211results about How to "Improve interference" patented technology

The invention discloses a CO2 detection device based on an optical fiber F-P cavity sensing and embedded system, which comprises a signal detection and processing module, an embedded control module, a power supply voltage stabilization module and a sensing module, and is characterized in that the sensing module comprises an incident light source, an incident conduction optical fiber, a photoelectric sensor, a reflective optical fiber patch cord and an optical fiber F-P cavity CO2 sensor; a control signal of the embedded control module supplies energy to the incident light source after being stabilized by the power supply voltage stabilizing module, incident light enters the optical fiber F-P cavity CO2 sensor through the incident conduction optical fiber, and an emergent light signal is transmitted to the photoelectric sensor by the reflective optical fiber patch cord to be converted into an electric signal which is fed back to the signal detecting and processing module to be processed. The device can realize intelligent and automatic detection, can effectively improve the detection efficiency and accuracy, reduces manual intervention and errors, and is low in cost and easy to industrialize.

The invention discloses an advanced data encryption protection technology for DNA information storage, and the specific technical scheme of the technology comprises the steps: coding to-be-encrypted data into small information blocks and transcoding the small information blocks into a DNA sequence through a coding and encryption scheme, after the DNA sequence is synthesized, introducing interference information through random errors or fractures, the interference intensity of random combination interference noise is effectively improved; the method for decrypting the DNA library carrying the encrypted information comprises the steps of sequencing, fragment assembly, information small block reconstruction, reordering and decoding, and finally decoding the original information based on an erasure coding scheme. The technology has a wide application prospect in the field of data storage and protection.

The invention provides an on-orbit calibration method for a space target pose measurement system based on vision and laser fusion, which is used for solving the problems that in the existing calibration method based on a checkerboard and other special calibration objects, the calibration objects are poor in setting universality for specific use scenes; or the existing calibration method based on a common scene has the technical problems of large calculation amount and relatively high requirements on calculation resources. According to the vision and laser fused space target pose measurement system on-orbit calibration method provided by the invention, firstly, a joint calibration model of a vision sensor and a laser range finder is established, and on the basis, a traditional orthogonal iteration algorithm is improved from two aspects of calculation of an iteration initial value and simplification of the iteration algorithm, so that the calibration precision of the visual and laser fused space target pose measurement system is improved. In order to solve and optimize the pose conversion relation between a laser coordinate system and a visual coordinate system, only one part of a spatial non-cooperative target is used as a laser projection target, and in-orbit high-precision calibration of a visual sensor and laser range finder combined pose measurement system can be realized.

The invention relates to the field of battery cooling equipment, in particular to a V-shaped fin liquid cooling system which comprises a liquid cooling plate, battery packs, cooling modules and heating modules, the upper surface of the liquid cooling plate is fixedly connected with the battery packs at equal intervals, and the cooling modules and the heating modules are arranged in the battery packs. A liquid inlet is formed in the outer wall surface of one end of the liquid cooling plate in a penetrating mode, a liquid outlet is formed in the outer wall of the end, away from the liquid inlet, of the liquid cooling plate in a penetrating mode, a plurality of rectangular straight flow channels are formed in the inner bottom wall of the liquid cooling plate, and the problem that a conventional liquid cooling plate cannot meet the high-discharge-rate heat dissipation requirement of a battery pack is solved; meanwhile, the power consumption of the battery pack is reduced, and electronic expansion valves of the cooling module and the heating module are controlled. The working temperature of the battery can be dynamically controlled within a proper working temperature range, when the temperature of the battery pack is proper, the thermal management system is closed, the power consumption of the battery pack is reduced by adopting natural cooling, and dynamic balance is realized.

The invention relates to the technical field of breeding monitoring, and provides a multi-water-area integrated unmanned breeding monitoring platform and a control method thereof. The platform takes a primary-secondary machine as a basic platform, and comprises a primary machine basic frame, a feeding bin device, a main body cabin, a secondary machine recycling lifting device, an underwater balance base device, a water area parameter monitoring secondary machine device, an underwater garbage recycling secondary machine device and a water surface floating object cleaning secondary machine device, according to the multi-water-area integrated unmanned breeding monitoring platform and the control method thereof, the functions of underwater environment monitoring and maintenance, feed feeding, oxygenation, multi-water-area self-adaption and the like are integrated, and intelligent and integrated breeding is achieved by integrating various modules needed by breeding.

The invention relates to the technical field of tunnel construction, in particular to tunnel segment defect nondestructive testing equipment based on an impact echo method and a testing method.The testing equipment comprises a knocking detection device, the knocking detection device comprises a fixed mounting plate, a double-blade cam, a driver, a knocking rod, an energy storage part and a detection device, the fixed installation plate is installed on the carrying sliding platform in a sliding mode, the double-blade cam and the knocking rod are rotatably installed on the fixed installation plate, the first end of the knocking rod makes contact with the double-blade cam, the second end of the knocking rod is used for knocking a tunnel segment, and the driver drives the double-blade cam to rotate. The elastic piece is connected between the fixed mounting plate and one end, close to the double-blade cam, of the knocking rod, and the detection device is connected to the fixed mounting plate and is used for collecting sound wave signals. According to the invention, the technical problem that engineering application and accurate detection of internal defects of tunnel segments cannot be realized in the prior art is solved.

The invention provides a muscle elasticity modulus detection system and method based on a flexible wearable sensor. The muscle elasticity modulus detection system comprises a flexible wearable ultrasonic sensor array, an ultrasonic pulse generation receiver and a data acquisition system / PC terminal, receiving a pulse signal generated by the ultrasonic pulse generation receiver through the flexible wearable ultrasonic sensor, converting the pulse signal into an ultrasonic signal, transmitting the ultrasonic signal to a detected object, and feeding back an ultrasonic echo signal to the ultrasonic pulse generation receiver; and the ultrasonic pulse generation receiver sends the ultrasonic echo signal to the data acquisition system / PC terminal for analysis and processing to obtain the corresponding muscle elasticity modulus. The flexible wearable ultrasonic sensor has certain flexibility, and the sensors are connected through a snakelike circuit, so that the stretchable and bending performance of the flexible wearable ultrasonic sensor array is ensured, the sensor array is more closely connected with the human body, and the wearable purpose is achieved.

The invention relates to the technical field of machine learning, in particular to a keyboard key pressure identification method and device based on deep learning. The method comprises the following steps: acquiring initial pressing stroke data of a single key of a keyboard; performing pressure measurement on a single key corresponding to the initial pressing stroke data by using a pressure sensor to obtain keyboard key pressure measurement data; keyboard pressing effective stroke screening is carried out on the initial pressing stroke data of the single key of the keyboard through the keyboard key pressure measurement data, and keyboard pressing effective stroke data are obtained; obtaining keyboard key structure data; and performing three-dimensional pressing depth identification on the keyboard key pressure measurement data according to the keyboard pressing effective stroke data to generate keyboard three-dimensional pressing pressure identification data. According to the method, through a multi-step technology of accurate pressing stroke screening, three-dimensional pressure distribution identification, mistaken touch correction, contact optimization and the like, the accuracy and error interference of keyboard key pressure identification are improved.

The invention discloses an intelligent wave recording and traveling wave ranging integrated design method and system, relates to the technical field of power systems, and comprises heterogeneous data fusion processing including hardware isolation and independent acquisition, a clock synchronization mechanism, data caching and alignment, data format standardization, algorithm optimization and fusion, and data verification and correction. Large-capacity and multi-backup mixed storage is achieved, and combination of multiple storage media, data redundancy backup and improvement of data reliability are included; starting based on multivariate information collection, wherein the starting comprises analog quantity starting, switching value starting, GOOSE message starting and station control layer message starting; the comprehensive distance measurement algorithm comprises establishment of an electromagnetic-circuit framework model, construction of a comprehensive distance measurement model, solving of an objective function through an optimization algorithm, correction of a preliminary fault point coordinate estimation value, realization of a power fault accurate positioning system and support of computer equipment and a storage medium. The problem of single recorded data is solved, and the data processing speed and the flexibility of the system are improved.

The invention relates to a preparation method of a high-performance soybean protein emulsion film, which comprises the step of carrying out fibrosis denaturation on soybean protein by adopting an acetic acid-water binary solution, so that the hydrophobic property, the film-forming ability and the environmental interference resistance of the soybean protein can be remarkably improved. And further starting from a protein self-assembly fibrosis principle and a protein film forming theory, a soybean protein-based biological membrane with certain mechanical strength and flexibility is constructed through self-assembly fibrosis denaturation of protein molecules in an aqueous solution, so that the problems of poor mechanical property and high solubility of the existing protein-based biological membrane are effectively solved.

The utility model relates to the field of air conditioner air outlet devices, in particular to a central air conditioner air outlet noise reduction device which comprises an air receiving barrel, a bucket-shaped air expanding cover, an air outlet cover, a noise reduction frame, an air direction guiding frame and a protection frame. A hopper-shaped air expanding cover is arranged at the lower end of the air receiving barrel, an air outlet cover is arranged at the lower end of the hopper-shaped air expanding cover, an embedded groove is formed in the end, away from the hopper-shaped air expanding cover, of the air outlet cover, and a noise reduction frame, an air direction guiding frame and a protection frame are sequentially arranged in the embedded groove. According to the air conditioner air outlet device, noise generated by airflow resonance is effectively reduced through the air expansion flow guide design of the hopper-shaped air expansion cover and the sound absorption cotton plate filling of the sound attenuation and noise reduction frame, airflow can be evenly and softly blown to a room through the design of the air direction guide frame, and the problems that airflow in a traditional air conditioner air outlet device is too strong and uneven in distribution are solved.

The invention discloses a sacrificial anode state monitoring and service life prediction method for ship cathode protection, belongs to the technical field of sacrificial anode state monitoring and service life prediction, and solves the problems that the anode protection current generated by a sacrificial anode can be reduced to a pA level at the initial stage of corrosion, and the service life of the sacrificial anode cannot be predicted under the condition that strong interference exists in an actual environment. The performance of an existing sacrificial anode detection technology cannot meet the high-precision detection requirement, so that the requirement for predicting the service life of a sacrificial anode cannot be met, and the requirement for judging the early corrosion trend cannot be met. The method comprises the steps that a plurality of node monitoring detection boxes are used for detecting the anode potential of a sacrificial anode connected with a cathode, and protection current between the cathode and the sacrificial anode is detected in a multi-range mode and stored; and the monitoring substation is used for monitoring the state of the sacrificial anode and predicting the service life of the sacrificial anode based on the anode potential and the protection current detected by the plurality of node monitoring detection boxes. The method is used for sacrificial anode state monitoring and life prediction.

The invention discloses a satellite cognitive radio low-interference communication method based on multi-objective optimization, and relates to the technical field of wireless communication, and the method comprises the steps: 1, collecting the coordinate position of a satellite, the coordinate position of a primary user, the coordinate position of a secondary user, and the number and initial positions of unmanned aerial vehicles used for executing beam forming; 2, establishing a target function; 3, solving an optimal solution for the target function according to an improved multi-target dragonfly optimization algorithm, and obtaining an optimal satellite transmission power, an optimal unmanned aerial vehicle antenna excitation current weight, an optimal position of the unmanned aerial vehicle and a secondary user sequence number; and 4, all the unmanned aerial vehicles fly to the optimal positions and then establish communication connection with the satellite, the satellite sends signals to the unmanned aerial vehicles according to the optimal transmission power and normally communicates with the primary user, and the antenna excitation current weight of the unmanned aerial vehicles is set to be the optimal weight and then the received satellite signals are forwarded to the secondary user. The method has the characteristic of improving the communication efficiency.

The invention provides an acousto-optic combined detection method for multi-angle inclined surface damage of an underwater facility, and belongs to the technical field of tests and tests. The method has the advantages that the problems that the influence of water quality on multi-angle inclined surface damage of underwater facilities is large when an optical means is adopted, the detection efficiency is low, and the detection resolution is low when a traditional image sonar means is adopted are solved. An acousto-optic combined detection module composed of a side-scan sonar, a ranging sonar and an underwater optical camera is carried in the detection direction of an underwater carrier, and a rotation steering engine in the pitching direction controls the overturning angle of the detection module, so that the detection direction of acousto-optic detection equipment is perpendicular to surfaces to be detected at different inclination angles. And the rotation of the side-scan sonar array is controlled by a rotary steering engine in the rolling direction relative to the pitching angle, so that the installation direction of the array is consistent with the maneuvering direction of the underwater carrier, and the maneuvering state of the carrier is controlled according to the data of the distance measurement sonar to keep the consistent distance from the detection mechanism to the surface of the facility, so as to obtain a high-resolution acoustic image. A suspected area where the damage is located is judged through an acoustic image, and after the carrier approaches, an optical camera and an underwater light source are adopted for sequentially photographing and judging suspected positions, so that the surface damage condition of the underwater facility is determined.

The invention discloses a copper-silver alloy, a copper-silver alloy plated optical fiber, an optical fiber panel and a preparation method and application of the copper-silver alloy plated optical fiber. The sum of the content of copper and the content of silver in the copper-silver alloy is larger than or equal to 99.995 wt%, and the content of silver ranges from 92 wt% to 95 wt%. The technical problem to be solved is that optical fiber monofilaments plated with copper-silver alloy layers are used for preparing the optical fiber panel, so that the optical crosstalk resistance and the electromagnetic interference resistance of the prepared optical fiber panel are remarkably improved, and high-resolution image transmission of the optical fiber panel with optical crosstalk resistance and high-frequency electromagnetic pulse resistance is realized.

The utility model relates to a balance weight and air supply inflation integrated base which is suitable for an air costal membrane structure and an inflatable membrane structure such as an inflatable dome and an inflatable fortress. During use, a membrane body structure to be inflated is connected to the inner membrane base, air is supplied to the inner membrane base and the membrane body structure through the air inlet mechanism, and inflation in the membrane body structure is achieved. Filler is placed in the outer film base through the filling hole, and weight balancing is achieved. The inner space of the outer film base is divided into a plurality of partition spaces, so that on one hand, the filler is separated from the air inlet mechanism, and interference of the filler on the air inlet mechanism is avoided; and on the other hand, the filler is convenient to take and place, and uniform distribution of the filler is achieved. Therefore, the balance weight, inflation and air supply are integrated, operation is easy, the attractiveness is high, and it can be guaranteed that the inflatable membrane structure is effectively fixed.

The invention provides a multi-component thin film evaporation film thickness monitoring and adjusting device and method and application, the multi-component thin film evaporation film thickness monitoring and adjusting device comprises a conveying module used for conveying a substrate so as to deposit a film layer on the moving substrate through a first evaporation source and a second evaporation source, the conveying module is provided with a position sensing module, and the position sensing module is used for sensing the position of the film layer on the moving substrate through the first evaporation source and the second evaporation source; a trigger signal is generated when the substrate enters the set area; the first film thickness monitoring module is used for acquiring the deposition speed of the first evaporation source and the thickness of a film layer deposited on the substrate; the second film thickness monitoring module is used for acquiring a reflection spectrum of the second evaporation source in a set area; and the control module is used for obtaining the film thickness according to the reflection spectrum of the second evaporation source through an inversion algorithm, and independently feeding back and controlling the real-time evaporation working condition. The thickness of materials with different evaporation rates is subjected to differential parallel monitoring, a closed-loop feedback system is formed, the film thickness monitoring accuracy is improved, and long-time operation of monitoring work is guaranteed.

The LED controller casing with the side cover comprises a main body, the side cover and a line pressing assembly, the main body is provided with an installation groove with an upward opening, the side cover is detachably installed at the opening of the installation groove, the side wall of the installation groove is provided with a first line pressing sub-notch, the side wall of the side cover is provided with a second line pressing sub-notch, and the first line pressing sub-notch and the second line pressing sub-notch are detachably installed at the opening of the installation groove. The first sub-wire pressing gap and the second sub-wire pressing gap form a wire pressing gap; a wiring terminal is arranged in the mounting groove, and a wiring port of the wiring terminal faces the wire pressing gap; the main body is provided with a wire pressing assembly, and the side cover is also provided with a wire pressing assembly. According to the utility model, when a user pulls out the side cover upwards, the wiring terminal and the connecting wire can be quickly connected, and then the side cover is pressed downwards, so that the side cover is buckled at the opening of the mounting groove. At the moment, the wire pressing assembly can rapidly abut against the connecting wire, the assembling process of the connecting wire is completed, the assembling efficiency is improved, and the problem that a traditional LED controller shell is connected and fixed through screws, and assembling is difficult is solved.

The invention discloses a copper-silver-zinc alloy, a copper-silver-zinc alloy optical fiber, a light cone and a preparation method and application of the copper-silver-zinc alloy optical fiber and the light cone, the sum of the content of copper, the content of silver and the content of zinc in the copper-silver-zinc alloy is larger than or equal to 99.99 wt%, the content of copper ranges from 15 wt% to 50 wt%, the content of silver ranges from 10 wt% to 80 wt%, and the content of zinc ranges from 5 wt% to 40 wt%. The technical problem to be solved is that optical fiber monofilaments plated with copper-silver-zinc alloy layers are used for preparing the light cone, so that the optical crosstalk prevention and electromagnetic interference resistance of a prepared light panel are remarkably improved, and high-resolution image transmission of the light cone capable of preventing crosstalk and resisting high-frequency electromagnetic pulses is achieved.

The invention discloses a radiation field SLAM (Simultaneous Localization and Mapping) pose optimization method based on motion constraint, which belongs to the technical field of SLAM positioning calculation, is used for pose optimization, and comprises the following steps: acquiring an IMU (Inertial Measurement Unit) data frame, an RGB (Red, Green and Blue) image data frame and a depth image data frame, carrying out pre-integration to obtain a motion increment, recording a Jacobian matrix of the motion increment relative to zero offset, and carrying out SLAM positioning calculation. Recording a noise covariance matrix of the motion increment; initializing gravity and speed; residual errors of estimated values and measured values of the pose and the speed are calculated, the pose, the speed and the zero offset are optimized in combination with radiation field loss, and a Jacobian matrix is used for updating the motion increment. According to the method, the optimization function is provided to optimize interference of IMU zero offset, and compared with a radiation field SLAM method fusing IMU data, motion parameters and gravity parameters are initialized, so that the radiation field SLAM has higher robustness for IMU data quality, and a solution is provided for practical application of the radiation field SLAM.

The invention discloses an insulator for a direct current converter transformer and an online monitoring method thereof, and relates to the technical field of insulators. According to the insulator for the direct current converter transformer, the insulator body and the connection structure are internally integrated with a multi-stage state monitoring system comprising a first-stage sensing unit, a second-stage sensing unit and a third-stage sensing unit; according to the invention, three types of key state data of internal mechanical stress of the insulator, the connection state of the first metal accessory and the second metal accessory and the overall temperature distribution can be synchronously acquired in real time, real-time monitoring of the insulator and multi-dimensional comprehensive information acquisition are realized, early warning is realized based on the multi-dimensional comprehensive information, and the reliability of the insulator is improved. According to the method, preliminary early warning can be carried out according to single data, meanwhile, time-space alignment and comprehensive judgment can be carried out on multi-dimensional data, high-level fault early warning is carried out, the stability and reliability of the insulator are improved, and sudden faults are reduced.

The invention discloses a satellite communication interference confrontation method and system based on the game theory, and the method combines a motion model and a communication interference model of a satellite, solves a Nash equilibrium point, optimizes the operation power and electromagnetic interference confrontation strategies of attackers and defenders, and improves the performance of the satellite communication interference confrontation when constructing a performance index. An attacker performance index and a defender performance index of each satellite are respectively established, and the operation power of a physical layer and the channel capacity of a network layer are considered, so that an attacker can reduce the communication capacity of a communication party to the greatest extent while keeping low-power operation, and the purpose of efficient interference is achieved; the channel capacity is related to the relative distance between the satellites, so that the interference satellites can utilize interference signals with lower power to achieve the attack purpose through control, the energy consumption of the satellites can be reduced, and the adaptability can be improved; and the communication party can increase the own communication capacity to the greatest extent while keeping low-power operation, so that the anti-interference purpose is achieved.

The utility model provides a wire harness with a flame-retardant structure, which relates to the field of wire harnesses, and comprises an inner core assembly, the main body of the inner core assembly is a cable, and the outer side of the inner core assembly is wrapped with an inner insulating layer. In order to solve the problem that a corresponding flame-retardant structure needs to be arranged due to the fact that the cable is fractured when being worn and is easily subjected to detonation and other potential safety hazards when being connected with an external conductor at a high temperature, multiple insulation and shielding structures, namely an inner insulation layer, an outer insulation layer and a shielding layer, are arranged, so that the insulation performance and the anti-electromagnetic interference capability of the wire harness are greatly improved, and the service life of the wire harness is prolonged. The electric leakage risk caused by poor insulation is effectively reduced, the potential safety hazard of detonation caused by connection between high temperature and an external conductor is avoided, the safety performance of a traditional wire harness is innovatively improved, and the design of the spiral tensile assembly enhances the tensile capacity of the wire harness.

The invention provides a NORD flash memory, a layout and a writing method and a reading method thereof, a common source line is arranged between adjacent word lines in the layout of the NORD flash memory, a first control gate and a second control gate which are parallel to each other are arranged on two sides of each word line, and only one of two bit cells sharing the word line is used, so that the layout of the NORD flash memory is changed, and the layout of the NORD flash memory is changed into the layout of the NORD flash memory. Although only half of the bit units are utilized, the area of the active region can be reduced by half by applying voltage between the source line and the bit line, so that the area of the device is basically unchanged; due to the fact that the pressurization mode of writing, erasing and reading data of the NORD flash memory is changed and only one of the two bit units sharing the word line is used, writing interference and reading interference of the NORD flash memory are improved, and the reliability of the NORD flash memory is greatly improved.

The application discloses a polyethylene pipe surface micro-defect real-time detection system based on machine vision, and relates to the technical field of monitoring and analysis, which comprises the following steps: constructing a dynamic deformation matrix by associating timestamp, three-dimensional height difference data and temperature drop gradient data; calculating the relative deformation rate corresponding to the unit temperature drop gradient data according to the dynamic deformation matrix; generating a bubble determination result when the three-dimensional height difference data represents inward collapse displacement, the relative deformation rate nonlinearly decreases with the increase of the temperature drop gradient data, and the difference value of the last section is lower than a preset stop threshold; generating an impurity determination result when the three-dimensional height difference data represents outward protrusion displacement, and the fluctuation variance of the relative deformation rate falls within a preset rigidity interval; establishing a corresponding relationship between the bubble determination result, the impurity determination result and the initial coordinates of the suspected target, and outputting detection result data. The application has the effect of improving the real-time detection efficiency of polyethylene pipe surface micro-defects.

The utility model provides a kind of multifunctional flip type mobile phone shielding box, including box and soundproofing module, box includes outer storehouse body and the flip of rotation connection in outer storehouse body, soundproofing module is composed of outer soundproofing layer and inner soundproofing layer, and inner soundproofing layer is equipped with the mobile phone accommodating cavity of accommodating, mobile phone accommodating cavity bottom is equipped with the sound interference collection recording interference module of interference sound and wireless charging module;Outer storehouse body inside is equipped with storage chamber, and soundproofing module is arranged in storage chamber, the inner wall of the chamber box is equipped with shielding component, when the flip is covered on outer storehouse body, the shielding component and soundproofing module hinder mobile phone accommodating cavity and the sound and signal transmission of outside. The utility model reduces the sound pickup effect of mobile phone to outside sound by recording interference, electromagnetic shielding and independent soundproofing module cooperation, improves the recording interference and electromagnetic shielding effect of shielding box.

The invention relates to the technical field of oil and gas fracturing flow-back fluid treatment, and particularly discloses a boron-doped diamond electrode electrooxidation system and method for treating fracturing flow-back fluid. According to the method for treating the fracturing flow-back fluid, before the fracturing flow-back fluid enters an electrooxidation unit, sulfide is efficiently removed, poison and competitive oxidation of the fracturing flow-back fluid to an electrode are eliminated, a BDD electrode is protected, a foaming agent is synchronously removed through aeration enrichment, the problem of foam interference in the flow-back fluid treatment process is fundamentally solved, and continuous and stable operation of a system is guaranteed; after the electro-oxidation unit, active chlorine (pypocholoride / chlorine) is effectively removed and recycled, so that waste is treated by waste, the oxidation efficiency is improved, and the dosage of chemicals is reduced. Through the collaborative design, the interference and foam problems of sulfide with high chloride ion content and strong reducibility are collaboratively optimized, and the treatment efficiency, economical efficiency and stability of the system are integrally improved.

The present application relates to the technical field of machine learning, and more particularly to a classroom student behavior detection optimization method based on deep learning, which extracts features from each frame of image of student video in the classroom, obtains corresponding high-dimensional semantic feature vectors, reduces the dimension of the high-dimensional semantic feature vectors of each frame of image, maps the three-dimensional feature vectors of each frame of image to a three-dimensional clustering space, obtains corresponding data points, filters the data points in the three-dimensional clustering space to obtain a target clustering space, and according to the multi-frame continuity similar features and the similar light environment features of any two data points, the distance measurement of any two data points in the target clustering space is enhanced and clustered, and a plurality of image categories are obtained, the deep learning model is trained using the plurality of image categories, and a trained deep learning model is obtained, which is used for classroom student behavior detection, and improves the interference ability and detection recognition accuracy of the deep learning model.

The utility model relates to fluid control technical field especially relates to a flow path control valve and fluid control device, including the support plate, the flow guide plate, the frame board and the bottom plate that form the overall structure in turn layering together along the thickness direction, the frame board has the opening part, still include the valve plate, the valve plate has the deformation part and the junction part that connect, the valve plate is fixed in the bottom plate through the junction part, and the outer peripheral edge of deformation part exists a part as the free area, and the deformation part can produce the elastic deformation under the pressure effect, the utility model discloses flow path control valve is advantageous to further promote the thin degree of the air leakage valve on one hand, the other hand makes the overflow passage formed after the deformation of deformation part not depending on the hole structure of deformation part can realize the communication of inlet hole and flow guide hole, and therefore deformation part can realize maintaining the overall structure, can better inhibit the risk that deformation part appears the deformation even the cracking because of setting the hole structure, ensure the stability and reliability of control valve work, improve the service life.

The present invention relates to the technical field of three-dimensional point cloud reconstruction, and specifically to a method for reconstructing three-dimensional point clouds of highly reflective surface objects, including obtaining left and right multi-step phase shift fringes of left and right cameras, generating left and right unfolded phase maps based on the phase shift fringes, and positioning left and right camera ROI areas based on the left and right unfolded phase maps; calibrating the screen extrinsic parameters and camera extrinsic parameters of the binocular camera system, and unifying the coordinates of the left and right cameras; based on the positioned left and right camera ROI areas and the coordinates of the left and right cameras after the unified coordinates, performing epipolar correction on the object images with reference point features obtained by the left and right cameras to obtain initial 3D coordinates; based on the initial 3D coordinates, obtaining the corresponding surface normal vectors of each pixel point on the left and right camera views; calculating each normal vector gradient based on the surface normal vector, calculating the height by integration, and obtaining the surface 3D point cloud coordinates. While improving the calculation accuracy, the present invention maintains reasonable calculation efficiency and meets actual use requirements.