109results about How to "Avoid Missing Cases" patented technology

An unmanned aerial vehicle remote power transmission line intelligent inspection system and method are disclosed. The system is composed of a remote control center, an unmanned aerial vehicle nest and an unmanned aerial vehicle, the remote control center sends a formulated inspection task GPS inspection route to a flight control module, and the unmanned aerial vehicle makes preparations before takeoff; the battery state of the unmanned aerial vehicle is read, and a landing point and a charging plan are formulated according to the GPS routing inspection route; the unmanned aerial vehicle takes off remotely and autonomously according to the inspection task; an image acquisition module uploads the acquired images to the cloud through a wireless communication module for intelligent identification and analysis of inspection image defects; and the unmanned aerial vehicle flight control module generates an allowable flight route, the remote control center sends a termination command to terminate task execution, landing and return operations are performed, and the unmanned aerial vehicle flies to a target nest after inspection. According to the invention, manual intervention operation is greatly reduced, the error rate is reduced, the inspection time is saved, the inspection efficiency is improved, the workload of manually identifying defect images is greatly reduced, and the inspection efficiency and quality of the power transmission line are improved.

The invention relates to the technical field of network virus detection and discloses a block-based virus detection method in a P2P (peer-to-peer) network. The method comprises the following steps: a step (1): locally performing network p2p data packet capture through a network card and performing preliminary filtration according to Winpcap (windows packet capture); a step (2): performing pretreatment on P2P network data packet information obtained by capture from the network card in the step (1), and enabling the data packet information to be used for virus detection; and a step (3): detecting whether data information after treatment contains virus or not, and detecting whether the data information in the p2p data packet information of an application layer, which is obtained after treatment in the step (2), contains the virus or not according to virus characteristic codes in a virus library. Compared with the prior art, by adopting the method, the missing detection situation caused by negligence of P2P characteristics and the P2P transmission file mechanism of existing detection software can be avoided.

The invention discloses an intersection vehicle queuing length detection method based on an aerial video. The intersection vehicle queuing length detection method comprises the following steps: step 1, registering the aerial video; step 2, setting a vehicle queuing length detection region at an intersection entrance path; step 3, detecting a red light state; step 4, when a lane is at the red light state, detecting moving vehicles in the pre-set vehicle queuing length detection region by using a three-frame difference method; step 5, determining a vehicle queuing region; step 6, detecting the quantity of static vehicles of the vehicle queuing region and calculating an actual vehicle queuing length by using an actual distance Scale represented by a calibrated image unit pixel; and step 7, repeating the steps to realize continuous vehicle queuing length detection. According to the intersection vehicle queuing length detection method, the vehicle queuing length detection is carried out by the aerial video and mutual shielding and size changes of the vehicles in the monitored video are overcome, so that an intersection vehicle queuing condition is directly represented and the detection precision is extremely improved.

The invention relates to a multiprocessor-embedded image acquisition and processing method and device which have the advantages of simple algorithm, high efficiency and low cost. The invention comprises image processing hardware structure, an image segmentation method, an image processing algorithm and an image synthesis method. The device comprises an image sensor, wherein the image sensor is connected with an image sensor control circuit, and also connected with an image acquisition and segmentation module; the image acquisition and segmentation module is connected with a plurality of subprocessors; the subprocessors are respectively connected with corresponding memories; and the image sensor control circuit is also respectively connected with the image acquisition and segmentation module, the subprocessors and the corresponding memories. Compared with the system with a multi-image sensor, the system provided by the invention is more convenient to install and use, and the cost is lower.

The invention provides dual-target site inverse-transcription fluorescent PCR primers, probes and kits for detection of 2019 new coronavirus (2019-nCoV). Nucleotide sequence alignment is performed onthe whole genome of the 2019-nCoV, 3 specific nucleotide sequences W1, W2 and W3 are found in N genes of the virus, and are shown in SEQ ID No. 1-3, multiple pairs of primer probes are designed separately for three target sites, and it is found that when a primer probe combination which is designed for the target sites of W1 and W2 and has sequences shown in SEQ ID No. 4-6 and SEQ ID No. 7-9 is applied to double-target site inverse-transcription fluorescent PCR, 2019-nCoV can be detected specifically and sensitively with detection sensitivity of within 10 copies; and no non-specific amplification of 22 clinically-common respiratory pathogens are caused, false-negative results caused by single target spot mutation can be avoided effectively, and good detection ability of specimens is achieved.

The invention relates to a high-speed image acquiring and processing method and device. The method has the advantages of simple algorithm, efficiency and low cost and comprises a hardware structure for image processing, an image segmentation method, an image processing algorithm and an image synthesis method. The device comprises an image sensor, wherein the image sensor is connected with an image sensor control circuit; the image sensor is connected with an image acquisition and segmentation module; the image acquisition and segmentation module is connected with an interface circuit; the interface circuit is connected with a computer; and the image sensor control circuit is also connected with the image acquisition and segmentation module and the interface circuit respectively. Compared with the system with multiple image sensors, the system is more convenient to install and use and has lower cost.

The invention belongs to the technical field of nondestructive testing, and relates to a circular dual-frequency excitation eddy current probe and a method for detecting thick-wall deep crack defects.The circular dual-frequency excitation eddy current probe comprises dual-frequency excitation probe assemblies, a detection probe assembly and a fixing frame; two sets of excitation probe assembliesare provided, and are distributed in bilateral symmetry with the detection probe assembly as a center; each excitation probe assembly consists of an excitation coil and a excitation coil mounting vertical column; the detection probe assembly consists of a detection coil and a detection coil mounting vertical column; the fixing frame consists of a vertical column fixing plate and a scanning component connecting component which are fixedly connected; the bottom end of the excitation coil mounting vertical column and the bottom end of the detection coil mounting vertical column are mounted on thevertical column fixing plate; the excitation coil is a circular eddy current coil and is excited by adopting a sinusoidal dual-frequency AC current. After adoption of the circular dual-frequency excitation eddy current probe, the penetration depth of an eddy current and the deep crack detecting ability of the probe can be improved, and the penetration depth of the circular dual-frequency excitation eddy current probe can reach about 21mm.

The invention provides a smart and safe campus big data safety early warning platform, which comprises a monitoring front end, a sensor and warning devices, wherein the monitoring front end, the sensor and a first warning device are connected with a cloud server through a first codec respectively; the cloud server is connected with a central control system; the central control system is connectedwith a monitoring terminal and a second warning device through a second codec; the first codec is connected with an inspection data acquisition system; the inspection data acquisition system comprisesa data acquisition box and an RFID label; an instruction key and a display screen are arranged outside the data acquisition box; the data acquisition box is internally provided with a card reader anda central processor; and the card reader is connected with the RFID label. The acquired data can be transmitted to the monitoring terminal through the codec, the manual inspection condition can be inputted to the inspection data acquisition system through an instruction input device and is then transmitted to the monitoring terminal, campus safety early warning is realized accurately, and real-time management on a related responsible person is carried out.

The invention discloses an examination task generating method, system and server wherein the method comprises the following steps: obtaining the information of materials in a production procedure system; according to the information of the materials, calling the examination plan information corresponding to the information of materials; receiving the attributes of a target sample chosen and aimed to the information of the materials by an examiner; if the information of materials and the attributes of the target sample meet the attached condition in the examination plan information, determining the examination requirement information for the attributes of the target sample according to the examination plan information; and based on the target examination project and the examination requirement information corresponding to the target examination project, generating an examination task. With the task generating method of the invention, people do not have to fill an examination paper manually, which prevents the missing examination of some projects due to human causes and increases the accuracy for afterwards examinations.

The invention discloses an outdoor parking lot unoccupied parking space detection method based on deep learning, and the method comprises the steps that a picture is collected through a high-altitude camera located above a parking lot, a PSPNet network model is improved for extracting a parking space line in the image, and the linear detection and contour extraction are carried out, the positions of all parking spaces in the image are obtained in combination with a corresponding parking space line filling method, a YOLOv3 network model is improved to detect all vehicles in the image, the positions of the vehicles are recorded, the parking spaces and the vehicles are subjected to Hungarian matching, and whether the parking spaces are occupied is judged according to whether the parking spaces are successfully matched and the distance between the two successfully matched parties; and the number of unoccupied parking spaces in the image is counted. The number and the positions of the unoccupied parking spaces of the outdoor parking lot can be accurately judged, the robustness is high, and a vehicle owner can conveniently find the parking spaces for parking.

The invention discloses a rapid prediction system and method for bacterial drug resistance. The system comprises bacterial genomic DNA extraction equipment, bacterial genomic fingerprinting equipment,bacterial drug resistance phenotype determining equipment, bacterial drug resistance phenotype data collection and analysis equipment, an intelligent terminal, a cloud server and a bacterial databaseserver, wherein the bacterial genomic DNA extraction equipment is connected with the bacterial genomic fingerprinting equipment, the bacterial genomic fingerprinting equipment and the bacterial drugresistance phenotype data collection and analysis equipment are connected with the intelligent terminal through circuits or a wireless signals, the intelligent is connected with the cloud server through the Internet, and the cloud server is further connected with the bacterial database server. By utilizing the Internet and a safe cloud computing storage technique, the tolerance degrees of to-be-tested bacteria to different antibiotics can be accurately judged, and the heterogeneous drug resistance of the bacteria can be comprehensively judged according to information of genotypes and drug-resistant phenotypes.

The invention relates to an enrichment culture solution for the detection of Salmonella in poultry eggs, belonging to the technical field of the enrichment culture solution which can realize the selective enrichment of sublethal injured Salmonella of which amount is extremely low in poultry eggs. 1L of the enrichment culture solution comprises the following raw material components by weight: 7-11g of peptone, 5-8g of sodium chloride, 7-9g of disodium hydrogen phosphate, 0.5-1g of potassium dihydrogen phosphate, 6-8mg of brilliant green, 0.5-1g of bile salt, 10-14mg of nalidixic acid and 15-40g of skimmed milk powder. The enrichment culture solution can realize the selective enrichment of sublethal injured Salmonella of which amount is extremely low; and the enrichment processing steps of Salmonella in foods are reduced, the defect that the leak detection of a small amount of sublethal injured Salmonella in a sample can be overcome and the detection reliability and accuracy of Salmonella in foods can be increased.

The invention discloses a three-dimensional detection method for the defects of a sealing surface. The three-dimensional detection method comprises the following steps: obtaining pictures with scratches on the sealing surface; obtaining the length and the initial width of each scratch according to the pictures with the scratches; determining a scanning route according to the initial width; planning a scanning track according to the scanning route; and scanning the scratches along the scanning track, thus determining the depth and the final width of each scratch. Compared with the prior art, the three-dimensional detection method for the defects of the sealing surface has the advantages that the pictures with the scratches on the sealing surface are obtained so as to avoid a leak detection condition of the small scratches; then the length and the initial width of each scratch are obtained according to the pictures, the scanning route is determined according to the initial width, the scanning track is further planted according to the scanning route, and finally, the scratches are scanned along the scanning track so as to determine the depth and the final width of each scratch, so as to realize the three-dimensional detection of the defects of the sealing surface. Meanwhile, the measurement precision is improved and the size quantification detection on the three-dimensional defects of the sealing surface is realized.

The invention discloses a circulation tumor cell separation and enrichment micro-fluidic chip and an enrichment method thereof. The circulation tumor cell separation and enrichment micro-fluidic chip comprises a sample treatment chamber and a separation purification chamber, wherein the separation purification chamber comprises a micro channel, a magnetic field adsorption part, a magnetic collection chamber, a cell collection chamber and a filtering membrane; a sub-channel is formed inside the micro channel; the magnetic field adsorption part is arranged on the inner wall of the sub-channel; one end of the micro channel is communicated with the sample treatment chamber; the other end of the micro channel is communicated with the magnetic collection chamber and the cell collection chamber in sequence; the magnetic collection chamber is used for collecting magnetic beads which are not adsorbed by the inner wall of the sub-channel; the filtering membrane which is capable of retaining circulation tumor cells is arranged inside the cell collection chamber; and a liquid outlet from which a liquid filtered by the filtering membrane flows out is formed in the cell collection chamber. The circulation tumor cell separation and enrichment micro-fluidic chip is efficient, rapid, good in specificity and high in accuracy.

The invention relates to an intelligent identification method and system for appearance defects of substation equipment. The method comprises the following steps: determining an inspection point map of inspection equipment according to the three-dimensional map information of a substation, the information of key equipment, and the point information of key parts of the key equipment; patrolling byutilizing patrolling equipment according to a patrolling path, acquiring first image data of each key equipment in the patrolling range in real time, and performing image recognition analysis on the first image data to acquire defect type information of each key equipment in a patrolling task; acquiring second image data of the key equipment in the monitoring range by using fixed video camera equipment, and performing image discrimination analysis on the second image data to determine state information of each key device in the monitoring range; and determining defective key equipment according to the state information, and determining final defect type information of the defective key equipment according to the defect type information and the state information of the defective key equipment.

The invention belongs to the technical field of electromagnetic non-destructive testing, and relates to an interlink-type vortex probe and a method for detecting deep cracks of opening fatigue and stress corrosion. The probe comprises an excitation component, a detection component and a fixing frame (9). The fixing frame (9) comprises a scanning frame connecting portion (91) and an component mounting portion (92), and the component mounting portion (92) is provided with a component mounting surface (921); the excitation component comprises a large exciting coil (1), a large exciting coil winding column (6), a small exciting coil (2) and a small exciting coil winding column (7); the detection component comprises a detection coil (3), a detection coil winding column (8), a magnetic field shielding plate (4) and a magnetic field shielding cylinder (5); the large exciting coil winding column (6), the small exciting coil winding column (7) and the detection coil winding column (8) are linearly arranged in sequence, and the bottom ends are vertically and fixedly mounted on the component mounting surface (921). The probe has a large detection depth that can be up to 25mm.

The invention discloses a self-adaption detection method for safety belt buckling of a driver and a passenger in the front row of a motor vehicle. The method includes steps of reading a road checkpoint picture and determining a vehicle front windshield area primarily through positioning a license plate; dividing the windshield area into a main pilot area and a copilot area; positioning the windshield areas of the pilot seat and the co-pilot seat precisely; calculating a pilot safety belt detection area and a copilot safety belt detection area and recording as Mbelt and Sbelt respectively; performing canny edge extraction on the Mbelt and Sbelt respectively and filtering away background noise; extracting safety belt suspected inclined lines; performing connection treatment on the safety belt suspected inclined lines. According to the invention, machine vision automatic detection replaces human eye detection, detection of safety belt buckling of drivers and passengers in a large quantityof checkpoint pictures and video images is implemented, so that a large amount of labor and material are saved and smart city construction is promoted.

The invention relates to an early fire detection method which is used for detecting a fire and comprises the following steps: S1, acquiring a real-time monitoring video; S2, providing a single-frame detection network model, and using the single-frame detection network model to detect whether flame and / or smoke exists in the image at the current moment or not so as to obtain a first detection result; S3, providing a time sequence detection network model, and detecting a flame level and / or a smoke level in the video sequence before the current moment by using the time sequence detection networkmodel to obtain a second detection result; and S4, calculating a fire risk level by combining the first detection result and the second detection result. Thus, missed detection of fire disasters is prevented, and the accuracy of fire detection is improved. The invention also provides a computer readable medium. The invention also provides a detection system.

The embodiment of the invention provides an object state detection method and device. The method comprises the following steps; acquiring a to-be-detected image aiming at a preset detection scene, extracting a movement foreground region from the to-be-detected image and extracting a movement foreground object from the movement foreground region; acquiring the first actual height H1 of the movement foreground object and judging whether the first actual height H1 is higher than the second actual height H2 of a preset reference plane; if the first actual height H1 is higher than the second actual height H2, judging that the movement foreground object is in the standing state when the first actual height H1 is higher than the second actual height H2 and the width of the movement foreground object satisfies the preset condition. Visibly, the actual height of the movement foreground object is acquired by using the depth information thereof through the scheme provided by the embodiment of the invention, and whether the movement foreground object is in the standing state is judged according to the actual height, and the standing state detection is realized without needing the detection on the stand-up action, the missing detection situation caused by short action duration time is avoided, and the accuracy rate of the object state detection is improved.

The invention provides a patrol task issuing device and belongs to the field of a power system. The device comprises an environment monitoring sensor, a circuit load sensor, an environment judgment server, a load judgment server, a master control computer, a patrol task issuing server and a communication base station. The environment monitoring sensor is connected with the patrol task issuing server through the environment judgment server, the circuit load sensor is connected with the patrol task issuing server sequentially through the load judgment server and the master control computer, and the patrol task issuing server is connected with the communication base station. By means of the patrol task issuing device provided with the patrol task issuing server, numbers and related parameters of power supply circuits needing to be patrolled are issued in a unified mode so that nearby patrol personnel can selectively receive patrol tasks according to self task schedule. Environment conditions and load conditions of the power supply circuit are taken into consideration in the issued patrol tasks. Thus, missing detection is avoided, and patrol timeliness is further improved.

The invention relates to a vulnerability detection method and system, and the method comprises the steps: configuring test parameters based on a test item, at least comprising a to-be-tested target domain name, a test end browser IP address, a proxy server port, a vulnerability and a detection strategy thereof; sending a function point test request data packet to a test target through a proxy server port based on the test end browser, and receiving a returned response data packet; enabling the automatic detection tool to obtain a mirror image data packet of the test request, and carrying out the vulnerability detection on the function points one by one according to the configured vulnerabilities and the corresponding detection strategies; and in response to meeting the ending condition, ending the vulnerability detection. According to the method and the system, the leak detection condition of the function points is avoided, the safety engineers can share the detection experience and strategy, and the dependence on the safety engineers in the detection process is remarkably reduced, so the influence of human factors is reduced.

The embodiment of the invention discloses a parcel detection method and device, equipment and a storage medium. The parcel detection method comprises the following steps: acquiring a current frame image in a monitoring video stream corresponding to a monitoring area for transmitting a parcel; obtaining a first reference area image of the first preset area when no parcel exists in the monitoring area and a second reference area image of the second preset area when no parcel exists in the monitoring area, and judging whether the image of the first preset area in the current frame image is matched with the first reference area image or not; if the image of the first preset area in the current frame image is not matched with the image of the first reference area, judging whether the image of the second preset area in the current frame image is matched with the image of the second reference area; and if the image of the second preset area in the current frame image is not matched with the second reference area image, performing parcel cutting on the current frame image to obtain at least one parcel image to be stored. The invention can ensure that parcels are cut, and meanwhile, the parcel missing detection condition is avoided.

The invention relates to a small target detection method based on an improved Fast RCNN, which is realized by executing an improved Fast RCNN algorithm instruction by a processor, and comprises the following steps: receiving a scene picture containing a small target and extracting a first feature map F; obtaining a predicted anchor point frame a (x, y, w, h) according to the first feature map F1; obtaining a second feature map F2 with the same size as the first feature map F1 according to the first feature map F1 and the predicted anchor frame a (x, y, w, h); and obtaining a detection result of the scene picture according to the second feature map F2 and the predicted anchor frame a (x, y, w, h). According to the invention, modification is carried out based on a Fast RCNN algorithm framework, and the RPN network under the Faster RCNN algorithm framework is replaced with the adaptive anchor frame network, so that the generated anchor frame can be better matched with targets with different scales, thereby avoiding missing detection caused by unreasonable size of the anchor frame, and improving the detection accuracy.

The embodiment of the invention discloses a file risk detection method and device, and the method comprises the steps: obtaining a webpage file when detecting that the webpage file is abnormal; performing file label analysis on the webpage file to obtain multi-dimensional file feature data of the webpage file; performing global information integration on the multi-dimensional file feature data ofthe webpage file according to pre-acquired whole network file feature statistical information to obtain multi-dimensional file risk information; and when the multi-dimensional file risk information reaches a preset requirement, determining that the webpage file is a malicious webpage file. According to the embodiment of the invention, the malicious degree of the file can be analyzed without depending on the file content, whether the webpage file is malicious or not is effectively confirmed, the missed detection condition of the webpage file is avoided, the detection accuracy is improved, and meanwhile, better real-time performance is achieved.

The invention relates to the field of communications, particularly relates to a method and a device for realizing DPI identification. The method is used for improving the service identification rate of DPI equipment. The method comprises the following steps: using a first set of traffic features to perform service type identification on the first N data packets in the target data stream; after failure, using a second set of traffic features to perform supplementary service type identification on data packets at specified positions after the first N data packets in the target data stream. Therefore, for the target data stream of the data packet with the business feature appearing at the rear position, accurate service type identification can be realized, so that the service identification rate of the DPI equipment is effectively improved, and the detection missing condition is avoided. Meanwhile, the DPI equipment only analyzes the data packet at the specified position and does not needto analyze all the data packets in sequence when the supplementary service type is identified, so that the consumption of the DPI equipment is effectively reduced, the overload of the DPI equipment is avoided, and the service performance of the DPI equipment is further ensured.

The invention discloses a method for detecting large-sized inclusions in high carbon steel. The method comprises the following steps: (1) taking a sample of the high carbon steel product within 3 daysafter production, and then breaking the sample; (2) observing the fracture of the sample to determine the positions of white points on the fracture; and (3) detecting and analyzing the size and composition of inclusions at the white point positions. The method can quickly locate the positions of large-sized inclusions by breaking, facilitates the detection and analysis of the inclusions, determines the size and quantity of inclusions at the maximum level of the sample, determines the inclusion control level of steel products, and guides the improvement of the steelmaking process. The method greatly shortens the time for searching for the inclusions in a detection process, and avoids the condition of frequently missed detection in a conventional detection method.

The invention discloses an intelligent personnel management and control method and system for epidemic prevention and control, and computer equipment, and belongs to the field of data processing. According to the intelligent personnel management and control method for epidemic prevention and control, basic information of a to-be-passed object can be obtained at an entrance of a target area, the basic information is verified, and then first face data of the to-be-passed object passing verification and a verification result are stored in an associated manner, so that the health state of the passing object in the target area is identified, and the management and control efficiency of entrance verification is improved; in the target area, the image in the target area is collected, and the passing information of the passing object in the image is identified according to the first face data stored in the storage unit, so that whether the passing object in the target area is abnormal is judged; and the abnormal passing object is marked and an alarm signal is generated to provide guarantee for health and safety of the passing object in the target area, so that missing detection is avoided and the workload of prevention and control personnel is reduced.

The invention discloses a reading recognition method of a pointer-type meter based on machine vision, which comprises the following steps: extracting all straight lines of an instrument panel in an image to be recognized by using an LSD straight line detection algorithm; using length constraint and center constraint to filter out the non-pointer line segment to get pointer line; taking the lengthof pointer line and the distance between pointer line and the center point of circle as weighted values, the deflection angle of instrument pointer in the image to be recognized is obtained by linearweighting and taking the artificial virtual 0 DEG standard line as a reference; and the deflection angle of instrument pointer in the image to be recognized is obtained by linear weighting. The deflection angle of the instrument pointer in the template image is obtained by executing the steps for the template image, and the pointer-type instrument reading in the image to be recognized is obtainedby combining the known parameters of the template image. The method of the invention can realize the reading recognition of the pointer type instrument in the intelligent patrol inspection system of electric power, chemical industry, petroleum industry and the like, and has the advantages of good practicability, strong real-time performance and high accuracy rate of reading recognition of the pointer type instrument.

A goaf fireproof monitoring system based on infrared thermal imaging comprises an infrared thermal imaging acquisition system, a 5G transmission system, an image storage and processing system, a remote monitoring system and an alarm system, wherein the signal output end of the infrared thermal imaging acquisition system is connected with the signal input end of the image storage and processing system through the 5G transmission system; the signal output end of the image storage and processing system is connected with the signal input end of the remote monitoring system through a communication cable; the signal output end of the image storage and processing system is connected with the signal input end of the alarm system. According to the disclosed system, high-precision real-time omnibearing monitoring of all goaf areas can be realized, monitoring differences caused by different proficiency degrees of employees are effectively avoided, missing detection conditions are effectively avoided, along with increase of the number of the goaf areas, the system can greatly reduce the cost of manual detection, the workload of detection personnel is greatly reduced, the detection efficiency is improved, and unified management of the goaf is facilitated.

The invention relates to a capsule real-time inspection method with double detection, which comprises the following steps: 1) providing a capsule real-time test platform with double detection, which includes a near-infrared detection mechanism, a visible light detection mechanism and a main controller, wherein the near-infrared detection mechanism and the visible light detection mechanism respectively perform defect detection to every capsule and the main controller is respectively connected to the near-infrared detection mechanism and the visible light detection mechanism for determining a capsule rejection strategy according to a defect detection result from the near-infrared detection mechanism and the visible light detection mechanism; and 2) performing inspection with the test platform. According to the method, there are two detection methods are available to detect the capsules, thereby improving reliability of capsule test.