644results about How to "improve detection accuracy" patented technology

The invention relates to an unmanned vehicle real-time positioning method based on laser reflection strength. The method comprises the following steps of S1, photographing two sides of a city road by an on-vehicle laser radar, obtaining a plurality of frames of point cloud data, extracting road edge points and converting the road edge points into a current vehicle coordinate system; S2, selecting the road edge points of which the z-axis coordinate value is in a preset range and obtaining coordinates of the road edge points in a GPS coordinate system, using the current GPS coordinate point of the unmanned vehicle as an original point, dividing a coordinate space for obtaining a grid map; S3, matching the gridded high-precision map with the grid map which is obtained in the step S2, thereby obtaining position of the unmanned vehicle on a high-precision map; and S4, predicating vehicle posture by means of a Kalman filter. Compared with prior art, the unmanned vehicle real-time positioning method can realize real-time accurate positioning in a complicated environment and can effectively improve driving safety of the unmanned aerial vehicle.

The magnetic sensor is fabricated such that a magnetic sensor chip, having a one-chip structure in which MRE bridges and a comparator are included, is mounted onto a lead frame using an adhesive material, and then the magnetic sensor chip mounted on the lead frame is encapsulated by molding in a molded material. The magnetic sensor includes a magnetic-field generating portion formed by magnetizing at least one of the chip mounting member, the adhesive material, and the encapsulating material.

There is provided a laser radar including laser beam generating means, photo detecting means, a mirror, light deflecting means, and rotation driving means. The laser beam generating means emits a laser beam having an axis thereof. The photo detecting means detects a reflected laser beam that is reflected back by an object. The mirror includes a through-hole that passes the laser beam and a reflecting surface that reflects a reflected laser beam reflected back by the object toward the photo detecting means. The light deflecting means deflects the laser beam toward a measuring region and reflects the reflected laser beam from the object toward the mirror. The rotation driving means rotates the light deflecting means so as to direct the laser beam toward the measuring region.

To provide a method for fabricating a light-emitting device using flexible glass which is capable of withstanding a process temperature higher than or equal to 500° C., and the light-emitting device. A second substrate is attached to a support substrate using an adsorption layer. The second substrate is bonded to a backplane substrate provided with a transistor and a light-emitting element. The backplane substrate includes a separation layer and a buffer layer. A first substrate is separated from the backplane substrate by separation between the separation layer and the buffer layer. A flexible third substrate is bonded, using a second adhesive layer, to a surface of the buffer layer exposed by the separation. The support substrate is separated from the second substrate by separation between the second substrate and the adsorption layer.

The invention discloses a cascaded convolutional neural network based human face occlusion detection method. The method comprises the following steps of 1) obtaining a video frame image; 2) performing normalization processing on the image, and copying and storing two images; 3) graying the image 1 and performing histogram equalization processing on the brightness imbalance image; 4) performing human head detection in a multi-scale sliding window form by adopting a three-level cascaded network, and storing window coordinates and sizes which meet the conditions; 5) performing clustering analysis on the window coordinates to obtain a target window position; 6) according to the obtained data, intercepting a human head region in the image 2, and performing normalization processing and brightness adjustment; and 7) performing eye and mouth detection in the multi-scale sliding window form by adopting a two-level eye/mouth cascaded network regionally, and if a set condition is not met, judging that the eyes/the mouth is occluded and triggering alarming. The method is strong in illumination and posture robustness, suitable for various occlusion types, and relatively high in detection precision.

The invention discloses a driver posture recognition method based on depth images and virtual data. The method includes driver joint detection: the driver depth image and the joint label are used as the input of a deep learning framework Caffe, a deep convolution neural network model is trained, and the joint position of a driver in the image is detected through a deep learning model; virtual driver head posture data set establishment: a driver model is constructed through the modeling function of the three-dimensional modeling software, the head movement of the driver is set by utilizing theanimation function, the driver images is rendered in batches by utilizing the rendering function, the head segmentation processing is carried out on the driver images, and the virtual driver head posture data set is constructed; driver head posture analysis: a deep migration learning model is trained by means of a migration learning method according to the virtual driver head posture data obtainedby the virtual driver head posture data set construction process, and the driver head posture in the target domain image is estimated and the facial feature point position is detected.

The invention discloses a pedestrian flow monitoring method based on pedestrian detection and tracking. The method includes: acquiring and decoding a camera video stream to obtain single-frame RGB images, performing pedestrian detection on each image frame to obtain a group of pedestrian positions, calculating similarity, matching the pedestrians of every two adjacent image frames so as to track the pedestrians and obtain motion trajectory of each pedestrian, setting detection lines in the monitoring video, and judging pedestrian flow in different directions through the obtained pedestrian trajectories. The method is based the latest developments of pedestrian detection in computer vision, high in detection accuracy, fast in detection and promising in development prospect. By combining the method with a fast tracking method based on similarity and using a multi-scale detection method from sparsity to density, detection speed is further increased, fast pedestrian tracking is realized, detection and tracking speed can reach 10FPS on a current common computer, and practical level is reached.

The invention discloses an electrical impedance scanning detection system for the withdrawal of real-time multi-information and the method thereof, wherein, the system comprises a PC machine, a data transmission interface module, a microprocessor, a control module, a data acquisition module, a preceding stage detection module, an excitation source and an exciting electrode; wherein, a detecting electrode array withdrawals a detection signal and sends the detection signal to the preceding stage detection module, the detection signal is sent to an A/D converter and a data buffer memory module corresponding to a data acquisition unit after passing through a current and voltage conversion module through a multiple way switch; the date buffer memory module enters into the microprocessor; the PC machine performs the data transmission, delivers a control command or revives date through the transmission interface module and a microprocessor module; a software with supporting data analysis, curve description, image display and information prompt is arranged in the PC machine; thereby obtaining the real-time electrical impedance information of a measured target body, the multiple frequency impedance information and the impedance property. The invention has the advantages that the detection accuracy is high, the processing speed is rapid, and the noise resistance ability is strong.

A voltage detection device includes a processor that has a first input port, and a second input port grounded at a first ground potential part, a first Schottky barrier diode that has a cathode connected to a junction of an electric wire between the first input port and a voltage detection object part, and an anode grounded at a second ground potential part, and a second Schottky barrier diode that has an anode connected to a junction of an electric wire between the second input port and a third ground potential part, and a cathode connected to a positive terminal of a power supply. When the processor detects a voltage at the voltage detection object part, the processor detects a value obtained by adding a voltage applied to the second input port to a voltage applied to the first input port as the voltage.

In a motor, an sensor magnet is attached to a shaft and three hole sensors for detecting the position of the sensor magnet are mounted on a circuit board. By using neodymium-iron-boron base rare earth magnet having radial anisotropy as a sensor magnet, the detection accuracy for the position of a field magnet via the sensor magnet by the hole sensor can be improved.

A parameter memory of an object detection apparatus retains a plurality of parameters used for a detection process for each of a plurality of detection conditions. A parameter selector selects a parameter from amongst the parameters retained in the parameter memory, according to an existing detection condition. Then an object detector performs the detection process of detecting an object approaching a vehicle, based on a captured image captured by a camera out of a plurality of cameras disposed at different locations of the vehicle, by using the parameter selected by the parameter selector.

The invention discloses a rapid and comprehensive tunnel measurement system which comprises a chassis truck, a sensor bracket, a sensor, a sensor synchronizing and signal processing device, a data acquiring and processing device as well as a power supply device, wherein the sensor bracket is mounted on the chassis truck through a horizontal rotating shaft, and a plurality of area array cameras, LED lamps and thermal infrared imagers are mounted on the sensor bracket respectively; the area array cameras acquire image data of the section of a tunnel; the thermal infrared imagers acquire the temperature information of the inner wall of the tunnel; the synchronizing and signal processing device provides synchronous signals for the thermal infrared imagers and the area array cameras; the data acquiring and processing device acquires signals of the thermal infrared imagers and the area array cameras. The rapid and comprehensive tunnel measurement system can acquire data of various tunnel defects such as lining cracks, lining water leakage, section deformation, a disengaging defect and the like simultaneously and rapidly, the detection efficiency and the detection accuracy are improved greatly, manual intervention is not required in a detection process, and the safety is good when the system is used for detecting the lining water leakage and disengaging defect of the tunnel.

The invention discloses a method and device for detecting vehicle flow speed, as well as a method and system for controlling traffic lights, in order to overcome the defect that the traditional vehicle flow speed detecting technology or traffic light control technology has of low accuracy or inconvenient operation and the like during acquiring vehicle flow speed. The method for detecting vehicle flow speed comprises the following steps of: acquiring video images in an acquisition range in a vehicle driving direction in real time; comparing a first video image and a second video image which are continuous in time; and obtaining vehicle flow speed in the vehicle driving direction by determining a position of a certain vehicle which is in the second video image in the first video image. Compared with the prior art, the vehicle flow speed detecting technology can quickly acquire the vehicle flow speed and reduce the realization cost and has the advantages of high detection accuracy and convenient operation.

The invention provides a vehicle door opening early warning auxiliary system and belongs to the field of vehicle safety. The vehicle door opening early warning auxiliary system comprises a rearview camera used for detecting pedestrian information and transmitting the pedestrian information to a processing unit through a CAN bus, a radar probe used for detecting vehicle information and transmitting the vehicle information to the processing unit through the CAN bus, the processing unit used for calculating the received vehicle information and the received pedestrian information, judging whether collision danger exists in the door opening process or not and sending out a command to a vehicle body control module and further used for receiving information of an owned vehicle, making a judgment and sending out a command for activating/closing the vehicle door opening early warning auxiliary system, and the vehicle body control module used for receiving the command of the processing unit and starting/stopping the vehicle door opening early warning auxiliary system or sending out an alarm. By the adoption of the vehicle door opening early warning auxiliary system, various motor vehicles or pedestrians can be accurately and effectively detected, and the collision danger caused by the fact that a driver is distracted when opening a door is avoided.

The invention discloses a vehicle safety belt detection method and device based on image analysis. The vehicle safety belt detection method and device based on image analysis is applied to the image analysis field of intelligent transportation. Firstly, a vehicle widow area is positioned; the position of a safety belt is detected primarily through a line segment detection method; the safety belt is accurately positioned according to the angle and the length of the line segment. The vehicle safety belt detection method and device based on image analysis can store the information of vehicles with passengers not belted up, and provide sufficient, reliable and strict evidences for detection of the traffic offence of not fastening the safety belt.

The invention discloses an array substrate, a signal line detection method, a display panel and a display device. The array substrate comprises a plurality of signal lines and a detection circuit, wherein the detection circuit includes a control circuit, a plurality of transistors, a plurality of test lines and a plurality of receiving lines; each test line is connected with one end of at least one signal line; and a first electrode of each transistor is connected with one end of one signal line, a grid electrode is connected with an output end of the control circuit, and a second electrode is connected with one receiving line corresponding to the signal line. Different test signals are applied to the signal line in the array substrate, and the corresponding relationship between the receiving signals and test signals is tested, so that the detection of a short circuit and an open circuit of the array substrate can be realized, the test cost is lowered, and the detection precision of the array substrate is improved.

The invention discloses a vehicle front obstacle detection method based on YOLO. The objective of the invention is to solve the problems of low detection accuracy and low detection speed of an existing target detection algorithm. The method comprises the following steps of 1 acquiring a data set, and labeling each target in a training set image by using a labeling box; 2 initializing a convolutional neural network, and inputting the training set marked by the marking box into the convolutional neural network; 3 preprocessing the images of the training set; 4 dividing into 7*7 grids; 5 selecting an initial candidate box by the grid; 6 obtaining the category confidence of the target category M; 7 setting the output of the convolutional neural network according to the category confidence coefficient to obtain a final prediction box; 8 obtaining a final weight and the trained convolutional neural network; and 9 testing the images of the test set by using the trained convolutional neural network, and determining barriers in front of the vehicle. The method is applied to the field of vehicle front obstacle detection.

The invention provides a real-time vehicle flow detecting and tracking method based on aerial photography data. The method comprises the following steps that 1, a pre-training model of a YOLO networkis trained on the basis of a pre-training part of weak supervised learning by using a weak supervised learning mode; and 2, the pre-training model of the YOLO network is improved on the basis of a real-time vehicle flow detecting part of aerial photography data by adopting a full convolutional neural network and a multi-target box detection method with prior information to obtain a YOLO detectingnetwork. The method has the advantages that improvement is achieved on the basis of a YOLO algorithm, the full convolutional neural network and the multi-target box detection method with the prior information are adopted, training is conducted by effectively utilizing the multi-view and multi-resolution image characteristics of an aerial photography data set of an unmanned aerial vehicle, the detecting accuracy rate of the algorithm is increased on the condition of not losing too much detecting time, and the balance of the accurate rate and the time is achieved.

The invention discloses an anti-overflow device for an electric cooking utensil. The anti-overflow device comprises a distance detection module and a controller; the distance detection module is disposed in an upper over of the electric cooking utensil and is used for detecting a distance between the lower surface of the upper cover and a level of food materials cooked in an inner pot of the electric cooking utensil; the controller is connected with the distance detection module and is used for controlling the electric cooking utensil to stop heating or reduce power of heating so as to prevent overflowing of the electric cooking utensil, when the distance between the lower surface of the upper cover and the level of food materials cooked in the inner pot is less than or equal to a preset distance. The anti-overflow device is capable of accurately detection an event wherein thin gruel is about to overflow from the electric cooking utensil during cooking, thereby avoiding causing cleaning and maintenance problems. The invention further discloses the electric cooking utensil and an anti-overflow control method for the electric cooking utensil.

A detection camera includes an imaging optical unit and a light receiving unit that acquire image data, a first detection processing unit that detects a specified substance, a second detection processing unit that acquires distance information from the imaging optical unit to the specified substance, and a display control unit that outputs display data in which information regarding the specified substance is combined with the image data in a case where the distance information is in a detection target distance range.

The invention provides a network anomaly detection method based on an artificial immunity principle belonging to information safety field. The invention implements the network anomaly detection method capable of fast detecting with high detection accuracy by simulating biological immune system response to external antigen, training antigen data collection and artificial immune system study, and detecting network anomaly. The network anomaly detection method has superior characteristics of a non-linearity of the biological immune system, cloning selection, immune network and immunological memory, solves the problems of low detection speed and low detection accuracy in the present network anomaly detection. The invention also applies to the pattern recognition, machine learning fields and has wide application prospect.

The invention discloses a high precision voltage transient event detecting and wave recording method, which comprises synchronizing a voltage signal collected by a digital signal processing (DSP) plate to a voltage signal of 50Hz via a software, and frequency doubling every frequency current voltage signal by four times; judging the voltage effective value deviation of sampled data, subsequently determining the effective value in a sequence of 1 frequency current, 1/2 frequency current, 1/4 frequency current, and 1/8 frequency current if the voltage effective value deviation exceeds 2%, recording voltage extreme value, accumulated duration time and corresponding voltage waveform of an event that the voltage effective value deviation exceeds 10% via an event recording program, making a time mark, and calculating event time; and writing the recorded event and corresponding waveform thereof into an event chain table, carrying out waveform recording via a micro control (MCU) plate, and storing in a hard disk memory. By adopting software synchronization and software frequency doubling technique, the high precision voltage transient event detecting and wave recording method can accurately the voltage transient event (not smaller than 5ms) and record wave, with event duration time precision up to 2.5ms.

The invention relates to the technical field of food safety, in particular to a pesticide detection technology. An electrochemical pesticide residue rapid detector comprises a pesticide detection module and a microprocessor system which is connected with the pesticide detection module, wherein the pesticide detection module comprises an acetylcholinesterase biosensor of which the signal output end is connected with the signal input end of the microprocessor system; and the microprocessor system is connected with a display. Due to the adoption of the technical scheme, a detection system performs detection according to variation in current generated by reaction of an enzyme catalysis substrate and is not interfered by pigments in vegetables and fruits, so that a sample is relatively easy to extract and treat; and the detector has the advantages of high flexibility, intuitive display of a detection result, automatic detection and the like and is easy to operate when used for detecting pesticide residues.

The invention discloses a visible detection method of the welding quality of a circuit board based on a support vector machine, which comprises the following steps: (1) acquiring an image of a weldingpart of the circuit board; (2) extracting a gray histogram of the welding image; (3) adopting an entropy-based image segmentation method to divide the welding image into three chroma strips; (4) taking three variables of area, barycentric ordinate and secondary moment from each chroma strip as characteristic quantities of the welding quality detection of the circuit board; and (5) taking the characteristic quantities as inputs of a support vector machine classifier, judging that the welding quality belongs to normal soldering, excessive soldering or insufficient soldering by three output values of the support vector machine classifier, and finally obtaining a welding quality result. The invention can overcome the problems of low manual detection efficiency, slow speed, low precision, highomission ratio, easy fatigue of detection personnel, and the like, and improves the automation degree of electronic manufacturing and production as well as the product quality.

The invention provides a parking space detection method based on a geomagnetic sensor. The method includes the steps that geomagnetism disturbance signals are acquired, smoothing and/or filtering are/is conducted on the acquired signal data, a state model is built according to the geomagnetism disturbance signals to generate a state analysis graph, and behavior feature analysis is conducted according to the state analysis graph to judge vehicle behavior states; according to the vehicle behavior states, use states of parking spaces are judged, geomagnetic signals of a parking space area static background and the geomagnetism disturbance signals caused by target detection vehicles are acquired in real time through the geomagnetic sensor, data which are useless, invalid and capable of influencing accurate analysis of the disturbance signals are classified and eliminated, and data which are precise and can effectively reflect disturbance signal changes are left and used for subsequent building of the state model. The behavior state model can accurately reflect vehicle running states, use conditions of the parking spaces can be deduced according to the vehicle running states, the design principle is reasonable, detection precision is high, detection speed is high, and parking space detection can be well conducted.

The invention discloses a gas permeability test device. The gas permeability test device comprises a vacuum pump and a permeation pool, wherein the permeation pool is divided into a test chamber and a test gas chamber; the vacuum pump is connected with the test chamber and the test gas chamber respectively through pipelines; a valve I and a valve II are respectively arranged on the pipelines through which the vacuum pump is connected with the test chamber and the test gas chamber; the test chamber is connected with a pressure sensor I; the test gas chamber is connected with a pressure sensor II; the test gas chamber is connected with the gas source of the test gas through a pipeline III; and a valve III is arranged on the pipeline III. The gas permeability test device has the advantages that the structure is simple; the gas permeability test device is reliable in use and convenient in application; the test chamber is vacuumized independently; the test sample and the test chamber are effectively sealed through the pressure difference between the test gas chamber and the test chamber; the test chamber is vacuumized completely; and the test process can be controlled completely and multi-chamber test can be controlled independently.

The invention discloses to a sowing machine based on an automatic control system and relates to a high-precision automatic control system of the sowing machine. The sowing machine solves the problems that a full-mechanical sowing machine can not detect the sowing condition in real time in the sowing process, and the sowing planting distance and the fertilizing amount can not be changed flexibly. A power output shaft of a driving motor of a seed-metering device and a power input shaft of the seed-metering device are coaxially connected, a measuring hole in a photoelectric coder is coaxially connected with a rotating shaft of a ground wheel, a power output shaft of a driving motor of a fertilizing box is coaxially connected with a power input port of the fertilizing box, a seed monitoring sensor is arranged at an outlet of the seed metering device; a rotating speed signal output end of the photoelectric coder is connected with a rotating speed signal input end of a controller, a counting signal output end of the seed monitoring sensor is connected with a counting signal input end of the controller; a driving signal output end of the seed metering device of the controller is connected with a driving signal input end of the driving motor of the seed-metering device, and a driving signal output end of the fertilizing box of the controller is connected with a driving signal input end of the driving motor of the fertilizing box. The automatic control system is suitable for being used by various mechanical sowing machines.

The invention relates to the technical field of lithium battery production equipment, and particularly relates to vacuum injection equipment of a flexibly-packaged lithium battery. The equipment sequentially comprises a material loading device, a weighing and detecting device, a flat fixture for clamping air bags, an injection device, a first standing room, a second standing room, a third standing room, a sealing device, a weighing and detecting device, a blanking mechanical arm and an unloading mechanism from left to right. The equipment provided by the invention can be used for improving the production efficiency, the detection precision, the judgment accuracy of the product quality, the sealing quality and the product quality, reducing the electrolyte waste, effectively avoiding the phenomenon of a loading direction error and reducing the working strength of workers.

An optical device includes a telecentric optical system, a variable wavelength interference filter, and a detection section, the variable wavelength interference filter includes a first reflecting film, a second reflecting film provided to a movable section, and an electrostatic actuator adapted to displace the movable section, an effective measurement area capable of transmitting a light with a wavelength, which is within a predetermined allowable range centered on a measurement center wavelength when an amount of the displacement of the movable section takes a maximum displacement value, is set in the first reflecting film and the second reflecting film, and the telecentric optical system guides the incident light to the variable wavelength interference filter so that a principal ray of the incident light is parallel thereto and perpendicular to the first reflecting film, and at the same time, collects the incident light in the effective measurement area.

The invention provides an Android application safety analysis method based on sensitive behavior identification. The Android application safety analysis method comprises the following steps: 1) obtaining and analyzing a source code; 2) obtaining sensitive behaviors; 3) extracting a UI (User Interface) text; 4) processing sensitive behavior data; 5) extracting feature values; 6) forming a training data set; and 7) analyzing safety. A relationship between the UI text and a sensitive API (Application Program Interface) is taken as the characteristics of machine learning, and accuracy for detecting the malicious applications of an Android platform is effectively improved.