1170 results about "Channel data" patented technology

The invention discloses a resource allocation method and system for 5G communication in an underground coal mine, and relates to the technical field of communication information processing, and the method comprises the steps: obtaining data information of an underground coal mine roadway, building a digital twin communication model through combining a synchronous positioning and map building algorithm and a wireless fingerprint matching algorithm, generating a channel profile map according to real-time data, and transmitting the channel profile map to a server; extracting channel data of each space node; constructing a computing resource pool for the 5G base station and the edge computing node, and packaging different service types into a plurality of network slice containers; according to the channel data, carrying out initial allocation on each container through a resource composer, and carrying out dynamic task migration among different MEC nodes to obtain a resource allocation state; dividing scheduling time according to a channel profile and a resource allocation state, and dynamically adjusting a duration proportion and bandwidth occupation of each window according to a task migration state; according to the invention, through digital twinning and dynamic scheduling, the problems of 5G communication coverage and interference in the underground coal mine are solved.

The invention relates to the technical field of Bluetooth communication, and discloses an optimization method for secure communication of Bluetooth chip equipment. The method comprises the following steps: identifying a device type identifier and a protocol version feature, extracting a hardware capability parameter and generating an initial security policy set; collecting channel data, and calculating a stability index and a dynamic noise baseline; then, screening and sorting security policies to generate an execution queue, adjusting an encryption strength threshold according to a noise baseline to perform iterative optimization, and outputting final policy configuration; and when the target Bluetooth chip establishes the communication connection, loading the final security policy configuration and monitoring the channel state in real time, and triggering a policy dynamic switching mechanism when the channel state change exceeds the preset tolerance so as to optimize the security communication effect of the Bluetooth chip equipment. According to the method, the security policy can be accurately matched with the device and the channel state, the policy is dynamically adjusted, and the security and the stability of communication of the Bluetooth chip device are effectively improved.

The invention discloses a 16-channel high-precision synchronous data acquisition system based on a multi-parameter model, which belongs to the technical field of data acquisition and comprises a multi-parameter model driving module, a self-adaptive channel scheduling module, a closed-loop calibration compensation module and a collaborative synchronous optimization module. The multi-parameter model driving module establishes a multi-dimensional optimization model fusing the sampling rate, the channel load and the error trend to output optimization parameters; the adaptive channel scheduling module performs real-time quality evaluation and dynamic resource allocation according to the optimization parameters to generate a scheduling instruction; the closed-loop calibration compensation module monitors a synchronization error in real time and calculates a compensation parameter; the collaborative synchronous optimization module realizes inter-channel parameter sharing and collaborative control to generate a synchronous control signal, and feeds back an optimization result to the multi-parameter model driving module to form a closed loop, and through deep coupling and closed loop feedback of the four modules, high-precision synchronization of multi-channel data acquisition is realized, and the synchronization precision reaches + / -35ns.

The invention relates to the field of financial science and technology, and particularly discloses a cross-platform financial data integration and real-time analysis system and method. The system is characterized in that a distributed heterogeneous data acquisition module accesses multi-source data in parallel through a pluggable protocol adapter; the streaming data cleaning engine realizes field-level normalization based on a dynamic template and semantic mapping; the hybrid storage architecture forms a three-level system by a memory database, a time sequence database and distributed file storage; the stream processing core engine adopts an elastic sliding window to execute CEP complex event processing; and the dynamic analysis decision module integrates online machine learning to generate a real-time transaction signal. The method comprises the steps of multi-source concurrent acquisition and microsecond-level timestamp injection, context sensing data standardization, dual-channel data storage, event time window aggregation, dynamic risk calculation and block chain audit evidence storage. The defects of low data fusion efficiency, high real-time analysis delay and the like in the prior art are overcome, and cross-platform data processing timeliness and decision accuracy are remarkably improved.

The invention discloses an airborne laser radar sounding data processing method and system, and relates to the technical field of marine surveying and mapping and earth observation, and the method comprises the following steps: collecting original full-waveform data of an airborne laser radar sounding system, the original full-waveform data comprising infrared laser channel data and blue-green laser multichannel data; joint denoising processing is carried out on the original full-waveform data to obtain denoised waveform data, and the joint denoising processing comprises background noise modeling and removing based on machine learning and random noise filtering based on frequency domain low-pass filtering; a collaborative joint denoising strategy is formed by fusing background noise modeling based on machine learning and frequency domain low-pass filtering based on signal-to-noise ratio optimization. According to the method, complex background noise can be accurately estimated and removed, random noise can be adaptively filtered out, weak underwater echo signals are reserved to the maximum extent, and the problem of water depth information loss caused by excessive smoothness is effectively avoided.

The invention discloses a rotor acoustic anomaly detection method, system and device based on an auto-encoder and wavelet packet energy entropy and a medium, and belongs to the technical field of hydroelectric generating sets, and the method comprises the steps: collecting an original acoustic signal, carrying out the noise reduction of the original acoustic signal, and carrying out the multi-channel data fusion to obtain an integrated acoustic signal; inputting the integrated acoustic signal into a wavelet packet for transformation processing to obtain wavelet packet coefficients under different scales; calculating energy values of different frequency band signals based on wavelet packet coefficients to form an energy entropy feature vector; constructing an implicit feature model of the hydroelectric generating set rotor in a normal state according to the energy entropy feature vector; and carrying out feature reconstruction on the energy entropy feature vector, calculating a reconstruction error, carrying out dynamic statistical analysis, and carrying out real-time monitoring and abnormity judgment on the operation state of the hydroelectric generating set rotor. According to the method, the detection precision and the response speed are improved in actual hydroelectric generating set rotor acoustic anomaly detection, and automatic identification and dynamic threshold adaptive adjustment of irregular perturbation are realized.

The invention relates to the technical field of radar reflectivity factor calibration, and discloses a terahertz cloud measurement radar calibration method based on channel perception, which comprises the following steps: S1, acquiring two types of input data; s2, constructing a multi-scale channel model; s3, constructing a distance error estimation model; s4, constructing a dynamic time correction model; and S5, constructing a calibration module. According to the method, a high-precision physical channel model is constructed to independently predict path attenuation of signals, the concepts of gain deviation and path attenuation factors are put forward, the influence of thermal noise attenuated by radar on radar waves in the atmospheric environment is reflected respectively, and a Kalman filtering algorithm is used to estimate the path attenuation of the signals. And filtering errors generated in the time direction, and finally, calibrating the radar reflectivity factor by using the channel data by means of a calibration formula.

The invention discloses a kilometer deep well roadway toughness regulation and control supporting system and a dynamic design method, and belongs to the field of deep well roadway supporting. The supporting system comprises a multi-layer collaborative structure and an intelligent regulation and control module, the multi-layer collaborative structure sequentially comprises a shape memory alloy energy absorption layer used for resisting impact loads and a gradient pore foam metal buffer layer used for energy grading dissipation from inside to outside, and the multi-layer collaborative structure is connected with roadway surrounding rock in an anchoring mode through an anchor rod anchoring structure; the intelligent regulation and control module comprises a distributed optical fiber sensor array used for monitoring surrounding environment signals of surrounding rock in real time, an electrostriction driver used for regulating and controlling the rigidity of the multi-layer cooperative structure and a multi-channel data fusion terminal used for achieving underground and surface data transmission synchronization. By the adoption of the kilometer deep well roadway toughness regulation and control supporting system and the dynamic design method, the problems that a traditional supporting mode is too high in material rigidity, lacks dynamic regulation and control capacity and is insufficient in impact toughness can be solved.

The invention discloses a visible light hyperspectral measurement method based on multispectral sensing, and belongs to the technical field of spectrum detection. According to the method, visible light eight-channel data is collected through a multispectral sensor, after optical crosstalk correction and filtering preprocessing, key principal components are reserved through PCA dimension reduction, a multispectral and hyperspectral mapping model is constructed through an SVR algorithm, and finally the hyperspectral reflectivity is reconstructed through PCA inverse transformation. The problems that traditional hyperspectral equipment is high in cost and poor in portability are solved, low-cost and high-precision visible light hyperspectral measurement is achieved, and the device and the method can be widely applied to food freshness detection, agricultural product quality evaluation and other scenes.

The invention relates to the field of real-time translation, and provides a real-time duplex translation method based on multichannel parallel processing and a corresponding product, and the method comprises the steps: collecting multipath voice signals of at least two user groups in real time through a group of audio collection modules; dynamically adjusting beam forming parameters of each audio acquisition module in one group of audio acquisition modules based on a sound source positioning result, and feeding back the beam forming parameters to the corresponding audio acquisition modules; monitoring the voice activity of each audio acquisition module corresponding to each audio channel; when it is monitored that the voice activity of any audio channel reaches a preset condition, automatically activating the translation processing flow of the audio channel and keeping the monitoring state of the other audio channels; a parallel processing mechanism is adopted for voice signals of the activated audio channels, and meanwhile real-time translation of the currently activated audio channels and voice activity monitoring of the other audio channels are executed; and transmitting a translation result of the current speaking user to other users participating in dialogue in the user group to realize synchronous coordination of multichannel data.

The invention relates to the technical field of signal processing, and discloses a multichannel data acquisition system and method based on an FPGA (Field Programmable Gate Array), an acquisition module drives a global counter by using a global synchronous clock, writes acquired data into an annular buffer area with a physical address and a counter value in a linear mapping relationship, and establishes implicit time index storage. When a trigger event occurs, the system broadcasts the locked global trigger timestamp, and the acquisition module backtracks and reads historical data according to the global trigger timestamp, and packages the historical data into a sparse matrix type data packet in combination with the channel validity mask. And after receiving the data packet, the data processing module directly calculates a memory mapping address by using global time information in the data packet, and writes a data load into a corresponding position of the waveform reconstruction buffer area. According to the method, through strict binding of the physical address and the absolute time, independent time label redundancy is eliminated, high-precision synchronization of distributed multiple channels is ensured, and out-of-order automatic in-situ recombination and efficient waveform reconstruction of data of a receiving end are achieved.

The invention discloses a test verification platform and a test verification method for a motion compensation gangway ladder of an offshore wind power operation and maintenance ship. The platform comprises a system calculation and control center, an environment simulation vibration platform, an active motion compensation execution mechanism, a high-precision sensing and sensing module, a multi-channel data acquisition and storage system, an offshore wind turbine target platform module and an optical motion capture module. The platform reproduces six-degree-of-freedom motion of the operation and maintenance ship through integrated cooperation and high fidelity, simulates active compensation of a gangway ladder and dynamic characteristics of a hydraulic system, and accurately measures key parameters such as poses, forces / torques and the like by using multiple sensors and an optical capture technology. The method comprises the steps of motion instruction synthesis, platform motion reproduction, compensation action simulation, data synchronous acquisition and performance evaluation. According to the invention, high-precision land simulation verification of ship-gangway ladder-fan dynamic coupling is realized, the problems of high cost, high risk and poor repeatability of offshore actual measurement are effectively solved, and a reliable and efficient solution is provided for performance verification of a motion compensation system.

The invention discloses a cloud edge cooperative multi-channel converged communication control method and system, belongs to the technical field of wireless communication networks, and aims to improve the high efficiency, reliability and real-time performance of multi-channel converged communication in a complex scene. The method specifically comprises the following steps: constructing a hierarchical collaborative architecture of a cloud end and an edge end; the edge end collects network state parameters of various communication channels, outputs a channel quality evaluation value based on a time sequence prediction model, and performs multi-channel data transmission through an extended communication protocol when triggering channel switching; the edge end executes network parameter configuration, flow priority scheduling and running state data acquisition operation, and the cloud end generates a soft routing optimization instruction based on running state data and issues the soft routing optimization instruction to the edge end to execute and adjust network transmission performance parameters; and monitoring the container operation state and the equipment state of the edge end in real time, when a monitoring result is matched with a preset abnormal rule, generating alarm information by the edge end, returning the alarm information to the cloud end to generate a global security policy, and issuing the global security policy to the edge end to update the preset abnormal rule.

The invention relates to the technical field of cable fault positioning, in particular to a cable fault accurate positioning method based on multi-channel synchronous acquisition, which comprises the following steps: S1, constructing a multi-channel adaptive acquisition system which comprises a main controller, a multi-channel acquisition module, a synchronous control module and a data transmission module; s2, multi-channel data synchronous acquisition is carried out through the synchronous control module, wherein the multi-channel data comprises impedance data, temperature data, partial discharge signals and insulation resistance data; during use, through a multi-channel adaptive acquisition design, monitoring channels can be flexibly selected according to cable materials, operation years and environmental conditions, different aging fault characteristics are comprehensively covered, the fault omission ratio and the false alarm rate are reduced, different types of low-voltage cables are adapted, a synchronous scheme of constant-temperature crystal oscillator and hardware triggering is adopted, and the reliability of the system is improved. The method does not need to depend on a GPS / PTP time service module, is favorable for saving cost input, and eliminates environmental interference and single parameter false triggering.

The invention discloses a charging pile data management method and system. The method comprises the steps of generating a session fingerprint and a sampling jitter plan at a charging pile end, and writing the session fingerprint and the sampling jitter plan into an under-chain hash structure; time alignment is realized through phase correlation and dynamic time warping based on multi-channel records; constructing an energy sequence and calculating an energy reconciliation distance to generate a session consistency label; according to the unified time axis and the rate rule, generating a microaccounting state machine and outputting an accounting deviation; a session consistency score is calculated by fusing multiple indexes, and diagnosis judgment is completed; and generating a compliance abstract, executing signature evidence storage, and providing a session result for the operation and maintenance interface and the settlement interface. Through multi-channel data acquisition and time alignment, energy reconciliation, accounting modeling and anomaly detection in a trusted execution environment, high-precision reconciliation and compliance intelligent evaluation of the charging pile session data are realized.

The invention relates to the technical field of electric digital data processing, in particular to a shock absorber data processing method and system based on digital twinning, and the method comprises the steps: obtaining a multi-channel data signal of a shock absorber, and constructing a panoramic vibration data tensor; extracting a discrete digital sequence on a panoramic vibration data tensor time dimension, executing a variational mode decomposition algorithm, and separating and extracting a transient impact feature vector flow; constructing a double-time-axis sliding mapping window, and mapping the transient impact feature vector flow to a future node of a prediction response time axis based on a front and rear wheel preview mechanism of a vehicle wheelbase; operating the time sequence recurrent neural network model, and predicting and generating a target state data matrix of future nodes; and in the numerical constraint space, a prediction type damping adjustment instruction data packet is solved and generated through a rolling optimization algorithm. Through prospective state prediction and optimization solution, the problem of data timeliness caused by calculation delay in digital twinning is solved, and real-time, predictive and safe adjustment of the shock absorber is realized.

The invention discloses an MES-based aluminum alloy casting production management method and system, and relates to the technical field of production management, and the method comprises the steps: firstly, synchronously collecting time sequence data streams of core process parameters such as pressure, speed, position and the like, and carrying out the multi-channel data alignment; and a deep learning auto-encoder model is introduced to carry out nonlinear dimension reduction and feature extraction on high-dimensional data, and reconstruction errors are utilized to realize real-time evaluation and anomaly detection of the quality of the casting process. And for the abnormal state exceeding the preset threshold value, key influence factors causing quality fluctuation are further traced through an intelligent diagnosis mechanism, autonomous decision is made based on a diagnosis result, an optimal parameter adjustment strategy is formed, and finally a closed-loop control instruction is generated to drive the production equipment to perform adaptive adjustment. In this way, the problem of unstable quality caused by parameter fluctuation in the casting process can be effectively solved, and the stability and the intelligent level of the aluminum alloy casting process are remarkably improved.

The invention discloses infrared night vision AR (Augmented Reality) intelligent glasses based on multispectral fusion, and relates to the technical field of infrared intelligent application. High-precision scene perception in a low-light / light-free environment is realized by fusing dual-channel data of low-light imaging and infrared thermal imaging and combining cross-modal feature matching and dynamic pose compensation; an environment detail structure is restored by a low-light image enhancement technology, thermal feature targets such as organisms are accurately recognized through thermal imaging target segmentation, and a three-dimensional environment model with space geometric information and target attributes is constructed through heterogeneous feature fusion; and dynamically adjusting the display transparency of the key target in the AR view field through a virtual information layer adaptive mixing mechanism driven by a hot target attribute, and finally outputting a virtual-real fused enhanced view, thereby breaking through the illumination limitation of the traditional AR glasses, and improving the visual effect of the AR glasses. And immersive augmented reality experience with panoramic detail perception and hot target accurate positioning is provided for the user in a dark environment.

The invention discloses a method and system for constructing an interferometric phase unwrapping model based on discrete wavelet transform, and relates to the field of image phase unwrapping, and the method comprises the specific steps: constructing an InSAR synthetic data set based on an SRTM DEM, a two-dimensional Gaussian curved surface and an InSAR measurement error source; an interferometric phase unwrapping model based on discrete wavelet transform is constructed, the interferometric phase unwrapping model comprises an encoder network, a bottleneck layer, a decoder network and an output layer which are connected in sequence, and the encoder network comprises a plurality of cross-stacked dual-channel data down-sampling modules and pyramid pooling modules; and training the interferometric phase unwrapping model based on the InSAR synthetic data set, and optimizing model parameters by minimizing a loss function to obtain a final interferometric phase unwrapping model. According to the method, the interferometric phase unwrapping of the low-coherence region is realized under the condition of lightweight parameters, the phase continuity hypothesis is not depended on, and the interferometric measurement precision of the synthetic aperture radar is improved.

The embodiment of the invention provides a real-time collection method and system for welding heat input, and the method comprises the steps: sensing the displacement movement of a welding gun or a welding wire through a diffuse reflection photoelectric switch array installed on the surface of a pipeline, and generating a corresponding pulse trigger signal; current data and voltage data of a welding loop are sensed through a Hall sensing system, and corresponding analog electric signals are generated; synchronously acquiring the pulse trigger signal and the analog electric signal through a multi-channel data acquisition device, and converting the analog electric signal into a digital signal with a timestamp; real-time welding speed data are calculated according to the time sequence of the pulse trigger signals and the preset space geometrical relation of the diffuse reflection optoelectronic switches; performing multi-stage filtering processing on the current data and the voltage data; a unified timestamp is determined, and time sequence matching is conducted on the filtered current data, the filtered voltage data and the filtered welding speed data based on the unified timestamp; and calculating to obtain a welding heat input value.

The invention provides an explosion-proof electric appliance overcurrent and temperature rise wireless test system and method based on ultrasonic communication, and relates to the technical field of explosion-proof electric appliance temperature rise tests. The system comprises a temperature acquisition device, an ultrasonic through-wall communication device, an overcurrent protection time acquisition device, a data receiving and storing device and a digital display operation table, wherein the output end of the temperature acquisition device is connected with the input end of the ultrasonic through-wall communication device, the output end of the ultrasonic through-wall communication device is connected with the input end of the data receiving and storing device through wireless communication, and the output end of the overcurrent protection time acquisition device is connected with the input end of the data receiving and storing device; and the data receiving memory is connected with the digital display console through a bidirectional data channel. According to the invention, the anti-interference performance and the test precision of temperature measurement are enhanced, and the multi-node distributed monitoring capability is provided through multi-channel data transmission, so that the overall efficiency of the internal temperature and overcurrent protection monitoring of the explosion-proof electric appliance is greatly improved.

The invention relates to the technical field of geological engineering, in particular to a dam anti-seepage effect detection method and system. The method comprises the following steps: obtaining dam structure data; performing anti-seepage structure performance evaluation according to the dam structure data to determine anti-seepage weak points; performing seepage anomaly analysis based on the anti-seepage weak points to obtain seepage anomaly data; performing free surface offset identification according to the seepage abnormal data to obtain free surface offset data; performing seepage channel positioning based on the free surface offset data to obtain seepage channel data; performing seepage field numerical simulation according to the seepage channel data to obtain a seepage field simulation numerical value; dam foundation stability detection is conducted according to the seepage field simulation numerical value, and dam foundation stability data are obtained; and on the basis of the dam foundation stability data, instability critical state prediction is conducted, and instability critical state data are obtained. According to the invention, based on the geological engineering technology, the accuracy of dam anti-seepage abnormity identification and the timeliness rate of safety early warning are improved.

The invention belongs to the technical field of high-power microwave effect tests in laboratories, and relates to a lossless reinjection data synchronization method and device for a multi-mode test body, and the method comprises the steps: S1, data collection and transmission: collecting electric field data in the multi-mode test body, and transmitting the electric field data out of a shielding cabin of the multi-mode test body; s2, an edge processing step: sending the electric field data transmitted to the outside of the shielding cabin into an edge GPU node, and carrying out voxelization processing and WebGL body rendering; s3, a data synchronization step: synchronizing the three-dimensional rendering data stream to a plurality of user side browsers in real time through WebRTC and HTTPS (Hyper Text Transfer Protocol over Secure Socket) protocols; s4, time synchronization is carried out, a unified high-precision time reference is provided through a Whit-Rabbit clock, and sub-nanosecond alignment of multi-channel data is achieved; and S5, a data storage step: generating a hash abstract for the synchronous data and writing the hash abstract into the block chain. By means of the method, the safety, the real-time performance and the reliability of data transmission can be effectively improved.

The invention discloses a vehicle-mounted CAN bus anti-interference communication system and method based on photocoupling, and the system comprises an anti-interference communication system frame, the anti-interference communication system frame comprises a CAN bus receiving and transmitting module, a photocoupling isolation module, and a signal processing and separating module, the CAN bus receiving and transmitting module is used for receiving and transmitting CAN bus signals, and the photocoupling isolation module is used for receiving and transmitting the CAN bus signals. The photoelectric coupling isolation module is used for realizing transparent conversion of data between an optical signal and an electric signal and realizing electrical isolation of a physical layer while maintaining the integrity of a CAN protocol, and the signal processing and separation module is used for constructing a cooperative processing mechanism of time sequence calibration, differential judgment, logic reconstruction and signal synthesis. According to the invention, based on a physical layer signal isolation mechanism of optoelectronic isolation, a bidirectional independent transmission path of a CAN bus TXD / RXD channel data stream is established, and TXD mapping data and CAN bus effective data in an RXD channel are accurately separated, so that the system reliability in an industrial-grade communication scene can be obviously improved.

The invention relates to an unmanned ship dynamic channel modeling method in an offshore water surface environment, and the method comprises the steps: constructing a three-dimensional offshore water surface simulation scene which comprises a sea surface, a wharf facility, an unmanned ship cluster and a communication node; establishing a wireless channel propagation model based on a ray tracing method, and simulating a propagation mechanism of direct, reflection and diffraction paths; establishing a motion characteristic model of the unmanned ship under the action of waves, and obtaining six-degree-of-freedom pose data of the unmanned ship; generating multiple frames of scene copies in batches by adopting an intelligent batch processing dynamic reconstruction method; dynamically correcting an antenna pattern in each frame of scene copy according to the pose data so as to couple the influence of hull motion on antenna radiation characteristics; importing the corrected antenna pattern into the wireless channel propagation model to generate a dynamic channel response; simulation is executed, and channel data obtained through simulation are collected and analyzed. According to the method, the motion characteristics of the unmanned ship in waves and the wireless channel propagation mechanism can be accurately coupled, and the accuracy of dynamic channel prediction is improved.

The invention discloses a data consistency verification method for dual-network communication of a digital substation, and relates to the technical field of operation data monitoring. The digital network A and the digital network B are arranged, time correction pulses are sent to the digital network A and the digital network B, then time synchronization of the digital double networks is completed, a plurality of collection cores are arranged in the digital network A and the digital network B, multiple fault-tolerant mechanisms are set for the collection cores, and the time synchronization of the digital double networks is completed. The message data in the operation process of the transformer substation is acquired through the acquisition core, the acquisition process is monitored through a multiple fault-tolerant mechanism, structured analysis is performed on the message data, and then a corresponding SV message detection data table and a GOOSE message detection data table are generated; and sequentially comparing the SV message detection data tables of the digital network A and the digital network B with each channel data in the GOOSE message detection data, and judging an abnormal digital network according to a comparison result.

The invention discloses a method for improving data transmission bandwidth through an FPGA. The method comprises the steps that a data packaging module of a detector main control board sends data to be transmitted to a multi-channel GTP sending end in parallel; the data acquisition card receives the data of the corresponding channels through the multi-channel GTP receiving end and recovers Data and KCOD of each channel; an FIFO read-write control module of the data acquisition card writes data recovered by multiple channels into asynchronous FIFOs corresponding to the channels through write logic, and data buffering is carried out; the FIFO read-write control module simultaneously reads cache data of a plurality of asynchronous FIFOs through read control logic to realize cross-clock domain synchronization of multi-channel data; performing CRC (cyclic redundancy check) and processing on the read multi-channel data to obtain synchronous effective transmission data; and a single-channel GTP is upgraded to a dual-channel GTP or a multi-channel GTP, so that the data transmission bandwidth is improved, and normal operation of the FPGA chip is ensured.

The application relates to the technical field of image recognition, in particular to a bearing surface defect automatic detection system based on image recognition, which comprises an image acquisition and preprocessing module, a bearing surface image is acquired, the image is cropped and filtered to remove noise and adjust contrast, and a processed image is obtained; the processed image is subjected to gray scale conversion to generate a preprocessing result. In the application, environmental noise interference is reduced through image cropping and filtering, the contrast is adjusted to optimize image quality, and the distinguishing degree of a defect area and a background is improved. The gray scale conversion standardizes image channel information, keeps the calculation precision of feature extraction consistent, and avoids multi-channel data interference analysis results. An image pyramid method constructs different scale image levels, in the local contrast and texture feature extraction process, the adaptability to fine cracks and large-area peeling defects is improved, and the stable recognition ability to different size defects is enhanced.