26results about How to "Solve the low detection accuracy" patented technology

The invention discloses an intelligent low-altitude unmanned aerial vehicle cluster air-ground cooperative detection fire extinguishing system and method, a patrol plan is made by determining the patrol time interval of a patrol area of an unmanned aerial vehicle and the density of patrol routes, the system can dynamically adapt to the change of seasons and human activities, and the system and method are more accurate and more efficient. Fire missing detection can be effectively avoided; identifying a fire source by adopting a multi-source data fusion technology, determining whether the fire source exists in the forest region through the infrared image and the actual image of the forest region, and determining the intensity and the position of the fire source according to the smoke concentration when determining that the fire source exists in the forest region. The problem that the accuracy of infrared thermal imaging is affected and the detection precision is low due to the temperature difference of different vegetation, the change of the sunlight irradiation angle, water vapor in the atmosphere and the like can be solved. The fire trend prediction module and the fire extinguishing scheme generation module are adopted to automatically realize fire extinguishing of the forest area, the prediction result is efficient and accurate, and missing of the optimal fire extinguishing opportunity can be avoided.

The invention relates to a high-voltage interlocking detection method, a high-voltage interlocking detection system and computer equipment. The high-voltage interlocking detection system comprises a controller of a vehicle and all high-voltage components. The controller and the low-voltage interlocking wire harness form a low-voltage interlocking loop; the high-voltage components are connected in series in the low-voltage interlocking loop through low-voltage interlocking wire harnesses; and the controller is respectively connected with each high-voltage component. The method comprises the following steps: performing low-voltage detection on a low-voltage interlocking loop to obtain a first detection state; and acquiring a voltage state corresponding to each high-voltage component, and determining a high-voltage interlocking detection result according to the voltage state and the first detection state. Through the method and the device, the problem of low detection accuracy caused by false alarm in related technologies is solved, the robustness of high-voltage interlocking detection is enhanced by combining the voltage state corresponding to each high-voltage component with the first detection state of low-voltage detection, and the driving safety is improved.

The application discloses a multi-modal false news detection method and device, and relates to the field of multi-modal cognitive computing. The method is used for providing in-depth mining and analysis of news information, realizing semantic enhancement of multi-modal information, and mining and application of news sentiment features. According to the BERT model, the first image caption information is subjected to feature coding to obtain a first caption representation and a first empirical representation. A first text representation is subjected to a self-attention mechanism, a full connection layer and averaging to obtain a sentiment representation of comprehensive analysis of the first text representation by all experts and a sentiment tendency evaluation value of comprehensive analysis of the first text representation by all experts. According to the attention weights of a plurality of different sentiment news processors and the first news processing features, news aggregation features for true or false classification are obtained. According to the news aggregation features for true or false classification and a false news detector, a prediction label of the first news is obtained.

The invention belongs to the technical field of artificial intelligence, and relates to a water immersion identification method and device, computer equipment and a storage medium, the method comprises the following steps: receiving real-time water immersion data collected by water immersion data collection equipment, the real-time water immersion data comprising real-time sensor data and real-time image data; performing state estimation processing on the real-time sensor data according to a Kalman filtering algorithm to obtain a water immersion state estimation value; performing visual verification processing on the real-time image data according to an image recognition technology to obtain visual water immersion confidence; performing fuzzy reasoning processing according to a fuzzy rule, the water immersion state estimation value and the visual water immersion confidence to obtain a fuzzy reasoning result; performing defuzzification processing on the fuzzy reasoning result to obtain an accurate alarm confidence coefficient; and confirming a water immersion identification result according to the accurate alarm confidence coefficient. According to the invention, through fusion analysis and intelligent reasoning of multi-source information, accurate identification and reliable early warning of the water immersion state are realized.

This invention discloses a method and apparatus for inspecting wafers. The inspection method includes: acquiring grain registration marks contained in the wafer and the wafer to be inspected; determining a grain registration path based on the wafer to be inspected and the grain registration marks; determining a detection path based on the wafer to be inspected; acquiring a grain registration image of the grain row while the wafer moves along the grain registration path; acquiring a grain registration mark template image; determining the deviation between the actual position and the theoretical position of the grain registration mark in each wafer row based on the grain registration image and the grain registration mark template image; when the wafer moves along the detection path and the wafer to be inspected image of the grain row is acquired, determining the region of interest image of each wafer based on the deviation; and determining the defect detection result of the wafer based on the region of interest image. The accuracy of wafer inspection can be improved by using the above inspection method.

The present application provides a method for detecting federal privacy leakage with anti-poisoning based on local personalized learning, which includes: the IoT device determines the logical adjustment loss value based on the total number of categories supported by the initial detection model, the weight of each category supported by the initial detection model, the correction vector of each category, the predicted label and the true label; the network parameters of the initial detection model are adjusted based on the logical adjustment loss value to obtain a candidate detection model; the reference model parameters are obtained from the candidate detection model, and the reference model parameters are sent to the central server; the central server determines the initial global model parameters based on multiple reference model parameters; the central server determines the target global model parameters based on the initial global model parameters, and sends the target global model parameters to the IoT device, so that the IoT device obtains the target detection model based on the target global model parameters. Through the technical solution of the present application, it is possible to accurately know whether the privacy data is leaked.

The invention discloses a component porosity lossless prediction method and device, a storage medium and electronic equipment. The method comprises the steps that a to-be-detected target component is placed in a detection platform to be subjected to frequency detection, a frequency detection result is obtained, and a laser excitation probe and a three-dimensional laser vibration meter are placed on the two sides of the detection platform; determining a highest frequency signal in the frequency detection result; determining an ultrasonic signal corresponding to the highest frequency signal through a laser excitation probe, and receiving the ultrasonic signal through a three-dimensional laser vibration meter; and processing the ultrasonic signal to obtain a non-linear zero-frequency wave signal, determining a target deep learning network model based on the non-linear zero-frequency wave signal, and predicting the porosity of the target component through the target deep learning network model. The problem that in the prior art, the detection accuracy of the porosity of the additive manufacturing component is low is solved.

This utility model provides a contact wire wear detection device, relating to the field of wear detection technology. It includes a busbar with a cleaning mechanism at its bottom. The cleaning mechanism includes a detection housing, a cleaning box fixedly connected to the outer surface of the detection housing, a cleaning rod rotatably connected to the inner surface of the cleaning box via a bearing, a cleaning brush fixedly connected to the outer surface of the cleaning rod, and two limiting rods fixedly connected to the outer surface of the cleaning box. Sliding blocks are slidably connected to the outer surfaces of the two limiting rods. This utility model achieves automatic cleaning of the contact wire inside the busbar before detection. Through the coordinated action of the cleaning brush and scraper, it can efficiently wipe away stains on the wire surface and promptly clean the dirt attached to the cleaning brush, avoiding secondary contamination. This solves the problem of low detection accuracy caused by stains on the wire surface in existing detection devices, significantly improving the reliability of wear detection data.

Embodiments of the present application disclose a kind of odometer detection method, device, electronic equipment and storage medium, method includes: the odometer associated data of odometer and the measurement source associated data of measurement source are acquired;Wherein, the odometer associated data includes odometer data and odometer state;The measurement source associated data includes measurement source data and measurement source state;In the case where it is determined that the odometer state is odometer delay effective and the measurement source state is measurement source effective, the deviation data of the odometer data and the measurement source data is calculated;According to the deviation data, the detection state of the odometer is determined.The technical scheme of the present application embodiment can improve the accuracy of odometer state detection.

The invention relates to the technical field of radar signal processing, in particular to a radar adaptive detection method based on improved large-dimensional precision matrix estimation. The problem that existing radar detection is low in accuracy is solved. The method comprises the steps of collecting radar data and preprocessing the radar data to obtain preprocessed radar data; determining a to-be-detected unit and a reference unit according to the preprocessed radar data; constructing a sample covariance matrix according to the reference unit, and then inversing the sample covariance matrix to obtain a sample precision matrix; obtaining an improved precision estimation matrix according to the sample precision matrix; and judging whether a target exists in the distance unit of the place where the unit to be detected is located according to the improved precision estimation matrix. By using the method, the detection performance of a radar system under the condition that the dimension N and the sample size K are in the same order of magnitude can be remarkably improved. The problem of low radar detection accuracy caused by poor detection performance of a self-adaptive detector in a large-dimensional asymptotic system in the existing method is solved.

The invention discloses a driving fatigue detection system based on abnormal actions of a driver, and relates to the technical field of driving fatigue detection, the system comprises a video acquisition unit, a video preprocessing unit, a feature extraction unit, a fatigue detection algorithm unit and a fusion detection unit; the video preprocessing unit is used for preprocessing each hand and head action image to obtain preprocessed hand and head action images; the feature extraction unit is used for carrying out key feature extraction on the preprocessed hand and head action images to obtain hand and head action features, and the fatigue detection algorithm unit is used for carrying out fatigue detection according to all the hand and head action features to obtain an action recognition result; and the fusion detection unit determines a final fatigue detection result of the driver according to the action recognition result. The driving fatigue detection precision is improved.

The utility model provides a lithium battery hot pressboard flatness and parallelism detection device and system, and belongs to the field of lithium batteries. The device comprises an upper pressing plate, a lower pressing plate and a detection device, the upper pressing plate and the lower pressing plate are arranged in parallel in a spaced mode, the detection device is arranged between the upper pressing plate and the lower pressing plate and comprises a sliding rail mechanism, a sliding rod and a displacement sensor, the sliding rail mechanism comprises two stand columns and a first sliding rail, the stand columns are arranged perpendicular to the lower pressing plate, and the two stand columns are arranged in a spaced mode; the first sliding rail is erected on the two stand columns and is parallel to the lower pressing plate, the sliding rod is arranged on the first sliding rail in a sliding mode, and the displacement sensor is arranged on the sliding rod and enables the probe direction of the displacement sensor to be perpendicular to the lower pressing plate. By adopting the lithium battery hot pressboard flatness and parallelism detection device and system provided by the embodiment of the utility model, the problem of low detection precision in the prior art can be solved.

The invention provides a text image direction detection method based on deep learning, and belongs to the field of computer vision, and the method comprises the steps: cutting an original text image according to a preset size, and obtaining a preprocessed text image; inputting the preprocessed text image into an image direction detection model to obtain a direction confidence data set; detecting whether the maximum confidence data in the direction confidence data set meets the confidence threshold requirement, and when the confidence threshold requirement is met, taking the direction corresponding to the maximum confidence data as the text detection direction; and when the confidence coefficient threshold requirement is not met, height-width ratio data of the original text image is adjusted, and re-detection is performed through the image direction detection model. The method is based on the image direction detection model, the detection precision of the character direction in the image is improved, and the attention of the model on the character content characteristics in the image is improved by adjusting the aspect ratio of the original text image, so that the method can be suitable for complex image scenes to meet the requirements of downstream task processing.

The invention belongs to the technical field of outlier detection / abnormal point detection of industrial production data, and discloses an outlier detection method based on data acquisition and knowledge filtering and a related device. The outlier detection method based on data acquisition and knowledge filtering comprises the following steps: acquiring industrial production data; based on the collected industrial production data, performing classification detection by using an outlier detection model based on knowledge filtering to obtain an outlier detection result; wherein the outlier detection model based on knowledge filtering comprises an encoder, a decoder and a knowledge filter. According to the technical scheme provided by the invention, the problem of low outlier detection precision caused by lack of priori knowledge utilization in the prior art can be solved, and compared with the prior art, the outlier detection precision can be remarkably improved.

The application relates to the technical field of oil filter detection equipment, in particular to an oil filter internal pressure detection device, which comprises a shell, a end cover located at the top of the shell, an oil inlet and an oil outlet arranged on the end cover, a filter core arranged in the shell, a detection assembly arranged at the bottom of the inner wall of the shell and used for detecting the internal liquid pressure of the filter core, and a liquid buffer assembly arranged in the filter core and used for buffering and guiding the internal liquid of the filter core to the detection assembly. The liquid buffer assembly is arranged to buffer the internal liquid of the filter core, so that the internal liquid of the filter core does not directly rush to the detection assembly, the problem of low detection precision of the detection assembly caused by uneven distribution of the liquid pressure value entering the filter core is solved, and the pressure detection precision of the detection assembly is improved.

The utility model discloses a detection device is used in machine tool accessory processing relates to accessory processing detection tool technical field, including the main part, the main part upper end is equipped with the connecting block, the connecting block upper end is fixed with the spacing block for installing spacing, the lower extreme of main part is equipped with the threaded hole, and the threaded hole is connected with the connecting column in screw thread, and the bottom of connecting column is equipped with the platter, and the platter is equipped with the adjusting mechanism, the utility model discloses the cooperation of the rotating ring of adjusting mechanism and extension block, and it is convenient for flexible expansion platter detection distance, has improved the stability of device when detecting large accessory, and further can realize the function of reliable detection of large accessory, and finally solved the low problem of the detection precision of existing accessory detection device.

The application provides an electroencephalogram fatigue detection method, device and computer readable storage medium. The method comprises the following steps: obtaining original sample data of an electroencephalogram, performing division processing on the original sample data of the electroencephalogram to generate positive sample data and negative sample data; inputting the original sample data, the positive sample data and the negative sample data into a multi-scale time sequence feature coding model to extract corresponding multi-scale time sequence features, performing graph construction processing on the original sample data, the positive sample data and the negative sample data according to the corresponding multi-scale time sequence features to obtain corresponding sample graphs, inputting each sample graph into a graph attention network model, training the graph attention network model, and obtaining a fatigue detection model; obtaining electroencephalogram data of a person to be detected, performing detection processing on the electroencephalogram data by using the fatigue detection model, and outputting a fatigue detection result. The problem of low detection accuracy of electroencephalogram fatigue detection in the prior art is solved.

The invention provides an online ore pulp multi-element grade analysis system. A filter assembly is used for filtering liquid in ore pulp to form ore-containing filter blocks; the material cake drying assembly is used for drying the ore-containing filter block to obtain a dried filter block; the cake crushing assembly is used for crushing and drying the filter blocks to obtain mineral powder; the mineral powder pressing assembly is used for pressing mineral powder into sample cakes for detection, the cake crushing assembly comprises a crushing shell, a crushing disc, a crushing driving mechanism and a feeding mechanism, the feeding mechanism is used for conveying dried filter blocks to the crushing disc, and the crushing disc is arranged in the crushing shell in a rotatable mode; the crushing mechanism is used for crushing the dried material blocks conveyed by the feeding mechanism so as to crush the dried material blocks into mineral powder; the crushing driving mechanism is used for driving the crushing disc to rotate. According to the invention, direct contact detection can be carried out without a detection window, and the problem of low detection precision caused by direct contact of ore pulp for element detection in the existing grade instrument detection is solved.

This application provides a testing fixture for a thermistor, relating to the technical field of electrical component testing equipment. The thermistor testing fixture includes a support platform, a conductive heating band, a clamping mechanism, a power supply circuit, a heating band temperature sensor, and a controller. The power supply circuit heats the conductive heating band and places it over the test position on the support platform. Simultaneously, the clamping mechanism presses the test piece placed at the test position against the conductive heating band. The conductive heating band rapidly heats up, transferring the temperature to the test piece for detection. The controller controls the heating power of the conductive heating band based on the temperature detected by the heating band temperature sensor, thereby enabling accurate testing of the test piece. The thermistor testing fixture provided in this application has a fast response speed, is more suitable for practical scenarios, and solves the problem of low detection accuracy in existing thermistor testing fixtures.

The invention discloses an anomaly detection method and device for lighting equipment, a storage medium and an electronic device.The method comprises the steps that the target brightness level of each piece of lighting equipment in a target object is determined according to the ambient light brightness in a target area, and the target area comprises the target object; under the condition that each lighting device is lightened in sequence based on the target brightness level, respectively acquiring a first video image corresponding to lightening each lighting device; and according to each first video image and the abnormal light spot template image, performing anomaly detection on each lighting device so as to determine whether stains exist in the target object. According to the method and the device, the problem that the detection accuracy of water stain detection is low due to the fact that a water stain detection method in the prior art depends on manual inspection or passive image analysis can be solved.

The present invention discloses a method and device for detecting deformation defects, the method comprising: obtaining a target image with a deformation defect image; inputting the target image into a preset deformation defect detection model; outputting the target position of the deformation defect image in the target image and / or the target defect type corresponding to the deformation defect image through the deformation defect detection model; wherein the target position and the target defect type are displayed through a prediction box output by the deformation defect detection model, the target defect type is one of at least one preset defect type, and the loss function of the deformation defect detection model includes a deformation constraint, and the deformation constraint is used to constrain the deformation of the prediction box. The method disclosed by the present invention can solve the problem of low accuracy of low-contrast deformation defect detection in the prior art.

The invention relates to the technical field of new energy power generation power prediction, particularly provides a new energy power generation power prediction system and method, and solves the problems of long-term and short-term dependency between data, slow model convergence speed and relatively high input space dimension. The system comprises a data module, a prediction model construction module and a test evaluation module. The method comprises the following steps: acquiring real-time environment sensing data, weather forecast and historical power generation power data, and combining to form a fusion data set; inputting the data set into a new energy power generation power prediction model, and predicting a time period and power generation power; the new energy power generation power prediction model performs dimension reduction on data features, constructs a convolutional neural network according to the data features and timestamp information, captures spatial features of the data features, and captures a long and short term dependency relationship by using a long and short term memory network layer. According to the method, the new energy power generation power prediction model is optimized, so that the new energy power generation power prediction accuracy is improved.

The invention relates to the technical field of casting devices, and discloses a zinc ingot quantitative casting device and a method for using the same. The zinc ingot quantitative casting device comprises a base and a mounting plate arranged on one side thereof, a storage barrel is arranged above the base, the storage barrel is connected to the top of the base through a mounting frame, a casting barrel rotatably connected to the base is arranged below the storage barrel, and the casting barrel is communicated with the storage barrel through a connecting hose, a flip component for driving the casting barrel to rotate is arranged at the top of the base, and a quantitative component is arranged inside the casting barrel; the quantitative component at the top of the casting barrel can adjust the space inside the casting barrel, so that the zinc liquid cast each time is the same and the casting amount can be adjusted according to actual conditions, thereby ensuring the casting quality of the zinc ingot, and the quantitative casting is controlled by the quantitative component, and compared with the traditional pressure sensor or liquid level sensor for detecting the injection amount of the zinc liquid, the structure is more stable, not easy to be damaged, and the problem of low detection accuracy can be solved.

The invention discloses a street garbage detection method based on improved YOLOv8s. The street garbage detection method comprises the following steps: making a street garbage data set and preprocessing the data set; on the basis of a C2f module in the YOLOv8s model, a FocudLearAttention self-attention mechanism is fused to form a C2f-FLA module, and the feature fusion effect of the model and the expression ability of the model to global context information are improved; an AIFI module of RT-DETR is introduced into a check layer in YOLOv8s, the AIFI module enhances the feature representation capability through a multi-head self-attention mechanism and a feedforward network layer, global context information and detail features can be captured more effectively, and the detection precision of a small target is improved; a CIoU loss function used in YOLOv8s is improved into a Focal-EIoU loss function, so that the problem of inaccurate positioning of street garbage detection is solved. The improved YOLOv8s model algorithm enhances the expression capability of the model for global context information, so that shallow information and deep information in the image are fully fused, and the image information can be processed more effectively. According to the method, the problem of low detection precision in a street garbage detection scene with complex background and multiple small targets can be well solved, and a relatively high detection speed is kept at the same time.

The present invention relates to the technical field of abnormal alarm, and specifically discloses an abnormal cover opening intelligent alarm system for underground mine maintenance, comprising: a live detection module: responsible for detecting the electrical state of equipment and judging whether the equipment is live; a deviation determination module: setting a standard dust concentration value and obtaining a critical value of the live detection module under the concentration; setting a plurality of different dust concentration values, collecting critical value data of the live detection module under the dust concentration value, and establishing a data set; using the data set to train a neural network model to obtain a simulation measurement model; a calibration module: real-time monitoring of the dust concentration of the current environment, and calculating a simulation critical value using a simulation measurement model; by comparing the difference between the current simulation critical value and the standard critical value, calibrating the standard critical value to obtain a calibration critical value; an alarm module: combining the live condition and cover opening state of the equipment to monitor whether the equipment has an abnormal cover opening.