80 results about "Animal identification" patented technology

The invention discloses a wildlife detection method fusing an unmanned aerial vehicle thermal infrared image and a visible light image, and belongs to the field of small target wildlife identification, and the method comprises the following steps: S1, obtaining a preprocessed TIR-RGB image pair set; s2, an FDM-YOLO double-source target detection model improved based on YOLOv81 is constructed, and the improved FDM-YOLO double-source target detection model is trained based on the preprocessed TIR-RGB image pair set obtained in the step S1; and S3, inputting an image acquired in real time into the improved FDM-YOLO double-source target detection model trained in the step S2, and outputting a wild animal detection result. By adopting the wild animal detection method fusing the thermal infrared image and the visible light image of the unmanned aerial vehicle, high-precision, real-time and robust detection of a small target of a wild animal in a complex field environment is realized by improving the FDM-YOLO model.

A marine zooplankter identification method based on in-situ images and deep learning is used for processing a multi-stage series deep learning architecture constructed for all in-situ images, and comprises the following steps: quickly positioning and framing zooplankter individuals under a complex background based on a YOLOv8 skeleton construction model; performing pixel-level fine segmentation on individuals in the in-situ image by using a U-Net model, and extracting an accurate contour to obtain a high-quality individual image; and finally, inputting the image into a PlanktonNet network innovatively designed based on a ViT architecture, and through adaptive small-size slice embedding and introduction of a convolution word embedding layer, improving small-scale target feature extraction capability, and realizing high-precision and fine-grained classification of genera and species. The method has the advantages that target detection, semantic segmentation and recognition tasks are organically fused, the problems of low recognition efficiency and low automation degree in the prior art are effectively solved, and marine zooplankton can be quickly and accurately recognized from in-situ images on a large scale.

The invention relates to the technical field of vehicle intelligent control, and provides an animal repelling method and device, electronic equipment and a storage medium, and the method comprises the steps: obtaining multi-modal data on a vehicle driving path; based on the multi-modal data, identifying whether a target animal hindering passage exists on the vehicle driving path, and if yes, determining attribute information and scene parameters of the target animal according to the multi-modal data; based on the attribute information and the scene parameters, generating a repelling signal for the target animal; and executing corresponding expelling operation based on the expelling signal. According to the method, the accuracy and the robustness of animal identification are improved by comprehensively utilizing the multi-modal data on the vehicle driving path, and a more targeted and more intelligent self-adaptive expelling signal is generated based on deep understanding of animal attributes and scene parameters, so that the success rate and the effectiveness of expelling can be greatly improved.

The invention relates to the field of backfat thickness detection, in particular to a backfat thickness detection system and method. The method comprises the following steps: setting a channel detection region, collecting animal image data and animal identity information through the channel detection region, analyzing the animal image data and the animal identity information, and constructing a three-dimensional target model; extracting the three-dimensional target model to obtain a three-dimensional target animal area, and identifying the three-dimensional target animal area to obtain a backfat detection area; processing the backfat detection area to obtain a backfat perspective model and a back section point set; in combination with a backfat perspective model, analyzing the back section point set to obtain a back boundary point set; and obtaining a detection thickness value according to the back boundary point set. According to the invention, the backfat thickness detection efficiency can be improved.

The application discloses a kind of aquaculture water body in aquatic animal identification method, system, storage medium and equipment, it is related to image recognition technical field, the method is by camera module deployed in water body interior and is gathered multi-view image construction image set, and by floating type communication device synchronization space-time metadata stable transmission to cloud end;In cloud end, sequentially execute the unsupervised screening based on image content features, color self-adaptive correction according to the optical characteristics of water body and the frequency domain enhancement processing combined with semantic guidance, form three-level preprocessing pipeline, significantly improve image quality;Finally, the optimized image is input into pre-trained underwater target detection model, and the high-precision aquatic animal identification result is output.The method effectively solves the problems of poor data transmission stability, low target feature recognition and insufficient real-time identification in complex water environment.

The invention provides a multi-species wild animal video tracking method based on deep learning, and belongs to the technical field of wild animal recognition. The problem that tracking task modeling is feature matching between a template and a current frame image is solved, a motion model does not need to be preset, and the method can more flexibly adapt to field scenes with irregular motion and complex environment. The system is mainly composed of a feature extraction and encoder module, a detection branch, a tracking branch, an ID matching and updating module, a trajectory reconstruction module, a tracking trajectory compensation detection mechanism, a triple association strategy and a trajectory unification strategy, by adopting the tracking method based on time sequence correlation modeling, the interference problem of a complex environment and an animal movement mode in wild animal tracking can be effectively solved, and the method is more suitable for target tracking in a field dynamic scene.

This invention relates to the field of animal management technology and discloses an AI-based animal identification and management method based on unmanned aerial vehicles (UAVs). The method includes: responding to the generation of a heat map of suspected animal locations, a second UAV cluster performs a pinpoint confirmation task on the heat map area, collecting detailed image data of the animals; inputting the detailed image data into an animal identification model; based on species category and behavioral state, calling the corresponding animal movement prediction model from a model library and combining it with real-time environmental data to predict the animal's movement trajectory within a future time window; and controlling a third UAV cluster to generate a virtual electronic fence according to the predicted movement trajectory and a user-defined range. The method in this invention not only enables rapid animal species identification and population statistics but also generates a virtual electronic fence based on the predicted animal movement trajectory and a user-defined range, comprehensively considering both animal activity trends and human management needs.

This application relates to a vehicle risk avoidance method and electronic device, belonging to the field of vehicle technology. The vehicle risk avoidance method includes: when the vehicle is parked and an animal recognition event is detected, acquiring animal status information and personnel location information, wherein the animal recognition event indicates the presence of a dangerous animal target within a preset area around the vehicle, and the animal status information is determined based on the state of the dangerous animal target; the personnel location information is used to represent the relative position between the personnel associated with the vehicle and the vehicle; determining a target risk avoidance strategy based on the animal status information and personnel location information; and controlling the vehicle to execute the target risk avoidance strategy. This application establishes a complete risk avoidance process from risk perception to targeted risk avoidance strategy formulation, and then to automatic execution of the risk avoidance strategy, enabling vehicles to provide users with proactive warnings and risk avoidance assistance in outdoor scenarios such as camping, thereby reducing the safety risks posed by dangerous animals in outdoor environments.

The present disclosure relates to an animal monitoring system (100) for an animal cage, the animal monitoring system (100) comprising: a physical platform divided into a grid of cells (106) wherein each cell (106) comprises at least one weighing sensor (108); an animal identification system; a processing unit configured to: identify at least one animal on the physical platform using the animal identification system; continuously obtaining a load signal from the load cell (108); the identified position and / or change in position of the at least one animal on the physical platform is tracked based on the load signal from the weighing sensor (108). The present disclosure also relates to a method of investigating an animal in an animal cage.

The invention discloses an animal identification and positioning method and system based on sound events, and relates to the technical field of biological acoustic monitoring, and the method comprises the steps: collecting multiple paths of original audio data to form a target sound event set; and performing animal type identification processing on the set, extracting and fusing acoustic features of each channel of audio, and generating an animal type determination result. Meanwhile, sound source space positioning processing is executed, and the estimated space position of the animal sound source is calculated based on the sound wave arrival time difference and the intensity difference. And performing data fusion and association on the type judgment result and the estimated position to generate an animal activity record containing the animal type and the position. And performing space-time continuity analysis on the record, performing association matching with historical animal activity records, and correcting or updating position or category information in the record. And finally, according to the animal activity record, generating an animal activity thermodynamic diagram or an animal trajectory diagram of the monitoring area. According to the method, the accuracy and data continuity of animal monitoring in a complex environment are improved.

A hand-held applicator for applying an identification tag to an animal, comprising a body (10), a first handle (11), a second handle (13), a first jaw (141), a second jaw (142), a tag loading area, a tag application area (101), a magazine for holding a plurality of interconnected animal tags (18), a tag loading device for loading the interconnected animal tags from the magazine into the applicator body, a tag cutting and pushing element (50) for separating a single animal tag from the plurality of interconnected animal tags and moving the single animal tag from the tag loading area to the tag application area, and an application assembly. Moving the tag cutting and pushing element (50) forwards and backwards and moving the first handle (11) and second handle (13) together causes the application assembly to apply the single animal tag to an animal.

A method for animal identification and tracking includes, during a dark cycle: illuminating an interior of an animal cage with light outside of a visible light range; and capturing a first image of the interior of the animal cage. The method also includes: detecting a first identification tag in the first image; extracting a first set of tag features from the first image; identifying a first animal identifier of a first animal, tagged with the first identification tag, based on the first set of tag features; interpreting a first animal position of the first animal occupying the animal cage based on the first image; appending a cage log with the first animal identifier, a first timestamp corresponding to a time of capture of the first image, and the first animal position.

A milking system (100) is designed to extract milk from an animal (101). The system (100) includes several components for efficient milk sampling and biomarker detection. It is characterized by a feed cassette (140) holding multiple vials (150), at least one of which is prepared for biomarker detection. During milking, a milk sampler (300) transfers some milk into the selected vial (150). The system also has an animal identification device (116) for identifying the animal (101) and an information tagging device (370) for tagging the vials with relevant details. A controller (120) is connected to the identification device (116), the milk sampler (300), and the information tagging device (370). The system identifies the animal, determines whether biomarker detection is required, and, if necessary, signals the milk sampler to transfer milk into the appropriate vial and signals the information tagging device to tag the vial with the animal's identity. The system streamlines the process of collecting and analyzing milk samples.

The invention relates to the technical field of safety production in the civil explosive industry, in particular to a civil explosive industry production site animal identification and driving method and a patrol robot, and the method comprises the steps: obtaining the image information of a to-be-identified animal, the image information comprises the features of the to-be-identified animal and the position coordinates of the to-be-identified animal; according to the image information, obtaining to-be-identified animal types and corresponding animal position information in the production site; determining a driving strategy according to the type of the to-be-identified animal, and determining a driving route according to the animal position information; and driving the to-be-identified animal out of the production site according to the driving strategy and the driving route. The patrol robot provided by the invention can identify and position animals in a patrol area, formulate a driving strategy and a driving route, and dynamically drive foreign animals away. And the problem of potential safety hazards caused by foreign animals to the production area of the civil explosive industry is effectively solved.

A animal husbandry system, where a group of animals can move about freely in a shed, a plurality of sensors being provided, suitable for monitoring respective different regions of the shed, the sensors being provided in such a way that, collectively, they are suitable for monitoring substantially the complete shed. The animal identification system provided is suitable for identifying and locating individual animals in the shed, and includes processing means operatively connected to the sensors and to the animal identification system, the processing means being programmed for determining for each individual animal a urinating and / or defecating pattern or frequency. In this way, a highly useful system is realized. Firstly, in order to be able to monitor the health status of animals, urinating and / or defecating patterns or frequencies are highly meaningful. Secondly, they render an efficient cleaning of the shed possible.

The invention relates to the technical field of contour detection and recognition, and discloses a human and animal recognition method and system based on contour detection, and the method comprises the steps: executing a contour rectangle fitting strategy, fitting an external rectangle of an object according to an object point set, and outputting the geometric features of the object; executing a contour axial change detection strategy, calculating a main shaft of an external rectangle, and calculating a change value of geometric features in real time; executing an adjoint relation detection strategy, calculating displacement vectors of the human body point set and the object point set, and identifying an object owner of the object based on a set ratio threshold value; executing a surrounding-main back-and-forth motion judgment strategy, and detecting a displacement vector of the object point set to judge a motion mode; executing an active cooperative motion judgment strategy, and detecting displacement vectors of the object point set and the human body point set to judge a motion mode; and a holding judgment strategy is executed, the motion mode that the object point set is attached to the human body point set is recognized, whether the object is a pet or not is detected, the vehicle recognition capacity of the elevator is improved, and the misjudgment risk is reduced.

A system (100), and method (200) thereof, is provided for identifying and forecasting animal activity. The method (200) may comprise receiving (202) the images (112) from the camera (110) to define image specific raw data (114); assimilating (204) the image specific raw data (114) with contemporaneous weather data (140); determining (206) a mean rate of animal sightings for each of a plurality of prior time intervals associated with the assimilated image specific raw data using a first analytics model; minimizing (208) a difference between observed sightings and predicted sightings based at least in part on outputs from the first analytics model using a second analytics model; and determining (212) an amount of likely animal sightings for each of a plurality of future time intervals based at least in part on the outputs from the first analytics model and outputs from the second analytics model using a third analytics model.

Presented is an animal identification tag (1) with a male component (2) and a female component (3) placed on either side of an animal's ear. These components are joined to secure the tag (1) to the ear. The male component (2) has a first panel part (20) and a pin (21), while the female component (3) has a second panel part (30) and a receiving part (31) for the pin. The first (20) and second (30) panel parts have distal and proximal ends. The pin (21) extends through the ear's piercing hole, and the receiving part (31) cooperates with the pin. At least one distal end has an extended part (22, 32) to prevent the corresponding panel part from passing into the ear's piercing hole.

Small animal trap (100) comprising a first trap (111) and a second trap (112); a trap housing (120) wherein the first and the second trap (111, 112) are arranged on the trap housing (120), in particular on opposite end sections (120a) of the trap housing (120); at least one identification device (130) for providing a small animal identification signal based on an identification parameter; a power source compartment (140) arranged on the trap housing (120), in particular in a central section (120b) of the housing (120), between the first and the second trap (111, 112); a control unit (150) configured to process the small animal identification signal and to determine whether the first and / or the second trap (111, 112) has caught a small animal, and to provide a trap status signal based on this determination; a base plate (122) which is attached to an underside of the trap housing (120);wherein the control unit (140) and / or the at least one identification device (130) are provided in a space between the base plate (122) and the underside of the housing (120); and a cover (121) coupled to a top of the trap housing (120).

The invention discloses a dynamic dome camera automatic tracking animal shooting system and method based on an AI edge. The system comprises an edge AI visual perception module, an edge decision and tracking control module and a dynamic dome camera execution module. The data storage and communication module is used for collecting video streams in real time by integrating a high-performance AI chip and a camera; a lightweight animal identification model is constructed and operated at the edge end, animal information is identified, when a target animal is detected, dome camera control information is calculated, a control instruction is generated, a control instruction of an edge decision and tracking control module is received, corresponding control operation of a dome camera holder and a camera is executed, local storage is carried out on edge equipment, and the target animal is identified. And information is transmitted to the cloud server through the low-power-consumption communication sub-module. According to the scheme, an automatic tracking animal shooting method on the edge side is provided, and a millisecond-level closed-loop control process from sensing to execution is realized.

A system (24) for assigning an unassigned RTLS tag (6) to an animal identity, represented by an RFID tag (4) on a specific animal individual (2'). Real-time location data is obtained from a real-time location system / RTLS (18). Animal identification data is obtained from an animal identification system (14) comprising a first RFID reader device (10). The RTLS (18) covers a monitored area (22) including the first RFID reader device (10). A control arrangement (26) is configured to obtain a first timestamp associated with a reading of the RFID tag (4) by the first RFID reader device (10), obtain a second timestamp associated with the unassigned RTLS (18) locating the unassigned RTLS tag (6) in a first predefined area (40) associated with the first RFID reader device, and establish a correlation for assigning the unassigned RTLS tag (6) to the animal identity if a predetermined relationship between the first timestamp and the second timestamp is fulfilled, and assign the unassigned RTLS tag (6) to the animal identity of the specific animal individual (2') based on the established correlation.

The invention concerns a method for manufacturing a part of an animal identification tag, comprising: - obtaining (81) a transponder (2) comprising: - an air coil antenna (21), and - an integrated circuit (22), - obtaining (82) a reception body (1) comprising: - a base (12) comprising: a lower side (121) and a peripheral wall (122) defining a housing, - filling (83) at least partly the housing of the reception body (1) with a resin (3), - inserting (84) the transponder (2) in the housing of the reception body (1), with the integrated circuit (22) located between the air coil antenna (21) and the bottom of the housing, - curing (85) the resin (3) encapsulating the transponder (2).

A system includes a memory storing computer-readable instructions and at least one processor to execute the instructions to obtain images, by at least one image capture device, of animal behavior information associated with an animal, compare the animal behavior information with a machine learning model, each machine learning model related to a particular animal health condition, determine that the animal behavior information indicates a potential health condition for the animal, obtain animal identification information for the animal with the potential health condition, and transmit a notification indicating the animal identification information for the animal with the potential health condition.

The application discloses a marine plankton identification method based on in-situ images and deep learning, which constructs a multi-stage deep learning architecture in series for processing of full in-situ images, and comprises the following steps: a model is constructed based on a YOLOv8 skeleton to quickly locate and frame plankton individuals in a complex background; a U-Net model is used to finely segment the individuals in the in-situ images at a pixel level, extract accurate contours and obtain high-quality single images; finally, the images are input into a PlanktonNet network based on an innovative design of a ViT architecture, the self-adaptive small-size slice embedding and the introduction of a convolutional word embedding layer are adopted to improve the small-scale target feature extraction capability, and high-precision and fine-grained classification of genera and species is realized. The application has the advantages that the target detection, semantic segmentation and identification tasks are organically integrated, the problems of low identification efficiency and low automation degree in the prior art are effectively solved, and large batches of marine plankton can be quickly and accurately identified from in-situ images.

A robot milking device includes a milking site, an animal identification device, robot milking means, a feeding device for providing feed to the dairy animal during the milking based on a feed allowance, and a controller for the robot milking system and for updating feed allowance. The method includes selecting a future time period, determining a free capacity C of the robot milking system, determining a sub-group of the dairy animals for which the controller may continue to provide feed after the milking, milking and feeding the dairy animals which visit the milking system, and additionally feeding a dairy animal from the sub-group depending on the determined free capacity C, if the dairy animal still has a feed allowance for milking.

The application relates to the field of image recognition classification, and discloses a wild animal species identification method suitable for a low-illumination environment, which comprises the following steps: target image acquisition, acquiring a target image in which wild animals exist in real time; image processing, performing image processing on the acquired target image through wavelet transform; image recognition, using deep learning to learn and acquire weights of the processed image, performing Fourier transform on the weights, acquiring Fourier domain information, combining image meta-information, performing vectorization on the meta-information through a nonlinear embedded multilayer full-connection neural network, and combining the information after vectorization to perform image recognition. Through the improved method of superimposing frequency domain information on space domain information, combining time information and geographical position information of animal appearance, and under the premise of increasing a small amount of parameters, the outdoor wild animal recognition capability is effectively improved, and especially the wild animal fine classification capability under low illumination and low resolution is improved.

This invention discloses a low-power analog front-end circuit and control method for animal tags. The circuit includes: a radio frequency (RF) unit, a power management unit, and an antenna. The power management unit includes a rectifier and a low-power power management circuit. The rectifier is connected to the antenna. The low-power power management circuit supplies power to the RF unit and digital circuits. The RF unit includes a load modulation and limiting circuit, a demodulation circuit, and a clock generation circuit. The demodulation circuit receives the demodulated envelope signal sent by the rectifier and outputs digital demodulated data to the digital circuits. The clock generation circuit extracts a clock signal from the antenna signal generated by the antenna and outputs the clock signal to the digital circuits. The load modulation and limiting circuit receives the limiting reference voltage sent by the rectifier and the uplink data output by the digital circuits. In this embodiment, an animal identification tag adapted for wireless communication achieves a balance between low power consumption and small area while ensuring communication reliability.

The invention relates to the technical field of transformer substation protection, in particular to a laser anti-small animal expelling device based on visual identification, which comprises a multispectral video acquisition module used for acquiring video image data of a transformer substation monitoring area in real time; the small animal identification and positioning module is used for identifying the type, the position and the movement track of the intruded small animal through a built-in multi-target small animal identification and tracking algorithm model and generating a small animal positioning signal containing coordinate information; the central control module is used for generating a random multi-mode expelling instruction aiming at a specific small animal variety according to a preset intelligent expelling strategy model; and the multi-mode expelling module is used for receiving the multi-mode expelling instruction. According to the method, the repelling means, the repelling strength and the action time are randomly combined, so that the repelling stimulation is unpredictable, and small animals can be prevented from generating adaptability to a fixed repelling mode, so that the long-term effectiveness of the repelling effect is ensured.

Milk extracting system (100) comprising a quarter milking point (500) with four milk meter devices (561, 562, 563, 564) for measuring milk yield and / or time duration and / or milk flow rate from each teat (111, 112) of an animal (101) during a milking session (311a), and an udder milking point (130) with four teat cups (131, 132, 133, 134), each fitting on a teat and comprising a liner (220a, 220b) and a shell (230a, 230b) to create a pulsation space (225). The system includes a pulsator device (210) with independent channels (221, 222) connected to the pulsation spaces, controlled by a controller (180). It also comprises an animal identification device (120a, 120b) and a memory device (190) for storing and utilizing milk data of the quarter milking point (500) to adjust pulsation settings for optimized milking when the animal is milked on the udder milking point.