34results about How to "Guaranteed real-time detection" patented technology

The invention discloses a living fingerprint verifying method based on a thin-plate spline deformable model, which collects and preprocesses the fingerprint sequence of a finger for verifying, registers the fingerprints before and after deformation, selects corresponding minutiae point pair to serve as the criterion point pair for estimating the thin-plate spline model, employs improved thin-plate spline model to describe deformation of fingerprints, computes the deformable energy of the fingerprint for verifying in each direction on the basis of model parameter to an overall bending energy vector, trains respectively on a true and a false fingerprint deformation base, obtains space distribution of the bending energy vector, respectively computes similarity from the bending energy vector for measuring to an eigen fuzzy set of true and false bending energies, and selects proper threshold value to obtain the verifying result. The invention can identify false fingerprint from the living fingerprint, effectively realizes living fingerprint verification in real-time, increases safety of automatic fingerprint verifying systems, enlarges the application range of automatic fingerprint verifying systems and has important application value in the field of biological character identification.

A method based on annular laser vision sensing for automatically locating the welded seam includes such steps as detecting annular laser tracings by CCD planar array, imaging on image sensing surface of camera, obtaining the expression equations for the X and Y coordinates of the characteristic points of said welded seam in conic coordinate system of laser, finding out the depth value of said characteristic point according to circular radius-depth relation algorithm, creating the equation for the Z coordinate of the characteristic point to be detected, finding it out, and converting the found out 3D coordinates from said conic coordinate system to the coordinate system of robot itself. Its advantage is high efficiency.

The invention provides a 3D target detection method for point cloud screening based on image semantic features. The method comprises the following steps: segmenting image data by a 2D semantic segmentation method to obtain semantic prediction; projecting the generated semantic prediction to an LIDAR point cloud space through a known projection matrix, thereby enabling each point in the point cloudto obtain the semantic category attribute of the corresponding image position; extracting related points of vehicles, pedestrians and riders from the original point cloud and forming a view cone; taking the view cone as the input of a depth 3D target detector, and designing a loss function conforming to the characteristics of the view cone to perform network training. According to the invention,a 3D target detection algorithm for point cloud screening based on image semantic features is designed, so that the 3D detection time is greatly shortened and the calculation requirements are greatlyreduced. Finally, the performance of the method on a reference data set KITTI of 3D target detection shows that the method has good real-time target detection performance.

The invention relates to a liquid near-infrared transmitted spectrum component online detection apparatus. According to the apparatus, near-infrared light emitted by a near-infrared light source can sequentially penetrates through a first condensation member, a first window sheet, a liquid to be detected, a second window sheet and a second condensation member, the transmitted light penetrating through the second condensation member is conveyed into a detector module, and the detector module analyzes to generate the component information so as to ensure the achievement of the product detection function; and on the basis, a pressure-bearing circulating pool module is set based on the pressure container standard so as to provide the pressure bearing capacity. According to the present invention, the limitations that the detection process of the existing detection apparatus can be only performed in the laboratory and the separate detection pipeline system is required to be arranged when the components of the high pressure liquid to be detected is analyzed are overcome, such that the product use is universal and convenient.

The invention provides a phase changing heat storing peak valley heating system and method. The phase changing heat storing peak valley heating system is provided with a phase changing heat storing device, an electric heating pipe, a circulating water pump, a circulating water network pipeline and a heating control system; the phase changing heat storing device comprises a medium and low temperature phase changing material installed in a cabinet and a heat preserving shell; the electric heating pipe is controlled by the heating control system to be started at the time of valley electricity atnight, and heat is stored for the phase changing heat storing device; according to the value of heat in the phase changing heat storing device, the power output of the electric heating pipe is dynamically controlled, and the electric heating pipe stops at the moment of peak electricity; and through the circulating water network pipeline, heat is released from the interior of the phase changing heat storing device to heating of a user port, and the problems of non-concentrated type heating for rural areas in the north in winter, phase changing heat storing electric power peak-load shifting, andintelligent and accurate control over the electric heating power are solved.

The invention discloses a fast and accurate single-stage target detection method and a device thereof, which discard the improved method of complex bottom layer and high layer network structure, and enhance the coordination and unity between the multi-layer regression characteristic diagrams of the high layer network only by optimizing the main stream structure and adding a lightweight split structure (i.e. Shunt structure). The invention uses the high-level network to carry out the maximum pooling and asynchronous convolution operation on the characteristics of the output of the bottom network, which is beneficial to the transmission of the spatial related information while reducing the dimension of the characteristic map, can improve the problem of the missing of the spatial related information and promote the diversity and the difference of the characteristics. After the shunting structure is added and the mainstream structure is optimized, the average accuracy of target detection is 80.5%, is 3.3% higher than SSD and 1.9% higher than DSSD321. At the same time, the average processing speed of 30fps is achieved on a 1080ti graphics card.

The invention relates to a potato disease detection method based on image recognition, which comprises the following steps that: step 1, a client acquires pictures of potato leaves, rhizomes and potato blocks to be recognized, and uploads the pictures to a server; step 2, the server preprocesses the picture, and then sends the preprocessed picture to a deep learning model; step 3, the deep learning model extracts high-dimensional features from the preprocessed image and then identifies and calibrates the types and positions of the potato scab contained in the image; And step 4, the identifiedscab position and the category thereof on the picture are superposed as marks to form a detection result picture for downloading and displaying by the client.

The invention relates to a corn kernel crush rate online detection method and device based on machine vision. The method comprises the following steps: obtaining corn kernels distributed in a single layer by a box body structure; collecting corn kernels images by an industrial CCD camera and carrying out sharpening, de-noising and edge detection to obtain the corn kernel edge contour feature; separating the crushed corn kernels and intact corn kernels by employing threshold segmentation and binarization processing; analyzing the images by a fuzzy clustering method to extract the crushed corn kernel characteristics; and matching the template with a standard crushed corn kernel template to obtain the quantity of the crushed corn kernels and calculating the crush rate, and displaying the crush rate on a vehicle-mounted terminal. The device comprises a detection box body unit, an image acquisition unit, an image processing unit, a data transmission unit, and a vehicle-mounted control and display unit. The units are matched one another to detect the corn kernel crush rate online in real time. The device is convenient to mount and maintain and is easily mounted in a locomotive grain tank, and meanwhile has the advantages of being high in detection speed, high in stability and relatively high in precision.

The invention relates to a fault detection and reconstruction system and method based on a dynamic-static hybrid strategy. Firstly, according to the current work mode of a control system, the fault detector and sensor configuration schemes are subjected to combined optimization; on the premise of ensuring the fault detection performance, the system energy consumption is possibly reduced; then, when the fault is detected, the real-time accurate recognition on the fault type is realized through solving the multi-model selection problem; finally, according to the recognized fault type, through the design of the reasonable switching control scheme, the control allocation and the system sterilization can be completed; the autonomous fault detection and reconstitution of the control system under various possible fault conditions can be realized. By using the technical scheme, the static fault monitoring and the dynamic fault type recognition and control system reconstruction are combined. Compared with a conventional fault detection and fault-tolerant control scheme, various kinds of potential faults can be effectively treated; the real-time performance is improved; the energy consumption is reduced; the system and the method can be applied to the high-reliability control problem of control objects of astrovehicles, mechano-electronics, chemical engineering and the like.

The invention relates to a network time delay collecting method based on distributed deployment, which aims at improving the instantaneity and the accuracy of network time delay collection. The accuracy of detection is ensured by selecting an appropriate detection algorithm and increasing the detection frequency, and the instantaneity of time delay data detection is ensured in a distributed deployment way. The method comprises the following steps of: selecting the distributed deployment way to increase the detection frequency; performing the time delay detection algorithm including a simple network and a redundant network; detecting at the frequency of multiple times every 5 min; and finally performing computation to increase the accuracy of detection. With the technology disclosed by the invention, while the instantaneity in detecting large amounts of chain time delays is guaranteed, the accuracy of time delay collection is increased through multiple detections in 5 min.

The invention discloses an on-line detection device for a shaft-sinking suspension steel wire rope, belonging to on-line detection and monitoring equipment. The on-line detection device of a shaft-sinking suspension steel wire rope comprises a plurality of force transducers, an A / D conversion circuit, an upper singlechip, a transmission circuit, a receiving circuit, a lower singlechip, a digital display circuit, a printing circuit, an alarming circuit and an executive circuit, wherein the force transducers are installed on a hanging scaffold or a suspension template platform; the A / D conversion circuit converts a tension value analog signal into a digital signal; the upper singlechip is connected with the A / D conversion circuit; the transmission circuit is connected with the upper singlechip; the receiving circuit is installed on the ground; the lower singlechip is connected with the receiving circuit via an RS232 interface; and the digital display circuit, the printing circuit, the alarming circuit and the executive circuit are connected with the lower singlechip. After the tension value of each steel wire rope is processed by the upper singlechip and sent into the transmission circuit via each force transducer, the transmission circuit sends the signal to the ground; the signal is sent to the lower singlechip via the receiving circuit to be correspondingly processed; and then, the signal is output by the digital display circuit, the printing circuit, the alarming circuit and the executive circuit. Thus, the system has the advantages of high antijamming capability, high accuracy and complete protective functions, is easy to operate, and can ensure that the stress of the suspension wire rope is detected in real time.

The invention provides a method and device for detecting CPU utilization ratio, which relates to an embedded system and solves the problem of inaccurate detection of the CPU utilization ratio. The method comprises the steps of: when a triggering condition is detected, determining a discrete time part and a sliding time part; respectively acquiring the CPU holding times of the discrete time part and the sliding time part; and computing to acquire the current CPU utilization ratio according to the CPU holding times of discrete time part and the sliding time part. The technical scheme provided by the invention is suitable for embedded systems.

An embodiment of the invention discloses a human body sitting posture supervision method. The human body sitting posture supervision method comprises an initial stage and a real-time supervision stage. In the initial stage, initial key point positions are determined through calculation of a plurality of images in a human body key point calculation network; and in the real-time supervision stage, the positions of key points in each frame of image are detected in real time on the basis of the initial key point positions, and whether the sitting posture of a human body meets the standard or not is judged according to the position relation between the key points and the relation between the same key points in a front frame and a rear frame. The human body sitting posture supervision method provided by the embodiment of the invention is simple and convenient to use, has a certain tracking function, and can provide a more accurate sitting posture judgment result.

The invention provides an optical waveguide. The optical waveguide comprises an optical waveguide sheet, a coupling-in optical grating, a turning optical grating, a coupling-out optical grating and detectors. The coupling-in optical grating is used for coupling light emitted by an external image source into the optical waveguide sheet; the turning optical grating is used for receiving the light coupled into the optical grating, and the turning optical grating is located on one side of the coupled-in optical grating; the coupling-out optical grating is used for receiving light of the turning optical grating, the coupling-out optical grating is located on one side of the turning optical grating, the connecting line of the center of the coupling-in optical grating and the center of the turning optical grating is perpendicular to the connecting line of the center of the coupling-out optical grating and the center of the turning optical grating, and the coupling-out optical grating couplesthe light out of the optical waveguide sheet; and at least one detector is located on the side, far away from the coupling-in optical grating, of the turning optical grating and is used for receivinglight which does not enter the coupling-out optical grating in the turning optical grating, and at least another detector is located on the side, far away from the turning optical grating, of the coupling-out optical grating and is used for receiving light which is not used for imaging in the coupling-out optical grating. According to the invention, the problem that the optical waveguide in the prior art can hardly realize real-time detection and the imaging quality is influenced is solved.

The invention discloses a measuring system for acetylene sludge slurry parameters for desulfuration. The measuring system comprises a measuring barrel, wherein the measuring barrel comprises a cushion section and a measuring section, the cushion section is positioned below the measuring section, the cushion section is formed by encircling an inverted cone shell, the measuring section is formed by encircling a columnar shell, the inner diameter of the measuring section is appropriately equal to the large-diameter end of the cushion section, and a density gauge and a pH gauge are mounted on the measuring section; the small-diameter end of the cushion section is communicated with a slurry storage tank of a desulfuration tower through a pipeline, the top of the measuring section is communicated with one end of a measuring pipeline, and the other end of the measuring pipeline extends into a pit; the inner diameter of the large-diameter end of the cushion section is three to five times of the inner diameter of the small-diameter end of the cushion section; the height of the top of the measuring barrel is lower than the liquid level inside the slurry storage tank of the desulfuration tower. A measuring pump is cancelled, the measuring system can measure flow rate reduction of slurry by means of flowing of acetylene sludge slurry, so that the erosive wear for a lining of the density gauge and a probe of the pH gauge is reduced, and the cost for changing the density gauge and the pH gauge regularly every year is reduced.

The embodiment of the invention provides a data exception detection method and device. The method comprises the following steps: carrying out de-noising division on each data set according to attributes, and calculating the weight of each attribute according to the information entropy of the divided data set; calculating cosine distances between different data objects in each data set according to the weight of each attribute, and calculating an initial abnormal value of each data object according to the cosine distances between each data object and other data objects; inserting the data sets with the calculated initial abnormal values of all the data objects into an ordered sequence according to a time sequence; updating the initial abnormal value of the data object of any data set according to the abnormal states of the data set at the previous moment and the next moment of any data set in the sequence; and detecting the data of each data set, and judging the data of which the abnormal value exceeds a specified threshold as abnormal. According to the embodiment of the invention, the data on the third-party storage medium of the block chain can be detected, and the accuracy and the recall rate are relatively high.

A method based on annular laser vision sensing for automatically locating the welded seam includes such steps as detecting annular laser tracings by CCD planar array, imaging on image sensing surface of camera, obtaining the expression equations for the X and Y coordinates of the characteristic points of said welded seam in conic coordinate system of laser, finding out the depth value of said characteristic point according to circular radius-depth relation algorithm, creating the equation for the Z coordinate of the characteristic point to be detected, finding it out, and converting the found out 3D coordinates from said conic coordinate system to the coordinate system of robot itself. Its advantage is high efficiency.

The invention discloses a road condition detection method based on radar communication integration, and the method comprises the steps that 1, a main station radar transmits radar communication integration signal pairs s1 (k) and s2 (k) at the moment k, and obtains the detection information I (k) of vehicles and obstacles in a road segment according to echo signals; 2, at the k + 1 moment, the main station radar transmits radar communication integrated signal pairs s1 (k + 1) and s2 (k + 1), and vehicle and obstacle detection information I (k + 1) in the road section is obtained according to echo signals; the detection information I (k) at the moment k is modulated into s1 (k + 1) and s2 (k + 1) by changing the width of each pulse in s2 (k + 1); 3, the vehicles in the road section receivethe radar communication integrated signal pairs s1 (k + 1) and s2 (k + 1), and the detection information I (k) at the last moment is obtained from the radar communication integrated signal pairs; according to the method, a radar signal used for detecting a target and a communication signal used for transmitting detection information are realized by using a group of equal-period pulse signal pairs,so that electromagnetic resources are saved on the basis of realizing a road condition detection function.

The invention provides a vehicle front road surface accumulated water depth detection system based on a vehicle sensor, and relates to the technical field of vehicle driving safety. The vehicle front road surface accumulated water depth detection system based on the vehicle sensor comprises a vehicle and a processing unit. The vehicle comprises a sensor assembly, and the underwater depth and the water surface height at the position with the preset distance in front of the vehicle can be obtained through the sensor assembly. The system for detecting the depth of accumulated water on the road surface in front of the vehicle based on the self-vehicle sensor can meet the technical requirements of detecting the depth of accumulated water on an irregular road surface profile in front of the vehicle based on a sensor assembly carried by the vehicle; and the vehicle can detect the depth of the accumulated water on the irregular road surface contour in real time, and the driving safety is ensured under the condition of the accumulated water on the road. Meanwhile, compared with an existing method for detecting the depth of accumulated water on the irregular road surface contour, the method has the advantages of being low in cost, compact in structure and low in power consumption, a long preview distance can be achieved, and the measurement precision can be guaranteed.

The invention discloses an intercept valve, and relates to the field of intercept valves. The intercept valve comprises a valve body, a valve cover, a power supply system, a fire extinguishing system, a motor and a valve; the two ends of the valve body are provided with a first pipe body and a second pipe body; the valve is arranged in the first pipe body; the first pipe body is sleeved with the fire extinguishing system; the valve cover and the valve body are connected through a screw; the motor is fixed to and electrically connected to the power supply system; the motor can drive the valve to be switched on and switched off. The intercept valve has the independent power supply system and fire extinguishing system, and can still automatically operate to switch off the valve after outage in the outside caused by fire, and meanwhile the fire extinguishing system further can extinguish fire around the valve body to ensure the safety of the valve body.

The invention relates to a wireless energy transmission system and method for electromagnetic echo excitation amplification. The system comprises a transmitting end and a receiving end, which are usedfor forming an adaptive wireless energy transmission channel. The wireless energy transmission channel constrains the energy of electromagnetic echoes transmitted back and forth within a set lateralrange. Compared with the prior art, the system and method, which are provided by the invention, have the advantages of long-distance movement, human safety, high power transmission and the like.

The invention relates to a network time delay collecting method based on distributed deployment, which aims at improving the instantaneity and the accuracy of network time delay collection. The accuracy of detection is ensured by selecting an appropriate detection algorithm and increasing the detection frequency, and the instantaneity of time delay data detection is ensured in a distributed deployment way. The method comprises the following steps of: selecting the distributed deployment way to increase the detection frequency; performing the time delay detection algorithm including a simple network and a redundant network; detecting at the frequency of multiple times every 5 min; and finally performing computation to increase the accuracy of detection. With the technology disclosed by the invention, while the instantaneity in detecting large amounts of chain time delays is guaranteed, the accuracy of time delay collection is increased through multiple detections in 5 min.

The invention discloses a living fingerprint detection method based on a thin-plate spline deformation model. The fingerprint sequence of the finger to be tested is collected and preprocessed; the fingerprint images before and after deformation are registered, and the corresponding detail feature point pairs are selected as the estimated thin-plate spline The reference point pair of the model; the improved thin-plate spline model is used to describe the deformation of the fingerprint; the deformation energy of the fingerprint to be tested in each direction is calculated based on the model parameters, and the overall bending energy vector is formed; respectively trained on the deformation database of true and false fingerprints to obtain The spatial distribution of the bending energy vector; respectively calculate the similarity between the bending energy vector to be tested and the true and false bending energy fuzzy feature sets; select an appropriate threshold to obtain the detection result. The invention can detect false fingerprints from living fingerprints, realize real-time and effective detection of living fingerprints, improve the security of an automatic fingerprint recognition system, broaden its application range, and have important application value in the field of biometric identification.

The invention relates to a fault detection and reconstruction system and method based on a dynamic-static hybrid strategy. Firstly, according to the current work mode of a control system, the fault detector and sensor configuration schemes are subjected to combined optimization; on the premise of ensuring the fault detection performance, the system energy consumption is possibly reduced; then, when the fault is detected, the real-time accurate recognition on the fault type is realized through solving the multi-model selection problem; finally, according to the recognized fault type, through the design of the reasonable switching control scheme, the control allocation and the system sterilization can be completed; the autonomous fault detection and reconstitution of the control system under various possible fault conditions can be realized. By using the technical scheme, the static fault monitoring and the dynamic fault type recognition and control system reconstruction are combined. Compared with a conventional fault detection and fault-tolerant control scheme, various kinds of potential faults can be effectively treated; the real-time performance is improved; the energy consumption is reduced; the system and the method can be applied to the high-reliability control problem of control objects of astrovehicles, mechano-electronics, chemical engineering and the like.

The present disclosure relates to a potato disease detection method based on image recognition, comprising the following steps: Step 1, the client acquires pictures of potato leaves, roots and tubers to be identified, and uploads them to the server; The above pictures are preprocessed, and then the preprocessed pictures are sent to the deep learning model; step 3, the deep learning model extracts high-dimensional features from the preprocessed pictures, and then identifies and calibrates the category and category of the potato lesions contained in the pictures location; step 4, superimposing the identified lesion location and its category as a mark on the image to form a detection result image for the client to download and display.

The invention provides a method and device for detecting CPU utilization ratio, which relates to an embedded system and solves the problem of inaccurate detection of the CPU utilization ratio. The method comprises the steps of: when a triggering condition is detected, determining a discrete time part and a sliding time part; respectively acquiring the CPU holding times of the discrete time part and the sliding time part; and computing to acquire the current CPU utilization ratio according to the CPU holding times of discrete time part and the sliding time part. The technical scheme provided by the invention is suitable for embedded systems.

A leak detection system for a pressure gas pipeline, including a hub, several spokes, a number of sensing units, and a console; wherein, the hub moves in the pipeline with the pressure gas, and is used to analyze and process the data collected by the sensing units to form real-time pressure data or pressure Gradient data, and the pressure data or the alarm signal obtained by analysis and processing are transmitted to the console; one end of the radial finger is fixedly connected with the central core (1), which is used to make the central core move with the pressure gas in the pipeline, and several radial fingers are distributed symmetrically with the central center ;The sensing unit is installed on the spoke finger to sense the pressure value of the pressure gas in the pipeline and transmit the collected data to the center; the console and the center are connected by wireless or wired to receive the signal transmitted by the center and analyze and process and send out instruction. The invention can more accurately judge whether there is leakage in the pressure gas pipeline.

The invention provides a grain transportation vehicle grain loading state detection method and a device based on a visual image. The method comprises steps: image acquisition is carried out on the grain transportation vehicle through a color digital camera, and a planar color image for a grain box port and the grain is acquired; smoothing, edge, line detection and other filters are used for processing the planar color image, and a complete edge contour of the grain box port is acquired; an image processing method such as color space conversion and threshold segmentation is used for acquiring the area of the grain in the grain box in the image; a two-dimensional convex hull algorithm is used for acquiring convex points representing edges of the grain area, and the linear distances between the convex points and the four edges of the grain box port are calculated respectively; according to the distances measured in real time, the loading state of the grain at each distribution position of the grain box port is judged through a preset threshold, and detection on the grain loading state when a reaper unloads the grain to the grain transportation vehicle can be realized; and finally, the grain state and the corresponding operation instruction are displayed on a detection terminal in real time. The method and the device have the advantages of convenient mounting and maintenance, easy implementation, quick detection speed, high stability and high precision.