32results about How to "Intelligent detection" patented technology

The invention discloses a gesture identification method based on a deep residual error network. The gesture identification method based on a deep residual error network includes the steps: 1) acquiring the original data information of gesture, and marking N points to obtain the 2N-dimensional original tag data; 2) preprocessing the original data information and the 2N-dimensional tag data; 3) taking the preprocessed original data information and the 2N-dimensional tag data converted in the hdf5 format as the original training data, inputting the original training data to a deep residual error network to train network parameters, and obtaining a gesture identification model; 4) performing N-point marking being identical to the step 1 on the gesture to be identified to obtain the 2N-dimensional tag data to be identified; and 5) preprocessing the gesture data to be identified and the 2N-dimensional tag data to be identified, inputting the processed gesture data and the 2N-dimensional tag data to be identified converted in the hdf5 format to the gesture identification model for identification, and then obtaining an identification result. The gesture identification method based on a deep residual error network effectively solves the gradient dispersion problem and the network accuracy problem.

The invention discloses a method and a system for directionally enhancing sound. The method comprises the following steps: detecting a sound source which initiates a control instruction; positioning the sound source; adjusting the signal channel delay of each loudspeaker in the array loudspeaker according to the relative position relationship between the sound source and the array loudspeaker, sothat the sound wave of each loudspeaker reaching the sound source generates in-phase wavefront interference. The system comprises a directional enhanced sound control device and a terminal, wherein the directional enhanced sound control device is connected with the terminal. The problem of unclear acquired sound caused by non-centralized propagation of sound energy is solved, and centralized propagation of the sound energy in a specific direction is realized so as to meet the requirement of improving the sound sharpness in different scenes.

The invention belongs to an inspection device, in particular to a comprehensive inspection system for motor vehicles in workhouses or work yards. The invention has the technical proposal that a comprehensive inspection system for motor vehicles in work yards and the structural key points thereof are that the invention consists of a control switch, a power supply, an auxillary switch, a plurality of data collectors and a data receiving and processing device; after inspecting and collecting data from a to-be-inspected vehicle, data collectors send the relative information to the data receiving and processing device for storage and processing. The invention has the advantages that, with strong expandable space, excellent practicality, convenience, safety, durability, adaptability, development; the invention can realize the intelligent inspection, improve the inspection accuracy and efficiency, and can guarantee the fairness of the test results. In addition, the invention also can complete inspections for all types of vehicles listed in the catalog of motor vehicles in workhouses or work yards, which is made by AQSIQ, such as forklifts, movers, stacking vehicles, pilling cars, loaders, grabs and so on.

The invention discloses an aircraft bearing surface rivet detection method based on a U-net network, particularly relates to the field of machine vision detection, and aims to solve the problems thatrivet assembly detection of an aircraft bearing in the prior art can only be achieved through manual detection, and the detection efficiency is low. The method comprises the following steps: collecting an image of a rivet with standard assembly as a standard template of a rivet image, and collecting an image of a rivet on the surface of an aircraft bearing as a training sample; constructing a U-net network; inputting a training sample into the U-net network for training to obtain a trained U-net model; inputting a to-be-detected image, and carrying out the feature recognition of the to-be-detected image through the trained U-net network; performing circle identification on the segmentation result by using Hough transform, and outputting the radius of the identified circle; performing rivetinstallation judgment; according to the invention, manpower and material resources are saved, the detection speed is improved, and the detection process is more automatic and intelligent.

The invention relates to the technical field of environment monitoring, specifically to a system and method for detecting air cleanliness of a living environment. The system comprises an air mixing element, a sensor element, and a touch-screen display, wherein the sensor element comprises a sensor fixing block and a sensor module. The sensor module includes temperature and humidity sensors distributed at intervals for detecting the temperature and humidity of room air, an odour sensor for detecting odour of room air, a PM2.5 sensor for detecting fine particulate matters in air, a formaldehyde sensor for detecting the content of formaldehyde, and an oxygen sensor for detecting the concentration of oxygen. According to the technical scheme of the invention, the system for detecting air cleanliness of the living environment is simple to operate and convenient to use, and comprehensive detection can be performed.

The invention relates to the technical field of image and video processing and particularly relates to a detection method for vulgar adult content of an Internet video image. The detection method comprises the following steps: reading color components R, G and B of each pixel in the Internet video image; carrying out white balance processing on the color components R, G and B of each pixel by utilizing data of changes of white blocks due to temperatures in the Internet video image, and adjusting the Internet video image into a standard-color colorful image; learning a background sample in the colorful image through a multi-level and multi-characteristic image deep learning algorithm so as to obtain a foreground image in the colorful image; and comparing an anchor image in the foreground image through an offline-acquired anchor skin nakedness receiving operation curve ROC, and determining whether the content is the vulgar adult content or not. The detection method provided by the invention has a relatively rapid detection and filtering speed, and has relatively good filtering precision and relatively low system expenditures; and the working amount of manual judgment can be extremely reduced so that the detection of the vulgar adult content of the Internet video image is efficient and intelligent.

The invention discloses an automatic temperature measurement and early warning device for power equipment in a substation, which comprises a temperature acquisition module, a platform server, a mobile terminal and an early warning robot, wherein the early warning robot comprises a wheeled robot body, a main controller, a navigation system, a video navigation probe, a fire extinguishing device and a power device; the fire extinguishing device comprises a powder fire extinguisher; the top end of the powder fire extinguisher is provided with a cover plate and a motor; the cover plate is provided with an electromagnetic valve; one end, far away from the motor, of a transmission shaft is provided with a supporting frame; a fire extinguishing nozzle is arranged on the supporting frame; the powder fire extinguisher is communicated with the fire extinguishing nozzle via a pipeline connected with the electromagnetic valve; and the supporting frame is also provided with an adjustment device for adjusting the height of the fire extinguishing nozzle. A turntable is also included, and the turntable is provided with an opening matched with the pipeline. The device of the invention has the beneficial effects that the accurate position of the temperature sensor can be quickly judged, and the fire safety conditions of the power equipment in the substation can be effectively improved.

The invention discloses a passive night vision intelligent mine detection system and an intelligent mine detection method. According to the system, a prime lens is connected with a low-light-level camera through an F bayonet; the low-light-level camera is placed on a support frame; an FPGA development board is connected with the low-light-level camera through a data transmission line and is connected with a display screen through a data line; a lead storage battery is sequentially connected with the low-light-level camera, the FPGA development board and the display screen through data lines; and all the parts are placed on an instrument loading vehicle. The method comprises the following steps of: constructing a passive night vision intelligent mine detection network model by using a YOLOv2 algorithm, constructing a data set of scattered landmines, training, evaluating and optimizing the passive night vision intelligent mine detection network, and embedding the trained passive night vision intelligent mine detection network model into the FPGA development board; constructing a passive night vision intelligent mine detection and distance measurement model by applying a geometrical optics imaging principle, and embedding the passive night vision intelligent mine detection and distance measurement model into the FPGA development board. According to the invention, intelligent and rapid detection of thrown landmines at night is realized.

The invention relates to the field of biomedical engineering, and discloses a micro-needle pressing type electrochemical sensor based on a mobile phone platform. The electrochemical sensor comprises apressing micro-needle sensor device with a reverse iontophoresis function and a three-electrode electrochemical detection function, and a hardware interface circuit for carrying out Bluetooth data exchange with a mobile phone, wherein the hardware interface circuit is connected with a circuit of the sensor device. According to the invention, a micro-needle punctures a skin to extract biomarkers by reverse electroosmosis, so that detection sensitivity is improved; and loss of bioidentification molecules can be prevented, so that a maximum utilization rate of the bioidentification molecules inin-vivo detection can be realized, and meanwhile, an intelligent detection process can be realized. The invention also provides a detection system of the micro-needle pressing type electrochemical sensor based on the mobile phone platform.

The invention discloses an online real-time voltage balance management system and method of a storage battery pack. The method comprises: through a voltage detection module, obtaining voltage signals of each storage battery; when original measurement voltage signals exceed a preset threshold or deviations among the original measurement signals exceed the preset threshold, triggering a selective charging function; the selective charging function selecting one or more storage batteries needing to be charged in case of voltage imbalances, and performing constant voltage charging; after time is set, through the voltage detection module, obtaining the voltage signals of each storage battery again; and when neither the original measurement voltage signals exceed the preset threshold nor the deviations among the original measurement signals exceed the preset threshold, determining that the voltage of the whole storage battery pack is equalized, such that online real-time voltage balance management of the storage battery pack is realized. The system and method provided by the invention can improve system operation safety and economic benefits, and reduce safety hidden risks of uncertainties caused by decreased storage battery performance.

The invention provides an air conditioning unit, a detection method and a detection device of the air conditioning unit and a web server. The detection method of the air conditioning unit comprises the following steps that the number of the air conditioning units is acquired in the detection area; the outdoor environment temperature detected by each air conditioning unit after the corresponding air conditioning unit operates for the preset time in a preset mode is acquired; and whether the heat dissipation of an outdoor unit of each air conditioning unit is abnormal or not is judged accordingto the number of the air conditioning units in the detection area and all outdoor environment temperatures. According to the detection method of the air conditioning unit, whether the heat dissipationof the outdoor unit of each air conditioning unit is abnormal or not is judged according to the number of the air conditioning units in the detection area and the outdoor environment temperature which is detected after each air conditioning unit operates for the preset time in the preset mode, so that the uncertainty of manually judging the heat dissipation effects of the outdoor unit of the airconditioning units is avoided, then the air conditioning units can be conveniently installed at the proper positions, then the heat dissipation effects of the outdoor units of the air conditioning units are guaranteed, and the labor cost is reduced.

The invention provides a fast detector of Clenbuterol hydrochloride residues which belongs to fast detecting instrument technology. The installation of all the components of the fast detector takes an operation table as the basis. A communication socket is arranged at the center of the front upper part of the operation table. A detection system is inserted in the communication socket. Guide seats are arranged at the two ends of a diagonal of the detection system. Driving components are arranged at the front center of the operation table, directly below the communication socket and the detection system. A control system is arranged at the back center of the operation table, at the due back of the driving parts and is connected with the driving parts. A detection circuit is arranged at the middle-lower back part of the operation table. After all components are arranged on the operation table, the operation table is fixed at the outside opening of a shell with an outer cover covering above the circumference of the operation table, thereby forming a complete enclosed device with the shell. The invention provides a fast detector of Clenbuterol hydrochloride residues which can be used in laboratory situation and non-laboratory situation and has the advantages of small size, field and simple operations and high detection sensitivity of a 1ppb detection limit.

A tire inner side defect detector based on machine vision and a deep learning algorithm relates to the technical field of tire quality detection and comprises a handle, a machine body, an automatic focusing camera, a searchlight, an embedded alarm single chip microcomputer, an embedded deep learning algorithm development board and a display screen. The embedded deep learning algorithm development board is provided with an image preprocessing module, an improved algorithm module based on classification optimization of a convolutional neural network and a support vector machine, and a defect identification and positioning module. According to the invention, automatically acquired tire inner side image information can be automatically interpreted to realize efficient and accurate identification of tire inner side defects, position information is automatically extracted, defect boundaries are drawn, defect area sizes are calculated, and alarm sounds are sent out, so that the detection and identification work of the tire inner side defects is simplified, automatic, intelligent and efficient, and the time and labor costs of tire inner side defect detection are reduced.

The invention discloses high-speed precision subway tunnel concrete lining crack detection equipment. The equipment comprises an image acquisition device, a shock-proof transportation device, a positioning device and a power supply control device. In this way, according to the high-speed precision subway tunnel concrete lining crack detection equipment, the image acquisition device is mounted on the shock-proof transportation device, a camera is mounted on a telescopic vertical frame, the camera performs image acquisition work while advancing along the tunnel direction, then an image signal acquired by the camera is converted into a digital signal which can be recognized and processed by an industrial personal computer through an image acquisition card, and finally, the digital image is processed to obtain a crack damage grade, safe operation of the tunnel is guaranteed. The equipment has the advantages of firm and stable overall structure, rapidity, high in precision, reliable detection and comprehensive and intelligent detection, and meanwhile has wide market prospects in application and popularization of tunnel disease detection.

The invention provides an oil-water separation device, which includes a filter residue pool, a separation box, partition plates, and oil guide pipes. A feed pipe is inserted into the filter residue pool, the separation box is internally provided with three uniformly spaced partition plates, the partition plates divide the separation box into four independent sedimentation tanks, the front side wall of each sedimentation tank is connected to an oil guide pipe, the tails of the plurality of oil guide pipes are welded to an oil guide main pipe and the plurality of oil guide pipes are communicated with the oil guide main pipe. Each partition plate is provided with a cut-through transverse hollow pipe, the left end of each transverse hollow pipe is welded with a vertical hollow pipe communicated with the same, the bottom of a front side wall on a rightmost sedimentation tank is equipped with a through-hole, and a discharge pipe is fixed on the through-hole. The oil-water separation device provided by the invention carries out repeated precipitation separation, and achieves thorough separation, the hollow pipes disposed on the partition plates realize flow and separation simultaneously, the oil-water separation device has fast treatment speed, realizes intelligent control and detection, and is in line with industrial production, and the arranged funnel and discharge pipe valve are convenient to observe whether complete separation is achieved, if the separation is incomplete, the valve can be closed for further separation.

The embodiment of the invention provides an industrial machine fault detection method and device based on machine vision, and the method comprises the steps: carrying out the image processing on an original image collected by a camera, and obtaining an image of a to-be-detected industrial machine; extracting features of each contour in the image of the industrial machine to be detected; and inputting the features of each contour in the image of the to-be-detected industrial machine into the fault detection model, and outputting a fault detection result corresponding to the to-be-detected industrial machine, wherein the fault detection model is obtained by training sample data of contour features of the mechanical structure image and a predetermined fault type label. According to the industrial machinery fault detection method and device based on machine vision provided by the embodiment of the invention, the fault detection result corresponding to the industrial machinery to be detected is obtained according to the feature of each contour in the image of the industrial machinery to be detected, a more accurate fault detection result can be obtained, and the fault diagnosis efficiency of industrial equipment can be effectively improved.

The invention relates to intelligent sports equipment capable of exercising abdominal muscle and measuring weight. The intelligent sports equipment comprises a reinforcing base, a weighing platform, afoot limiting support, a measuring display and a position detector, wherein the upper surface of the reinforcing base is connected with the weighing platform through a plurality of fixed supporting rods; a limiting hole is formed in one side surface of the weighing platform; a soft mat is laid on one side surface of the weighing platform; the limiting hole is fixedly connected with the foot limiting support; one side surface of the weighing platform is fixedly connected with the measuring display through a support frame; and the surface of the weighing platform on one side of the measuring display is fixedly connected with the position detector.

The invention belongs to the field of disaster detection, and discloses a Multi-life-body intelligent detection device and method based on a CPPWM-MIMO radar. The device comprises a CPPWM signal generator, a plurality of power amplifiers, an array antenna, a plurality of low-noise amplifiers, a multi-channel high-speed hybrid sampler, and a data processing system, wherein the array antenna comprises a plurality of transmitting antennas and a plurality of receiving antennas, the CPPWM signal generator simultaneously generates multiple paths of orthogonal CPPWM detection signals and multiple paths of CPPWM reference signals corresponding to the CPPWM detection signals, each CPPWM detection signal is connected with a transmitting antenna through the power amplifier, the output end of each receiving antenna is connected with the multi-channel high-speed hybrid sampler through the low-noise amplifier, each CPPWM reference signal is connected with the multi-channel high-speed hybrid sampler, the multi-channel high-speed mixed sampler is connected with the data processing module, and the data processing module is used for calculation and analysis to obtain a life body detection result. The device has the advantages of being not affected by environment temperature, noise, field visibility and external electromagnetic interference and the like.

The invention provides a detector conveying system with a state sensing function, which comprises a pipeline conveying device, a ball receiving and sending cylinder, a ball receiving and sending grabbing device and a control device, and is characterized in that the ball receiving and sending grabbing device is arranged above the pipeline conveying device; the ball receiving and sending grabbing device is connected with the ball receiving and sending cylinder; the control device is connected with the pipeline conveying device, the ball receiving and sending cylinder and the ball receiving and sending grabbing device. The conveying system can be used for an unpowered pipeline detector or a powered pipeline detector, and detection blind areas are avoided. A plurality of sensors such as a displacement sensor, a tilt angle sensor, an infrared transmitter, an infrared receiver, an infrared distance measuring sensor and a distance sensor are adopted, the state sensing function is achieved for the working state of each mechanism, automatic and intelligent pipeline conveying and detection can be effectively achieved, and the labor cost is reduced.

The invention discloses a multi-sensor fused real-time monitoring device for the bending forming state of a pipe fitting. Comprising an in-pipe fixing device, a fast disassembly and assembly universal joint and bending state detection modules, the first bending state detection module is connected with the in-pipe fixing device through the fast disassembly and assembly universal joint, and the other multiple bending state detection modules are connected end to end through the fast disassembly and assembly universal joint to form an open chain structure and are limited and fixed in a pipe through the in-pipe fixing device. A linear displacement sensor, a metal thickness sensor and a gyroscope sensor of the bending state detection module are used for monitoring the section shape change, the wall thickness change and the springback information of the pipe fitting correspondingly, and the rapid disassembly and assembly universal joint can select the number of the bending state detection modules according to the length of the bending section of the pipe fitting to achieve rapid disassembly and assembly. The device is used for monitoring changes of pipe wall shapes, wall thicknesses and axis information of all positions of a pipe fitting bending section in the pipe fitting bending process in real time, pipe fitting bending dynamic information can be obtained, the bending test process is shortened, and the production efficiency is improved.