product; 3) extracting three masks by using the auxiliary optical image and confused pixels around the high-reflection area; and 4) performing mask processing on the continuous raft frame product by using an extracted result to obtain a spatial distribution grid data set of the laver culture raft frame. According to the invention, the optical image is prevented from being limited by a cloud layer and a turbid water body, the time sequence is integrated through a maximum value synthesis algorithm, and the complete spatial distribution of the raft frame submerged by the high tide level is captured.

The invention relates to a bimodal detection device for an underwater bridge pile and a detection and repair robot system, and belongs to the technical field of underwater structure detection and repair robots, the bimodal detection device comprises a turbidity sensor used for continuously sampling a working water area to determine the turbidity; the main controller is used for controlling the forward-looking sonar to scan the bridge pile to obtain an acoustic image when the turbidity is greater than or equal to a preset value; the data processing unit is used for identifying a suspected disease area based on the acoustic image; the main controller is also used for controlling the laser scanner to obtain three-dimensional point cloud of the suspected disease area and controlling the visible light camera to obtain a visible light image of the suspected disease area; and the data processing unit is also used for carrying out defect identification on the bridge pile based on the three-dimensional point cloud and the visible light image. According to the bimodal detection device for the underwater bridge pile, through a bimodal detection mode combining acoustics and optics, accurate recognition of apparent defects of the bridge pile in a muddy water environment is achieved.

The invention relates to a non-contact high-precision pipe network damage detection system, which relates to the technical field of underground pipe network detection, and comprises a bimodal sensing unit, a data processing unit, a self-adaptive optical unit, a protective shell and a power management module. Through cooperation of the infrared thermal imaging module and the laser scanning module, millimeter-level precision can be realized, a visible light camera or a near-infrared camera can be automatically switched according to environmental turbidity through arrangement of a light filtering switching mechanism, clear imaging in a turbid water body can be ensured, and through a built-in FPGA chip of the data processing unit, the data processing unit can be used for realizing high-definition imaging of the visible light camera or the near-infrared camera. Through cooperation of the bimodal sensing unit, the data processing unit and the adaptive optical unit, temperature field and point cloud data can be synchronously acquired, after fusion, + / -3mm damage positioning precision can be realized, and when the turbidity is greater than 200NTU, high reliability can still be maintained, the detection efficiency can be improved, and the detection precision can be improved. And the risk of manual underwater operation is thoroughly avoided.

The invention discloses a coastal tidal flat large-range remote sensing automatic extraction method, and belongs to the technical field of remote sensing tidal flat mapping, and the method comprises the steps: calculating a turbid water index pixel by pixel based on a cloud removal time sequence remote sensing image, selecting a comparison wave band according to the turbid water index, and calculating a self-adaptive mixed water index in combination with a green light wave band; for each image determined by the self-adaptive mixed water body index, calculating a value domain histogram of the image, and taking a self-adaptive mixed water body index value corresponding to a derivative mutation point in the value domain histogram as a self-adaptive land and water segmentation threshold value; calculating an improved normalized difference water body index scene by scene, extracting a maximum water surface range based on the improved normalized difference water body index and a self-adaptive water and land segmentation threshold value, extracting a tidal flat exposure frequency image based on each scene image and the self-adaptive water and land segmentation threshold value, and further extracting initial tidal flat spatial distribution; and generating a tidal flat spatial distribution visualization result after connectivity processing. According to the method, rapid, accurate and stable extraction of tidal flat space distribution in a large-range and complex environment can be realized.

This application provides an intelligent optical sensing system and method for imaging in turbid underwater environments, belonging to the field of underwater optical imaging. It solves the problem of low efficiency in identifying underwater objects due to the difficulty of obtaining clear images in turbid water with existing equipment. The system includes an optical imaging module, an image acquisition module, and a computing device mounted on a diving device. The computing device runs computer programs for an image preprocessing module, a depth sensing module, and a neural network denoising and recognition module. The optical imaging module generates a structured light field carrying an orbital angular momentum mode based on a spiral phase plate. The image acquisition module acquires the reflection information of the target object after being illuminated by vortex light. The image preprocessing module enhances the acquired image. The depth sensing module obtains the current water pressure using a pressure sensor, and the computing device decomposes the current water depth. The neural network denoising and recognition module performs real-time image recognition and denoising. This application is applied to underwater imaging in turbid water.

The invention relates to the technical field of submarine cable detection, in particular to a submarine cable detection method and system based on an underwater robot. The method comprises the following steps: acquiring original turbid image frames in a seabed underwater environment by using an underwater robot along a preset robot route to obtain an original turbid image sequence; solving transmissivity and ambient light parameters of the original turbid image sequence, and constructing an image enhancement function to correct the original turbid image frame to obtain an enhanced image frame; performing image texture refinement and detail recovery operation on all the enhanced image frames by using the image deblurring enhanced convolutional neural network model; and extracting the cable position, the cable trend and the cable edge contour in the clear image frame sequence by using the cable identification model, and feeding back to the underwater robot control system to update the robot route in real time. The method is used for improving the precision and stability of the underwater robot for identifying and detecting the submarine cable in a high-turbulence turbid water area with suspended silt.

This invention discloses a wastewater treatment control device, belonging to the field of underground coal mine wastewater treatment technology. It includes a vibrating screen mechanism and a water tank. The vibrating screen mechanism is used for primary solid-liquid separation of coal slurry water. A rotating mechanism driven by a linear drive mechanism is provided between the vibrating screen mechanism and the water tank. This rotating mechanism includes a partition section for dividing the water tank into a turbid water zone and a clear water zone, and for secondary separation of fine coal slurry particles, enabling the discharge of coal slurry from the turbid water zone and the underground utilization of the water from the clear water zone. This invention utilizes a multi-stage treatment process consisting of vibrating screening and dynamic zoned sedimentation to complete the entire process from coarse particle separation to fine particle settling and clarification underground, achieving on-site reuse of clear water and significantly reducing hoisting costs and surface treatment load. The adjustable partition plate, as a secondary treatment unit, can dynamically adjust the volume of the clear and turbid water zones, ensuring water quality stability and system shock resistance. The stirring and scraping mechanism can automatically clean the water tank and screen, reducing the burden of subsequent manual sludge removal.

To provide a turbid water treatment apparatus capable of realizing an optimum treatment method (pH range) using magnetic particles and a method for suppressing the use amount of a neutralizing agent when turbid water is mixed with cement and is alkaline, and a turbid water treatment method.SOLUTION: A muddy water treatment apparatus 1 according to the present invention includes a reaction tank 4 that coagulates suspended solids in cement muddy water to which an inorganic coagulant and magnetic particles are added, a coagulation tank 5 that is provided downstream of the reaction tank 4 and coagulates the suspended solids in the cement muddy water to which a polymer coagulant is added to form magnetic floc, a sedimentation tank 6 that is provided downstream of the coagulation tank 5 and sediments the coagulated magnetic floc to obtain supernatant water and sludge containing the magnetic particles, a recovery unit 8 that recovers the magnetic particles from the sludge, and a supply unit 9 that supplies the recovered magnetic particles to the cement muddy water.SELECTED DRAWING: Figure 11

The invention relates to an underwater positioning and posture monitoring device of a desilting robot for a pool and a use method of the underwater positioning and posture monitoring device, which belong to the technical field of underwater positioning of robots and comprise a desilting pool, a desilting robot, a desilting robot platform deck, a pressure sensor, a pressure signal processing and converting system, a high-pressure jet pump, a high-pressure nozzle, a binocular camera, an image processing system and a base mark, according to the method, the underwater high-pressure jet technology and the visual identification technology are adopted, high-precision monitoring of the position and the posture of the underwater robot is achieved through the reasonable device and the implementation method, and the problem of high-precision positioning of the underwater robot in a semi-closed space and a turbid water body can be effectively solved.

The invention discloses a construction method of an unmanned aerial vehicle multi-spectral remote sensing basic model based on agricultural application, and relates to the technical field of intelligent agricultural equipment, and the method comprises the following steps: S1, collecting a multi-band remote sensing image of a rice region through a multi-spectral camera carried by an unmanned aerial vehicle; s2, calculating an anti-flooding vegetation index, and generating a vegetation coverage rate initial value according to the anti-flooding vegetation index; s3, extracting a rice plant area image, and inputting the rice plant area image into the trained rice state recognition model; s4, judging whether the flooding stress probability is greater than a flooding threshold value or not; and S5, determining the rice growth potential grade based on the vegetation coverage rate initial value or the corrected vegetation coverage rate. According to the method, through the anti-flooding vegetation index and the self-adaptive calculation algorithm in the turbid water body, the strong absorption interference of the water body in the near-infrared band is effectively stripped, so that the initial value of the vegetation coverage rate can truly reflect the survival state of underwater vegetation, and the situation that part of flooded healthy rice is misjudged as poor growth or death is fundamentally avoided.

The application provides a device and method for testing the plugging mechanism of a pressure-reducing well under the condition of bidirectional and alternate clear water and turbid water, which comprises a test tank base, a plexiglass tank fixed to the test tank base, a sample bottom support plate arranged at the bottom of the plexiglass tank, a multi-hole well pipe installed on the sample bottom support plate, a filter layer fixing frame installed on the outer wall of the multi-hole well pipe, and a filter layer sample fixed in the filter layer fixing frame. The filter layer sample is divided into at least three parts by a partition slot. The sample bottom support plate is supported on the test tank base by a bottom support plate annular support, and a bottom water tank is formed between the sample bottom support plate and the test tank base. The bottom water tank is connected to clear water through a bottom water inlet pipe. The top of the multi-hole well pipe is covered by a sample cover plate, and the top of the plexiglass tank is covered by a top fixed cover. A top turbid water tank is formed between the sample cover plate and the top fixed cover, and the top turbid water tank is used for connecting to turbid water. The application can realize the test and quantitative analysis of the plugging evolution process of the filter layer of the pressure-reducing well.

The invention discloses a sea cucumber accurate fishing robot suitable for a shallow sea culture area, and relates to the technical field of aquaculture equipment.The sea cucumber accurate fishing robot comprises a rack, an equipment cabin, a collecting cabin, a power supply cabin and a plurality of propellers are installed on the rack, and a fishing mechanical arm matched with the collecting cabin is installed on the front side of the rack; a precise fishing framework is arranged in the equipment cabin, the precise fishing framework identifies the positions of sea cucumbers in the shallow sea culture area and controls the fishing mechanical arm to fish the sea cucumbers and then put the sea cucumbers into the collecting cabin, and a power supply unit is arranged in the power supply cabin to supply power to the precise fishing framework and the fishing mechanical arm. Through fusion perception, turbid water body image enhancement and accurate positioning, full-process autonomous accurate fishing of the sea cucumbers in the shallow sea high-turbidity environment is achieved, manual diving operation is replaced, the fishing efficiency and the product quality are greatly improved, and intelligent upgrading of the sea cucumber culture industry is promoted.

The invention discloses a multi-scale heterogenous remote sensing data collaborative water quality inversion method, computer equipment and a medium, and belongs to the technical field of water quality remote sensing monitoring. S3, an inversion model is constructed, and the step S3 comprises the steps of S31, wave band combination optimization and correlation analysis; s32, constructing a multi-scale heterogeneous remote sensing data collaborative fusion inversion model; and S33, multi-source data collaborative wave band selection and model determination. According to the method, high-frequency and high-spatial-resolution water quality inversion is realized, the spatial-temporal resolution and precision are improved, a more comprehensive and accurate water quality inversion model is constructed, the robustness of the model in a turbid water body and an eutrophicated water body is remarkably improved, the adaptability to a complex water body is enhanced, and the method is suitable for large-scale popularization and application. An efficient and economical technical means is provided for water quality inversion in a complex water body environment, and the method has remarkable low cost and universality.

The invention discloses a net cage fouling degree intelligent identification system and method based on electromagnetic-optical fusion perception, and belongs to the technical field of net cage fouling degree intelligent identification. The system comprises a mobile scanning platform, an integrated multi-mode sensing unit and a central processing unit; the integrated multi-mode sensing unit comprises an optical imaging module, an electromagnetic characteristic detection module and a positioning auxiliary unit. The method comprises the steps of cooperative scanning and multi-modal data acquisition, data fusion and contamination intelligent identification, accurate positioning and coordinate mapping, global contamination mapping, quantitative evaluation and cleaning decision support. According to the invention, through optical and electromagnetic multi-mode perception fusion, the identification precision of the types of the fouling organisms and the quantification accuracy of the coverage rate are greatly improved, and especially in a turbid water body, the limitation of a traditional single vision method is effectively made up; based on the global fouling distribution map generated by the fusion positioning technology, the operation mode change from blind traversal to accurate targeting is realized.

The application discloses a kind of phytoplankton photosynthetic activity turbidity correction device and method, belong to phytoplankton monitoring technical field, the device includes excitation emission optical structure, light source driving module, signal detection module and main control module;The application constructs water turbidity scattering and initial fluorescence relative increment relationship by measuring 750nm water turbidity scattering and combined light-induced fluorescence kinetics curve, by calculating the correlation coefficient of both, the influence of water turbidity scattering on phytoplankton photosynthetic activity detection is corrected into photosynthetic activity correction model.The device structure is simple, and the measurement period is short, and it is non-invasive, and the underwater in-situ detection of phytoplankton photosynthetic activity of turbid water can be carried out, which provides a new way for rapid and accurate measurement of phytoplankton photosynthetic activity.

The utility model provides an irrigation monitoring device of a drip irrigation system, which relates to the field of irrigation monitoring and comprises a main body, a valve body and a metering component which are sequentially arranged on the inner side of the main body as well as a mounting piece 1 and a mounting piece 2 which are arranged at two ends of the main body, and a filter element is arranged in the middle of the main body and positioned between the valve body and the metering component. According to the utility model, the bypass valve, the filter, the flow meter and the pipeline are built into a whole, so that the problem that a dripper is blocked due to turbid water or insufficient dissolution of fertilizer during irrigation is solved, and the phenomena of blockage and abrasion of the micro flow meter connected behind the filter are eliminated; the service life of the micro flow meter is effectively prolonged, the water flow monitoring accuracy is effectively improved, in addition, due to the integrated design, installation is facilitated, the risks that traditional connectors are increased, and the water leakage amount is increased are reduced, and the convenience and practicability of the device are improved.

The invention belongs to the technical field of ocean engineering, particularly relates to an underwater mobile robot based on an underwater acoustic communication technology, and provides the following scheme aiming at solving the problems that a fish-imitating underwater robot is easy to wind in a complex turbid water area, a sensor is easy to lose efficacy due to sediment adhesion, and the maneuvering operation capability of the fish-imitating underwater robot is seriously hindered. One side of the fish body shell is fixedly connected with a fish head, one side of the fish head is provided with a motorized auxiliary module, the motorized auxiliary module comprises a mounting plate, and the tail of the fish body shell is provided with a tail swinging mechanism. The underwater mobile robot based on the underwater acoustic communication technology has the advantages that entanglements can be actively removed through the movable cutting blade of the maneuvering auxiliary module, system faults are prevented, meanwhile, an automatic cleaning function is achieved, a sensor can be kept clean, it is ensured that the image quality is stable, and the maintenance requirement is remarkably reduced.

The invention belongs to the technical field of solid-liquid separation, and particularly discloses winding pipe filler and a sedimentation tank vertically provided with the winding pipe filler. The winding pipe filler comprises at least two mutually-wound pipelines and pipe hoops which are fixed outside the mutually-wound pipelines and enable the relative positions of the pipelines to be stable, and when the winding pipe filler is vertically placed, the inner wall and the outer wall of each pipeline form a gradient larger than 30 degrees and smaller than 70 degrees with the horizontal plane. A plurality of the winding pipe fillers are vertically arranged in the tank body of the sedimentation tank, and the inner wall and the outer wall of each pipeline of the winding pipe fillers in the tank body and the horizontal plane form a gradient greater than 30 degrees and less than 70 degrees. By vertically arranging a plurality of winding pipe fillers in the tank body, the specific surface area of precipitation contact can be obviously increased, so that suspended matters are precipitated on the inner wall and the outer wall of a pipeline when muddy water passes through the winding pipe fillers, and slide to the bottom of the tank body along a slope under the action of gravity; the device has the characteristics of simple structure, high precipitation efficiency, strong adaptability and good sludge discharge effect.

The invention provides a sturgeon rollover detection method and system based on deep learning and ellipse fitting. The method comprises the following steps: acquiring a sturgeon image and preprocessing the sturgeon image; performing semantic segmentation on the preprocessed sturgeon image by using a pre-trained deep learning model; according to the semantic segmentation result, binary masks of the back area and the belly area of each sturgeon are generated, and noise areas in the masks are removed through morphological processing and connected domain analysis; respectively performing ellipse fitting on the denoised back and belly region pixel sets of each sturgeon, extracting long axis direction angles of the back and belly ellipses of each sturgeon, further calculating a rollover angle, and judging whether the sturgeon has rollover or not and the rollover degree according to the rollover angle; according to the method, high-robustness and high-precision automatic monitoring can be realized in complex environments such as strong light reflection and turbid water quality, so that the manual inspection cost is remarkably reduced, the culture efficiency is improved, and sturgeon health and caviar production quality are guaranteed.