22results about How to "Reduce inaccuracy" patented technology

The invention provides a children's ability assessment method and device based on multi-modal data fusion, and the method comprises the following steps: S1, collecting the voice data of parent communication, and collecting the video data of children; s2, converting the voice data obtained in S1 into text data, and preprocessing video data; s3, inputting the text data and the preprocessed video data into a multi-modal large language model; and S4, inputting a questionnaire into the multi-modal large language model, and generating a child ability evaluation result by the multi-modal large language model. The beneficial effects of the invention are that by acquiring the child video data and the parent voice data and performing evaluation through the artificial intelligence multi-mode large language model, the subjective deviation of the evaluation personnel is reduced, and the child video data form is convenient for parents to acquire and reduces the inaccurate evaluation result caused by child tension.

The invention relates to the technical field of aerial image detection and the technical field of small target detection, in particular to a small target detection model in an unmanned aerial vehicle aerial scene, which is characterized in that an aerial image data set is acquired, an improved YOLOv8s model is trained by adopting the aerial image data set, and the improved YOLOv8s network model comprises a backbone feature network, a detection head and a loss function; an SPD-Conv module is introduced into a P2 layer, a P3 layer, a P4 layer and a P5 layer of the backbone feature network, a Dynamic Head module is introduced into a detection head, a specific small target detection head (SODH) is added, Shape-IoU is used as a loss function, and a to-be-detected aerial image is input into a trained improved YOLOv8s model for detection; the SDSS-YOLOv8 not only can execute a small target detection task in an aerial image shot by an unmanned aerial vehicle, but also has good performance in the aspect of small target detection in an unmanned ship image, the model has strong generalization ability, and although the image environment and target distribution have changes, the SDSS-YOLOv8 can maintain high detection performance in different scenes.

The invention relates to a regulation and control method and system for mine fireproof doors, a terminal and a medium, and belongs to the technical field of mine fireproofing. The regulation and control method comprises the steps that sensing data sent by sensors around each fireproof door are received; according to the multiple sets of sensing data, whether a fire occurs in the mine or not is judged; if yes, determining a target fireproof door according to the fire spreading area; acquiring personnel position information of workers around the target fireproof door; based on the personnel position information, whether a worker is on a fire spreading path and is located between the target fireproof door and a fire spreading boundary is judged; if yes, the closing degree of the target fireproof door is controlled according to the spreading speed of the fire; if not, the closing degree of the target fireproof door is controlled according to the fire behavior of the fire. The intelligent monitoring management system has the beneficial effect that intelligent monitoring management of the fireproof door can be conveniently realized according to the field environment.

The invention belongs to the field of vehicle-mounted tire pressure monitoring, and discloses a vehicle-mounted tire pressure monitoring platform and method based on the Internet of Things. Comprising the following steps: collecting comprehensive state data of each tire, and constructing a joint structure chart of the tires; calculating the difference data of the joint structure chart at the current time point and the previous time point to form a difference packet, and transmitting the difference packet by using an improved difference technology; building and training a monitoring and early warning model, and importing the difference packet into the trained monitoring and early warning model to obtain early warning information; and the early warning information is displayed through a vehicle-mounted interface, and the vehicle is adjusted, so that intelligent, high-efficiency and high-quality monitoring of the vehicle-mounted tire pressure based on the Internet of Things is realized.

The invention discloses a method for carrying out simultaneous transmission call translation by using an earphone, relates to the field of call translation, and solves the problem that in the process of simultaneous transmission call translation, if only the paraphrase of original voice is rigidly transmitted in a target language, the connotation and context of an original text cannot be fully captured and expressed, and the interpretation of the original text cannot be fully captured and expressed. The problems of serious information distortion and connotation loss are often caused; by simulating the mood of a speaker, the quality of simultaneous transmission call translation can be remarkably improved, the emotion expression is enhanced, the information accuracy is improved, the communication effect is enhanced, and the user experience is improved; in the aspect of technology implementation, the speech synthesis and sentiment analysis technology can provide powerful support, so that tone simulation is more natural and real; according to the method, voice data of different accent, different speakers, different background noise and different speech speeds are collected, a more robust and accurate voice recognition model can be trained, the method is suitable for diversified voice input, and the generalization ability of the method in different scenes is improved.

The embodiment of the invention provides a rotor magnetic steel temperature estimation method and device, equipment, a medium and a product. The method comprises the steps that firstly, if a current motor of the vehicle is in a first working state, the current motor rotating speed, a first counter electromotive force peak value and the first stator winding temperature are obtained; then, according to the current motor rotating speed, the first counter electromotive force peak value and the mapping relation, the first rotor magnetic steel temperature of the current motor is obtained; then, calculating a first temperature difference value between a first rotor magnetic steel temperature and a first stator winding temperature of the current motor; and finally, when the current motor is in a second working state, determining a second rotor magnetic steel temperature according to the first temperature difference value and a second stator winding temperature of the current motor. By means of the method, estimation of the temperature of the rotor magnetic steel is achieved, and the accuracy and reliability of calculation of the temperature of the rotor magnetic steel are improved.

The invention discloses an intelligent closestool cleaner performance test system, an intelligent closestool cleaner comprises a spray rod and a nozzle, the nozzle is arranged on the spray rod, and the intelligent closestool cleaner performance test system comprises a controller, an industrial personal computer, a manipulator, an image acquisition module, a cleaning water consumption measuring module, a cleaning force measuring module, a cleaning area measuring module and a nozzle self-cleaning force measuring module; the nozzle stretching time measuring module is used for measuring the stretching time and the retracting time of the nozzle; the cleaning water consumption measuring module is used for measuring the water consumption of one-time cleaning of the closestool cleaner; the cleaning force measuring module is used for measuring the cleaning force of the closestool cleaner; the cleaning area measuring module is used for measuring the cleaning area of the closestool cleaner; according to the full-process test, image recording, test process semi-automation, test result program judgment and test report automatic generation are carried out. Inaccuracy of test data caused by human factors in the test process is reduced, the whole test process is traceable in the whole process, and the test efficiency is greatly improved.

The invention relates to the technical field of data processing, in particular to a data acquisition and processing method for Actiwatch, which comprises the following steps of: obtaining a sleep state level at each moment according to data in all index data sequences, and obtaining a sleep time period of each small fragment sequence; according to the fluctuation condition of data in the sleep time period, the sleep abnormality degree of each small fragment sequence is obtained, according to the data of each sleep quality parameter of each small fragment sequence, the user exercise amount of each small fragment sequence is obtained, a sleep curve is obtained through fitting, and a low-sleep-quality sequence and a high-sleep-quality sequence are obtained; obtaining a sleep quality index of each small fragment sequence; and scoring the sleep quality of each small fragment sequence according to the sleep quality index of each small fragment sequence, and obtaining an evaluation parameter most suitable for the user. According to the method, the most suitable evaluation parameters of the sleep quality are obtained in combination with the physical conditions of different users, and the degree of misjudgment is reduced.

The application discloses a motion control system of a robot flexible joint module based on double encoders, and the flexible joint module comprises a drive controller, a motor, a first encoder and a second encoder. The drive controller comprises a communication module, a control module and an intelligent drive module which are sequentially connected. The intelligent drive module is electrically connected with the motor. The first encoder and the second encoder are both connected with the control module through the communication module. The intelligent drive module detects three-phase current and bus voltage of the motor and sends them to the control module. Meanwhile, the intelligent drive module receives and amplifies the control signal output by the control module to drive the motor to move. The control module runs a flexible joint motion control algorithm according to the position information of the motor end and the load end, the three-phase current of the motor and the bus voltage, and outputs the control signal to the intelligent drive module. The application has the advantages of small structure size, easy use and high positioning accuracy of the joint.

A light curtain detection device includes a light curtain transmitting unit, a light curtain receiving unit, the light curtain receiving unit is arranged opposite to the light curtain transmitting unit, and receives the light beam emitted from the light curtain transmitting unit; and a transparent protective shell, including a first protective shell and a second protective shell respectively sealed, the light curtain transmitting unit and the light curtain receiving unit are respectively arranged inside the first protective shell and the second protective shell; a detection channel for the product to pass through is formed between the opposite surfaces of the first protective shell and the second protective shell. Compared with conventional photoelectric switches, the light curtain can detect a wider range; the transparent protective shell can reduce the probability of inaccuracy when the light curtain sensor structure is affected by external factors such as water droplets splashing.

The invention discloses a self-testing device for an ultrasonic transducer array, relates to the technical field of transducer array testing, and solves the technical problem that the development speed of a transducer array integration process is too slow due to insufficient transducer array testing accuracy and efficiency. According to the technical scheme, the magnitude of a to-be-measured capacitor is obtained by calculating the rising time and the falling time of charging and discharging of the to-be-measured capacitor, and measurement of array elements in a transducer array is achieved; and meanwhile, a reference capacitor is introduced to reduce current errors and inaccuracy caused by circuit parasitic capacitance. By means of the self-testing device, measurement of electrical parameters of all mainstream transducers can be achieved, and the development speed of the transducer array integration process is greatly increased.

The present invention discloses a Chinese medicinal material seedling bed with air permeability detection, relates to the technical field of Chinese medicinal material planting, comprises a seedling rack and a seedling tray, the top of the seedling rack is slidably connected with the seedling bed, the bottom of the seedling bed is provided with two symmetrical groups of four sliding racks, the bottom of the sliding rack slides on the seedling bed, the two groups of sliding racks are rotatably connected with movable rotating rods, the two movable rotating rods roll between the seedling bed and the seedling rack, and the ends of the two movable rotating rods that are far away are fixedly connected with a handle. The seedling tray can be vibrated synchronously by the vibration of the placement rack, so that the compacted soil in the seedling tray is vibrated and loosened, reducing the steps of manually loosening the soil one by one, and improving the efficiency of loosening the soil in the seedling tray. In addition, loosening the soil by vibration can also avoid the existing manual loosening of the soil causing the soil to leak out of the seedling tray, reducing the steps of additional soil cleaning, and avoiding waste.

The invention discloses an ultrasonic atomization dye liquor dispersing device and a slide dyeing system. The ultrasonic atomization dye liquor dispersing device comprises an operation table and a spray dyeing device, a spray dyeing area is arranged on the operation table, and the spray dyeing area is provided with a plurality of slide carrying tables; the spray dyeing device comprises a liquid injection unit and an atomization unit, the atomization unit comprises a first adjusting part and an atomizer, the atomizer is movably arranged on the operation table, the first adjusting part drives the atomizer to be close to or away from the liquid injection unit, and the liquid injection unit can store and inject dye liquor; the liquid injection unit is communicated with the atomizer, and the atomizer atomizes the dye liquor and sprays the dye liquor to the slide on the slide carrier.

The invention discloses an equipment fault detection method based on Internet of Things and data driving. The method comprises the following steps: deploying and installing a production line environment monitoring module, an equipment state monitoring module, a fault detection module, a control module, a plurality of environment sensors and a plurality of equipment state sensors in a target monitoring area; the production line environment monitoring module obtains environment data around the equipment; the equipment state monitoring module obtains equipment state data; the production line environment monitoring module and the equipment state monitoring module respectively pre-process the acquired data and then transmit the data to the fault detection module; the fault detection module performs fault detection on the received data to obtain a detection result, and transmits the detection result to the control module; the control module adjusts equipment on the production line in real time according to the detection result. According to the method, efficient production line equipment fault detection is realized through the synergistic effect of the sensor, the monitoring module, the fault detection module and the control module.

The application provides a fault detection method and device, electronic equipment and storage medium. The method comprises: obtaining running data corresponding to a plurality of detection points of a target device; inputting the plurality of running data into a pre-trained fault detection model, wherein the fault detection model comprises at least an input layer, a multi-head attention mechanism layer, a plurality of feature extraction layers, a full connection layer and an output layer according to the transmission order of the running data; calculating the weight of each running data through the multi-head attention mechanism layer, and fusing the plurality of running data based on the weight; and extracting and classifying the fused running data through the plurality of feature extraction layers and the full connection layer, and outputting the fault probability of the target device through the output layer. The application solves the technical problem of excessive model parameters of the fault detection model in the prior art, and reduces the complexity of the fault detection model.

This invention discloses a method and system for determining the spatial relationship of potential hazards based on hierarchical early warning, belonging to the field of computer data processing technology. It includes acquiring historical risk data and real-time environmental data, delineating virtual electronic warning zones, and generating dynamic virtual warning zone data; collecting video stream data and identifying and locating potential hazard targets, generating hazard target identification data, and performing spatial matching analysis in conjunction with the dynamic virtual warning zone data to generate an intrusion determination result; if an intrusion occurs, calculating the three-dimensional spatial clearance distance between the hazard target and the live conductor, and outputting a hierarchical alarm result based on multiple key risk factors; if no intrusion occurs, analyzing and predicting the target movement status and location based on the hazard target identification data to generate an out-of-zone early warning result. This invention employs a data processing method that combines dynamic zoning with quantitative analysis and trend prediction of potential hazard targets both inside and outside the zones, enabling intelligent, precise, and forward-looking risk early warning.

The invention provides a business data analysis and additional recording system and method, and relates to the technical field of data processing, and the method comprises the steps: obtaining an additional recording form, and inputting the additional recording form into an additional recording item analysis model, so as to recognize at least one to-be-additionally-recorded power business attribute item in the additional recording form; for each first power service attribute item to be additionally recorded, determining at least one recorded second power service attribute item associated with the first power service attribute item from an additional recording form, and determining additional recording content for the first power service attribute item according to each second power service attribute item; and filling the corresponding first power business attribute item based on each piece of additional recording content information. Therefore, the deep learning model based on the time attention mechanism is used for business data analysis and additional recording, and an efficient, accurate and reliable power grid business data additional recording solution is provided.

The invention belongs to the technical field of big language model reasoning, and particularly relates to a big language model reasoning enhancement method and system based on logic chain verification. The method comprises the steps that historical question and answer pairs are classified into progressive question and answer pairs, multiple logic chains are constructed based on the progressive question and answer pairs, key nodes and secondary nodes are distinguished according to the distribution situation of the nodes in the multiple logic chains, and then a key node evaluation score and a secondary node evaluation score are calculated according to historical evaluation data; and when to-be-verified inquiry content is received, constructing a to-be-verified logic chain for response content and generating a reasoning answer, and then performing weighted summation on evaluation scores corresponding to key nodes and secondary nodes in the chain to calculate a verification score. According to the method, double verification is carried out on the structural importance and historical performance of the nodes in the logic chain, low-credibility results are filtered, and the accuracy and reliability of reasoning answers are improved.

A blade pitch control system for a cycloidal marine propulsion system is provided. A cycloidal marine vessel propulsion system includes a rotating wheel and a plurality of rotating blades attached to the rotating wheel and individually rotatable relative to the rotating wheel. The blade pitch control system includes: at least one rotating wheel position sensor configured to provide a sensor signal indicative of a measured position of a rotating wheel; a wheel position estimation unit configured to determine an estimated position of the rotating wheel based on a sensor signal of the at least one rotating wheel position sensor; and a blade pitch determination unit configured to determine a blade pitch set point of at least one of the plurality of rotating blades based on the estimated position of the rotating wheel. Determining the estimated position of the rotating wheel includes data processing of the sensor signals.

The invention relates to the technical field of building material detection, in particular to a building component strength testing device and method.The building component strength testing device comprises a workbench and a downward pressing mechanism, the downward pressing mechanism comprises a first mounting plate, the first mounting plate is fixedly connected to the workbench, a second mounting plate is fixedly connected to the first mounting plate, and the second mounting plate is fixedly connected to the second mounting plate; a hydraulic cylinder is fixedly connected to the second mounting plate, a pressing block is fixedly connected to a piston shaft of the hydraulic cylinder, a pressure sensor is fixedly connected to the end, away from the hydraulic cylinder, of the pressing block, an anti-sputtering mechanism is arranged on the hydraulic cylinder, and the anti-sputtering mechanism comprises a left shielding box, a right shielding box and a fixing assembly. The left shielding box and the right shielding box are installed on a piston shaft of the hydraulic cylinder through the fixing assembly. The device has the effects that the sputtering radius is reduced when the concrete test block is broken, and the probability that the sputtered concrete test block pollutes the workbench is reduced.

This application discloses a video decoding method, a video encoding method, and related equipment, belonging to the field of video processing technology. The video decoding method of this application includes: when the decoding end determines that the target information does not meet a first preset condition, using the Position Correlated Prediction Combination (PDPC) method to obtain the first pixel prediction value of each pixel point within the encoding unit to be decoded, wherein the target information includes information for determining whether to use the PDPC method to obtain the pixel prediction value; the decoding end decodes the encoding unit based on the first pixel prediction value.

The invention relates to the technical field of signal transmission, in particular to a paperboard batch conveying method and system for packaging box processing, and the method comprises the steps: starting a batch conveying unit, deploying a conveying sensor and a conveying controller, carrying out the data collection of paperboards, calculating the difference degree of the paperboards, and classifying the paperboards according to the difference degree of the paperboards. The batch processing unit, the paperboard difference degree and the paperboard group are used for obtaining a batch sequence, the cloud processing unit is used for calculating a real-time processing load, if the real-time processing load is larger than a real-time processing threshold value, the real-time processing load is adjusted, and if the real-time processing load is not larger than the real-time processing threshold value, the paperboard conveying efficiency is calculated; and if the paperboard transportation efficiency is smaller than the minimum transportation efficiency, transportation adjustment is executed, and if the paperboard transportation efficiency is not smaller than the minimum transportation efficiency, paperboard conveying in batches is completed. The paperboard conveying device can improve the paperboard conveying efficiency.