4306 results about "Video streaming" patented technology

Disclosed in the present invention are an edge-deployed semi-supervised anomaly detection method and system for a railway track foreign object. The method comprises the following steps: an edge device encoding and decoding a video stream captured by a camera to obtain an image frame sequence, and performing frame extraction; and using a semantic segmentation model to perform image segmentation on a certain image frame obtained by means of frame extraction, to obtain a railway track region segmentation image. The use of a single image as input may generate an expert model result having a high weight value; however, the determination based on a single image is not stable, multiple consecutive images of the task scene need to be inputted, the frequency of each expert model obtaining the highest weight is computed, and the expert model corresponding to the highest frequency is the final solution. The present invention supports scene-adaptive foreign object detection algorithm automatic selection, and a user can perform selection on the basis of prior knowledge, or selection may be performed by a scene-adaptive automatic algorithm selection method; the user only needs to provide a batch of image data of the current scene, and the optimal algorithm selection can be evaluated.

The invention discloses a digital human interaction method and device based on multi-modal sentiment analysis and a medium, and relates to the field of artificial intelligence, and the method comprises the steps: collecting multi-modal data of a user in real time through a multi-source sensor device; the multi-modal data comprises face video stream data, voice audio stream data and text dialogue data; calling data analysis engines corresponding to different modalities, and extracting corresponding modal feature sequences; according to the current interaction scene, the modal feature sequence and the historical dialogue context features are fused, and a comprehensive emotion evaluation result is generated; outputting a corresponding multi-modal response data packet based on the modal feature sequence through an interactive response engine corresponding to a comprehensive emotion evaluation result; and executing the multi-modal response data packet. And after feature fusion is carried out in combination with the current interaction scene, the generated response can more accurately fit the current emotion demand and communication context of the user, so that the digital human can be more easily fused into various scenes needing emotion interaction.

The invention discloses a cross-platform virtual-real fusion scene construction method and system based on AI space calculation, and relates to the technical field of artificial intelligence and space calculation, and the method comprises the steps: carrying out the multi-scale feature fusion based on a received cross-modal conversion instruction set, and generating an initial image sequence; carrying out implicit field coding on a target object by combining a three-dimensional reconstruction algorithm to obtain an initial parameterized model; performing space-time alignment on the multi-view video stream, loading a digital scene asset package in combination with physical sensing data and a preset spatial index structure, and establishing a bidirectional data channel between a virtual scene and a physical sensor; performing rendering and illumination parameter adjustment on the initial parameterized model to obtain an optimized parameter model; performing differential coding processing on the optimization parameter model to obtain a target virtual-real scene fusion model; and distributing the target virtual-real scene fusion model to a preset terminal. The invention provides a virtual-real fusion construction method for end-to-end collaborative optimization, which is suitable for cross-platform live broadcast or dynamic interaction scenes.

The invention relates to the field of image processing, and discloses an object monitoring method and device based on multi-source data and a storage medium, and the method comprises the steps: synchronously collecting multi-channel video streams, environment parameters and equipment position information, and carrying out the time alignment and data association processing, and forming associated data; performing feature matching on the multiple paths of video streams, fusing position information and environment parameters, and establishing a mapping relation between video feature points and a unified space coordinate system; splicing the multiple paths of video streams in real time according to the mapping relation to generate a panoramic video stream; performing dynamic scene analysis based on the panoramic video stream and the environmental parameters, and identifying a target type and a state to obtain an analysis result; and generating an equipment regulation and control strategy based on the analysis result, and generating and issuing a regulation and control instruction for adjusting the working parameters of the multi-view image acquisition device based on the equipment regulation and control strategy. According to the invention, the splicing precision, the real-time performance and the target identification accuracy of panoramic monitoring can be improved, and the dynamic adaptive regulation and control of the equipment can be realized.

The invention discloses a video stream adaptive low-delay real-time transmission method and system based on edge calculation. The method comprises the following steps: receiving a real-time video stream from a network camera, creating a pipeline queue, adding timestamp information for each video frame, and setting a queue protection mechanism; the coded video frames are taken out from the input queue, and the frames in the video are processed through hardware acceleration decoding; the resource use condition of the system is monitored in real time; executing a self-adaptive frame skipping decision according to a performance monitoring result; timestamp generation: dynamically calculating a timestamp interval according to an actual processing frame rate; receiving the decoded original video frame and the corresponding timestamp information, accelerating decoding by using hardware, and executing a video coding operation; and packaging and transmitting the coded video data, and providing a standard protocol interface to be connected with a client for playing. According to the scheme, stable low delay and relatively low resource occupation can be kept, and meanwhile, the video quality is remarkably improved.

The invention discloses an electric power operation risk early warning method and system based on knowledge enhancement and multi-modal fusion. The method comprises the steps that video monitoring data, sensor monitoring data and service system data are collected in real time through multi-source sensing equipment deployed on an electric power operation site; the method comprises the following steps of: extracting entities and relationships from unstructured texts such as regulation documents and job logs by utilizing a natural language processing technology based on deep learning, extracting behavior characteristics from video streams by adopting a computer vision algorithm, and constructing an electric power security knowledge graph with dynamic updating capability; designing a multi-modal feature fusion algorithm based on an attention mechanism, and effectively integrating visual features, text features and sensor data; a graph neural network is adopted to train a dynamic risk prediction model to carry out risk prediction, intelligent research and judgment of electric power operation risks are realized, accurate management and control of the risks are realized through a grading early warning mechanism, and closed-loop management from risk perception to early warning treatment is formed.

The invention relates to the technical field of photovoltaic power station operation and maintenance, in particular to a photovoltaic power station intelligent inspection system based on AI vision, which comprises a video acquisition module, a geometric reference construction module, a tremor offset resolving module, a coordinate inverse correction module and a defect fine calibration module, the video acquisition module is used for acquiring a real-time video stream of an unmanned aerial vehicle polling photovoltaic array, and performing time-space synchronization calibration on the video stream to generate an original image sequence. According to the invention, the computer vision technology is utilized to calculate a current frame blanking point in a video picture as a tiny offset of a reference object relative to a reference position in real time, the displacement is deducted from a GPS coordinate, and a coordinate inverse correction module is utilized to inversely calculate a visual axis offset into a ground projection error. The shake amount and the shake direction of the camera at each moment can be accurately calculated, and then the GPS coordinates are corrected in turn, so that the geographic accuracy of defect positioning is improved, and the operation and maintenance personnel can accurately find a fault component.

The invention relates to the technical field of intelligent monitoring, in particular to a multi-thread low-power-consumption intelligent monitoring system based on an AI processor. The method has the advantages that aiming at the problems of unbalanced computing power and power consumption, high multi-task processing delay and strong hardware dependence in the prior art, the NPU module of the RK3588 processor is combined with the INT8 quantitative model, so that the power consumption is lower than 10W under the 6TOPS computing power; a dynamic multi-thread scheduling mechanism is designed, parallel processing of more than eight paths of video streams is supported through binding of a priority queue and an NPU core, and end-to-end delay is compressed to be within 200 ms; a zero-copy video stream architecture is constructed, data transfer is eliminated through memory mapping, and preprocessing time consumption is reduced by 90%; an energy efficiency control module is integrated, the NPU voltage frequency is dynamically adjusted according to the load, and the energy efficiency ratio reaches 0.83 TOPS / W; space-time alignment of multi-model reasoning results is realized by adopting a frame ID synchronization technology, and the mismatching rate is lower than 0.1%.

The invention relates to the technical field of biomedical engineering and computer vision, in particular to a non-contact physiological signal extraction method and system based on frequency self-adaption and illumination noise perception.The method comprises the following steps of multi-mode video stream collection and spatio-temporal data preprocessing, illumination-noise perception mask generation and feature filtering, multi-mode video stream collection and spatio-temporal data preprocessing, illumination-noise perception mask generation and feature filtering, and non-contact physiological signal extraction. Frequency adaptive gating and frequency domain feature enhancement, depth time attention feature re-calibration, physiological signal regression and closed loop optimization; the method has the beneficial effects that a lightweight end-to-end deep learning network architecture is constructed by systematically fusing three core modules of illumination-noise perception mask, frequency adaptive gating and depth time attention, and the defects that a traditional physical model depends on artificial prior and is poor in anti-interference performance and high in reliability are overcome. And the one-sidedness caused by high calculation complexity and difficulty in distinguishing the signal and noise of the existing deep learning model is avoided, and the weak physiological signal can be recovered from the face video more accurately and robustly.

The invention belongs to the technical field of computers, particularly relates to a digital twin power plant infrastructure multi-source heterogeneous data real-time fusion method, and aims to solve the problems of high data fusion delay, semantic segmentation and poor system adaptability in the prior art. The method comprises the following steps: constructing a unified space-time reference frame to realize nanosecond-level time synchronization and space coordinate normalization; the method comprises the following steps: accessing and preprocessing multi-source data such as a building information model, an Internet of Things sensor, a construction log and a video stream, and generating a standardization unit with space-time metadata; performing semantic analysis and cross-modal feature alignment based on the power plant infrastructure ontology knowledge base; and millisecond-level dynamic fusion is realized by adopting an event-triggered streaming engine. According to the scheme, real-time fusion within 100 milliseconds is realized, the semantic alignment precision is 98% or above, the state confidence is 90% or above, the system throughput is improved by three times by relying on a cloud edge collaborative architecture, and precise twin mapping and intelligent decision making of the whole process of power plant infrastructure construction are comprehensively supported.

The invention provides a driving behavior intelligent identification method and system based on video analysis, and the method comprises the steps: obtaining a driver face video stream and a road environment video stream collected by a vehicle-mounted camera, and reading the driving information recorded by a whole vehicle communication network; recognizing an eyelid closing state, a sight line direction and a head posture in the driver face video stream based on a posture recognition model, and performing fatigue distraction analysis to obtain driver state information; performing motion trail analysis on the road environment video stream and the driving information, and performing driving risk assessment in combination with the driver state information to obtain driving assessment information; and performing early warning construction according to the driving evaluation information, generating early warning prompt information, and synchronously writing the early warning prompt information, the driving evaluation information and the driver state information into a safety data protection memory. The fatigue and distraction states of the driver can be recognized more accurately, and the accuracy of state judgment is improved.

The invention relates to a method and system for identifying a highway event by using a video large model, and the method comprises the steps: employing a three-stage processing architecture, firstly carrying out the real-time target detection and preliminary event judgment of a highway monitoring video stream through employing a YOLO algorithm, and generating an event candidate set; inputting the candidate events and the video clips thereof into a specially trained visual large model for deep semantic analysis and secondary reasoning; and finally, a reasoning result is rechecked through a rule engine, and false alarms are filtered by applying illusion suppression and a space-time association rule. According to the method, the real-time performance of traditional target detection and the deep reasoning capability of a visual large model are fused, so that the problems of high false alarm rate and high missing report rate of a traditional method are effectively solved, the accuracy and reliability of event identification in a complex traffic scene are remarkably improved, and meanwhile, the real-time processing capability of a system on multiple paths of high-definition video streams is ensured.

The invention relates to a multi-mode privacy area shielding switching household camera monitoring method and system, and the method comprises the steps: obtaining a real-time monitoring video stream collected by a household camera, and carrying out the recognition processing of a scene, and obtaining scene type information and user behavior feature information; determining a privacy protection demand level corresponding to the current monitoring environment according to the information; selecting a matched privacy shielding mode from a preset multi-mode privacy shielding strategy library based on the level; dynamically adjusting the privacy shielding range and shielding strength of the selected mode according to the user behavior feature information; performing privacy shielding processing on a specific area in the real-time monitoring video stream; and continuously monitoring the change of the user behavior feature information, and if the change is significant, re-evaluating and switching the privacy shielding mode. According to the scheme, the privacy shielding strategy can be dynamically adjusted according to different scenes and user behaviors, and the privacy protection effect is improved.

The invention relates to the technical field of computer vision, in particular to an intelligent conference video frame dynamic coding method based on multi-modal semantic understanding, which comprises the following steps: acquiring a video stream sequence and a synchronous audio stream in a conference scene in real time; performing semantic analysis and decoupling on the video stream sequence, and extracting key frames and subsequent frames; extracting a sparse motion field from a subsequent frame, and segmenting a video frame into candidate visual areas including a face, a mouth shape and a background; extracting audio semantic features, executing cross-modal semantic correlation analysis, calculating semantic correlation between the sparse motion field distribution features and the audio semantic features, and positioning a pronunciation area highly related to the voice content; and calculating a quantization offset value of each candidate visual area according to the semantic relevancy, applying the quantization offset values in different areas, and packaging the quantization offset values into a variable-code-rate video code stream. According to the invention, the multi-mode semantic understanding model is constructed to carry out deep semantic analysis on the video frame content so as to realize the dynamic coding of the conference video frame.

The invention relates to the technical field of multi-modal large language model training, in particular to a prompt construction method and system for a multi-modal large language model, computer equipment and a medium. The method comprises the following steps: extracting a key frame set from an input video stream; executing a motion reconstruction process on the video stream to generate motion track information; and performing visualization processing on the motion track information to generate a track visualization graph. Performing space-time correlation coding on the key frame set and the motion track information to generate an enhanced key frame; a multi-modal prompt is constructed in a mode of integrating visual input and text input, and the multi-modal prompt is input into a preset multi-modal large language model for spatial reasoning. Through the mode, the technical problem that an existing prompting method is difficult to give consideration to the spatial reasoning precision and the calculation efficiency is solved, efficient and accurate spatial reasoning of the multi-modal large language model is achieved, and the calculation efficiency, the reasoning precision and the environmental adaptability of the model are improved.

The invention discloses a mutual cooperative door lock complementary acquisition method and system, and the method comprises the steps: extracting the structural feature data of a target person according to an event that a self door lock recognizes a non-white list person, generating an assistance request containing a target ID and the feature data, and transmitting the assistance request to an adjacent door lock; according to the assistance request received by the adjacent door lock, target feature matching is carried out, and after matching succeeds, a target video stream is collected and selectively coded; according to the time points when the target appears and disappears, generating an association starting pointer and an association ending pointer, and synchronously storing the pointers between the door locks; and analyzing the association pointer according to the playback request, and automatically starting stream playing of the association video streams of the plurality of door locks to realize multi-view synchronous display. By utilizing the embodiment of the invention, continuous automatic tracking and multi-view synchronous playback of cross-equipment target activities can be realized through active cooperation and data association between door locks, and the continuity and efficiency of security monitoring are improved.

The invention relates to a six-eye camera panoramic video real-time splicing method and system and computer equipment, and the method comprises the steps: synchronously carrying out the image collection through a six-eye camera array, and adding a timestamp to each video frame image; performing real-time feature extraction on the video frame image, and establishing a feature matching relationship in a view field overlapping region of adjacent cameras; based on the feature matching relationship, mapping each video frame image to a unified splicing coordinate system through projection transformation; performing splicing processing on the video frame images after projection transformation through multi-band fusion, and eliminating splicing traces at the boundary of the field of view; and reconstructing the panoramic frame images obtained after splicing processing according to a time sequence, and outputting a continuous panoramic video stream so as to realize real-time and seamless splicing and correction of six paths of high-definition video streams and finally output a global panoramic video stream with no visual fracture and consistent time and space.

The invention relates to the technical field of industrial internet of things operation and maintenance, and particularly provides a remote maintenance auxiliary method integrating video monitoring and three-dimensional modeling. The method comprises the following steps: acquiring engineering graphic data and point cloud scanning data of maintenance equipment, and collecting video stream data of a maintenance equipment site; the video stream data is used for describing the operation state of maintenance equipment; the video stream data comprises a plurality of video frames; matching the point cloud scanning data with the engineering graphic data, and constructing a watertight three-dimensional grid model according to a matching result; mapping texture features of the maintenance equipment in a target video frame to the surface of the watertight three-dimensional grid model to obtain a target three-dimensional model; and receiving a first maintenance instruction marked in the target three-dimensional model by a remote expert, and sending the first maintenance instruction to a video picture of a client of an on-site maintainer. According to the technical scheme provided by the invention, the time consumption for positioning the overhaul part can be reduced.

The invention discloses a distribution-micro collaborative operation simulation optimization method, which belongs to the technical field of simulation optimization, and comprises the steps of preprocessing grid-connected point voltage data, tie line power data and communication time delay data, constructing a power distribution network power flow physical network following a Kirchhoff's law, outputting a source load power prediction curve by using a long short-term memory network algorithm, and calculating the distribution-micro collaborative operation according to the source load power prediction curve. And a distribution-micro collaborative simulation optimization model is obtained based on residual error rolling correction tie line impedance parameters, a delay penalty term is set in a target function in combination with the preprocessed communication delay data, a power regulation instruction is obtained by using a particle swarm optimization algorithm, and a dynamic simulation video stream is generated. According to the invention, through rolling correction of the tie line impedance parameters and setting of the delay penalty term positively correlated with the time delay, the problem of control failure caused by physical deviation caused by model parameter solidification and communication time delay accumulation is solved, and the defects of voltage deviation calculation distortion and inaccurate network loss evaluation are eliminated. And the simulation precision and the operation stability of distribution-micro cooperation are improved.

The invention relates to the technical field of industrial vision and intelligent monitoring, and discloses a loom control system operation state monitoring and fault diagnosis method based on machine vision, which comprises the following steps: acquiring a video stream in a loom shed area and constructing a two-dimensional space-time slice tensor; performing global motion compensation processing on the space-time slice tensor by using a homography transformation matrix, mapping a compensated dynamic texture feature sequence to a three-dimensional phase space by using a time delay embedding algorithm, and reconstructing a closed phase space trajectory representing periodic operation logic of the loom; the discrete Frechet distance between the phase space trajectory of the current operation cycle and the preset reference trajectory is calculated, and a control instruction is generated. The health degree of the sequential logic of the system is directly quantified on the premise that specific components are not recognized by using the invariant characteristic of the phase space manifold topology; the technical problems that small phase lag is difficult to perceive and nonlinear faults cannot be early warned in a strong noise environment are solved.

The invention discloses a self-adaptive multi-level digital watermarking method based on coding process optimization. The method comprises the following steps: firstly, performing deep preprocessing on an input video stream through a pre-trained multi-modal deep learning model, extracting space complexity, time dynamics and content significance features, and generating a uniform feature vector; and constructing a dynamic decision model based on the feature vectors, adaptively determining the watermark intensity, the embedding position and the type, and realizing accurate matching of watermark parameters and video content characteristics. A multi-level watermark hierarchical embedding mechanism is adopted, a robust invisible watermark, a fragile invisible watermark and a dynamic visible watermark are embedded into a video, and the security is enhanced by combining chaotic encryption or Hash chain encryption. During copyright verification, watermark information of each layer is recovered through an adaptive extraction algorithm, and data verification and infringement traceability are completed in combination with a block chain evidence storage system. According to the method, a full-process copyright protection system of embedding, coding, extraction and verification is constructed.

The invention relates to the technical field of image recognition, in particular to a fire behavior recognition method and system based on video monitoring. The technical problem that the early warning reliability of a video monitoring system on fire smoke is insufficient is solved. The method comprises the following steps: acquiring a monitoring video stream of a monitoring area and time sequence data acquired by a sensor in the monitoring area; determining an evaluation value of the to-be-detected area based on the image feature and the morphological change of the to-be-detected area; determining a confidence factor based on the form change trend of the to-be-detected region and the change trend of the time series data; determining a smoke authenticity value based on the evaluation value and the confidence factor; and adjusting an alarm threshold value of the smoke alarm based on the smoke authenticity value. The method is used for fire smoke early warning and monitoring scenes.

The invention provides a video monitoring abnormal behavior identification and tracking linkage method based on artificial intelligence, which relates to the technical field of artificial intelligence, and comprises the following steps: carrying out space-time registration on a multi-camera video stream, and establishing a unified coordinate system; in the system, static objects are detected, and suspected remnants are marked; constructing a time sequence backtracking window to determine the association between the article and the person in charge; establishing a topological graph model containing a camera switching probability and a spatial adjacency relation; and predicting a motion track of an uncovered area according to the current motion state, dynamically distributing a tracking weight and generating an alarm. According to the invention, the remnant tracking efficiency and accuracy are improved.

User authentication, spoofing and replay attack prevention, liveness detection, and user-and-document verification using a live video stream with spatial challenges. A camera of an electronic device captures and transmit a live selfie user-facing video, as part of a user registration process. The user is instructed to spatially move his body or face, such that his face would appear within a first particular on-screen shape; and to also, concurrently or simultaneously, spatially hold in his hand or move a particular an identification document such that it would appear within a second on-screen shape. Optionally, the on-screen shape moves on the screen, and the user is required to spatially move the relevant item to keep it within the boundaries of the moving on-screen shape. The system then analyzes the video via computerized vision, to determine whether the user complied with the spatial manipulation challenges.

The invention relates to the technical field of audio and video transmission, in particular to an internal and external network audio and video secure transmission method and system based on a cloud platform. By combining the load balancing technology with the real-time load state and the audio and video stream characteristics of the cloud platform, the encrypted traffic can be dynamically distributed to a plurality of back-end servers, and the calculation overhead in the encryption process can be effectively reduced through selection of the lightweight asymmetric encryption algorithm and the optimized key exchange process; in a handshake stage, by optimizing a key negotiation process and adjusting a key exchange process according to segmentation characteristics, the execution efficiency of a protocol is further improved, and by dynamically adjusting the execution opportunity of encryption operation, it is ensured that the encryption operation is executed at a proper opportunity, and the encryption efficiency is improved. By monitoring and adjusting the load distribution strategy and the encryption parameters in real time, the system operation mode can be dynamically adjusted according to the distortion degree and the transmission state of the audio and video streams, and the distortion degree of the audio and video streams is effectively reduced.

The invention discloses a video stream defogging method and system for 5G remote control, and relates to the technical field of image enhancement. The method comprises the following steps: acquiring a foggy video in real time; inputting the single-frame foggy video into a pre-trained monocular depth estimation neural network, and reasoning to obtain a depth map with the same size as the input single-frame foggy video; performing close-shot and long-shot region division on the scene of the current foggy video frame according to the depth map to obtain a region mask with the same size as the depth map; based on the region mask and the foggy video, calculating to obtain an atmospheric light parameter; calculating the transmissivity of each pixel according to the distance estimation result of each pixel in the depth map, and obtaining a transmissivity map with the same size as the depth map; based on the depth map, the atmospheric light parameters and the transmissivity map, adaptive calculation is performed on the foggy video to realize defogging, and a defogged clear video frame is obtained; the method can adapt to different depth-of-field fog effects, realizes different depth-of-field defogging, and outputs clear video frames.

The invention belongs to the technical field of building construction safety control, and particularly discloses a building construction site dangerous behavior recognition method and system based on machine learning, and the method comprises the steps: obtaining a single-view continuous video stream of a construction site, segmenting the single-view continuous video stream into a time sequence video frame sequence, and extracting an initial spatial feature sequence through a spatial feature encoder; a pre-trained virtual visual angle projection and feature compensation module encodes the visual angle implicit vector and maps the visual angle implicit vector to a visual angle invariant feature space, and a sequential context is combined to compensate occlusion missing features to generate an enhanced feature sequence; a time sequence memory alignment module captures long-term and short-term time sequence dependence and outputs time sequence consistency characteristics; and finally, outputting a current dangerous behavior classification result through the classification prediction head, and outputting a future dangerous intention probability curve through the time sequence prediction head. The method does not need an additional camera, can accurately cope with a shielding scene, predicts the danger in advance, and improves the construction safety monitoring efficiency and reliability.

In an example, a device may include logic to receive a plurality of video frames from a video source, logic to process the plurality of video frames, logic to provide at least some of the plurality of video frames to a video sink, and logic to reduce a display latency of each of the video frames provided to the video sink by changing the frame rate of some or all of the frames.

The invention discloses a video compliance monitoring system based on multistage event aggregation, which adopts a distributed two-stage architecture of an edge end and a center end, the edge end realizes video stream acquisition, dynamic sampling, multi-modal AI structured analysis and video segmentation buffer storage, and pushes an analysis result to the center end in real time, and the center end receives data and then sends the data to the edge end. Through a three-stage analysis system including atomic event extraction, session scene recognition and cross-time-period violation detection, violation detection is completed, a backtracking video with a superimposed mark can be generated, and a violation record and a video URL are pushed to a service system, so that efficient analysis and accurate backtracking of video compliance monitoring are realized; the invention further relates to a video compliance monitoring method based on multi-level event aggregation.

The invention relates to the technical field of fire early warning, in particular to an early fire early warning system based on AI image recognition, and the system comprises an image collection module which is used for obtaining the video stream data of a monitoring area in real time; an AI image analysis module which is in communication connection with the image acquisition module and is used for receiving the video stream data and carrying out real-time analysis on video frames based on a pre-trained fire identification model so as to extract visual features related to the fire; and the early warning judgment module is in communication connection with the AI image analysis module and is used for receiving an analysis result of the visual features. According to the early fire early warning system and method based on AI image recognition, through video image analysis, the system can recognize weak flame or smoke characteristics at the initial stage of a fire and when naked eyes do not obviously see the characteristics, and the delay problem that a traditional smoke-sensing and temperature-sensing detector needs to wait for physical parameters to be diffused to the detector to be triggered is solved.