9 results about "Animal data" patented technology

The invention discloses a cross-species animal posture estimation method based on an animal behavior monitoring technology, and belongs to the technical field of artificial intelligence and computer vision. In order to solve the problems that in the prior art, cross-species attitude estimation is poor in complex scene robustness, low in system integration level, insufficient in adaptability and real-time performance and the like, the invention provides a cross-species animal attitude estimation method based on an animal behavior monitoring technology, and the method comprises the following steps: collecting image data of a multi-species target animal; preprocessing the image data to obtain a processed animal image; performing cross-species key point configuration and labeling on the image data set to obtain a cross-species animal data set; and performing multi-level feature extraction and fusion on the cross-species animal data set, and realizing a posture estimation result of the target animal based on the fused features. The method is suitable for the fields of wild animal protection, livestock breeding management, animal behavior research, health monitoring, disease prevention and control and the like.

The invention provides a medicine abnormal toxicity inspection electronic experiment record management system. The system comprises an inspection method management module, an animal management module, an animal data management module and an inspection result management module. Wherein the inspection method management module can carry out design, auditing and version control on a drug abnormal toxicity inspection experiment recording template; the animal management module is used for recording full life cycle data of experimental animals; the animal data management module is used for recording drug abnormal toxicity inspection experiment data; and the inspection result management module is used for generating, auditing and publishing electronic experiment records of drug abnormal toxicity inspection. According to the embodiment of the invention, the system achieves the automatic and full-life-cycle management of the drug abnormal toxicity inspection experiment data, guarantees the primitiveness, authenticity, safety and traceability of the laboratory data, effectively improves the inspection work efficiency, and achieves the comprehensive, scientific, unified, ordered and efficient computerized system management of the inspection data.

The present disclosure relates generally to keeping animals (10) and in particular to a method for investigating eating behavior of animals (10) in a livestock area (30). The present disclosure also relates to a control device (100) and to a computer program for performing the method. According to a first aspect, the present disclosure relates to a method for investigating eating behavior of animals (10) in a livestock area (30). The method comprises monitoring animal data for a period of time after feed has been dispensed at a feeding station (31) in the livestock area (30), the animal data being collected using tags carried by the animals (10) and being indicative of activity of the animals (10) and / or position of the animals (10) relative to the feeding station (31). The method further comprises performing an action when the monitored animal data fails to meet one or more eating criteria, the one or more eating criteria defining normal eating behavior of the animals (10) in terms of activity of the animals (10) and / or position of the animals (10) relative to the feeding station (31). The present disclosure also relates to a control device (100) and to a computer program for performing the method.

The invention discloses a wild animal on-line monitoring method based on satellite communication, which comprises the following steps: when an infrared sensor detects a signal, a camera module acquires current visual data, and a positioning module acquires spatio-temporal information; the main controller combines the spatio-temporal information and the current visual data into original monitoring data, and performs identification processing on the original monitoring data to obtain optimized data; calculating a transit time window of the next satellite according to the broadcasted satellite ephemeris data; the main controller sends the optimization data to a satellite through a satellite communication module in a transit time window; the satellite sends the optimization data to a ground gateway station; the ground gateway station sends the optimization data to a monitoring center server; the monitoring center server receives and analyzes the optimization data; and the mobile terminal obtains the data. According to the invention, efficient, accurate and intelligent wild animal data monitoring in a public network-free environment is realized. The invention further discloses a system for implementing the method, electronic equipment and a computer readable storage medium.

The application relates to the technical field of animal monitoring, and particularly discloses an experimental animal monitoring system and method, which comprises the following modules: an acquisition module for acquiring basic animal data, monitoring stage data, monitoring stage labels, monitoring target data and animal monitoring data of an experimental animal to be monitored, and determining real-time monitoring data; a monitoring module for determining monitoring abnormal data; an analysis module for acquiring historical abnormal data and determining reason abnormal data of each abnormal reason; an abnormal module for determining a candidate abnormal reason set and an abnormal reason label of each real-time monitoring label of the experimental animal to be monitored; and a determination module for determining real-time animal monitoring data of the experimental animal to be monitored. The candidate reasons and core causes of abnormal animals can be accurately positioned, the integration of abnormal monitoring, reason tracing and real-time data optimization is realized, the monitoring accuracy and systematicness are improved, different monitoring requirements are adapted to, and the real-time animal monitoring data is ensured to be comprehensive and reliable.

The application discloses a kind of navigation method and system of breeding house inside inspection robot, it is related to intelligent agricultural technology field, the method is by dividing breeding house into several regions, in combination with the shortest path algorithm and depth-first search algorithm, the path of preliminary planning of the inspection robot is patrolled, and environment data and animal data are collected in real time using internal sensor group and intelligent ear tag, by preprocessing, obtain health data group, calculate the health state index of environment and the health state index of animal, further obtain comprehensive health index, and preset first comprehensive health threshold and second comprehensive health threshold are evaluated.In health evaluation triggers path planning, dynamic path optimization model is used to adjust the inspection path, and the area with higher health risk is preferentially covered until the health evaluation of the inspection task area reaches a good state.This method can effectively improve the health monitoring efficiency of breeding environment and animals, and ensure health management and risk control during breeding process.

The invention relates to a quadruped robot traffic inspection controller training method based on animal real data, and the method comprises the steps: collecting animal motion data, solving a pose sequence, obtaining a control strategy through simulation reinforcement learning in combination with an original level state, generating a data set, and building an original level control network output action; and constructing a parameterized complex environment scene, training an environment level control network to output potential vectors to an original level network, and realizing complex environment motion control. According to the method, animal motion data is redirected, and an original level network is constructed by combining simulation reinforcement learning and a variational auto-encoder, so that motion feature representation compression and knowledge multiplexing are realized; environment-level network targeted training is constructed, and the environment-level network and an original-level network cooperatively improve the specific scene control capability; a top layer decision module is introduced to deconstruct an environment into a typical sub-scene, the scene is recognized based on sensing data and an ontology state, a corresponding environment level network is activated, and self-adaptive control is achieved.

The application provides a mountain collapse prediction method and a collapse prediction system, and relates to the technical field of mountain collapse prediction. The method comprises the following steps: first, obtaining animal data of a target mountain area, screening out target resident animals, determining monitoring points according to the daily activity range of the target resident animals, and implementing monitoring. Then, collecting historical and real-time daily activity video data of the target resident animals, extracting behavior characteristics, comparing and obtaining distinguishing characteristics, and matching the distinguishing characteristics to obtain a matching score. When the matching score exceeds a set threshold, a geological sensor is controlled to collect geological data. If the data is abnormal, it is determined that a mountain collapse will occur, and warning information is generated in combination with the distinguishing characteristics. Early warning is realized by using the sensitivity of animals to geological changes, and accurate judgment is realized through geological data.

A system for offline-capable, multimodal animal identification and secure data management, consisting of: one or more Internet of Things (IoT) collar devices configured for attachment to livestock. Each device incorporates a variety of physiological sensors and a Bluetooth Low Energy (BLE) communication interface configured to capture and transmit physiological parameters of the livestock via the Bluetooth Low Energy communication interface; a mobile computing device configured to receive multimodal livestock data from one or more IoT collar devices, the captured multimodal livestock data including images of the livestock, images of the muzzle pattern, acoustic vocalization samples, and GPS location data;an artificial intelligence processing module within the mobile computer device, configured to process the multimodal livestock data offline locally, identify breed characteristics of the livestock based on the multimodal livestock data and assess the health status of the livestock based on the multimodal livestock data and physiological parameters, generating a unique livestock identification number (ULID) for each individual animal;a blockchain-based distributed ledger within the mobile device, connected to artificial intelligence processing and configured to store the unique animal identification number, owner information and health data of each individual farm animal, and ensuring tamper-proof verification of the stored information, wherein the blockchain-based distributed ledger enables the creation of a digital twin profile for each individual farm animal, with this digital twin profile being dynamically updated based on incoming multimodal farm animal data and physiological parameters;and an analytics dashboard interface connected to the artificial intelligence module and a blockchain-based distributed ledger, configured to display predictive insights into the health, milk productivity, disease risk, and reproductive cycles of farm animals.