325results about How to "Real-time processing" patented technology

The invention provides an intelligent networked automobile dynamics simulation method and system based on a physical engine, and relates to the technical field of intelligent networked automobiles, and the method comprises the steps: obtaining vehicle parameters and road condition information of an intelligent networked automobile, optimizing a multi-agent cooperative control model through a self-adaptive optimization algorithm based on the vehicle parameters and the road condition information, and carrying out the optimization of the multi-agent cooperative control model. Constructing a digital twinborn model, and mapping the optimized multi-agent cooperative control model to an actual intelligent networked automobile control system; the method comprises the following steps: establishing a distributed edge computing network of an intelligent networked automobile group, constructing a decentralized knowledge base by utilizing a block chain technology, forming a decision-making experience data set, designing a multi-level attention mechanism to extract target information based on the decision-making experience data set and multi-modal data of a current traffic scene, and generating a final vehicle control instruction. And transmitting the final vehicle control instruction to a vehicle execution system to realize dynamic adaptive group cooperative control of the intelligent networked vehicle.

The invention relates to the technical field of bird repelling, in particular to an intelligent bird recognizing, tracking and repelling device, which comprises a video monitoring module, a tracking module, a bird repelling module, a control module and a power supply module, and is characterized in that the video monitoring module is used for monitoring bird conditions in a no-fly zone in real time by using a camera; the laser bird repelling module is used for repelling birds by using a laser; the control holder module is used for installing the video monitoring module and the laser bird repelling module; and the core control module is used for collecting the video information transmitted by the camera in the video monitoring module through identification and analysis, and detecting and tracking the birds in the no-fly zone. According to the invention, multi-algorithm real-time processing of the STC target tracking algorithm, the ECO algorithm and the YOLO-V3 detection algorithm is realized, the operation efficiency of the system is improved, the YOLO-V3 detection algorithm can adjust the size of a tracking frame according to the size of an identified target, targets of different sizes can be better adapted, the tracking accuracy is improved, laser is used for repelling birds, and the tracking accuracy is improved. The bird repelling effect is improved, and meanwhile safety is considered.

The invention provides a multi-channel signal acquisition and analysis system for safe navigation of a ship, and relates to the technical field of signal acquisition, and the system comprises a signal acquisition layer which comprises signal acquisition equipment and a signal conditioning module which comprises a programmable filter, a gain amplifier and an analog-to-digital converter; the signal processing layer comprises a logic control module, and the logic control module adopts an FPGA (Field Programmable Gate Array) and a reconfigurable data flow pipeline to carry out synchronous signal processing; the data analysis layer comprises an edge processing unit, an analysis algorithm library and a real-time decision unit; and the control and communication layer comprises a system control unit, a data transmission module and a storage management unit, the system control unit performs working mode management according to an instruction of the upper computer and is provided with a resource scheduler and a power consumption manager, and the data transmission module is provided with a multiple access transmission controller, a data compression unit and an encryption transmission module. According to the invention, comprehensive integration from signal acquisition to data analysis is realized, and the safe navigation capability of the ship is greatly improved.

The invention relates to a data channel state sensing method based on multi-source time sequence data fusion and a topological structure. The method comprises the following steps: collecting multi-source time sequence data; cleaning is conducted; obtaining a multi-source fusion time sequence; constructing a knowledge graph structure, wherein the knowledge graph structure comprises data and operation conditions of an equipment link topological structure, an equipment body and topological association equipment; obtaining a trained long short-term memory network model; and inputting real-time data into the trained long-short-term memory network model, predicting the real-time data by the trained model, and dynamically sensing the running state of the equipment. According to the method, a knowledge graph structure is adopted, modeling is carried out on data and operation conditions of an equipment link topological structure, an equipment body and associated equipment, the complex association relation between the equipment can be expressed in the mode, the sensing capacity of the equipment state is improved in combination with the actual operation data, and the real-time performance of the equipment is improved. Therefore, the monitoring system can perform more comprehensive state analysis by using the existing professional knowledge and the association between the devices.

The invention discloses a speech recognition streaming feature extraction and batch reasoning method for a high-concurrency artificial intelligence real-time dialogue system. According to the method, a heterogeneous computing architecture in which a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit) cooperatively work is adopted, operations such as preprocessing, feature extraction and normalization of voice data are executed through the CPU, and batch feature reasoning is executed by utilizing the GPU; the method specifically comprises the following steps: after segmenting and preprocessing voice data at a CPU (Central Processing Unit) end, executing Mel filter bank feature extraction and cepstrum mean variance normalization to obtain a standardized acoustic feature vector; extracting feature vectors from the feature buffer area in batches at the GPU end, filling and aligning the feature vectors, and sending the feature vectors to a neural network model for parallel reasoning calculation; according to the method, streaming processing and batch reasoning in the speech recognition process are realized, and the concurrent processing capability and the real-time performance of the system are effectively improved.

The invention discloses a plant disease and insect pest detection method and system, and belongs to the field of forestry production. In order to solve the problems of poor precision and time consumption of existing plant disease and insect pest detection, the invention provides a plant disease and insect pest detection method which comprises the following steps: acquiring a visible light image of a plant in a nursery garden and three-dimensional depth information of the plant to obtain an RGB-D image; dividing the RGB-D image into a plurality of equal-size grids, preprocessing the grids, inputting the grids into the deep learning model for detection, and finally obtaining a plant disease and insect pest detection result. According to the detection method, the three-dimensional depth information of the plant is introduced, so that the position and the range of the pest and disease damage can be more accurately judged; meanwhile, the robustness of the deep learning model is enhanced, and the model can still accurately detect diseases and insect pests under the condition that the illumination condition is poor; the whole detection method not only improves the detection accuracy, but also improves the detection efficiency, so that real-time and automatic disease and pest detection can be realized in a field environment, and the method is suitable for large-scale popularization and application. The system is simple in structural composition, real-time processing and analysis are carried out while data are collected, the complexity of installation and maintenance is reduced, and the real-time performance of the whole detection process is enhanced.

The invention relates to the technical field of environment monitoring, in particular to a sewage monitoring method and system in a rainstorm scene based on deep learning. The method comprises the following steps: acquiring sewage quality data and meteorological data based on a water quality monitoring node, and generating an environment monitoring data sequence; performing cleaning and feature extraction on the environment monitoring data sequence to generate a water quality spatial-temporal feature matrix; inputting the water quality spatial-temporal characteristic matrix into a deep learning model, predicting a sewage quality change trend and an abnormal event, and generating a prediction result sequence; based on the prediction result sequence and historical environment monitoring data, constructing a multi-level decision tree model, and generating a multi-level regulation and control scheme; and executing the regulation and control scheme and feeding back an execution effect, and optimizing the deep learning model and the multi-stage regulation and control scheme through a federated learning updating strategy. According to the invention, sewage quality change and abnormal events can be accurately monitored and predicted under extreme weather conditions such as rainstorm, and the real-time performance and accuracy of sewage treatment are improved.

The invention discloses a DQN-based intelligent factory job shop scheduling method and system, and belongs to the technical field of production scheduling, and the method specifically comprises the following steps: obtaining real-time data generated by intelligent equipment, and extracting key shop system features; calculating a current workshop system state based on the key workshop system characteristics; based on the current workshop system state, using the DQN network to select and execute an optimal scheduling rule, and obtaining a scheduling result; and calculating and obtaining a reward function rt based on a scheduling result. And finally, updating the weight of the DQN network according to the workshop system state and the scheduling rule, obtaining the updated DQN network according to the updated weight of the DQN network, and selecting a better scheduling rule in future decisions by using the updated DQN network. According to the method provided by the invention, a large amount of data can be processed in real time, and the optimal scheduling rule is automatically explored and learned through deep reinforcement learning, so that the scheduling efficiency and the capability of coping with a complex environment are improved.

The invention discloses a water and soil conservation monitoring system and method based on the Internet of Things technology, and relates to the technical field of water and soil conservation monitoring. A data acquisition module of the Internet of Things technology is utilized to judge small, medium and large drainage basins according to the drainage basin length of a monitoring area; then real-time data of soil humidity, drainage basin correlation and water and soil conservation measures are collected in real time from a monitoring area, a calculation and evaluation module is used for respectively inputting, sequentially calculating and outputting a soil erosion rate TQ, a drainage basin monitoring area evaluation value LY and a water and soil conservation measure adjustment value SBF, and the water and soil conservation measure adjustment value SBF is calculated and output on the basis of the water and soil conservation measure adjustment value SBF; according to the method, the accuracy of monitoring and the pertinence to small, medium and large drainage basins are ensured, a closed-loop feedback mechanism is formed, an innovative monitoring method is realized, and the beneficial effects provide more scientific, comprehensive and effective support for water and soil conservation work.

The invention relates to a dynamic signal event rule engine processing method, system and device and a medium, and the method comprises the steps: writing preprocessed ship and aircraft data into message middleware; and an event rule is constructed, the event rule comprises a data source node, a service node and an output node, the service node is used for carrying out conditional judgment and logic processing on the dynamic and static data of the ship and the airplane, and the output node is used for outputting alarm information. And converting the event rule into an SQL statement of the Flink, and sending the SQL statement as a task to the Flink for operation. And when the Flink runs, reading real-time dynamic and static data of the ship and the airplane in the message-oriented middleware, judging based on an event rule, and writing the data meeting the requirement into an alarm theme. The problem that heterogeneous signals and data of different data sources are inconsistent when the event rule engine is applied to ship and aircraft dynamic signals in the prior art is solved.

The invention relates to the technical field of intelligent traffic, and particularly discloses a pedestrian movement track prediction method and system based on visual information and a storage medium, and the method comprises the following steps: unmanned aerial vehicle data collection, coordinate projection transformation, coordinate sampling and track aggregation, sampling of pedestrian ground real coordinates after projection transformation, and tracking of the pedestrian movement track. Aggregating the coordinate data into trajectory information of a plurality of pedestrians; spatial visual information and time information are combined, a global interactive directed topological graph is constructed, then spatial feature and time feature extraction is performed, time sequence features of pedestrian motion are efficiently extracted through a multilayer convolution structure, and finally the future motion trail of pedestrians is predicted. Through the method, pedestrian motion in a complex scene can be predicted more efficiently, and the application effect of the model in pedestrian flow management and public safety planning is enhanced. The future movement condition is predicted by analyzing the historical trajectory of pedestrians, the pedestrian flow management is obviously optimized, and the safety level of public places is improved.

The invention discloses an SKU intelligent classification and label generation method fused with a visual pre-training model. The method comprises the following steps: obtaining image data and text data of an SKU; extracting multi-modal features based on the image data and the text data; according to the multi-modal features, zero sample classification of the SKU is realized; generating an SKU label through an improved fractal condition variational auto-encoder based on a multi-modal feature; on the basis of newly added SKU data, continuous learning of the model is realized; carrying out compression and mobile terminal deployment on the model; through an innovative fractal attention encoder and a chaotic dynamics condition generation decoder, the classification accuracy and the label generation quality are remarkably superior to those of an existing method, the BLEU scores of 95.8% and 0.76 are achieved respectively, commodities can be classified and described more accurately, and the search and recommendation effects of a platform are improved; and due to the continuous learning ability and the dynamic category extension mechanism, the performance is excellent when emerging commodity categories are processed, and the recognition rate reaches up to 89.2%.

The invention discloses a multi-story building mapping method and system based on laser radar and IMU coupling and a medium, and the method comprises the steps: scanning and collecting environment information through a laser radar sensor to obtain point cloud data, carrying out the motion distortion removal of a laser radar point cloud through IMU input data, and carrying out the down-sampling of the point cloud data through a grid down-sampling algorithm; carrying out ground point cloud removal on the single-frame point cloud based on a Markov random field; constructing an optimized pose map based on a multi-plane ground constraint pose constraint algorithm; calculating to obtain a fusion error value by using an error fusion algorithm based on Kalman filtering, and compensating motion estimation; registering and fusing the point cloud data of the current key frame and the global map obtained by accumulation at the preorder moment through a point iteration ICP algorithm; according to the mapping method, the phenomena of mapping drifting and scene degradation are effectively restrained for corridor scenes, upstairs and downstairs and use scenes of multiple similar floors with similar indoor mapping.

The invention relates to the field of video inspection, in particular to an active video inspection system based on Internet of Things driving, which comprises a current analysis module, an image input module, an image input detection module, an image output detection module and an audio detection module. According to the method, the current of the conference room equipment is monitored, abnormal equipment is found in time, rapid fault diagnosis is achieved, the effectiveness of the video stream is effectively judged by obtaining and processing the video stream and analyzing the grayscale image, it is ensured that the input picture is always in an available state, the blank screen phenomenon and picture distortion are avoided, and the user experience is improved. The image output detection module verifies the display effect, ensures the accuracy of an output picture and determines an abnormal area when the output picture is abnormal, the audio detection module plays a test audio and collects feedback, the sound receiving effect of a microphone is effectively evaluated, the conference room equipment is automatically and actively inspected, the equipment state is continuously monitored and evaluated, and the sound receiving efficiency is improved. Potential faults can be found and prevented in time, and the sudden fault risk of equipment is remarkably reduced.

The invention discloses a pollution visual data processing system and method applying a cloud computing technology, and relates to the technical field of pollution data processing. Analyzing the correlation of the data between the monitoring stations, and determining the related monitoring station of each monitoring station; determining the contribution rate of a single monitoring station to all the monitoring stations and the contribution rate of all the monitoring stations to the single monitoring station based on the correlation of the data between the monitoring stations; determining the processing sequence of the atmospheric pollution data of the monitoring station by the cloud platform, generating a first processing result and a second processing result, and visually displaying the first processing result and the second processing result; the correlation of atmospheric pollution data among different monitoring stations is analyzed, part of monitoring station data is processed in real time, and the data processing burden of a cloud platform is reduced; real-time performance and safety are both considered, data of all monitoring stations are processed regularly, and the possibility that abnormal conditions are ignored at monitoring sites is reduced.

The invention provides an epitaxial growth rate calculation method and system based on a neural network. The method comprises the following steps: collecting reflectivity data in a wafer epitaxial growth process; and preprocessing the reflectivity data to obtain input data. According to the method, the complex nonlinear relationship among the reflectivity data, the refractive index value (n) and the extinction coefficient value (k) is constructed through the neural network model, and the preprocessed reflectivity data is input into the neural network model, so that the refractive index value (n) and the extinction coefficient value (k) of the nonlinear relationship can be quickly and accurately obtained; the epitaxial growth rate of the wafer is calculated according to the refractive index value (n) and the extinction coefficient value (k). The method is high in processing efficiency, the epitaxial growth rate can be rapidly and accurately calculated in real time, the neural network model can adapt to multiple types of epitaxial growth by adjusting weight parameters in training, and the method has multi-process universality and is wide in application range.

The invention discloses a key area intelligent monitoring system based on communication towers, and relates to the field of communication technology, IoT and intelligent monitoring, and the key area intelligent monitoring system is characterized in that each communication tower is provided with a front-end sensing layer; the front-end sensing layer is used for acquiring sensor data of a key area and preprocessing the sensor data to generate preprocessed data; the channel transmission layer is used for transmitting the preprocessed data to the data processing and analyzing layer; the data processing and analysis layer is used for generating a self-adaptive abnormal behavior recognition model based on a machine learning model library, processing the preprocessed data according to the self-adaptive abnormal behavior recognition model, analyzing abnormal behaviors and generating a data analysis result; the application service layer is used for transmitting the data analysis result to the mobile terminal and formulating an early warning strategy according to abnormal behaviors in different scenes; the early warning strategy comprises an early warning threshold value and a response rule. According to the invention, comprehensive coverage of a monitoring area, rapid monitoring response and timely data processing can be realized.

The invention discloses a data security management and control method and device based on classification and grading, a medium and a product, and relates to the technical field of data security management, and the method comprises the steps: obtaining a data confidentiality grade labeling strategy; performing grade labeling on the data stored in each local computer and the data center according to a grade labeling strategy; establishing a data classification grading and security control model based on a neural network by adopting a transverse federated learning architecture; performing data set division based on the desensitized data, different from the data confidentiality level of the local computer, in the data center and the data in the local computer to obtain a training set and a test set; according to the training set and the test set, training the data classification and grading and security management and control model to obtain a trained model; and based on the trained data classification and grading and security management and control model, carrying out classification and grading on data with unknown confidentiality levels. According to the invention, accurate classification and security control can be carried out on the data with unknown confidentiality levels.

The invention discloses an intelligent workshop digital twin modeling and simulation system, and belongs to the technical field of intelligent manufacturing. The system comprises a multi-dimensional workshop modeling and management module, an intelligent analysis and decision module and a data management module. The multi-dimensional workshop modeling and management module dynamically displays and generates a scheduling flow through two-dimensional and three-dimensional models which are highly consistent with the actual workshop layout in combination with a production scheduling scheme; the intelligent analysis and decision module processes and analyzes data by applying a scheduling algorithm, optimizes a production process and updates scheduling information in real time; the data management module is based on a server, transmits and integrates data of the multi-dimensional workshop modeling and management module and the intelligent analysis and decision support module, realizes seamless communication between the modules, and ensures consistency and real-time performance of the data. The combination of the three modules effectively solves the problems of insufficient real-time performance, difficulty in comprehensive reaction of data complexity, poor user experience and the like in the prior art.

The invention discloses an intelligent data compression method for a twin system. The method comprises the following steps: collecting video, audio and text data for preprocessing; constructing a Bootstrap model, and performing feature extraction and preliminary compression on the input data; the method comprises the steps that a Combined model composed of a Bootstrap model and a Support model is constructed, and probability distribution output by the Bootstrap model is combined with probability distribution output by the Support model; finally compressing the output probability distribution by using an arithmetic coding method so as to reduce the volume of data transmission, and transmitting the data to target equipment; the target device decodes and reconstructs the received data and restores the original data before compression; according to the method, the data compression ratio and the compression time efficiency are improved, so that the time delay of data transmission is effectively reduced, the method is suitable for various data types in a twin system, and real-time processing of the data is ensured.

The present invention relates to the technical field of sleep diagnosis, and in particular to a sleep staging method based on deep learning, the method comprising: collecting a number of target physiological signals, preprocessing each target physiological signal to obtain a number of standard physiological signals; synchronously aligning and integrating each standard physiological signal in turn to obtain a standard sleep data set; determining the actual correlation between each physiological signal to be integrated in the standard sleep data set, and determining an actual signal combination strategy according to the actual correlation; generating an actual sleep data set according to the actual signal combination strategy, and extracting features from the actual sleep data set to generate actual feature parameters; inputting the actual feature parameters into a deep learning model to determine the actual sleep state; analyzing the actual sleep state, and performing corresponding sleep intervention according to the analysis results. The present invention realizes low-cost non-intrusive sleep monitoring, thereby improving people's physical health.

The invention discloses a navigation method and system for an inspection robot in a breeding house, and relates to the technical field of intelligent agriculture. The method comprises the following steps: dividing the breeding house into a plurality of areas, and preliminarily planning an inspection path of the inspection robot in combination with a shortest path algorithm and a depth-first search algorithm; environment data and animal data are collected in real time through an internal sensor group and an intelligent ear tag, a health data group is obtained through preprocessing, an environment health state index and an animal health state index are calculated, a comprehensive health index is further obtained, and the comprehensive health index and a preset first comprehensive health threshold value and a preset second comprehensive health threshold value are evaluated. And when path planning is triggered by health assessment, a dynamic path optimization model is adopted to adjust an inspection path, and an area with relatively high health risk is preferentially covered until the health assessment of the inspection task area reaches a good state. According to the method, the breeding environment and animal health monitoring efficiency can be effectively improved, and health management and risk control in the breeding process are ensured.

The invention discloses a BSM-YOLO-based rat hole detection method, and belongs to the field of target detection, and the method comprises the following steps: introducing a BiFormer module and a Shuffle Attention mechanism into a YOLOv5 target detection model to construct a BSM-YOLO rat hole detection model; and inputting the acquired image data into the BSM-YOLO mouse hole detection model to obtain a mouse hole detection result. According to the invention, through the YOLOv5 deep learning algorithm, automatic detection of the mouse hole is realized, and dependence on manual detection is reduced.

The invention relates to the technical field of big data, and particularly provides a talent body fluid subsidy application and redemption method and device.Firstly, a user management module performs user identity verification and authority management of an applicant, an auditor and a manager, and a data acquisition module performs data exchange with an external system through a standard API interface; automatic collection of user information is realized; then, a data analysis module carries out deep analysis on the user data collected by the data acquisition module, and a subsidy matching module clarifies subsidy objects, conditions, quotas and application process information; and finally, the application processing module performs automatic processing of an application process and automatic data verification, and the fund redemption module establishes an instant payment system to realize payment of subsidy fund. Compared with the prior art, accuracy and efficiency can be improved, the process is simplified, efficiency is improved, and justice is ensured.

The invention relates to a display apparatus and a near-to-eye display device. The device comprises a processing control chip used for sending a control instruction; the display sensing chip comprises a display area and a sensing area arranged on the outer side of the display area, and the display area is used for displaying virtual reality / augmented reality content; the invisible light emitter is installed on the peripheral side of the display sensing chip and used for emitting invisible light according to the control instruction; the sensing area is used for receiving reflected invisible reflected light and acquiring an eye image of an operation body; the processing control chip is also used for analyzing the received invisible reflected light and the received eye image, and determining an adjustment behavior and a target position of the display area corresponding to the adjustment behavior; the display area is further used for displaying the virtual reality / augmented reality content at the target position according to the adjustment behavior. The size of the near-to-eye display device is reduced, and the portability of the near-to-eye display device is improved.

The invention provides a multi-position overlapping image privacy protection method and system based on compressed sensing, and relates to the technical field of compressed sensing and image privacy protection, and the method comprises the steps: obtaining an original image, and dividing privacy parts and non-privacy parts of different parts of the original image; generating a coincident privacy mask for the coincident privacy part, generating a privacy mask for the non-coincident privacy part, confusing all the privacy parts, and performing image reconstruction by using the confused privacy part and the original non-privacy part; sampling encryption of compressed sensing is carried out on the reconstructed image, a compressed sensing ciphertext is generated, corresponding watermark matrixes and watermark embedding matrixes are generated for different privacy parts according to a confusion matrix, the watermark embedding matrixes and the watermark matrixes are embedded into the compressed sensing ciphertext, and encryption of the compressed sensing ciphertext is achieved; and users with different levels of keys are utilized to decode and reconstruct different parts of images of the encrypted compressed sensing ciphertext, so that privacy protection of multi-position overlapped images is realized.

The invention discloses a multi-source intelligent analysis system and method for power plant area personnel safety management. According to the system, multiple technologies including big data analysis, a deep learning algorithm, UWB high-precision positioning, Beidou positioning and navigation, a pressure sensor, a smoke alarm and an attitude sensor are integrated into an intelligent helmet worn by a worker, and a complete safety management solution is formed. The system can collect position information, behavior states and environmental data of operators in real time, perform multi-source data fusion processing, and intelligently analyze the standardization of operation behaviors and the safety of the surrounding environment. Through multi-dimensional risk assessment, early warning of safety risks of operators is realized, early warning prompts are sent to managers, and intelligent management and control support is provided for personnel safety in a power plant. The system is mainly composed of a data acquisition module, a data fusion processing module, a behavior analysis module and an early warning and feedback module, and can significantly improve the safety management level of a power plant area and the protection effect of operators.

The invention discloses a high-speed denoising method and device suitable for an airborne area array Geiger avalanche diode radar, and the method comprises the steps: enabling a radar image collection control module to achieve the data collection of the area array Geiger avalanche diode radar, and carrying out the segmentation of effective pixels and noise pixels of a collected 64 * 64 image; self-adaptive denoising is added by analyzing a radar equation and equipment parameters; automatically setting a signal threshold according to avalanche diode data distribution in a single frame; the signal proportion is increased by using matched filtering; signals are extracted and noise is removed by using a time and space correlation method. According to the invention, real-time de-noising processing is carried out on acquired radar data by combining the inherent characteristics of the area array Geiger avalanche diode radar, data distribution in a single frame, a time-space correlation method and a matched filtering algorithm, so that the problem of data submerging of the Geiger avalanche diode radar under an extremely low synthesis frame number is solved; the background noise interference in the image can be effectively reduced, and the pixel signal centroid can be quickly obtained.

The invention provides a live broadcast content auditing method, device and equipment and a computer readable medium, and the method comprises the steps: extracting a cut frame from a live broadcast video stream, taking the extracted cut frame as a to-be-audited image, calling a text detection model to carry out text detection on the to-be-audited image, and cutting the to-be-audited image according to the detected text. The method comprises the following steps: acquiring a to-be-audited image to obtain at least one character image, calling a lightweight text recognition model to perform text recognition on the character image for each character image to obtain a text recognition result, and performing violation judgment according to all the text recognition results corresponding to the to-be-audited image to obtain a violation judgment result. And then processing the live broadcast video stream which is judged to be illegal and / or the live broadcast room of the live broadcast video stream according to a preset illegal live broadcast processing rule, thereby realizing automatic auditing and processing of text content in a live broadcast picture, and effectively preventing spreading of illegal content.

The invention discloses an image sensor detection method, an image dynamic detection method and a correction method, and the detection method comprises the steps: screening any abnormal pixel point in abnormal cluster points as a center, the number of the second bright spots in the rectangular array area containing the two layers of outer neighborhood single-pixel rectangular rings is smaller than the number threshold value, and the corresponding abnormal pixel points serve as dead pixels; wherein the abnormal pixel point is confirmed by judging whether the gray value of the corresponding pixel channel is greater than or equal to the abnormal pixel point reference threshold value; and screening the corresponding abnormal pixel points of which the outer neighborhood single-pixel annular region is less than three continuous second bright points by taking the corresponding abnormal pixel point reaching the preset number threshold as the center as defective pixels. According to the method, the spatial distribution characteristics of the positive samples and the negative samples are effectively utilized, the positive samples and the negative samples are precisely discriminated in combination with the spatial distribution statistical information, and a reliable correction basis is provided for dead pixel correction.