52 results about "Surveillance Methods" patented technology

The invention relates to the technical field of video monitoring, and discloses a camera cooperative monitoring method and system based on multi-modal perception, and the method comprises the steps: detecting an abnormal event signal in an environment through a non-visual sensor node, triggering the activation of a camera, and carrying out the visual capture of a target region; the camera performs relay tracking based on target feature matching to generate a continuous motion track; performing multi-view collaborative shielding on privacy sensitive information in the tracking target, generating desensitized monitoring data, and uploading the desensitized monitoring data; and optimizing the energy consumption of the camera in a non-event triggering period, and updating the behavior recognition model based on the desensitization data. According to the method, the contradiction between privacy protection and monitoring efficiency is solved, cross-regional model evolution and energy consumption reduction are realized, and the problems of response lag, data redundancy, privacy disclosure and high energy consumption of traditional monitoring are systematically avoided.

The invention relates to the technical field of community video monitoring, and discloses a community personnel abnormal behavior video monitoring method based on human factor engineering, and the method comprises the steps: dividing community monitoring video data based on a preset activity region, and obtaining the monitoring data of each sub-region, including a personnel track, environment interaction and the like; determining the abnormal level of each sub-region according to the movement frequency of the personnel trajectory data and the interaction strength of the environment interaction data; key monitoring nodes are calculated in combination with abnormal levels and the like, and the scanning sequence and the scanning duration are adjusted and determined according to a dynamic path; then generating observation anchor points and redundant anchor points at the boundaries of the sub-regions; and finally, associating and integrating to generate a monitoring path. According to the method, through multi-dimensional data processing and dynamic optimization, the accuracy, flexibility and environmental adaptability of community abnormal behavior monitoring are improved, and the method is suitable for community safety management.

The invention discloses a video monitoring system, a video monitoring method and a spherical camera. The system comprises an edge computing node cluster deployed in a monitoring area, and a heterogeneous computing unit integrating a GPU (Graphics Processing Unit) and a DSP (Digital Signal Processor), a 8K image sensor and a cache unit are arranged in the ultra-high-definition video acquisition module; the multi-type environment sensor array is connected with the edge computing node cluster through a high-speed interface and collects multi-dimensional data such as temperature, sound and vibration. Constructing a hierarchical transmission channel based on a 5G network slicing technology; a cloud distributed computing platform extracts video dynamic features through a space-time convolutional neural network, and constructs a fusion decision tree model driven by mutual information in combination with sensor spectrum analysis. Through edge-cloud cooperative computing, dynamic feature optimization and high-precision space-time calibration, the technical bottlenecks that a traditional monitoring system is poor in real-time performance and rough in multi-source data fusion are solved, the anomaly detection accuracy and response speed in a complex scene are remarkably improved, and the method is suitable for the fields of smart cities, industrial security and protection and the like.

The invention discloses a high-voltage transmission line insulator health monitoring method and system, and the method comprises the steps: collecting multi-modal data to construct a data set, constructing a model containing an ITCN time sequence branch and a double-branch U-Net space branch, achieving the cross-modal fusion through an LTSFM module, and carrying out the cross-modal comparison pre-training and the optimization of a multi-task loss function (DiceLoss + WIoULoss + TSCL), thereby achieving the health monitoring of a high-voltage transmission line insulator. And finally, outputting health state classification, fault segmentation masks and degradation trend parameters, and triggering fault early warning when a threshold condition is met. According to the invention, the high-precision real-time monitoring of the weak degradation characteristics of the insulator can be realized, and the edge deployment efficiency and the complex environment adaptability are both considered.

The invention discloses a video monitoring method based on multi-target tracking analysis, which relates to the technical field of computer vision, and comprises the following steps: acquiring original trajectory data of each target, extracting key data and associated features of each target according to the original trajectory data, and acquiring vector data; the method comprises the following steps: distributing a unique identifier to each target, generating accurate original track data, extracting key data and associated features of the target, matching vector data with a rule base to identify an abnormal behavior of each target, triggering classification early warning according to the abnormal behavior, and dynamically optimizing the rule base according to different scenes. The rule base is dynamically optimized according to different conventional scenes and unconventional scenes, the practicability and pertinence of monitoring are enhanced according to classification early warning and historical track labeling functions triggered by abnormal behaviors, and the safety management requirements of various scenes are better met.

The invention relates to the technical field of artificial intelligence and channel disease monitoring, and discloses a channel revetment multi-dimensional disease intelligent monitoring method and system based on multi-source fusion, and the method comprises the steps: obtaining multi-source data of a target revetment region; constructing a three-level index system; performing DERU evaluation on each third-level index, and calculating a risk category score corresponding to each second-level index according to an evaluation result and a second-level index classification result; dividing the secondary indexes into a disease class and an environment class according to influence types, calculating a total risk value according to the risk class score of the secondary indexes of the disease class, and calculating an environment correction coefficient according to the risk class score of the secondary indexes of the environment class; calculating a health value score of the target bank protection area according to the total risk value and the environment correction coefficient; and outputting a disease intelligent monitoring result and a maintenance strategy according to the multi-source data and the health value score. According to the scheme, the health state of the revetment structure can be comprehensively, timely and accurately mastered, and identification monitoring and intelligent early warning of various diseases are realized.

The invention provides a factory traffic AI video monitoring method and system, electronic equipment and a storage medium, and aims to solve the problems that existing video monitoring is poor in complex scene adaptability and single in function, and the method comprises the following steps: collecting factory traffic data, and carrying out data preprocessing and labeling to obtain a data set; a Mask-RCNN model is selected as a vehicle detection and segmentation network, an attention mechanism is introduced into the model, a target classification system and a recognition strategy are customized, a TensorFlow deep learning framework is utilized to train the model through a data set, and the training process is optimized and adjusted; performing target detection and segmentation on a to-be-detected factory traffic video by using the trained model, and automatically identifying a target; and data processing and analysis are carried out on an identification result, so that intelligent decision-making of factory traffic monitoring is realized. According to the invention, the intelligent level of factory traffic management is improved, and the factory traffic safety and logistics efficiency are ensured.

This invention relates to the field of nursing technology and provides an internet-based remote monitoring method and system for health care. The method includes: screening for deviation events and identifying risk monitoring nodes, extracting sampling data only for these risk monitoring nodes; performing comparative analysis on computational parameters in chronological order, comparing the magnitude of parameter fluctuations with preset fluctuation thresholds based on historical data statistics, and identifying key parameters with significant fluctuations; and generating predictive intervention nodes through a preset assessment model. This invention avoids indiscriminate processing of all data by locating effective data segments with high-risk characteristics from the raw data, reducing computational load and data transmission pressure, and improving the targeting and efficiency of risk assessment. By identifying fluctuation nodes where key parameters change, it can more sensitively capture the initial signs of potential risks and improve the accuracy of risk identification.

The invention discloses a multi-modal health monitoring method and system based on large model driving, a terminal and a storage medium, and the method comprises the steps: collecting video stream data, audio data and physiological monitoring data of a target object, and carrying out the preprocessing of the collected data, and obtaining multi-modal health data; feature extraction is carried out on the multi-modal health data, and multi-modal health data fusion processing is carried out on the extracted features based on a multi-modal health large model; according to the fused health features, a dynamic intervention strategy is generated by means of intelligent agent decision logic driven by a large model; and performing voice reminding and health video playing based on the dynamic intervention strategy, and dynamically outputting multi-modal health monitoring data through a multi-modal interaction interface. According to the method, deep fusion of non-contact health monitoring and personalized health management is realized through technologies such as multi-modal sensor fusion, large-model-driven intelligent agent decision and natural interaction design, and the autonomous decision and personalized capabilities of the health monitoring technology are improved.

The application discloses a human health monitoring method and system based on proportion philosophy and individual self-adaption. A multi-level nested proportion model is constructed by non-invasively collecting multi-dimensional signals of a human body. A golden section 0.618 is taken as a harmonious anchor point to calculate an individual exclusive health benchmark zero point and a stable interval. Through real-time proportion comparison and fine ladder deviation analysis, the structure harmony and function state are judged. The health trend early warning is realized by combining multiple retests. The application breaks away from the limitation of unified medical indicators, realizes personalized, global and original health evaluation, can early identify imbalance and sub-health state, and is suitable for scenes such as daily monitoring, physical evaluation and health intervention effect judgment.

A surveillance apparatus (101) includes a first acquisition unit (104), a second acquisition unit (105), and a correlation unit (106). The first acquisition unit (104) acquires first moving body information MD1 including the position of a held apparatus (107) held by a predetermined moving body MB. The second acquisition unit (105) acquires second moving body information MD2 including the position and the type of a sensor moving body SMB being a moving body MB detected based on sensor information SI generated by a sensor apparatus (103). The correlation unit (106) correlates the first moving body information MD1 and the second moving body information MD2 that are related to a common moving body MB, based on the position of the held apparatus (107), and the position and the type of the sensor moving body SMB.

The invention provides a monitoring system, a monitored device, a monitoring method, and a computer readable storage medium capable of efficiently setting an output condition of information related to a connection state. A monitoring system is provided with: a monitoring device that monitors network devices included in an IP network on the basis of an SNMP that is a protocol for managing the IP network; and the monitored device is a monitoring object of the monitoring device in the network equipment. The monitoring device transmits, to the device to be monitored, an output request for information relating to the connection state of the device to be monitored. When the destination information of the monitoring device which has transmitted the output request is not registered, the monitored device stores the destination information of the monitoring device in a storage unit. When a change in the connection state of the device to be monitored is detected, the device to be monitored transmits information relating to the connection state of the device to be monitored to the monitoring device identified by the stored destination information of the monitoring device.

A plurality of hierarchical information management devices (4a, 4b, 4c) have a first hierarchical information management section that manages first hierarchical information (50a, 50b, 50c) that is information that expresses the information of the devices in a hierarchical structure, and the plurality of hierarchical information management devices (4a, 4b, 4c) each manage the first hierarchical information (50a, 50b, 50c) of different devices. The monitoring device (2) has a second hierarchical information management section that generates hierarchical information that is second hierarchical information that connects the first hierarchical information (50a, 50b, 50c) in a hierarchical structure based on a plurality of different first hierarchical information (50a, 50b, 50c) acquired from the plurality of hierarchical information management devices (4a, 4b, 4c), a display section that displays information, and a display processing section that performs processing to display a plurality of different monitoring screens that correspond to the operation states of the display monitoring targets of each hierarchical level in the second hierarchical information on the display section by switching the monitoring screens.

The application provides a kind of old patient sign monitoring method and system based on multi-modal perception, it is related to intelligent health monitoring technical field, including: acquisition and fusion physiological and behavioral data, construct unified representation and modal correlation matrix.Baseline features are calculated and tensor fusion is carried out, abnormal fragments are identified and dominant influence factors are traced back by encoding reconstruction.Risk level is mapped by deducing data evolution trajectory and calculating deviation from health benchmark.Simulate the timeliness of intervention scheme, determine the target scheme and generate monitoring report.The application realizes the deep correlation analysis and early warning of abnormality of multi-modal data, can automatically determine the risk dominant factor and generate individualized intervention scheme, improves the accuracy and initiative of old health monitoring.

The invention relates to the field of video monitoring, in particular to a video monitoring method for intelligent medical patient nursing, which comprises the following steps: determining a personnel state based on a personnel proportion value and a non-affected characterization value of a monitoring area, and determining whether to carry out frame optimization or not according to the personnel state; according to a comparison result of the association density corresponding to the preset video frame and the preset association density, determining that a data uploading mode is to upload the preset video frame and the corresponding compensation video frame or upload the preset video frame and the corresponding video paragraph; constructing a region sequence based on the stable number of the nursing personnel in the reference range corresponding to each monitoring region, and determining whether sequence adjustment is performed or not according to a comparison result of a stable number difference value of the nursing personnel in the region sequence and a preset number difference value; and carrying out data uploading on the finally obtained region sequence according to a sequence number ascending sequence. According to the invention, the video monitoring data processing efficiency of patient nursing is improved.

The invention discloses a public opinion monitoring method and system based on multi-modal data fusion. The public opinion risk management and control capability and the emergency response level in a complex public opinion environment are improved. The method comprises the following steps: collecting multi-modal data, and carrying out edge preprocessing on the collected multi-modal data to generate structured input data; performing feature extraction and projection on the multi-modal data to generate a unified event embedding representation; identifying event types based on a public opinion classification model, and performing risk assessment on events in combination with set parameters; constructing a public opinion knowledge graph, and inputting cue words into the pre-training language model to generate a response strategy; and triggering early warning based on a hierarchical risk threshold, and performing reinforcement learning fine tuning and strategy optimization on the pre-training language model based on event feedback and response results.

The present application discloses a method and system for health emergency monitoring under heavy pollution weather, which involves health emergency, data processing, and artificial intelligence, including: establishing a map mark of a target area on a corresponding regional map according to the location of each health monitoring point; determining possible monitoring blind spots; configuring a drone flight path according to possible monitoring blind spots; obtaining health monitoring data covering the target area from health monitoring points and configured drones when the indicators of weather monitoring data exceed the prediction threshold; preprocessing the health monitoring data, and using an improved ConvLSTM model to predict the possible distribution of health events in the future based on the preprocessed health monitoring data; performing hierarchical marking based on the regional map according to the prediction results, and executing corresponding health emergency strategies for the hierarchically marked areas. The method of the present application can effectively improve the auxiliary management effect of health emergency.

The invention discloses a platanus orientalis wadding monitoring method based on an image recognition technology, and relates to the technical field of environmental monitoring, and the method comprises the following steps: selecting a road containing platanus orientalis during the centralized outbreak of platanus orientalis wadding every year, and collecting the floating wadding through a floating wadding collection container; identifying the floating flocculus area in the floating flocculus collecting container by using an image identification technology; obtaining the weight of the floating catkins in each floating catkins collecting container; fitting the floating wadding area and the floating wadding weight through a floating wadding area-weight algorithm formula; quantifying floated floc grades, and dividing the floated floc of the platanus orientalis into four grades of'no ', 'light', 'medium 'and'heavy'. According to the method, the flouring grade standard can be quantified, the flouring phenomenon can be converted from subjective description to measurable scientific indexes, the subjective deviation of manual observation is eliminated, the data comparability of different areas and different time is ensured, and a unified and objective reference basis is provided for urban flouring treatment management, public health protection and scientific research.

Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for retroactive event detection. The methods, systems, and apparatus include actions of obtaining an image captured by a camera at a current time, determining that the image depicts a change in a region from a previous image captured by the camera at a previous time, determining, based on determining that the image depicts the change in the region, whether the change depicted in the image is of a known object type, determining, based on the determination that the change depicted in the image is of a known object type, whether the change does not correspond to a previously detected event, and determining, based on the determination that the change does not correspond to a previously detected event, whether the images captured by the camera between the current time and the previous time depict an event.

The invention relates to an urban surveillance method, comprising a training phase (100), intended to create training data; a detection phase (200), in which at least one detection vehicle (20) equipped with at least one sensor is mobile in an area to be monitored (22), the at least one sensor records data relating to the area to be monitored (22), then the recorded detection data is processed by a computer detection device (28), configured to detect elements (24) present in the area to be monitored using the training data; and an analysis phase (300), in which the area to be monitored (22) and the elements (24) present in the area to be monitored (22) are analyzed by a computer analysis device (38); characterized in that the training phase (100) and the detection phase (200) are distinct;and in that during the analysis phase (300), the computer analysis device (38) establishes a score for the area to be monitored (22), taking into account various parameters, including the different elements (24) present in the area to be monitored (22). The invention also relates to an urban surveillance system. Abbreviated figure: 1;

The invention belongs to the technical field of disease prognosis prediction, and provides a prognosis prediction method for multidrug-resistant tuberculosis in order to solve the problem that an existing prognosis monitoring method for a multidrug-resistant tuberculosis patient has certain limitation. By integrating handmade radiology features and deep learning derived imaging features extracted from baseline, two-month and six-month continuous CT scanning, processing data by adopting a gated recurrence unit (GRU) network, and combining with a multi-drug-resistant tuberculosis prognosis prediction system, a treatment result of a multi-drug-resistant tuberculosis patient is predicted. According to the prediction method provided by the invention, the high-risk patient can be accurately identified in the early stage of the treatment process, and powerful support is provided for personalized treatment decision making, intervention opportunity optimization and reasonable public health resource allocation. Moreover, the kit has higher sensitivity and specificity, and the potential in the aspects of multi-drug-resistant tuberculosis treatment result prediction and risk stratification accuracy is improved.

The invention relates to the technical field of live pig breeding monitoring, in particular to a large-scale live pig farm monitoring method and system based on data analysis, and the method comprises the steps: obtaining original image data of a live pig farm, carrying out the denoising processing, size adjustment and image enhancement processing of the original image data, and obtaining target image data; target image data are input to a preset target area positioning model, a frame selection area image and coordinate frame data are output, the coordinate frame data comprise a first coordinate frame used for representing position information of a live pigtail hip area in the target image, and a second coordinate frame used for representing position information of a corresponding pigpen ground area in the target image; removing a non-ROI region in the target image data according to the frame selection region image and the coordinate frame data to obtain a feature map; and inputting the feature map into a preset target transfer learning model, outputting a judgment probability value, and realizing the prediction of the common epidemic diseases of the live pigs in the farm through the recognition of the diarrhea substances.

A surveillance apparatus (100) includes: a biometric information acquisition unit (102) that acquires biometric information about a target person present in a predetermined space; an image acquisition unit (104) that acquires an image including the target person and surroundings of the target person; a decision unit (106) that decides whether the biometric information satisfies a first criterion and a state of other person located around the target person satisfies a second criterion; and an output processing unit (108) that outputs predetermined information when it is decided that both of the first criterion and the second criterion are satisfied.

A surveillance apparatus (100) includes a detection unit (102) processing an image in which an area around a road is captured and detecting that the relative distance between a person and a vehicle included in the image is in a state of being equal to or less than a reference value, and an output unit (104) selecting a type of information to be output, by using an attribute of the person detected to be in the state, and outputting information of the selected type.

This invention discloses a primer design method and primer set. Through in-depth analysis of influenza surveillance network data nationwide, this application makes significant contributions to improving the accuracy and sensitivity of influenza virus identification. The primer revisions not only achieved significant results in experiments but also provided valuable insights for future influenza surveillance and research. A deeper understanding of the genetic characteristics and epidemiological patterns of influenza viruses is crucial for developing scientifically effective prevention and control strategies. In disease prevention and public health management, continuous improvement of surveillance methods and strategies will bring greater benefits to timely responses to infectious diseases such as influenza.

The present invention discloses a radar-guided video surveillance method, comprising: locating a target entering a monitoring area by means of a radar, determining whether the target is close to a warning area and calculating an estimated warning duration of the target, evaluating the warning level of the target and marking a monitoring point selection period, calculating a selection performance value of a video monitoring point, marking a first tracking monitoring point and adjusting an angle to track the target, comparing the selection performance values of various video monitoring points within the monitoring point selection period and determining whether to switch the first tracking monitoring point, generating an alarm signal and entering a video tracking period when the target enters the warning area, calculating a loss risk value of the target in real time and determining whether the angle of the first tracking monitoring point needs to be adjusted and switched, if the target is lost, retrieving the target through other video monitoring points and switching the first tracking monitoring point, and if the target cannot be retrieved for a long time, marking the lost area and generating a maintenance optimization signal.

The invention provides a monitoring device, a monitoring method, and a computer program product for monitoring. The monitoring device includes: a detection unit that detects a person from each of a plurality of chronological images generated by an imaging unit; a tracking unit that tracks the person in one or more images in which the detected person is displayed among the plurality of images; and a determination unit that, on the basis of the tracking result, determines whether or not the position of the detected person on the image is included in a first range on the image corresponding to the prohibited region over the entire first period. And determines whether or not the position of the detected person on the image is included in a second range on the image provided around the first range over a second period that is longer than the first period. When the position of the detected person on the image is included within the first range throughout the first period or within the second range throughout the second period, it is determined that the detected person has entered the prohibited area.

This monitoring system for monitoring a vehicle (10) or a monitoring target operating in the vehicle (10) comprises: a reliability management unit (1186) for managing a reliability indicating a security protection status of the monitoring target, in accordance with a vehicle event of the vehicle (10); and a function restricting unit (1163) for restricting at least some functions of the monitoring target in accordance with the reliability.