376results about How to "Real-time detection" patented technology

The invention relates to the technical field of photovoltaic solar energy, in particular to a photovoltaic solar support structure with a light following function and a light following method thereof.The photovoltaic solar support structure comprises a supporting base, an L-shaped supporting plate and a control system are sequentially and fixedly installed on the side face of the supporting base from top to bottom, and the upper end of the supporting base is in a T shape; a rotating groove is formed in the upper end of the supporting base, and the control system executes the following steps that the control system comprises a light following sensor, a receiving and transmitting module, a motor control module, a potentiometer, an analysis processing module and a comparison processing module. The motor shaft of the stepping motor drives the main gear to rotate, and the light-following rotating seat drives the photovoltaic solar panel and the light-following sensor to rotate through the photovoltaic support, so that the position of the photovoltaic solar panel is adjusted, the strongest light can be vertically irradiated on the photovoltaic solar panel, and the illumination efficiency is improved.

The invention discloses a remote sensing image target recognition method and system based on a visual language model, and relates to the technical field of image recognition. The method comprises the following steps: constructing an initial image detection model based on a real-time open vocabulary detector YOLO-World and a remote sensing visual language basic model RemoteCLIP; adding a double-attention fusion encoder and a multi-level text guide fusion decoder on the initial image detection model to obtain a target image detection model; performing knowledge self-adaptive adjustment and optimization on the basis of the pre-training weight of the target image detection model to obtain a final image target recognition model; and performing target identification on the remote sensing image based on the image target identification model. According to the method, a powerful visual language model is introduced, and an image feature extraction and target classification strategy is optimized, so that the accuracy and efficiency of target recognition in the remote sensing image are improved.

The invention discloses an ultra-large and ultra-long digital oil cylinder based on a novel key component. The ultra-large and ultra-long digital oil cylinder comprises an oil cylinder body structure, an oil cylinder V-group main seal, a knuckle bearing, a piston rod laser cladding coating, an oil cylinder state monitoring and adjusting unit, an oil cylinder bearing monitoring and evaluating unit, a piston rod coating monitoring and evaluating unit, an industrial personal computer, a display screen and an audible and visual alarm. The knuckle bearing is a domestic self-lubricating knuckle bearing. The V-group seal is a domestic seal and comprises a pressing ring, a sealing ring and a supporting ring. The piston rod laser cladding coating is made of domestic self-developed powder, and an ultra-high-speed laser cladding construction technology is adopted. Fine control over the pile driving barge pile frame main oil cylinder under different working conditions is achieved, and the use safety of the pile driving barge pile frame main oil cylinder is remarkably improved; online performance evaluation and state real-time detection of the knuckle bearing of the oil cylinder are achieved, and long-time safe use of the oil cylinder is guaranteed; the residual life of the knuckle bearing in the service process is predicted, the knuckle bearing is planned and replaced in advance, and the construction period is prevented from being delayed.

The invention discloses an infrared expressway foreign matter detection method based on a Mama framework, and the method comprises the steps: continuously carrying out the data collection of expressway scenes under different illumination conditions and different weather conditions through an infrared sensor, obtaining infrared image data, carrying out the preprocessing of the infrared image data through data augmentation, generating training data, and carrying out the detection of foreign matter in an expressway. Dividing the training data into a training set, a test set and a verification set in proportion; constructing a detection network based on a Mama framework, wherein the detection network comprises a student network and a teacher network; training the teacher network through back propagation based on the training set; training a student network based on the result of the teacher network and the real data to obtain a target detection network; and inputting a to-be-identified image into the target detection network to obtain a detection result. According to the invention, the method achieves the quick and accurate recognition of potential risks, guarantees the high detection precision through the combination of the infrared imaging technology and the powerful processing capability of a Mama frame, achieves the real-time response to foreign body threats, and remarkably improves the efficiency and safety of expressway monitoring.

The invention provides an ore particle granularity detection device and method based on a flexible vibration disk, and the device comprises the flexible vibration disk which is used for vibrating an ore material at a certain frequency and amplitude, so that the ore material is uniformly distributed and is in a proper posture; the image acquisition device is used for continuously shooting the ore materials on the flexible vibration disc to obtain ore images; the image processing device is used for carrying out preprocessing, segmentation, feature extraction and particle size analysis on the ore image and outputting the ore particle size distribution condition and an accumulative distribution curve; and the result output device is used for displaying the ore image and the granularity analysis result. According to the invention, while material accumulation is prevented, real-time, accurate and efficient detection of attitude distribution of ore materials is realized, and the method has the advantages of low cost of detection equipment, easy maintenance, strong interference resistance, high accuracy and stability of detection results, and the like.

The invention discloses a photo-thermal amplification and real-time PCR (polymerase chain reaction) detection integrated optical fiber system, and relates to the field of pathogen testing, the optical fiber system comprises an optical fiber probe, a photo-thermal photosensitizer and a micro reaction chamber; the micro-reaction chamber is used for extracting and discharging a to-be-detected sample by utilizing air pressure, the optical fiber probe is arranged in the micro-reaction chamber, and the optical fiber probe reacts with the to-be-detected sample in the micro-reaction chamber; the photo-thermal photosensitizer is arranged inside or on the outer surface of the optical fiber probe and is used for receiving heat generated by the heating light to heat the optical fiber probe and controlling the reaction temperature of the optical fiber probe and the to-be-detected sample; the optical fiber probe is also used for receiving the exciting light which excites the to-be-detected sample to generate a fluorescence signal, and the optical fiber probe is used for detecting the fluorescence signal. According to the application, sample collection, photo-thermal cycle amplification and real-time fluorescent quantitative detection can be integrated, and rapid and effective detection of a virus sample is realized.

The invention provides an automatic spraying cooling method for an air cooling island, and solves the problems that an existing air cooling island auxiliary cooling spraying device is complex in structure, and a spraying cooling control and adjustment method is complex, and according to the technical scheme, the cooling method comprises the following steps that an air cooling island automatic spraying cooling system is arranged on the air cooling island; checking equipment and a control system; opening water supply and opening a water inlet valve of a water supply system; adjusting pressure to start spraying; cooling is controlled, and a PLC of the master control cabinet judges according to a preset temperature threshold value; monitoring parameters; and observing the spraying effect. The system can automatically perform spray cooling according to the real-time temperature of the air cooling island, strengthen the cooling effect of the air cooling island, increase the air humidity, prevent corrosion and dust accumulation, cope with the environment temperature change, reduce the operation of cooling auxiliary equipment, relieve the thermal stress of the air cooling island, assist in improving the cooling efficiency of the air cooling island, improve the output of a generator set and save energy.

The invention belongs to the technical field of image processing, and provides a small-target lightweight substation equipment defect detection method and system, and the method comprises the steps: obtaining a substation equipment image; according to the acquired image and a preset small target lightweight defect detection model, completing defect detection of the substation equipment; wherein the preset small target lightweight defect detection model adopts a YOLOv8 target detection network which adopts an Officient Former V2 and a separation convolution attention mechanism as a backbone network, adopts scale time sequence feature fusion as a neck network and adopts a prediction head network of a shared convolution lightweight detection head, and image features are extracted based on the backbone network; obtaining feature maps of different scales; the neck network performs feature fusion on the feature maps with different scales to obtain a multi-scale feature map; and detecting the multi-scale feature map through the shared convolutional lightweight detection head, and performing defect positioning of the equipment image to complete defect detection of the substation equipment.

The invention provides an image sensor for online detection of oil quality, and the image sensor comprises a sensor body which is provided with an oil inlet, an oil outlet and a detection chamber, the detection chamber is arranged between the oil inlet and the oil outlet, and the detection chamber comprises a cavity and a transparent protection plate arranged at the top of the cavity; the magnetic adsorption assembly comprises an annular electromagnet arranged above the transparent protection plate; the image acquisition assembly is arranged on the upper surface of the transparent protection plate and comprises an imaging lens, a CMOS sensor and an annular light source; the processing circuit is used for preprocessing and analyzing the acquired optical image; and the control circuit is used for controlling the coil electrifying preset time so as to adsorb the abrasive particles and control image acquisition. According to the invention, real-time and accurate detection of the abrasive particles in the oil can be realized, and the morphology feature information of the abrasive particles can be obtained.

The invention relates to the technical field of manipulators, in particular to a punching machine three-axis manipulator which comprises a supporting mechanism, a moving mechanism, a clamping mechanism and a limiting mechanism. The supporting mechanism is used for mounting and fixing the moving mechanism and the limiting mechanism; the moving mechanism comprises an X-axis moving mechanism, a Y-axis moving mechanism and a Z-axis moving mechanism; the clamping mechanism comprises a plurality of clamping jaw assemblies and corresponding arm rods, the clamping jaw assemblies comprise the first clamping jaw assembly and the second clamping jaw assembly, the first clamping jaw assembly is used for clamping materials through vacuum adsorption, and the second clamping jaw assembly is used for clamping the materials through electromagnetic adsorption; the arm rod is used for fixing each clamping jaw assembly and adjusting the relative position between each clamping jaw assembly and the moving mechanism; and the limiting mechanism comprises a plurality of limiting supports and is used for limiting the moving position of the clamping mechanism in the third preset direction. The stability in the material moving process can be improved, and the materials can be stably and rapidly moved to the next stamping station.

The invention provides a YOLOv10-based reservoir area floating object identification method, device and equipment, a storage medium and a product, relates to the technical field of environment monitoring, and aims to solve the problems that a traditional floating object identification method generally depends on manual observation, is low in efficiency and high in cost, and is difficult to meet the requirement of efficient management. The method comprises the steps that real-time detection is conducted on floaters according to a flotage recognition model, information of the floaters is obtained, and the information of the floaters comprises the position and the type; the floating object identification model is obtained by training a YOLOv10 model based on image data marked with floating object positions and types; the information of the floating objects is sent to a central control system for analysis and display, and the positions, types and the number of the floating objects are obtained; whether the number of the floating objects exceeds a preset threshold value or not is judged; and when the number of the floaters exceeds a preset threshold value, a corresponding alarm mechanism is triggered.

The invention discloses a liquid density detection device and detection method based on a capacitive micromechanical ultrasonic transducer, and belongs to the technical field of ultrasonic detection and sensors. The detection device is mainly composed of a liquid detection tank, an aluminum block, a CMUT, an ultrasonic pulse generation receiver, an oscilloscope, a data processing module and a display storage unit. According to the detection method disclosed by the invention, the liquid density is calculated according to a preset frequency-density calibration curve by measuring the change of the resonant frequency of the CMUT in liquids with different densities according to the fact that the change of the liquid density can influence the diaphragm damping of the CMUT. The detection device has the advantages of being high in precision, high in portability, high in response speed, good in real-time performance and the like in liquid density detection, has the characteristic of being suitable for various liquids, and can be widely applied to the fields of industry, environment monitoring, biomedicine and the like.

The invention discloses a detection and tracking method and device for dense pedestrians, relates to the technical field of computer vision, and can solve the problem of low tracking accuracy during real-time detection and tracking in a scene with extremely dense pedestrians in the prior art. According to the specific technical scheme, firstly, a dense pedestrian detection data set and a tracking data set are obtained; an alpha-EIoU loss, a GAM attention mechanism and a Swin Transform module are used for improving a YOLOv5s algorithm, and a detection model is obtained; training and optimizing the detection model to obtain a target detection model; a ShuffleNetV2 module is introduced to improve a DeepSORT algorithm, and then training is performed to obtain a target tracking model; and detecting and tracking the pedestrian target video according to the target detection model and the target tracking model to obtain an experimental result of detection and tracking. The method is used for detecting and tracking dense pedestrians.

The invention provides a method for rapidly detecting various antibiotics in a water body. The preparation method comprises the following steps: firstly, preparing an enhanced substrate for surface enhanced Raman spectroscopy analysis, and enabling the substrate to have high SERS activity by a method for re-dispersing large aggregates through salt regulation and ultrasonic treatment; secondly, preparing a'glass-PDMS (Polydimethylsiloxane)-glass' micro-fluidic chip; the enhanced substrate and a sample to be detected are premixed and then pumped into the micro-fluidic chip for SERS (Surface Enhanced Raman Scattering) detection. According to the invention, ultralow-concentration detection of multiple antibiotics in multiple water bodies can be realized; the method has the advantages of no need of marking, short detection time and high sensitivity, and can be used for rapid and high-precision detection of various antibiotics in a water body.

The present invention discloses an infrared highway foreign body detection method based on Mamba architecture, comprising: continuously collecting data of highway scenes under different lighting conditions and different weather conditions through infrared sensors to obtain infrared image data, preprocessing the infrared image data through data augmentation to generate training data, and dividing the training data into training set, test set and verification set in proportion; constructing a detection network based on Mamba architecture, including a student network and a teacher network; training the teacher network through back propagation based on the training set; training the student network based on the results of the teacher network and real data to obtain a target detection network; inputting the image to be identified into the target detection network to obtain the detection result. Rapid and accurate identification of potential dangers is achieved, and the powerful processing capability of the infrared imaging technology and the Mamba framework is combined to ensure high detection accuracy, achieve real-time response to foreign body threats, and significantly improve the efficiency and safety of highway monitoring.

The invention relates to the technical field of nucleic acid detection, and discloses a multi-target micro-fluidic chip with a novel liquid distribution mode, which comprises a reaction layer, a sealing layer and a liquid distribution layer, the reaction layer is provided with a plurality of amplification reaction cavities, sample loading buffer bins and test paper storage grooves; a vertically arranged sample inlet, a plurality of sample shunting channels and a plurality of sample vertical runners are arranged in the liquid separation layer; the sample inlet and the sample vertical runners upwards and downwards penetrate through the top surface and the bottom surface of the liquid separation layer respectively; the sample vertical flow channels, the sample shunting channels and the amplification reaction cavity are correspondingly arranged in position, the sample shunting channels are respectively communicated with the sample inlet, the sample shunting channels are the same in size and are arranged at intervals along a conical surface, the sample vertical flow channels are also the same in size, and the sample vertical flow channels are communicated with the amplification reaction cavity. The sample can be completely equally divided, and a plurality of parallel reactions are promoted to be simultaneously carried out; and as many samples enter the amplification reaction cavity as possible, so that the liquid loss is reduced, and the detection accuracy can be further improved.

The invention discloses an intelligent necklace, and relates to the technical field of intelligent wearable device.The intelligent necklace comprises a necklace body and a pendant, the necklace body defines a wearing hole, and the necklace body can be worn outside a wearer through the wearing hole; the pendant comprises a pendant body, a heart sound sensor and a main control chip, the pendant body is connected to the chain body, a decorative face and a binding face are formed on the two opposite sides of the pendant body respectively, a containing cavity is formed in the pendant body, the heart sound sensor and the main control chip are both contained in the containing cavity, and the detection end of the heart sound sensor faces the binding face. According to the necklace, the heart sound sensor is arranged in the pendant of the necklace, so that the position aligned to the heart sound sensor does not need to be manually adjusted during heart sound detection, the accuracy of heart sound signal detection is improved, meanwhile, the heart sound sensor is perpendicular to the body surface of a wearer, all detected heart sound signals can be fully utilized, and the detection efficiency is improved. Therefore, the detection effect of the heart sound sensor and the accuracy of the detection result are improved.

The invention relates to the technical field of auxiliary driving systems, and discloses an intelligent control method and system for a sun shield, and the method is applied to a vehicle and comprises the following steps: S1, obtaining a relative position of the sun relative to the vehicle; s2, judging whether the irradiation light of the sun at the relative position irradiates the face of a driver of the vehicle or not; s3, if the irradiation light irradiates the face of the driver, calculating an actual ratio of the area, irradiated by the irradiation light, of the face of the driver to the area of the face of the driver, and judging whether the actual ratio is larger than a standard ratio or not; and S4, if the actual ratio is greater than a standard ratio, sending a prompt message to remind a driver to open a sun shield in the vehicle. By means of the mode, a driver is reminded to open the sun shield under the condition that the conditions are met, the sun shield can be automatically opened through the voice awakening function, and the driving safety and the driving comfort of the driver are well improved.

The invention provides a device and a method for measuring wafer eccentricity. The device comprises a plurality of ejector pin supporting seats distributed in the circumferential direction, a plurality of pressure sensors, a plurality of vacuum penetration pieces, a fixed bottom plate and a data processing module. One pressure sensor is arranged at the upper end part of each ejector pin supporting seat; the lower end part of each ejector pin supporting seat is fixed on the fixed bottom plate; the upper end portion of each vacuum through piece is connected with one pressure sensor in a sealed mode, and the lower end portion of each vacuum through piece penetrates through the fixing bottom plate and is connected to the data processing module.

The invention provides a target tracking compensation control method and system for a surgical robot in a dynamic breathing environment, and the method comprises the steps: collecting an ultrasonic image of a patient in real time, and carrying out the self-adaptive admittance control algorithm based on force feedback; force control output is adjusted in a self-adaptive mode according to the contact force between the ultrasonic probe and the skin surface and the change of the displacement and speed state of the ultrasonic probe, rigidity parameters of an admittance control model are adjusted in real time, constant contact force between the ultrasonic probe and the skin surface is kept, and it is ensured that the ultrasonic probe and the body surface keep stable contact force and meanwhile adapt to the dynamic change of the body surface; a focus is continuously identified and positioned by utilizing a real-time ultrasonic image through template matching and a dynamic template updating algorithm, real-time position information of an internal focus target spot is obtained and serves as a target pose, the motion compensation amount of a puncture mechanical arm is calculated, the track of a puncture needle point and the motion of the target spot are kept synchronous, and the target pose is accurately positioned. And target tracking and compensation control in a dynamic breathing environment are realized. Target tracking compensation control of the surgical robot can be carried out.

The invention provides a silicon carbide crystal cutting line fracture detection system and method and a medium, and relates to the technical field of silicon carbide substrate multi-line cutting. The system comprises a wire breakage alarm and an impedance detection rod, and the wire breakage alarm is used for triggering an alarm instruction if the impedance detection rod detects that the impedance value of the impedance detection rod to a target reference point is smaller than a preset impedance threshold value. According to the silicon carbide crystal cutting line breakage detection system, the line breakage alarm and the impedance detection rod are integrated, the impedance value of the impedance detection rod to the target reference point is compared with the preset impedance threshold value, and when the measured impedance value is smaller than the preset impedance threshold value, it can be judged that the cutting line is broken; according to the device, the cutting line state can be detected in real time and quickly responded, the production safety and efficiency are improved, the maintenance cost and the product reject ratio are reduced, and powerful support is provided for optimization, cost reduction and benefit improvement of the silicon carbide crystal cutting process.

The invention discloses a fluorescence sensor protein design method, a product thereof and a detection method, and belongs to the technical field of biology. The design method of the fluorescent sensor protein is characterized in that the distance between amino acids at two ends of an allosteric domain of the protein is less than 10, and the distance change is more than 2. Based on the design method, the invention also provides a fluorescent sensor protein, a gene thereof, a recombinant expression vector, a transformant, a fluorescent sensor and a method for high-throughput real-time detection of intracellular metabolites, and the fluorescent sensor protein comprises allosteric sterol O-acyltransferase and circulating fluorescent protein cpEGFP; the allosteric sterol O-acyltransferase has an amino acid sequence as shown in SEQ ID NO. 1; the circulating fluorescent protein cpEGFP has an amino acid sequence as shown in SEQ ID NO. 2. The fluorescence sensor protein is applied to concentration detection of intracellular sterol substances, detection is more real-time, and a more accurate tool is provided for analysis of related metabolic flow.

The utility model relates to the technical field of bearing beam shearing force detection, in particular to a building bearing beam shearing force detection device which comprises a workbench, supporting legs are fixedly connected to the periphery of the lower end of the workbench, a control keyboard is fixedly connected to the middle of the front end of the workbench, and a reinforcing frame is fixedly connected to the rear portion of the upper end of the workbench. The right portion of the inner upper wall of the reinforcing frame is fixedly connected with two guide rods, the ends, away from the reinforcing frame, of the two guide rods are fixedly connected with the upper end of the workbench together, the left portion of the upper end of the reinforcing frame is fixedly connected with a height adjusting device, and the height adjusting device is slidably connected with the two guide rods. According to the building spandrel girder shearing force detection device, the height adjusting device is guided through the guide rod, the building spandrel girder shearing force detection device can adapt to spandrel girders with different heights, and the shearing force stamping structure can apply shearing force to the fixed spandrel girders, so that the whole building spandrel girder shearing force detection device can detect the spandrel girders with different sizes, and the applicability and the detection accuracy of the device are improved.

The invention discloses a wearable C peptide detection device as well as a detection method and application thereof. The device comprises a detection chip, and the detection chip comprises a microneedle array, a reaction cavity and an injection part which are sequentially connected through a liquid outlet channel. According to the wearable C peptide detection device, rapid detection of C peptide can be achieved within 15 min, and meanwhile, the wearable C peptide detection device can be repeatedly used after elution, so that repeated, rapid and dense data collection can be achieved, and real-time monitoring of the C peptide is achieved.

The invention relates to the technical field of inverter fault diagnosis, in particular to an inverter fault diagnosis method, device and system, and the method comprises the steps: obtaining the three-phase current data of each moment outputted by an inverter in each period; obtaining a forward wave crest of any phase in each period; determining a wave crest abnormal coefficient of any phase of current in each period; determining the wave crest area ratio of each forward wave crest of any phase in each period; determining a waveform abnormal coefficient of any phase of current in each period; determining a current abnormal coefficient of any phase current in each period; determining the fault sensitivity of any phase of current in each period; and performing fault diagnosis on the inverter based on the fault sensitivity. According to the invention, the fault of the inverter can be detected in real time, and the sensitivity of fault diagnosis of the inverter is improved, so that fault detection can be carried out more accurately.

The invention provides a radio frequency front-end assembly and a working method thereof. A radio frequency front end assembly comprises a first amplification module, a second amplification module, a third amplification module, a fourth amplification module, a switch assembly used for realizing dual polarization switching, a first circulator, a second circulator, a third circulator, a fourth circulator, a duplexer A, a duplexer B and a power divider. The first amplification module and the second amplification module are connected with the first circulator and the second circulator in a one-to-one correspondence manner; the third amplification module and the fourth amplification module are connected with the third circulator and the fourth circulator in a one-to-one correspondence manner; the switch assembly is connected with the first circulator, the second circulator, the third circulator and the fourth circulator. The first circulator and the second circulator are both connected with the duplexer A, and the third circulator and the fourth circulator are both connected with the duplexer B; and the duplexer A and the duplexer B are both connected with the power divider.

The invention provides an optical fiber DAS phase demodulation method and demodulation device. The demodulation method comprises the following steps: S1, acquiring a digital signal of a sensing optical fiber under each pulse signal; s2, after sub-band extraction and matched filtering are carried out on the digital signal under each pulse signal and the digital signal under the pulse signal at a preset interval from the digital signal, adjacent differential rotation vector summation processing is carried out, and a complex signal is obtained; s3, solving a phase angle for each complex signal, and performing phase angle expansion to obtain a phase signal; s4, performing spatial differential processing on each phase signal to obtain a phase demodulation signal; s5, according to a preset optical fiber sampling position, extracting a sampling point corresponding to the optical fiber sampling position in each phase demodulation signal to obtain a vibration signal; windowing processing, FFT conversion, frequency domain compensation, IFFT conversion and windowing processing are sequentially carried out on the vibration signal, and a result is output. Accumulated errors can be eliminated, and a complete frequency response range can be reserved.

The invention relates to a rapid and high-selectivity viscosity fluorescent probe as well as a preparation method and application thereof, and relates to the field of fluorescent probes. Specifically, the probe disclosed by the invention can be used as a viscosity probe for detecting and measuring viscosity change in a biological model and a beverage. The probe can realize at least one of the following technical effects: the rapid and high-specificity viscosity response can be realized, and the quantitative analysis of the viscosity can be realized; the viscosity change in a biological system and a beverage can be detected; the viscosity change in the liver injury process can be tracked; beverage viscosity change caused by the thickening agent can be monitored; the synthesis is simple, the property is stable, and long-term storage and use can be realized.

The invention provides a pressure state detection method and system and a storage medium, and relates to the technical field of pressure detection.The detection method comprises the steps that a detection environment is selected, preset detection time is determined, and pressure signals in the detection environment within the preset detection time are collected based on the multi-sensor fusion technology; receiving the pressure signal, preprocessing the pressure signal according to a signal processing algorithm, and generating corresponding pressure data; obtaining a target pressure model, and determining a pressure model index of the target pressure model according to the pressure data; and inputting the pressure data into the target pressure model to obtain the pressure state in the detection environment within the preset detection time, so that efficient and real-time detection of the pressure state can be realized, the detection precision is relatively high, the accuracy and reliability of pressure detection are ensured, and meanwhile, the model is relatively high in applicability and flexibility, can adapt to complex and changeable application scenes, and is suitable for popularization and application. The change of the detection environment is reflected in time, and the algorithm complexity and the calculation cost are low.

The invention relates to a contact network static geometric parameter detection method based on a Swin Transform-CNN (Convolutional Neural Network), and aims to improve the precision and efficiency of contact network geometric parameter detection. The method comprises the following steps: firstly, acquiring static image data of the overhead line system through image acquisition equipment, and carrying out preprocessing, such as denoising, graying and normalization processing; then, the image data are input into a hybrid model based on Swin Transform and a convolutional neural network (CNN), the Swin Transform is used for extracting global geometric features of the overhead line system, and the CNN is responsible for extracting local detail features; by fusing the global and local features, the static geometric parameters of the overhead line system, including the height, inclination angle, tension, position offset and the like of the contact line, can be accurately detected. The method not only can adapt to complex environmental changes, but also can generate a visual detection report, provide real-time alarm and maintenance suggestions, and improve the maintenance efficiency and safety of the overhead line system.