120results about "Counting objects with random distribution" patented technology

The invention belongs to the technical field of computer images, and provides an object counting method and device for enhancing text guidance by utilizing frequency characteristics. The method comprises the following steps: inputting a query image to a visual encoder to obtain visual features; inputting the object description text to a text encoder to obtain text embedding; fusing the visual features and the text features by using a fusion network to obtain fusion features; the fusion features are input into a decoder to obtain a density estimation graph of the object, the decoder comprises K cascaded adaptive frequency selection modules and one output layer, and the kth adaptive frequency selection module comprises a first convolution unit, an adaptive frequency selector and a first up-sampling unit; the kth self-adaptive frequency selector carries out filtering processing on the amplitude spectrum and the phase spectrum of the output characteristic of the first convolution unit in the frequency domain, and converts a filtering processing result into a spatial domain; according to the method, related frequency components are dynamically emphasized in decoding, the space domain and frequency domain features are combined, and accurate object positioning and accurate counting are achieved.

The invention provides a slipper sorting and counting method and system based on AI in the technical field of slipper sorting management. The method comprises the steps that S1, a large number of historical slipper images are collected to construct a data set; s2, creating a slipper identification model; s3, training a slipper identification model by using the data set, and deploying the slipper identification model to a sorting robot; s4, the industrial personal computer obtains the production instruction list and analyzes the production instruction list to obtain SKU codes and the sorting number, corresponding induction counting tables are started, all the SKU codes, the sorting number and the number of the counting tables are bound to generate a binding relation, and the binding relation is sent to a sorting robot; s5, the sorting robot collects real-time slipper images and inputs the real-time slipper images into the slipper recognition model to obtain SKU codes, and the slippers are sorted to the corresponding induction counting tables based on the SKU codes and the binding relation; and S6, counting the slippers by the sensing counting table through a photoelectric sensor, and checking a counting result through a visual counting camera. The slipper sorting and counting device has the advantage that the efficiency and quality of slipper sorting and counting are greatly improved.

To provide a system for checking the number of bound steel materials and a method for checking the number of bound steel materials with which the number of steel materials in one bundle can be more accurately checked.SOLUTION: A system for checking the number of bound steel materials comprises: a counting processing apparatus that acquires a photographed image obtained by photographing the bundle of steel materials from an imaging apparatus, and determines a count number that is the number of steel materials included in a bundle to be collated through image processing on the acquired photographed image; and a collation processing apparatus that acquires the count number, acquires a registered number that is the number of steel materials to be included in the bundle to be collated, from a bundle information database in which information on the bundle of steel materials is registered, and performs collation processing as to whether the count number and the registered number match each other. When both the numbers do not match in the collation processing, the system causes the imaging apparatus to photograph the bundle to be collated again, and the collation processing apparatus performs the collation processing again on a count number determined from the photographed image obtained by photographing the bundle again.SELECTED DRAWING: Figure 1

To provide technology for accurately counting an object included in a laminate.SOLUTION: A counting method acquires image data on a lamination surface of a laminate, partitions the image data into a plurality of predetermined regions with respect to a lamination direction and a width direction, generates a signal profile indicating each thickness component in the predetermined region, acquires a thickness component estimated from each predetermined region, and counts an object, on the basis of the thickness component.SELECTED DRAWING: Figure 1

To provide an object recognition device and an object recognition method with which it is possible to appropriately estimate the object position or the number of pieces of material.SOLUTION: An object recognition device 10 according to the present invention is constituted to comprise: a data acquisition unit 11 that acquires two-dimensional moving-image data in which an object group 20 including a plurality of substances 22 to be recognized are imaged from multiple directions and a distance image of the object group which is synchronized with the two-dimensional moving-image data; a data analysis unit 12 that obtains three-dimensional point cloud data of the object group on the basis of the distance image of the object group; and an object detection unit 13 that estimates a boundary box BB of the substances to be recognized, for each frame of the two-dimensional moving-image data, projects boundary boxes of the same identification number to a representative image via a partial point cloud corresponding to the boundary box of the three-dimensional point cloud data, determines the circumscribing rectangle of each of two-dimensional point clouds of the projected boundary boxes, and integrates the circumscribing rectangles to detect the substances to be recognized.SELECTED DRAWING: Figure 6

The present invention relates to a fry counting device, comprising a fish supply device, a fishway and a counting box, wherein the fishway is tilted downward as a whole, a fill light is provided at the bottom, and a host is provided at the top, wherein the host comprises a camera arranged at the bottom for photographing the internal position of the counting box, a development board arranged at the inside, and a display screen arranged at the top, and is characterized in that the fish supply device can be a pump suction type, a siphon type or a pneumatic type, and the fishway comprises an initial step channel, a middle step channel, and a final step channel connected in sequence, and the angle between the fishway and the horizontal plane decreases step by step. The stepped channel is used to gradually ease the water flow, so that the fry can be evenly distributed at the shooting position, and through different modifications of the fish supply device, it plays a role in preventing the fry from being injured, reducing human operation, and speeding up the fry counting process on the basis of ensuring the counting accuracy.

The invention belongs to the technical field of machine vision and coal mine intellectualization, and discloses a coal mine water exploration and drainage drill rod counting method based on deep learning. According to the specific technical scheme, a coal mine underground video stream is obtained; carrying out preprocessing by utilizing the obtained drill site video stream; analyzing the to-be-detected drilling image by using a pre-constructed target detection model to obtain position information of the key area; calibrating by using the initial position and the end position of the drilling machine; constructing a drill rod counting model by using the calibrated position information so as to determine the rod feeding times of the drill site; according to the drill rod counting method provided by the invention, full-process monitoring and counting of drilling and rod withdrawing can be realized, the workload of workers is reduced, and the situations of false report, missing report and the like are prevented, so that the development efficiency is improved, and the personal safety of coal miners is ensured.

To provide a method of counting the number of round bars allowing for correctly counting the number of round bars irrespectively of the position of a rest surface.SOLUTION: A method of counting the number of round bars includes: acquiring height data in an upper-half outer periphery of a plurality of round bars arrayed at a constant interval in an array direction; taking a highest position of obtained height data as a reference position and identifying a position where height becomes minimal or minimum respectively on positive and negative sides in the array direction within a predetermined range determined from a diameter of the round bar with reference to the reference position as a boundary; adding a range from a negative minimal or minimum position over to a positive minimal or minimum position through the reference position, as an existence range of one round bar, to the number of round bars; taking a highest position of remaining height data excepting height data present in the existence range of the one round bar as a next reference position and identifying a position where the height becomes minimal or minimum on positive and negative sides respectively in the array direction within a predetermined range determined from the diameter of the round bar with reference to the next reference position as a boundary; and adding the same as a next existence range of a round bar to the number of round bars into calculation of the number of round bars.SELECTED DRAWING: Figure 3

The present invention belongs to the field of tea tree plucking, and specifically relates to a device for automatically identifying and counting tea tree buds, comprising a movable guide rail and an electric slider I, each movable guide rail being slidably connected to at least two electric sliders I. The present invention achieves this by inserting a lever and a baffle between tea trees, enabling the lever and the baffle to pry the leaves and buds on the upper portion of the tea tree. The lever then moves the leaves and buds on the upper portion of the tea tree, thereby preliminarily removing water droplets adhering to the leaves and buds on the upper portion of the tea tree. The baffle then moves again to pry the leaves and buds on the upper portion of the tea tree. The movement of the baffle is blocked by the leaves and buds on the upper portion of the tea tree, causing the baffle to bend. Simultaneously, the leaves and buds on the upper portion of the tea tree are also pushed by the baffle, causing them to bend. When the baffle gradually separates from the leaves and buds on the upper portion of the tea tree, the tea tree rebounds, thereby further removing water droplets adhering to the leaves and buds on the upper portion of the tea tree.

A hole counting device is provided with: an image acquisition unit that acquires one or more images obtained by capturing an image of a plurality of holes that capture a target substance; and a counting unit that counts the number of fluorescent holes among the plurality of holes for each of the one or more images acquired by the image acquisition unit, and that counts the number of fluorescent holes by excluding fluorescent holes among adjacent holes among the plurality of holes.

The present invention comprises: a complex transfer function calculation unit (30) which calculates a complex transfer function, regarding signals transmitted from N transmitting elements, from reception signals received by M receiving elements; a biological component extraction unit (40) which extracts biological information that is a variable component in a predetermined space; a correlation matrix calculation unit (50) which calculates a target correlation matrix from the biological information; a noise information storage unit (60) which records a noise correlation matrix; a first person count information calculation unit (61) which calculates first person count information that is an assumed value of the number of people existing in the predetermined space on the basis of a threshold value calculated from the noise correlation matrix and the target correlation matrix; a MUSIC spectrum calculation unit (70) which estimates candidates for a position in a living body by using the target correlation matrix, and outputs a likelihood spectrum in which the living body is present at the position; and a second person count information calculation unit (80) which estimates second person count information or position, which is the number of living bodies, with higher accuracy from position information based on the likelihood spectrum including a plurality of position candidates.

Properly manage the cleaning of ion exchange resins. [Solution] The ion exchange resin cleaning system 1 comprises a container 2 for cleaning the ion exchange resin 3, a cleaning water supply pipe L2 for supplying cleaning water to the container 2, a discharge pipe L4 for discharging the cleaning wastewater, and a measuring device 5 connected to the discharge pipe L4 for measuring the water quality of the cleaning wastewater. The measuring device 5 comprises an evaporation unit 52 that evaporates the cleaning wastewater to form residual particles, and a measuring unit 53 that measures the number of residual particles.

Embodiments for an AI-based system and method for feed monitoring in a land-based fish farm are described where the system may include: a feed camera mounted to an effluent pipe for capturing a video feed having images of objects that traverse a field of view of the feed camera when each image is acquired; and a special-purpose computer that executes a pellet-tracking algorithm that employs a region of interest (ROI) proposal module, an ROI classification module, an ROI tracking module, and a trajectory classification module for at least counting uneaten feed pellets. In at least one embodiment, an object enhancer may be mounted to the effluent pipe opposite to and in the field of view of the feed camera to provide a background so that the objects in the acquired images have a definable perimeter.

According to the method and the system for auditing the multimedia content of the PIS system in real time, the AI technology is deeply integrated into the existing PIS system architecture, so that the content of the PIS system is automatically audited and processed without delay. The invention provides a method for auditing multimedia content of a PIS system in real time. The method comprises the steps that 1, the PIS system generates the multimedia content; 2, judging whether the source of the multimedia content in the step 1 is a PIS system center side or not, if yes, entering a step 3, and if not, entering a step 4; step 3, judging whether the type of the multimedia content is a video stream or not, if so, entering step 5, and if not, entering step 6; and step 4, preprocessing the multimedia content to obtain a frame-cut picture, and auditing the obtained frame-cut picture in real time. And 5, auditing the video stream in real time. And step 6, auditing the text information in real time.

1. A method for operating a field crop transporting machine, in particular a field crop harvesting machine (2), comprising the steps of acquiring sensor data (3a, 3b), calculating mass data by an evaluation device (14) based on the sensor data, and providing yield data (16) calculated on the basis of at least the mass data, wherein the evaluation device (14) generates, based on at least the sensor data (3a, 3b), the mass data and / or the yield data (16), an adjustment signal (18) for adjusting at least one separating element of a separating device (20), the separating device (20) being configured to separate a first part of a harvested material (12) from another part of the harvested material (12).

To provide a person monitoring system, a person monitoring method, and a person monitoring processing program capable of quickly and accurately monitoring the number of persons and a person-to-person distance with a simple configuration regardless of the states of the persons existing within a predetermined monitoring region.SOLUTION: In a person monitoring system 1, a person measurement device 10 includes: a photographing unit 110 that photographs persons 500a to 500c; and a depth measurement unit 120 that receives reflected waves IL12 to IL32 obtained when indoor illumination light or the like is reflected by the persons 500a to 500c to measure a distance. Further, an information processing device 20 (personal computer 410) includes: a person region identification unit that identifies a person region from an image; a person number calculation unit that calculates the number of persons; a person region distance calculation unit that calculates a first distance that is the distance from the depth measurement unit 120 to the person region for each person; a person-to-person distance calculation unit that calculates a second distance that is the distance between respective persons; and a monitoring unit that determines whether information on the number of persons and information on the second distance satisfy predetermined conditions.SELECTED DRAWING: Figure 1

To provide a recognition system of a chip having a setup capable of recognizing the color of a chip according to an image and counting the quantity of chips under a different illumination environment.SOLUTION: In a chip recognition system 10, a chip W is configured to at least partially have a specific color 121 showing the value of the chip W. The chip recognition system 10 includes: a recording device 11 for recording a state of the chip W as an image by using a camera 212; an image analysis device 14 for recognizing at least two colors of the specific color 121, and a reference color 4a differing from the specific color 121 existing in the recorded image by analyzing the image; and a recognition device 12 at least including an artificial intelligence device 12a for identifying the specific color 121 of the chip W by using an image analysis result by the image analysis device 14. Multiple images of the chip W irradiated with light having different illuminance, and the reference color 121 are taught to an artificial intelligence device 12a of the chip recognition device 12 as teacher data.SELECTED DRAWING: Figure 1

The method for estimating the accumulated mass of solid microparticles on a surface comprises describing the variation of the accumulated mass of the solid microparticles on said surface over time with a model defined according to a differential equation.

Method and apparatus (1) for detecting passengers getting on and / or getting off a means of transportation (T), comprising a) an acquisition phase (P1), wherein at least one image is acquired which comprises at least one portion depicting at least one portion of a door of said means of transportation (T), b) a processing phase (P2), wherein it is determined whether said door is in a closed state or in an open state on the basis of said at least one image, and a detection phase (P3), wherein at least one passenger getting on or getting off said means of transportation (T) is detected when said door is in the open state.

The concentration of non-volatile dissolved substances in the water flowing through the water treatment system must be properly managed. [Solution] The water treatment system 1 has a measuring device 5 that introduces the water flowing through the water treatment system 1 as the water to be measured and measures the water quality of the water to be measured. The measuring device 5 has an evaporation unit 52 that evaporates the water to be measured to form residual particles, and a measuring unit 53 that measures the number of residual particles.

The utility model discloses an anti-reflection insect sticking plate data collection and detection device, and mainly relates to the field of intelligent agriculture. Comprising a box body with an opening in the top, an insect sticking plate is placed at the bottom in the box body, a telescopic rod is arranged on the outer side of one end of the box body, a fastening bolt used for fixing the length of the telescopic rod is in threaded connection with the telescopic rod, a transverse rod is arranged at the top end of the telescopic rod, and a camera electrically connected with the Raspberry Pi is arranged at the end, away from the telescopic rod, of the transverse rod. The beneficial effects of the utility model reside in that the insect sticking plate is placed at the inner bottom of the box body, the box body prevents external sundries from adhering to the insect sticking plate in the shooting process, the accuracy of data analysis is ensured, light around the insect sticking plate can be shielded, the problems of light reflection, light pollution and the like in the shooting process are avoided, the quality and the effect of shot pictures are ensured, and the shooting efficiency is improved. And then the collected pictures are transmitted to the inside of the Raspberry Pi, and corresponding types and numbers are quickly and accurately counted by using a deployed corresponding target recognition algorithm, so that the method is more practical.

The application relates to a crowd behavior attribute counting method, system and product, and belongs to the field of computer vision.The counting method comprises the following steps: acquiring a complex crowd image; constructing a multi-scale feature fusion attention mechanism crowd behavior attribute counting neural network trunk; training the multi-scale feature fusion attention mechanism crowd behavior attribute counting neural network trunk based on the complex crowd image; acquiring a complex crowd image to be counted; and inputting the complex crowd image to be counted into the trained counting neural network to obtain a counting result.The above scheme in the application is suitable for dividing the behavior attributes of large-scale high-density crowds and detecting the number of crowds, and effectively improves the accuracy of detecting the number of specific crowds.

The high-precision predatory mite counting device comprises a base, a temperature box, a stereoscopic microscope, a counting structure, a walking column and a dismounting assembly, a positioning seat is driven to move through cooperation of a moving motor, a threaded rod, a threaded ring and a threaded block, so that the counting structure enters the temperature box and is stored in the temperature box with the environment temperature being 36 DEG C for 2 h, and then the predatory mites are counted. After all the predatory mites climb out and are limited by an insect sticking plate to move, the counting difficulty is reduced, a counting plate is moved out of the temperature box through a moving motor, a dismounting column drives a reset column to move, the reset column drives a reset plate to move, and the reset plate drives a limiting column to move and extrudes a reset spring; the relative positions of the limiting columns and the limiting holes are controlled, so that the positioning seat and the dismounting seat are quickly dismounted, the counting structure is conveniently moved to the position of the stereoscopic microscope, manual counting one by one is carried out under the stereoscopic microscope, and the working efficiency is improved.

The invention relates to a surgical instrument counting device for counting one or more different types of tools (h) for use in a surgical procedure, which comprises: an image-capturing camera (2); an electronic device (3) connected to the camera (2) and linked to a processing unit (3.1) that processes the images captured by the camera and executes counting algorithms; software with image recognition algorithms that uses artificial intelligence and computer vision, which is installed in the processing unit (3.1) and is capable of detecting and counting one or more different types of tools (h), including groups or families of tools, and capable of self-learning; and a support (4) suitable for mounting and positioning the camera (2) on a surface (5) on which tools (h) forming the set of instruments necessary for a surgical procedure are disposed.

Exemplary methods and systems for identifying suspect cytological specimens containing target cells are disclosed. One exemplary method includes acquiring a first image associated with a first region of the cytological specimen through a first objective lens, acquiring a second set of one or more images associated with a first subregion of the first region through a second objective lens, identifying a first number of target cells, and determining whether the first number of target cells is greater than a threshold number. If the first number of target cells is greater than the threshold number, the exemplary method prompts a warning. If not, the exemplary method further includes acquiring a third set of one or more images associated with a second subregion through a second objective lens, and identifying a second number of target cells associated with the second subregion.

The invention relates to a neutron spectrum and proton spectrum distinguishing measurement method and device based on a multi-ball measurement system, and the method comprises the steps: obtaining a current neutron and proton spectrum through employing an SAND-II few-channel spectrum unfolding algorithm based on a neutron and proton response function matrix of the multi-ball measurement system and an original count obtained through a to-be-measured field test; obtaining an original deviation between the measured value and the calculated value; reducing the count of each channel of the current neutron spectrum according to a set proportion, and calculating based on the SAND-II few-channel spectrum unfolding algorithm again to obtain a new proton spectrum, a new proton count, a new neutron spectrum and a new neutron count, and obtaining a new deviation between the total count and the original count; obtaining a plurality of deviations by changing the set proportion for multiple times; and finally, taking the neutron spectrum and the proton spectrum corresponding to the minimum value in all the deviations as measurement results. According to the invention, the neutron and proton real-time distinguishing measurement with relatively wide energy range and relatively high precision in a high-altitude region of the earth or in other neutron-proton mixed fields is realized.

To provide an information processing device that can more accurately grasp the storage status. [Solution] The information processing device according to the present invention comprises: a distance measurement information acquisition unit that acquires distance measurement information of a space including a three-dimensional virtual space area of ​​a predetermined size by a distance measuring sensor mounted on a moving body that moves based on a self-position estimation means; a calculation unit that calculates the amount of space occupied by an object present in the three-dimensional virtual space area based on the acquired distance measurement information of the three-dimensional virtual space area and pre-stored topographic information of the space including the three-dimensional virtual space area; and a request information generation unit that generates request information requesting the replenishment of the object based on the time-series change in the amount of space occupied by the object.

To accurately count the number of insects.SOLUTION: A counting apparatus according to the present disclosure includes: an image acquisition section for acquiring an image of collected insects; an occupancy rate calculation section for calculating the occupancy rate, indicating the proportion of the region occupied by insects in the image; and a counting section for counting the number of collected insects. When the occupancy rate is lower than a first threshold, the counting section performs a first counting process to count the number of insects through the analysis of the image, while when the occupancy rate is higher than the first threshold, the counting section performs a second counting process to count the number of insects in accordance with the occupancy rate. By altering the counting process according to the occupancy rate, the number of insects can be accurately counted depending on the overlap between insects.SELECTED DRAWING: Figure 10

The invention discloses a tobacco insect image recognition and counting method based on YOLOv5 and a micro-service architecture, and the method comprises the following steps: S1, image acquisition, S2, image processing and data enhancement, S3, model training, S4, to-be-detected image input, S5, detection result writing into a text file, S6, statistics, and S7, packaging a statistical result in a system, and carrying out the data rendering of a front end. And S8, returning a statistical result. The method has the advantages that through fusion of deep learning algorithm optimization and distributed architecture design, efficient intelligent analysis service for tobacco insects is constructed. The system adopts a three-stage technical framework: firstly, performing weight training on a model level by using YOLOv5, reconstructing a backbone architecture through a lightweight feature extraction network, and enhancing a small target feature expression ability in combination with a cross-level attention mechanism; secondly, building a FastAPI-based micro-service, carrying out asynchronous processing on tasks, and realizing request distribution and dynamic scheduling of computing resources; and finally, realizing elastic expansion of algorithm service through a containerized deployment scheme.