119results about How to "Accurate count" patented technology

The invention belongs to the technical field of image analysis, and particularly relates to a multi-expert voting type crowd counting method based on scene understanding, which comprises the following steps of: extracting visual features from an input image and extracting text features from a text prompt by utilizing a backbone network comprising an image encoder, a text encoder and a feature enhancer; projecting the text features into an embedding space which is the same as the visual features for alignment; constructing a local subject expert used for local crowd target detection, a global context expert used for understanding the overall layout of the scene, a spatial relationship expert used for modeling the spatial relationship between crowds, and an attribute feature expert used for processing the appearance and attribute information of the target in the scene; a knowledge distillation strategy is adopted, and experts are trained in a fine-grained mode; and obtaining the output of each expert, and outputting a final counting result through a decoder. According to the method, the fine-grained crowd counting performance in a complex scene can be optimized by utilizing a voting mechanism and a knowledge distillation strategy of a multi-expert system.

The invention provides a cell analyzer and a cell analysis method.The cell analyzer comprises a chip layer and a runner layer, a liquid inlet, an impedance measuring unit and a liquid driving unit are arranged in the chip layer, the liquid inlet is communicated with a first surface and a second surface of the chip layer, the impedance measuring unit comprises an electrode plate exposed on the first surface of the chip layer, and the liquid driving unit is arranged on the first surface of the chip layer. The liquid driving unit comprises a driving piece exposed on the first surface of the chip layer, the runner layer is combined with the first surface of the chip layer and comprises a runner and a nozzle, the runner is formed in the face, facing the chip layer, of the runner layer, the liquid inlet, the runner and the nozzle are sequentially communicated, and an outlet of the nozzle is formed in the face, away from the chip layer, of the runner layer. At least one part of the electrode plate is located in the flow channel, and at least one part of the driving plate faces the outlet of the nozzle. The cell analyzer provided by the invention can finish accurate quantitative transportation of liquid to be detected and accurate counting of specific cells, realizes accurate measurement of the number and concentration of cells on a chip, and has the characteristics of miniaturization, automation and low cost.

The invention discloses a soybean plant pod-related phenotype extraction method, device, medium and equipment, and relates to the technical field of agronomy, and the method comprises the following steps: obtaining front and back images of a soybean plant in a to-be-detected area; constructing an improved twin network model; based on an original twin network model, replacing a feature extraction part with a compressed excitation residual network SE-ResNet50, increasing an attention mechanism, and adding a Dropout layer in a network classification layer; inputting the front and back images of the soybean plant into a target detection model to obtain coordinate information of stem nodes; according to the coordinate information of the stem nodes, the coordinate information of the pods is extracted, the Euclidean distance is adopted to obtain the distance between each pod and each stem node, and the pods and the stem nodes are matched; and after the pods are matched with the stem nodes, inputting the soybean plant image with the front and back images containing the pods into the improved twin network model, performing similarity judgment on the pods in the image, and obtaining the soybean plant pod related phenotypes.

The invention discloses a focus detection method and device based on deep learning, and relates to the technical field of deep learning, and the method comprises the steps: obtaining a capsule endoscopy image sequence, and carrying out the video frame extraction of the capsule endoscopy image sequence, so as to obtain a gall bladder endoscopy image; constructing a focus detection neural network for detecting the focus of the gallbladder endoscopic image, and performing multi-level feature extraction and multi-level feature fusion on the focus of the gallbladder endoscopic image based on the focus detection neural network to obtain a preliminary detection result; and constructing a deep calculation neural network for deeply extracting focus features and calculating the number of the focus, performing deep fusion feature extraction on the gallbladder endoscopic image according to the preliminary detection result based on the deep calculation neural network, refining repeated judgment of the focus, and generating a final detection result of the focus. And counting and analyzing the lesions according to the final detection result to obtain the number of the lesions and lesion information of each lesion. According to the invention, focus detection speed and accuracy are improved.

The invention provides a text-guided weak-annotation flotation froth segmentation and counting method and system, and relates to the technical field of artificial intelligence and computer vision. The method comprises the steps that an image of foam in a flotation tank is collected, preprocessing and point level labeling are carried out, and text description of flotation foam is collected and preprocessed; inputting the preprocessed image into a point heat map generation module, identifying a key point in the image, and forming a bounding box with a preset size by taking the key point as a center; inputting the key points and the bounding boxes into a hierarchical mask generation module based on weak annotation, and guiding the hierarchical mask generation module to generate integral, partial and subpart hierarchical masks; inputting the text description and the few-sample example image into a vision-language feature mapping module, and mapping the text description and the few-sample example image into the same embedding space for feature splicing to obtain classification guidance features; and extracting regional features by using a hierarchical mask, and carrying out similarity alignment on the regional features and classification guidance features, so that accurate segmentation and counting of specific types of foams in the image are realized.

The invention provides a method and system for counting paper through machine vision, and relates to the technical field of paper counting, and the method comprises the following steps: building an extraction algorithm for shooting stacked paper through an imaging system, analyzing image features, determining splicing feature points of paper stacked images, and building a relation model of paper thickness and pressure; combining the model with a gray projection method to realize paper counting; the machine vision technology can quickly, accurately and automatically count paper through an industrial camera and an image processing algorithm, so that the working efficiency is greatly improved; through an accurate image processing algorithm and high-quality camera shooting equipment, high-precision counting can be guaranteed through machine vision, and the situation of repeated counting or missed counting is avoided.

The invention discloses a compound dye for fluorescent staining of cells in a complex matrix. The formula of the compound dye comprises a buffer system with the pH value of 6.0-8.0, a dye solution with the pH value of 6.0-8.0 and a dye solution with the total volume of 100 mL, the fluorescent dye is Sybr Gold 2 * (40 *) or 2 to 80 mg of propidium iodide; and 0.1 to 10 mL of triton X-100, and 0.1 to 10 mL of a 0 to 24 mL of glycerol; 0 to 800 mg of EDTA (Ethylene Diamine Tetraacetic Acid)-disodium; the invention relates to a polyamide-amine dendritic polymer. According to the two preferable kit compound fluorescent dye formulas, efficient dyeing of cells in a complex matrix and accurate counting of the number of the cells can be achieved, and the dye has excellent fluorescence developing capacity. Compared with other dyes, the compound dye has the advantages of high dyeing rate (dyeing can be carried out within 30 seconds), stable dyeing, long-time illumination and difficulty in fluorescence quenching.

The invention discloses a counting system for products on a conveying belt, and particularly relates to an automatic counting system and method for different products on the conveying belt. The system comprises a visual sensor for acquiring 2D image data; the structured light sensor is used for collecting 3D point cloud data; a deep learning neural network algorithm is adopted to carry out target detection and classification on collected 2D image data and 3D point cloud data, a 2D image and 3D point cloud multi-modal registration algorithm is adopted to carry out accurate registration on the 2D image data and the 3D point cloud data, point cloud surface reconstruction and point cloud segmentation are carried out on the 3D point cloud data, the point cloud volume of a corresponding product is calculated, and the 3D point cloud volume of the corresponding product is calculated. And the calculation module is used for matching and counting different product volumes. The technical problem of accurately counting different products on the conveying belt is solved. According to the invention, unused products on the conveyor belt can be counted, whether the products have obvious volume defects or not can be distinguished, and qualified products and unqualified products can be counted separately.

The invention belongs to the technical field of medical instruments, and particularly relates to a drainage liquid antifouling extractor for tumor patient nursing, which comprises an extraction cylinder, an extraction needle head is arranged at the right end of the extraction cylinder, a piston rod is slidably connected to the middle of the left end of the extraction cylinder, a piston is fixedly connected to the right end of the piston rod, and the piston is slidably connected to an inner cavity of the extraction cylinder. The upper end of the connecting pipe is communicated with a drainage hose, the drainage hose is communicated with a drainage barrel through a liquid inlet pipe, the connecting pipe is provided with a one-way valve, and the drainage barrel is internally provided with an anti-blocking viscous hydrops crushing mechanism used for pretreating hydrops; according to the anti-blocking thick hydrops crushing mechanism, the multiple blades can rotate at the same time to crush thick hydrops until all the hydrops can pass through the through holes, then dispersed hydrops are obtained, during subsequent cytology examination, cell types can be accurately counted and recognized, and the difficulty of cytology examination is reduced.

The invention discloses a parasite egg detection method and system, and relates to the technical field of egg detection. Comprising the following steps: collecting a certain volume of a test sample, putting the test sample into a centrifugal tube, performing centrifugal operation, and reserving a precipitation part at the bottom of the tube; re-suspending the centrifuged precipitate part in normal saline, injecting the mixed liquid into a microfluidic chip, arranging a protruding electrode in the microfluidic chip, and coating the surface of the electrode with a specific antibody to capture eggs; current at different moments is calculated, and meanwhile microscope images of the electrodes are collected; establishing a Yo lov5s model, and training the Yo lov5s model by taking the contour of the parasite egg as a label; inputting a microscope image of the electrode into the Yo lov5s model, identifying a parasitic ovum contour in the image, and performing feature extraction on a contour image part; and jointly judging whether the parasitic eggs exist in the sample based on the change of the current at different moments in combination with the characteristic parameters, and if so, calculating the concentration of the parasitic eggs based on the change of the current and the number of the parasitic eggs.

The invention discloses an intelligent micro-plastic pollutant detection method and system based on image recognition, and belongs to the technical field of image recognition, and the method specifically comprises the following steps: setting a detection environment, sequentially putting different numbers of micro-plastic particles into a measured water body, and continuously collecting a water body image after each time of putting; a gray scale image of the water body is obtained, the number of gray scale steps and the number of pixel points of each gray scale step are obtained according to the gray scale histogram, the gray scale sum is calculated, and the gray scale sum corresponding to the order of magnitude of each micro-plastic particle is obtained; establishing a fitting function for fitting the relationship that the gray sum changes along with the magnitude order of the micro-plastic particles; acquiring a to-be-detected micro-plastic polluted water body, acquiring a gray sum of the polluted water body through image recognition, and substituting the gray sum into the fitting function to obtain a corresponding micro-plastic particle magnitude order; the method realizes high-efficiency detection of micro-plastic orders of magnitude in the polluted water body.

The present invention relates to the technical field of material counting, and in particular to a material counting device based on machine vision, comprising: an image acquisition module, a quality detection module, a detection and analysis module, a counting and determination module, and a material counting module; the image acquisition module acquires a surface image and an internal structure image of a material, the detection and analysis module processes the image, extracts the surface features of the material, and calculates a characterization parameter of a tendency of morphological abnormality, and the counting and determination module selects a counting and analysis method of a corresponding material according to the characterization parameter to determine how to adaptively count the material; the material counting device of the present invention can automatically identify the surface features and internal structure of a material by integrating machine vision technology, and counts and classifies the material according to preset standards, thereby effectively avoiding the processing of unqualified materials and reducing material waste.

The invention discloses a sweet potato and its closely related species centromere-specific repeat sequence and its application. The sweet potato is sequenced by second-generation sequencing and third-generation sequencing technology, and the sweet potato centromere-specific repeat sequence is obtained by analysis. After comparative analysis, identification and optimization, the tandem repeat sequence unique to the sweet potato centromere is obtained. After the sequence is designed, the fluorescent signal is marked and converted into a centromere probe, which proves that it can generate clear and bright centromere region hybridization signals in the FISH experiment of sweet potato and closely related species, and accurately display the position of the centromere on the chromosome. By counting the number of centromere signals, the accurate counting of the chromosomes of sweet potato and closely related species can be achieved.

This utility model provides a hinge opening and closing test fixture. The fixture table includes a fixed base, a rotating base, a fixed fixture, and a rotating fixture. Support feet are located at the bottom and rear. The fixed base mounts the fixture table, and the fixed fixture is connected to the fixed base. Hinge fixing pieces are secured with screws through fixing holes. The rotating base is connected to the fixture table via a rotating shaft, and the rotating fixture is connected to the rotating base. A clamping groove holds the hinge rotating piece, and the groove has an elastic clamping layer to prevent loosening. The fixture table contains a switching power supply, a motion controller, a servo motor, and a worm gear reducer to drive the rotating shaft. A counting sensor under the rotating base measures the number of rotations, which is displayed on the front side of the fixture table via a counter. This fixture is securely fixed, provides accurate driving and counting, is efficient and convenient, and is suitable for testing the opening and closing performance of hinges.

The invention discloses a hematuria analysis method, device and equipment based on artificial intelligence and a storage medium. The method comprises the following steps: acquiring a urine visible component graph of urine to be analyzed and an interference state of the urine to be analyzed; selecting a hematuria analysis network model corresponding to the interference state from a preset corresponding relation set of the interference state and the hematuria analysis network model; the urine visible component graph is input into the hematuria analysis network model, the number of red blood cells corresponding to the to-be-analyzed urine is determined through the urine visible component graph, the hematuria category corresponding to the to-be-analyzed urine is determined based on the number of red blood cells, and the hematuria category comprises pseudo hematuria or true hematuria. According to the interference state of the to-be-analyzed urine, by constructing a plurality of corresponding hematuria analysis network models, misjudgment of the red blood cells in the to-be-analyzed urine can be reduced, accurate counting of the red blood cells is achieved, and therefore the accuracy of true hematuria detection is improved.

Disclosed is a wide-temperature nanoparticle counter, which includes a gas intake flow-dividing structure, a saturator, a condenser, and an optical counter. The saturator, the condenser, and the optical counter are sequentially connected. The saturator, the condenser, and the optical counter each are provided with a temperature control unit; and a liquid working substance with a thermal diffusivity greater than the mass diffusivity is stored in the saturator. A to-be-tested gas is divided into two paths by the gas intake flow-dividing structure, where one path of gas enters the saturator after being filtered, and then carries working substance steam to enter the condenser; and the other path of gas enters the condenser from a center of a bottom of the condenser through a capillary or a microporous channel, and is mixed with the gas carrying the working substance steam. A working substance is chosen from a group consisting of alcohols, silicone oil, esters, and a solution formed by mixing at least two of the silicone oil, esters and alcohols. Different liquid working substances are selected based on the temperature of different sample gases to be tested, thereby enabling effective measurement of the number concentration of nanoparticles across a wide-temperature range.

The utility model relates to the technical field of dumbbells, in particular to a novel counter dumbbell capable of improving training efficiency, which comprises a dumbbell rod, connecting rods are fixed at two ends of the dumbbell rod, dumbbell pieces are mounted on the surfaces of the connecting rods, a connecting sleeve is connected to the surface of the dumbbell rod, and protective rings are arranged on two sides of the connecting sleeve. And the protective ring is sleeved on the surface of the dumbbell rod. The counting device has the advantages that accurate counting can be conveniently carried out on training, training efficiency is improved, a trainee can conveniently and rapidly observe and read numbers, and use is convenient; according to the dumbbell counter, the positioning blocks and the positioning grooves are connected in a fastening and inserting mode, the first magnetic attraction blocks and the second magnetic attraction blocks are connected in an attraction mode, and therefore the first connecting piece and the second connecting piece can be connected stably and detached conveniently, and the connecting sleeve and the counter body can be detached and installed on the surface of the dumbbell rod conveniently.

The application discloses an automatic counting method for the egg-laying quantity or hatching rate of Spodoptera exigua. The method comprises the following steps: collecting a multilayer focal plane image sequence of an egg mass, constructing a height field model of the egg mass, dividing egg grains into different physical layers based on the spatial position relationship in the vertical direction, and respectively counting the number of each layer. The method also introduces a shielding compensation calculation to correct the lower layer of the blocked egg grains, so as to accurately obtain the layered egg quantity data. The application solves the problem that the traditional two-dimensional image recognition cannot distinguish the hierarchy of the egg mass, which leads to a serious low count, realizes the accurate counting of the stacked egg mass, and provides a reliable technical means for entomological research and pest forecasting.

This application relates to the field of aquaculture monitoring technology, and provides a method and corresponding product for estimating fish population size based on forward-looking sonar data. The method includes: acquiring sonar monitoring data obtained from a forward-looking sonar device monitoring a target water area, and constructing an effective monitoring area in the sonar monitoring data according to the field-of-view parameters of the forward-looking sonar device; preprocessing the sonar monitoring data located within the effective monitoring area to enhance the distinguishability of fish targets; identifying fish targets located within the effective monitoring area based on the preprocessed sonar monitoring data, and generating an initial population estimate based on the number of identified fish targets; acquiring aggregation characteristics reflecting the aggregation state of fish within the effective monitoring area; and correcting the initial population estimate based on the aggregation characteristics of fish within the effective monitoring area to obtain a corrected fish population estimate.

The utility model discloses a kind of bag type packing material anti-continuous bag counting control device, including light sensing detection device, gravity sensor and the speed sensor being set on belt conveyor;The light sensing detection device includes light sensing generating device and light sensing probe receiving device;The light sensing generating device and light sensing probe receiving device are oppositely arranged;Belt conveyor both sides are respectively provided with the light sensing generating device and the light sensing probe receiving device;The bag type packing material anti-continuous bag counting control device further includes the counting controller for data processing according to light sensing detection device, gravity sensor and speed sensor;The belt conveyor is equipped with gravity sensor, and gravity sensor and light sensing detection device are at the same end of belt;The utility model realizes accurate judgment and accurate counting to bagged material continuous bag condition, solves the problem of inaccuracy of counting, high labor cost, low production efficiency caused by continuous bag when traditional bagged material loading counting.

The invention relates to the technical field of sports equipment, and discloses a body test rope skipping counting device which comprises a rope skipping line body, the outer surface of the rope skipping line body is rotatably connected with a shell, and the interior of the shell is fixedly connected with a telescopic counting mechanism. An auxiliary stretching mechanism is rotatably connected to the outer surface of the skipping rope body, a storage rotating mechanism is rotatably connected to the interior of the shell, the telescopic counting mechanism comprises a counting displayer, the counting displayer is fixedly connected with the shell, and a circuit board and a first bearing are fixedly connected to the inner wall of the shell; a first box body is fixedly connected to the interior of the first bearing, the clamping force is proper during normal rope skipping, the skipping rope is prevented from loosening and affecting the playing of a rope skipping person, and the skipping rope is prevented from being damaged due to excessive clamping; when people stumbles the skipping rope, the skipping rope can be quickly loosened to reduce the injury risk of a rope skipping person and the breakage possibility of the skipping rope; and the Hall element can accurately record the rope skipping number.

The present invention provides a method for counting protein aggregated particles in yogurt, comprising the following steps: dispersing a yogurt sample with a solvent so that the non-particle part in the yogurt sample is dissolved and the particle part is dispersed to form a mixed solution; taking a sample picture of the mixed solution using an imaging device; performing interference elimination processing on the sample picture in the imaging device; and partitioning and counting the sample picture after the interference elimination processing. The counting method of the present invention can effectively evaluate the severity of protein aggregated white particles in the production and sales process of yogurt products for quantitative analysis, and then formulate appropriate quality standards, providing a scientific basis for improving the protein aggregated particle problem of yogurt products.

This utility model relates to the field of PET cold-pressed sheet processing technology, and discloses a PET cold-pressed sheet stacking and counting device, comprising: a frame, including a frame plate and several support legs, the support legs being fixedly disposed below the frame plate; a counting component, including a weighing scale and a display screen, the weighing scale being fixedly disposed above the frame plate, the display screen being fixedly disposed on the front side of the frame plate, and a control board electrically connected to the weighing scale being built-in; an auxiliary packaging component, including a strapping groove and positioning rods, the strapping groove being disposed on the frame plate, and the positioning rods being fixedly disposed at the four corners above the frame plate; and a compaction mechanism, including a pressure plate and a support, the support being fixedly disposed on the rear side of the frame plate, with a movable plate rotatably connected to its upper end, one end of the movable plate being driven to rotate by a stepper motor fixedly disposed on the support, and the pressure plate being disposed below the extended end of the movable plate and being driven to rise and fall by a cylinder fixedly disposed on the movable plate. The advantages of this utility model are: accurate counting, small error, and easy packaging after counting.

The discharging device with the grading function comprises a vibration disc assembly, a conveying belt assembly, a sensor counting assembly and a divider platform, the output end of the vibration disc assembly is connected with the input end of the conveying belt assembly, and only one product is allowed to pass through the output end of the vibration disc assembly; the conveying belt assembly is used for conveying products, the output end of the conveying belt assembly is connected with the input end of the sensor counting assembly, the sensor counting assembly is used for counting the products, and the sensor counting assembly is provided with an output end capable of being opened or closed; a plurality of material boxes are arranged on the divider platform, the divider platform is in transmission connection with the material boxes, and the material boxes can move to the position below the output end of the sensor counting assembly. The magnetic steel product separating and counting device is simple in structure and convenient to operate, magnetic steel products can be efficiently separated, orderly arranged and accurately counted, it is guaranteed that packaging of final finished products can be accurately and rapidly completed in the subsequent packaging process, and production efficiency is greatly improved.

The present invention belongs to the field of biological detection technology, and discloses a test paper cutting device, including a punch seat, a support seat and a collection mechanism, wherein the support seat is arranged below the punch seat, the punch seat is provided with a punch array facing the support seat, the support seat is provided with a grid array that forms a punching cooperation with the punch array, each punch in the punch array corresponds to the position of the test strip on the whole test paper to be punched, the collection mechanism is arranged below the support seat, and a plurality of collection units are arranged in the collection mechanism, and each collection unit corresponds to a certain number of meshes in the grid array. The present invention provides a test paper cutting device, which can count quickly and accurately while completing the cutting of the test strips, which is convenient for subsequent bottling, and the device has a simple structure and is easy to use.

This invention relates to the field of auxiliary equipment technology for laying hen farming, and discloses an egg-blocking device for accurate egg counting. The device includes a box, an inclined plate, and a conveyor belt. The box is composed of multiple metal meshes, one of which can be quickly opened and closed, specifically using a pin or similar mechanism for opening and closing. A conveyor belt is located on one side of the box, and the bottom of the box slopes inwards towards the conveyor belt. An inclined plate is connected to the bottom of the box extending towards the conveyor belt. In actual operation, multiple boxes can be arranged parallel and symmetrically on both sides of the conveyor belt. A protrusion is provided on the outside of the conveyor belt to collect and transport the eggs inside the multiple boxes on both sides. This invention uses a motor to adjust the height of the baffle, which can effectively meet the blocking requirements of egg sizes at different cycles. When eggs need to be collected, the device quickly opens and releases the eggs, guiding them orderly to the conveyor belt for accurate counting.

The utility model relates to the technical field of safety display equipment, in particular to a safety warning board. The device comprises a warning board shell, a clamping device and an infrared sensing device, the warning board shell is provided with a clamping device used for clamping the warning board. An infrared sensing device used for detecting the flow of people passing through the warning board and counting the total time needed by people to browse the warning board is installed in the warning board shell. The infrared sensing device comprises an infrared sensor, a single-chip microcomputer, a power supply and a display device. The infrared sensing device and the single-chip microcomputer are connected with the power supply. The infrared sensor is connected with the single chip microcomputer; the single-chip microcomputer comprises a counter and a timer. The counter and the timer are both connected with the display device; the method has the advantages that a manager is assisted in judging whether each employee reads the warning board or not and whether the content on the safety warning board is seriously read or not.

The invention provides equipment based on a vector biological database and an application method, the equipment comprises a fixing assembly and an acquisition terminal, the fixing assembly comprises a storage table, a control support is arranged on the storage table, the acquisition terminal is arranged on the control support, and the acquisition terminal is provided with an identification module, a marking module, a counting module and a classification module; wherein a sample is firstly placed on the object placing table, then the position of the collection terminal is controlled through the control support so that the collection terminal can sample the sample, and finally information obtained through sampling is recognized through the recognition module, marked through the marking module and classified through the classification module. According to the invention, sample collection, image acquisition, intelligent identification, marking, accurate counting and classification can be realized, and the whole process is automatic and intelligent, so that the working time for screening the vector organisms from the large-flux sample is shortened; therefore, the timeliness, accuracy and intelligent level of public health monitoring, port import and export cargo customs clearance and vector biological prevention and control can be greatly improved.