57results about How to "Realize accurate detection" patented technology

The invention provides a pavement crack detection method and apparatus, and an electronic device. The method comprises the steps of obtaining a crack image and a crack prediction model obtained by pre-training; wherein the crack prediction model comprises a visual modeling model, an inference model and a dynamic multi-modal fusion network; performing hidden space global visual modeling on the crack image based on a visual modeling model to obtain hidden space global visual features; generating a semantic mask matrix guided by global scene semantics for the crack image based on an inference model; and inputting the hidden space global visual features and the semantic mask matrix into a dynamic multi-modal fusion network, and predicting to obtain crack information. According to the invention, pixel-level prediction of the crack can be carried out, accurate detection of the pavement crack is realized, and technical support is provided for pavement maintenance.

The invention discloses an infrared target detection method based on knowledge distillation and fast multi-scale fusion, and the method comprises the steps: constructing an infrared target detection model based on a teacher-student network, dividing an obtained infrared target image into a training set and a test set, inputting the data of the training set into the model for training, and obtaining a trained model; and inputting the test set for testing, inputting a to-be-detected infrared target image into the model after the test is completed, and outputting an infrared target detection result to complete infrared target detection. According to the method, a knowledge distillation technology is utilized, a high-complexity teacher network model is used for assisting student network model training, a fast multi-scale fusion module and a fast self-adaptive sensitivity strategy are designed for a teacher network and a student network, computer storage space and computing resources are saved, small target positioning is more accurate, and the method is more practical. And features of different levels of the small target are reserved, a large amount of storage space and computing resources are effectively relieved, and accurate detection of the infrared small target is realized.

The embodiment of the invention discloses a transformer winding short circuit fault detection method. The method comprises the following steps: acquiring magnetic flux leakage data between transformer windings; determining a short-circuit fault area according to the magnetic flux leakage data; constructing a finite element model of the transformer; simulating the finite element model of the transformer based on the short-circuit fault area to obtain a short-circuit database; and acquiring magnetic flux leakage data between the transformer windings in real time, and determining the short-circuit fault position of the transformer windings according to the short-circuit database. According to the method, accurate detection of the short-circuit fault of the transformer winding is achieved by obtaining the magnetic flux leakage data, determining the short-circuit fault area, constructing the finite element model, obtaining the short-circuit database through simulation and locating the fault in real time, the accuracy and efficiency of fault detection are improved, and powerful guarantee is provided for safe operation of the transformer.

The invention discloses a 3D target detection method and system based on binocular vision and open vocabularies, and relates to the technical field of target detection. The method comprises the following steps: calculating pixel features of each image by adopting an open vocabulary 2D detection model; pixel features of the left view and the right view are aggregated to each three-dimensional point; obtaining a parallax estimation map based on the pixel features of the left and right views; introducing a time aggregation guide parallax consistency constraint to obtain a depth parallax estimation map; converting the depth parallax estimation map into a 3D point cloud, and extracting features to obtain 3D features; and carrying out correlation integration on the text features, the 2D fusion feature vectors and the 3D features by utilizing object-level image guide detection loss, object-level cross comparison alignment and object-level self-comparison learning to obtain a three-dimensional target detection model, realizing three-dimensional target detection and distinguishing different target objects. According to the invention, accurate detection and distinguishing of the target can be realized, so that the flexibility and universality of 3D target identification are improved.

The invention discloses an intelligent positioning system capable of conveniently adjusting the standing posture of a patient and a method thereof, and relates to the technical field of intelligent medical treatment. Comprising the following steps of initial posture evaluation and skeleton modeling, gravity center and supporting point distribution calculation, spine curvature and inclination detection, intelligent laser-assisted posture adjustment, dynamic muscle activation feedback monitoring, pelvis and lower limb alignment correction, posture balance prediction and guidance, upper limb posture and shoulder blade stability correction and real-time adjustment of biological feedback driving. And carrying out long-term attitude optimization and habit remodeling, and carrying out system construction based on the method. According to the intelligent positioning system convenient for adjusting the standing posture of the patient and the method thereof, accurate detection, real-time adjustment and long-term optimization of the standing posture problem of the patient are realized by setting and combining multi-modal data acquisition, a dynamic feedback mechanism and biomechanical analysis. The problems of low static evaluation precision, lack of dynamic feedback and poor long-term effect in the prior art are effectively solved.

The invention discloses a target detection method and device based on multi-modal fusion, electronic equipment and a medium, and the method comprises the steps: obtaining a first image set and a second image set, the first image set comprises a plurality of visible light images, and the second image set comprises a plurality of infrared images; performing data alignment operation on the first image set and the second image set to obtain a first feature image set corresponding to the first image set and a second feature image set corresponding to the second image set; performing feature fusion on the first feature image set and the second feature image set through a preset dual-scale image fusion algorithm to obtain a fused image set; and performing target detection on the fused image in the fused image set through a preset deep learning model to obtain the object category of the to-be-detected object and mark the bounding box of the to-be-detected object. In the embodiment of the invention, the accuracy of target detection can be improved, and the detection precision is further improved.

The invention provides a fault detection circuit, system and method for buses and switch tubes between the buses, multiple groups of switch tubes are connected between positive and negative buses, the detection circuit comprises a switch module and a voltage-dividing sampling detection module, and the voltage-dividing sampling detection module comprises a power supply and a voltage-dividing resistance network. The divider resistance network comprises a detection resistor with a voltage dividing function, and the switch module is arranged to connect the two ends of the detection resistor to a positive bus and a negative bus respectively when the switch module is closed, so that the detection resistor is short-circuited when the positive bus and the negative bus are short-circuited. And the sampling voltage output by the voltage division sampling detection module is different from the sampling voltage output when the positive and negative buses are not short-circuited. The bus voltage short-circuit detection function and the switch tube short-circuit and open-circuit detection function between buses are achieved.

The invention belongs to the technical field of image processing, and discloses a nut flaw detection method and system based on image processing, and the method comprises the steps: positioning an internal thread effective detection region based on an internal thread region image of a to-be-detected nut; converting the area from a Cartesian coordinate system to a polar coordinate system, and generating a polar coordinate expanded image; extracting thread feature positions in the angular direction, and constructing a thread phase sequence; calculating adjacent phase differences and comparing the adjacent phase differences with an allowable range of a thread phase continuity reference model established based on qualified samples; and when adjacent phase differences of a plurality of continuous angular positions exceed an allowable range, determining that flaws exist. The method focuses on the phase continuity of a thread periodic structure, can accurately identify hidden defects which are qualified in size but failed in assembly, is high in anti-interference capability, is adaptive to a high-speed online detection scene, and effectively improves the nut detection precision and reliability.

The utility model relates to the technical field of robot positioning, and discloses a detection device for secondary positioning of a robot, which is characterized in that a rotating seat is arranged on a base, and a driving assembly is arranged in the base and comprises a rotating motor, a belt wheel I and a belt wheel II. The first belt wheel and the second belt wheel are connected through a belt, and the second belt wheel is connected with the rotating base through a rotating column. A turnover motor is arranged on the rotating seat, the output end of the turnover motor is connected with a worm, a turnover shaft is arranged above the worm, a worm gear is sleeved on the shaft, and the worm is matched with the worm gear. A mechanical arm connecting seat is arranged on the turnover shaft, and a mechanical arm is arranged in the mechanical arm connecting seat; the horizontal direction detection module and the vertical direction detection module can detect the horizontal rotating angle of the rotating base and the overturning angle of the mechanical arm correspondingly, and therefore accurate detection of the overall position of the robot is achieved. Meanwhile, the controller controls the driving assembly, the position of the robot can be automatically adjusted, and the positioning precision is further improved. In addition, the device is simple in structure, easy to implement and low in cost, and has high practicability and market popularization value.

This invention discloses an online inspection device and method for the light conversion film of photovoltaic modules. The online inspection device includes a conveyor roller, an inspection cover, multiple top excitation light source units, multiple top signal acquisition units, multiple edge excitation light source units, multiple edge signal acquisition units, and a processing unit. The conveyor roller is used to transport photovoltaic modules along a first horizontal direction; the inspection cover has an openable and closable inlet / outlet on its side wall along the first horizontal direction; the top excitation light source units and top signal acquisition units are both located on the top inner wall of the inspection cover and are used to project a first wavelength of light onto the photovoltaic modules inside the inspection cover; the edge excitation light source units and edge signal acquisition units are both located on the inner side wall of the inspection cover along a second horizontal direction and are used to collect the second wavelength of light projected by the photovoltaic modules inside the inspection cover. This online inspection device achieves accurate inspection of non-transparent photovoltaic modules without blind spots, breaking through the dependence of traditional through-type inspection on transparent photovoltaic modules.

This utility model belongs to the field of automated production line equipment, and particularly relates to a branch track merging system for a traditional Chinese medicine outer packaging production line. It includes: two independent conveyor belts, one merged conveyor belt, a linear motion mechanism, a vertically moving cylinder, a gripping stop, and two material detection photoelectric sensors. The two independent conveyor belts transport medicine boxes to be merged; the merged conveyor belt receives and transports the merged medicine boxes; the gripping stop is used to grasp and release the medicine boxes; the two material detection photoelectric sensors are respectively installed at the ends of the two independent conveyor belts to detect the presence of medicine boxes. This branch track merging system for traditional Chinese medicine outer packaging production line can achieve precise detection, pushing, and merging operations of paper boxes. The system can automatically complete the merging of branch tracks according to set parameters such as quantity, delay time, and speed, reducing paper box damage, improving production efficiency, and reducing production costs.

This invention provides a field-effect transistor (FET) type electronic nose, its fabrication method, and a gas identification method, comprising a sensor array consisting of multiple sensing units. Each sensing unit includes, from bottom to top, a back gate electrode, a dielectric layer, a two-dimensional material layer, and an organic semiconductor layer. An active electrode and a drain electrode are respectively disposed on both sides of the two-dimensional material layer. The organic semiconductor layers of different sensing units are made of the same material but have different thicknesses. By using the same material to form the same structure for each sensing unit, not only is the manufacturing process and cost reduced, but the device stability is also effectively improved. Simultaneously, by varying the thickness of the organic semiconductor layer in each sensing unit, the differential response characteristics of different thicknesses of the organic semiconductor layer to different gases are utilized to achieve accurate detection of different gases. This solves the problem of reducing manufacturing difficulty and improving stability while maintaining the sensitivity of the electronic nose.

The application discloses a downhole instrument creep detection method based on multi-source sensor fusion and relates to the technical field of oil and gas well engineering, which comprises the following steps: obtaining sensor data of a target instrument in a target downhole preset period and preprocessing the sensor data to obtain preprocessed sensor data; performing magnetic interference detection based on the preprocessed sensor data to obtain a first detection result; performing multi-source joint likelihood ratio test based on the first detection result and the preprocessed sensor data to obtain a second detection result; when the second detection result meets the requirements, performing accelerometer variance detection and fiber-optic gyroscope output energy detection based on the preprocessed sensor data to obtain a third detection result; when the third detection result meets the requirements, constructing a long-period observation window and obtaining a second variance based on the preprocessed sensor data; and determining a creep termination time based on the second variance and the long-period observation window. The application realizes sensitive capture of small creep movement of a downhole instrument and further obtains an accurate creep time of the downhole instrument.

The present application relates to air conditioning technical field, specifically disclose a kind of parking air conditioner voltage detection method, detection device and parking air conditioner, to solve the problem that the user experience is poor due to the large voltage detection error of existing parking air conditioner.For this purpose, the parking air conditioner of the present application includes indoor unit and power supply unit, detection unit is arranged on power supply unit, detection unit is connected between indoor unit at least by short-range wireless communication mode, the parking air conditioner voltage detection method includes the following steps: obtaining the connection state of short-range wireless communication mode;When short-range wireless communication mode is connected successfully, power supply unit is detected based on first detection logic;When short-range wireless communication mode is connected failed, power supply unit is detected based on second detection logic.In the present application, according to the connection state of short-range wireless communication mode, and according to different detection logic, the accurate detection of power supply unit is realized, so as to effectively improve the use experience of user.

The present invention relates to a blink category judgment method, comprising dividing all window sample data into a short blink sample data set, a long blink sample data set and a non-blink sample data set; using a K-Means algorithm to obtain a short blink centroid, a long blink centroid and a plurality of non-blink centroids; obtaining a minimum blink peak value and an average angle; obtaining real-time electrooculogram window data; respectively calculating the correlation coefficients between the real-time electrooculogram window data and the short blink centroid, the long blink centroid and a plurality of non-blink centroids, obtaining the category of the sample data set to which the centroid corresponding to the maximum correlation coefficient belongs, as an initial category label of the real-time electrooculogram window data; calculating the maximum peak value of the real-time electrooculogram window data, and comparing it with the minimum blink peak value; calculating the real-time specific angle of the real-time electrooculogram window data, and comparing it with the average angle, to obtain the actual category label of the real-time electrooculogram window data.

The utility model discloses a kind of support leg telescopic detection mechanism and aerial work vehicle, it is related to engineering vehicle technical field, to solve the problem that current technical means cannot accurately determine whether support leg is in extended state or retracted state, the detection mechanism is used to set on the oil cylinder of driving support leg telescopic, comprising: movable rod, movable by limiting piece is in oil cylinder, the moving direction of movable rod is consistent with the telescopic direction of the piston rod of oil cylinder, and the end of movable rod is connected with the extended end of piston rod by cross byte;Detection component is located in movable rod;Sensor is located on limiting piece, and sensor is used for real-time detection position change of detection component;So set, the telescopic state of support leg can be accurately detected, improve the safety of aerial work platform, ensure that movable rod and piston rod always maintain parallel state, ensure that detection mechanism can normally work under various working conditions, enhance the adaptability and reliability of equipment.

The application discloses a parking space detection method and related device, and relates to the technical field of intelligent driving, and comprises the following steps: acquiring a to-be-detected parking space image, generating a bird's-eye view feature map according to the to-be-detected parking space image, initializing semantic identities and initial position representations of a plurality of contour points by using a learnable query vector, sampling local features in the bird's-eye view feature map based on the initial position representations, interacting the local features with the query vector, generating corrected position representations of the plurality of contour points according to the query vector updated in the interaction process, returning the corrected position representations to the step of sampling the local features as the initial position representations, and until a preset iteration termination condition is met, taking the semantic identities and the last iterated corrected position representations of the plurality of contour points as a parking space detection result of the to-be-detected parking space. The application models a parking space edge as a plurality of contour points, and corrects position representations of the plurality of contour points through an iteration process, thereby realizing high-precision modeling of an arbitrary-morphology parking space.

The invention relates to the technical field of locks, in particular to a double-movable shifting shaft assembly of a motor lock body. The double-active shifting shaft assembly of the motor lock body comprises a shifting shaft which is used for rotating under the action of external force and is provided with at least one first driving structure; the unlocking structure is arranged on the shifting shaft in a sleeving mode so as to be in running fit with the shifting shaft, the unlocking structure is provided with at least one matching part, and the shifting shaft rotates under the action of external force, so that the first driving structure is connected with the matching parts, and the unlocking structure is driven to rotate to conduct mechanical unlocking action; and the second driving structure is connected with the motor, and the second driving structure is used for driving the unlocking structure to perform motor unlocking action. During mechanical unlocking, a clutch mechanism does not need to be disengaged by a motor speed reducer, the unlocking structure can be driven to rotate by directly shifting the shifting shaft, the unlocking action is completed, the lock body structure is greatly simplified, the manufacturing cost is reduced, meanwhile, the stability and reliability of the lock are improved, and the failure rate is reduced.

The present invention belongs to the technical field of polymer materials, and specifically relates to an environmentally friendly material for soil detection and a preparation method thereof. The material preparation method comprises the following steps: (1) preparing a carboxylated nanocellulose fiber dispersion; (2) preparing a mixed solution of polyethylene glycol and curcumin; (3) dropping the mixed solution into the carboxylated nanocellulose fiber dispersion; (4) placing the mixture in a light-proof acidic vapor environment for gelation, and freeze-drying to obtain an environmentally friendly material for soil detection. The gel can specifically detect and adsorb lead ions, and has the advantages of high sensitivity and no contact pollution. It is particularly suitable for the specific detection of lead ions in soil containing multiple metal ions. In addition, the material is in solid form, which is convenient for storage and transportation, and has broad application prospects.

The utility model discloses a high accuracy detection equipment with feeding mechanism, including base and detection seat, the top of base is fixed with detection seat, one side of detection seat is provided with the stand, the middle part welding of stand has a plurality of layers of partition board, the top of partition board places the raw board. This high accuracy detection equipment with feeding mechanism drives first screw rod to rotate through first motor, drives first sliding block to move along first sliding slot transversely, realizes the positioning of feeding mechanism in horizontal direction, drives second sliding block to move along second sliding slot vertically through the rotation of second screw rod driven by second motor, realizes accurate material taking and material discharging of feeding plate in vertical direction in combination with the telescopic action of electric telescopic link, can take out raw board from the partition board of stand automatically, and accurate conveying to detection seat carries out high accuracy detection, significantly improves detection efficiency and automation level, reduces manual intervention error.

This application relates to the technical field of bus communication technology, and in particular to a communication anomaly detection and self-recovery method in CANFD single-transmission mode. The method includes acquiring the buffer occupancy rate and error status of the CANFD controller's transmit FIFO; determining a communication anomaly state when the buffer occupancy rate reaches a preset full-load threshold for a preset time window, or when an ACK error is detected in the error status; switching the data transmission cycle from a preset communication cycle to a test cycle longer than the preset communication cycle, and sending test data frames to the CANFD bus according to the test cycle; and when a successful ACK response is detected for the test data frame, releasing the application layer protection lock and resuming writing service data to the transmit FIFO according to the preset communication cycle. This application can overcome cascading deadlocks caused by no response, perform cross-layer backpressure and frequency reduction detection, and achieve communication self-recovery.

This invention relates to the field of aluminum tube printing monitoring, and more particularly to a method and system for monitoring aluminum tube printing based on color difference component analysis. The method includes: acquiring a color image of the aluminum tube to be tested; converting the color image to a target color space and extracting basic color difference feature values ​​compared with a standard template; obtaining geometric confidence weights based on the equivalent surface curvature angle and texture complexity, and reconstructing the basic color difference feature values ​​using these geometric confidence weights to obtain geometrically compensated color difference features; using dynamic environmental compensation weights to perform global compensation of the geometrically compensated color difference features over time to obtain a final dynamic adaptive color difference score, and comparing the final dynamic adaptive color difference score with a preset safety color difference threshold to achieve monitoring and early warning. This invention enables accurate detection of the printing quality of curved, highly reflective packaging materials in complex and harsh industrial environments.

The invention discloses a multi-modal model-based target detection method and system, electronic equipment and a medium. The method comprises the steps of obtaining a to-be-detected image, a depth image and millimeter wave signal data; performing feature mapping on the to-be-detected image and the depth image through a modal mapping module of a multi-modal model to obtain first feature data, and performing matrix transposition operation on the millimeter wave signal data to obtain second feature data; performing feature fusion operation on the feature data through an adaptive fusion module to obtain fused features; performing feature coding on the fusion features through a global encoder to obtain a first feature map and a second feature map; inputting the first feature map and the second feature map into a target decoder for fusion coding operation, and outputting a category vector and a target frame vector; and marking the target object on the to-be-detected image according to the category vector and the target frame vector to obtain a detection result. According to the embodiment of the invention, the target recognition precision can be improved, and accurate multi-target detection is realized.

The invention discloses a multi-layer composite medium power distribution network signal distinguishing method based on an acoustic resonance spectrum, and belongs to the technical field of electric power, and the method comprises the steps: obtaining the historical detection data of a multi-layer composite medium; extracting initial acoustic resonance spectrum data from the historical detection data of the multilayer composite medium, and preprocessing the initial acoustic resonance spectrum data to obtain preprocessed acoustic resonance spectrum data; key feature points and feature parameters are determined; fitting the characteristic parameters and the material characteristics based on the key characteristic points to obtain a relation model of the characteristic parameters and the material characteristics; obtaining the relationship between the signal complexity and the defect type; and obtaining a to-be-detected ultrasonic signal, and respectively obtaining a defect size prediction result and a defect type prediction result based on the relation model of the characteristic parameters and the material characteristics and the relation between the signal complexity and the defect type. According to the invention, quantitative characterization of defect types and sizes is realized.

The invention discloses a preparation method and application of a hypochlorite photoelectrochemical sensing platform, and belongs to the technical field of analytical chemistry. According to the preparation process, adopted raw materials are easy to obtain, and reaction conditions are mild and easy to control. The sensing platform is designed and synthesized based on phenothiazine-chromone photosensitive micromolecules as a photosensitizer, and photoelectrochemical (PEC) sensing is realized. The sensing platform has the characteristics of high specificity, high sensitivity and easiness in preparation, quantitative photoelectrochemical signals of pypocholoride can be realized, and the PEC detection electrode is 1.64 nM.

This invention provides a core particle gravity flow deposition film, its preparation method, and its application. The preparation method includes the following steps: (1) grinding and screening core particles sequentially to obtain core nanoparticles; (2) mixing water and the core nanoparticles obtained in step (1) to obtain a suspension; (3) transferring the suspension obtained in step (2) to a solid matrix plate, allowing it to settle, and drying to obtain the core particle gravity flow deposition film. The core particle gravity flow deposition film provided by this invention can achieve accurate analysis of reservoir core permeation characteristics and has broad application scenarios in the field of reservoir wettability and permeation research.

The application discloses a kind of complex scenic spot road scene-oriented automatic driving obstacle identification method, comprising: collecting scenic spot road original image;Input multi-scale feature extraction network, generate multi-scale feature map;It is input into multi-branch decoder, respectively obtain obstacle region segmentation result, obstacle position bounding box coordinates and obstacle type classification probability;Combine the surrounding environment information of road geometric shape, weather and signal light state, using attention mechanism to its feature weighting, generate initial obstacle saliency map;According to the preset obstacle type priority level, detect and process the recognition result of spatial overlap in the map, obtain the obstacle recognition result after conflict resolution;It is consistent with road path segmentation map Space collation, and based on time series smoothing processing eliminates interframe jump, finally output recognition result.The application effectively solves the problem of high-precision, real-time obstacle identification for complex scenic spot road scene.