88results about How to "Guaranteed detection rate" patented technology

The invention discloses a method for detecting network attack behaviors, which comprises the following steps: (1) reconstructing a network traffic time sequence to be detected to a multi-dimensional phase space according to the complicated non-linear property of network traffic, and establishing a statistic distribution module according to a normal network traffic time sequence; (2) carrying out smoothing treatment on the network traffic time sequence to be detected, and dividing into sub-windows; and (3) obtaining a parameter sequence according to parameters of all the sub-windows of the network traffic time sequence after the smoothing treatment according to the calculation of the statistic distribution module, and establishing a comprehensive decision module according to the parameter sequence for detecting abnormality. The network traffic is re-constructed to the multi-dimensional space, thereby fully displaying information hidden in a one-dimensional space, reducing the computational complexity of each dimensional space and improving the running speed; and the method can improve the robustness and the accuracy of a system and is characterized by low computational complexity, high detection rate and low false detection rate.

The invention discloses a dual-aperture photoelectric imaging system-based unmanned aerial vehicle detection and recognition system. According to the dual-aperture photoelectric imaging system-based unmanned aerial vehicle detection and recognition system, a wide-angle camera and a long-focus camera are adopted as an unmanned aerial vehicle detection device and recognition device respectively for performing monitoring; and a monitoring image sequence is acquired, a target detection algorithm is adopted to detect an unmanned aerial vehicle suspected target, a pattern recognition algorithm is adopted to recognize the suspected target, and after the target is confirmed, the track is tracked, interfered and controlled. Photoelectric sensors are adopted as the unmanned aerial vehicle detection device and recognition device, and therefore, the dual-aperture photoelectric imaging system-based unmanned aerial vehicle detection and recognition system has the advantages of high reliability, low cost and the like; the suspected target is searched in a large space region by means of a wide-angle imaging system, and the target is confirmed and tracked by means of a long-focus imaging system which is a two-axis turntable, and therefore, requirements for high detection rate and high accuracy can be satisfied, the reliability of the system can be greatly improved, and therefore, the system can bring outstanding economic benefits and has high practical value.

The invention relates to an ultrasonic nondestructive testing device. The ultrasonic nondestructive testing device comprises a probe, a signal transmitting module, a signal receiving module, a processing module, a storage module and a control module, wherein the signal transmitting module is used for exciting the probe to generate ultrasonic waves; the probe is used for transmitting and receivingultrasonic signals; the signal receiving module is used for receiving the reflected ultrasonic signals from the probe; the processing module is provided with an amplifying circuit, a filter circuit, an analog / digital (A / D) conversion circuit, an envelope signal conversion unit, a scanning unit and a data fusion processing unit; the storage module is used for storing signals processed by the processing module; the scanning unit is provided with a software dynamic gate with continuously set time slots; and the software dynamic gate is used for monitoring ultrasonic signals of a sample to be tested at positions with different depth. The ultrasonic nondestructive testing device overcomes the contradiction between thickness direction subdivision and plate thickness direction coverage rate which cannot be overcome by other detection methods. For a nonuniform-thickness laminated board particularly, the ultrasonic nondestructive testing device guarantees detection rate and guarantees subdivision of the thickness direction.

The invention relates to the field of video monitoring and discloses a multiple-viewing-angle human face detecting method and a device thereof under complex scene. In the invention, at least one cascaded branchless node classifier is arranged in a cascading vector tree structure detector for carrying out human face detection to a child window to be detected and before at least one branch node classifier. A cascading classifier is led in between branch node classifiers and is utilized to carry out effective step-by-step removal to non-face child windows on all branches, the complexity of the branch node classifiers is reduced, the quantity of the contained weak classifiers is reduced, and the detecting speed is improved while the detection rate is ensured.

The invention relates to a driver fatigue detection method based on CNN eye state recognition. Driver fatigue detection can be completed by acquiring useful information of a face through the eye staterecognition. With the application of the method provided by the invention, eye states can be recognized and classified more accurately. As for the eye state recognition method via a convolutional neural network (CNN), a recognition rate for a circumstance that sunglasses are worn is improved via an infrared video sample; and finally, a plurality of constraint conditions are detected on the basisof fatigue / drowse physical quantity (PERCLOS) and blinking frequency, so that the driver fatigue states can be judged. Based upon experiments, it is proved that with the application of the method, theeye states can be recognized accurately in real time and fatigue driving behaviors can be early warned effectively.

The invention relates to a railway vehicle bolt looseness detection method and system. The railway vehicle bolt looseness detection method comprises the steps: S1, acquiring a plurality of railway vehicle key part images at bolts under different illumination conditions; S2, cutting each image to a set size; carrying out manual marking on each image; S3, establishing and training a key part positioning network; S4, setting a cutting layer; S5, establishing and training a bolt looseness detection network; and S6, performing image testing. The railway vehicle bolt looseness detection system comprises a multispectral time-sharing light source, an image acquisition device, an image processing module, a key part positioning network building and training module, a cutting layer setting module, a bolt looseness detection network building and training module and a storage and test module. The railway vehicle bolt looseness detection method and system can monitor the bolt state of a key part, can give the specific position information of the loose bolt, are high in the detection speed, can achieve real-time online detection, can improve the detection efficiency, and can improve the detection accuracy.

The invention discloses a Trojan horse communication feature fast extraction method based on network data stream clustering. The method comprises the steps that firstly, a captured network data packet is sorted according to a network conversation, wherein an IP address and a port of a monitoring object serve as a source IP address and a source port, and the data packet is subjected to conversation division according to equivalent tetrads; secondly, data streams are clustered into data stream clusters through a data stream clustering algorithm based on timestamps; lastly, Trojan horse communication features are extracted, wherein the Trojan horse communication features are extracted at the Trojan horse interactive operation stage. According to the Trojan horse communication feature fast extraction method, on the basis of network data stream clustering, the network data streams are processed with clusters as units, the difference between a Trojan horse communication behavior and a normal network communication behavior is analyzed, the difference between the two behaviors is dug deeply and the network communication features are extracted in combination with traditional statistic analysis, correlation analysis and other technologies, the false alarm rate is lowered while the detection rate is guaranteed, and the Trojan horse communication feature fast extraction method can be used for detecting a secret stealing behavior in a network.

The invention discloses a range hood capable of detecting the oil fume concentration. The range hood comprises a shell. A fan and a main control unit capable of controlling the rotating speed of the fan are arranged in the shell. The fan is mounted in a fan shell, and the fan shell is arranged in the shell and provided with an air inlet and an air outlet. The main control unit is further connected with a photoelectric oil fume sensor capable of detecting the oil fume concentration. The photoelectric oil fume sensor comprises a detecting light emitting device and a detecting light receiving device and is characterized in that the detecting light emitting device and the detecting light receiving device have an included angle after being mounted, and an intersection point or an intersection area of the straight line where the emitting direction of the detecting light emitting device is located and the straight line where the receiving direction of the detecting light receiving device is located is located 1-5 cm below the air inlet of the fan shell; and the detecting light emitting device and the detecting light receiving device further comprise protection structures, or, the detecting light emitting device or the detecting light receiving device further comprises a protection structure. The range hood is high in detection sensitivity, and the oil stain resistant effect of the oil fume sensor is good.

The invention provides an image retrieval method and device, and relates to the field of image retrieval. The method comprises the following steps: receiving query images sent by a user; according to an image classification method, obtaining the confidence coefficients, which correspond to each category, of the query images; according to a descending order of the confidence coefficients of the query images, retrieving image characteristics corresponding to the category of each confidence coefficient in a current confidence coefficient segment from an image characteristic library; according to the order of priority of retrieval, loading the retrieved image characteristics step by step; carrying out sorting processing on the loaded image characteristics; and sending a sorting processing result obtained by processing to a display interface step by step so as to bring convenience for the display interface to display. A situation that a result is output after all image characteristics in an established image database are retrieved is avoided, the time promptness and the stability of retrieval result feedback time are guaranteed, the problem that a true target is missed due to classification errors is unlikely to happen, indexes including a retrieval detection rate, a retrieval recall rate and the like are guaranteed, and the image retrieval method and device is more suitable for application scenes including security and protection video surveillance and the like.

The invention discloses a method for randomly accessing front check dynamic threshold value, comprising that descrambling input signal, descrambling frequency bias, and coherent non-coherent integration, obtaining the sign power time delay distribution of one front check mission at one access time space, finding the average noise horizontal evaluated value of the front check mission, using the evaluated value and the signal / noise ration set by superior, to interfere the integration length to dynamically real-time find the threshold value of single front check mission, finding the threshold values of all front check missions at the access time space, to select the maximum one as the threshold value of the front check. The invention via the noise level evaluation can dynamically real-time find front check threshold value, with simple, reasonable and effective threshold value setting, to avoid most false alarm, control the false alarm rate in allowed range, and confirm check efficiency.

The invention discloses a storage abnormality detection method based on artificial immunity, which adopts non-self detection mechanism of natural immunity system to judge the legality of data read / write request, realizes immunity abnormality detection on the data of storage element via monitoring the data of storage element formed by data read / write request, induces self learning and forgetting mechanism and the like in natural immunity system to refresh the weight value of a detector after each detection on user's read / write request and periodically eliminate or rebuild the detector according to the weight value, to adapt new non-self ones continuously appearing in storage system. Via the detector refresh mechanism, the method is different from prior storage abnormality detection technology, to realize actual intelligent abnormality detection, and the method has artificial intelligence as self learning and self adaptive characteristics, to effectively recognize new appeared abnormalities.

The invention discloses a layering single-class ship target false alarm eliminating method based on the intra-class difference. The method includes the steps that an optical remote sensing image is divided into a large ship slice, a small ship slice and a false alarm slice, a first-layer classifier is established based on large ship feature data, and a large ship to which great attention is paid in ship detection can be basically recognized through the first-layer classifier; when features of the large ship are obvious, the large ship can be detected through primary detection, and therefore the purpose of rapid large ship detection can be achieved; a large ship data set missing detection is trained, so that a second-layer classifier is formed, when the large ship is not recognized through the first-layer classifier, the large ship easily missing detection can be detected through the second-layer classifier, and therefore the detection probability can be increased; through screening of the first-layer classifier and the second-layer classifier, data with the large ship features in a small ship data set are eliminated, the data with the obvious small ship features are kept, a small ship mistakenly-distributed set is formed, and the detection probability of a small ship is increased.

The invention relates to a method and a system for detecting a fabric defect, a computer device and a storage medium. The method comprises the following steps: obtaining the surface image of a fabricto be tested and the surface image of a standard fabric; comparing the surface image of the fabric to be tested with the surface image of the standard fabric by using a first fabric defect detection algorithm; processing the surface image of the fabric to be tested and the surface image of the standard fabric when the defect on the surface image of the fabric to be tested is detected in order to obtain an image with prominent defect features; and detecting the obtained processing results by a second fabric defect detection algorithm in order to obtain the parameters of the defect. The method can effectively reduce noises of the defect fabric in the defect detection process and detect the existence and the parameters of the detect by adopting detection using two algorithms and image processing in order to ensure the detection rate.

The invention discloses an unmanned vehicle route data transmission system with breakpoint continuous transmission and method thereof. The system comprises a server and a vehicle terminal; the vehicleterminal receives route data sent by the server, and collects road condition information and vehicle condition information while a vehicle is traveling and uploads the information to the server; theserver and the vehicle terminal transmit and receive the data as a transmitting end and a receiving end in turn; the transmitting end comprises a data block module, a data transmitting module and a signal detection module; and the receiving end comprises a data receiving module and a data verification module. According to the unmanned vehicle route data transmission system with the breakpoint continuous transmission provided by the invention, the continuity and the real-time performance of route data transmission in the travelling process of the vehicle can be effectively ensured, the data transmission time is shortened, and the transmission efficiency is improved.

The invention discloses an airport runway foreign matter detection and classification method based on unmanned aerial vehicle high-low altitude joint scanning. The method comprises the steps of training a Yolo neural network model; obtaining a fine detection area by adopting a method of combining two rounds of high-low altitude thickness detection, photographing the fine detection area, and inputting a picture into the trained Yolo neural network model to detect foreign matters and classify the foreign matters, thereby realizing accurate detection and classification of the foreign matters.

The invention provides a pedestrian detection method, system, computer readable storage medium and electric device. The detection method includes training a pedestrian sorter according to pre-collected training sample set, wherein the pedestrian sort is used for judging whether a to-be-detected image is a pedestrian image or not; performing calibration on a monitoring image so as to obtain conversion matrix parameters from an image coordinate system to a world coordinate system or from the world coordinate system to the image coordinate system; according to the conversion matrix parameters, generating a mapping relation required for reconstructing an upright standing posture from a non-upright standing posture in the monitoring image and generating horizon detection coordinates; mapping the horizon detection coordinates into the monitoring image, cutting an image block from the to-be-detected image, performing reconstruction on the image block by utilizing the mapping relation, and utilizing the pedestrian sorter for sorting the reconstructed image block. Therefore, the pedestrian image is judged. The invention can solve problems quickly and conveniently and ensures the detection success rate of pedestrian detection in any monitoring scene.

The invention relates to an automatic safety control system for island detection of distributed energy resources. The automatic safety control system comprises a control terminal mounted on the side of a distributed power supply, wherein the control terminal is used for performing synchronous frequency measurement on the distributed power supply, marking a time scale on the distributed power supply and sending synchronous frequency measurement data to the side of a public power network through a 3G network. The automatic safety control system further comprises the control terminal mounted on the side of the public power network, wherein the control terminal is used for performing synchronous frequency measurement on the public power network, marking the time scale on the public power network and receiving the synchronous frequency measurement data sent from the side of the distributed power supply. The control terminal mounted on the side of the public power network computes a frequency difference between the sides of the public power network and the distributed power supply according to the time scales; if the frequency difference exceeds a setting action value, a tripping command is sent to the side of the distributed power supply after a given time delay is carried out on the frequency difference; if the frequency difference exceeds a setting start value but not exceeds the setting action value, an energy integral value (Kdf) of the frequency difference is computed; and if the value (Kdf) exceeds an energy integral setting value of the frequency difference, the tripping command is sent to the side of the distributed power supply.

The invention relates to a human visual characteristic compressive sensing-based grayscale image tampering and detection method, which comprises the following steps of: (1) generation and transmission of a watermark of an image: prior to the transmission of the image, a sending terminal firstly generates a Hash value of the image by using a human visual characteristic compressive sensing-based sparse base change and measurement matrix, then, transmits the Hash value through a secure channel, and transmits the image to a receiving terminal through a public channel; (2) image detection of the receiving terminal: the receiving terminal verifies whether the image obtained from the public channel is tampered or not by using the watermark of the image received from a trusted channel; and (3) tamper localization of the receiving terminal: after the condition that the received image is maliciously tampered is judged by the receiving terminal, a compressive sensing orthogonal matching pursuit reconstruction method is further used for restoring a difference value D, and therefore, an image E of a tampered position is obtained. Compared with the prior art, the human visual characteristic compressive sensing-based grayscale image tampering and detection method has the advantages of high rate in detection, low quantity in transmission, fastness in computation and the like.

The invention discloses an instruction vulnerability analysis-based SDC error detection method. The method comprises the following steps of: 1, carrying out a fault injection test on a program set soas to obtain a sample data set; 2, extracting instruction SDC vulnerability description features; 3, training a support vector regression-based instruction SDC vulnerability analysis model; 4, extracting instruction features of a target program, and predicting and analyzing instruction vulnerability; 5, carrying out instruction redundancy processing on the target program according to an instruction vulnerability value; 6, carrying out fault injection test on the target program after the redundancy processing, and detecting SDC errors during the operation of the program; and 7, assessing an SDCerror detection ability of the target program after the redundancy processing. The method is high in SDC error detection rate and low in performance overhead, and can be applied to detection of program SDC errors caused by transient faults under severe environment.

The invention relates to an all-zero-block detection method based on an energy conservation law, which is characterized in that an all-zero block can be detected in advance by predicting the block (all-zero block) with coefficient being zero both before the quantization and after the quantization, so that the quantization process of a residual block can be omitted. Due to the adoption of the method, while the detection efficiency at a maximal quasi peak (QP) is ensured, the effect for detecting the all-zero block in the video coding process at the minimal QP is remarkable, and the introduced calculation quantity in the detection process is small. Therefore, the calculation quantity in the quantization process can be effectively reduced, and the coding efficiency can be improved.

The invention discloses an ATM program updating safety protection method and system. The method comprises the following steps of: dynamically sensing whether an updating program is operated in an ATMterminal or not, and if the sensing result is positive, stopping the operation of the updating program and carrying out encryption and backup on all the pieces of information in the ATM terminal; after the encryption and backup are finished, releasing the operation of the updating program; creating a bait folder, and if a behavior for modifying a bait file by the updating program is sensed, judging that the updating program is malicious and stopping the operation of the updating program; if the behavior for modifying a bait file by the updating program is not sensed, automatically testing theupdated program and judging whether the updating program is malicious or not according to the test result; aiming at the condition that the updating program is malicious, deleting all the pieces of information created by the updating program and recovering the data which undergoes encryption and backup; and the updating program is not malicious, considering that the updating program is safe. The method and system are capable of effectively judging whether updating programs have potential threats or not.

The invention discloses a face image processing method and device and a readable storage medium. The method comprises the steps: carrying out the face recognition of an image, and determining a face image in the image; calculating a skin color probability template according to the face image; performing high contrast reservation processing on the face image, and performing binarization by using athreshold to obtain a binary template of spots; determining a spot area to be removed according to the skin color probability template and the binary template of the spot; and removing the spot area needing to be removed in a filling mode from outside to inside. By adopting the scheme, the spots in the face image can be quickly and conveniently removed.

The invention relates to a time-type network covert channel detection method based on secret sharing, which fully analyzes the communication characteristics of various types of time-type network covert channels and extracts robust channel intrinsic characteristics by preprocessing network flows. A channel is segmented by using a phase space reconstruction method to take the local time slot characteristics of the channel as sub secrets. A channel global entropy feature is used as a weight. A channel identifier is constructed by using a secret reconstruction strategy in a threshold secret sharing principle, mapping of channel features to the channel identifier is realized, traditional cryptography and network time hidden channel detection are well combined together, and the robustness of thedetection method is improved while the detection rate is ensured.

The invention relates to a pseudohypha detection method and device by increasing direction dimensions and a storage medium, and the method comprises the steps: inputting a to-be-detected image into anartificial intelligence detection model of pseudohypha, and obtaining a category label of the to-be-detected image. The method for generating the artificial intelligence detection model of the pseudohypha comprises the following steps: marking the artificial category of an original image, and training the original image marked with a category label to obtain the artificial intelligence detectionmodel of the pseudohypha. By constructing the training set of the pseudohyphae with large inter-class difference and small intra-class difference, convergence of the loss function is accelerated, andthe model training time is shortened.

The invention discloses a license plate detection method under a complex scene. The method includes the steps that input images are coarsely positioned, coarse positioning includes horizontal filtering and binarization processing, and a trained fine positioning classifier is used for carrying out license plate region detection on the images with coarse positioning regions. The method further includes the steps that a false detection decrease classifier obtained after branch training is used for removing the images with license plate regions in false detection out of the images with the license plate regions detected, and the final images containing the vehicle plate regions are obtained. The invention further provides a license plate detection device. Compared with existing methods, the license plate detection method and device have the prominent advantages that the false detection rate is low, and the speed is higher than that of the existing methods on the basis of guaranteeing the correct detection rate of license plates.

The invention discloses a method and system for non-complete flow detection and complete flow detection in a switchable mode. The method comprises the steps that firstly, a preset number of data packages on the front portion of a network data flow are obtained and restored and then virus detection is conducted on a part of the restored data; the data packages of the data flow are obtained continuously, if the data packages receiving fails or reception of the data packages of the data flow is accomplished, all the received data packages are restored, and virus detection is conducted on the received data packages. When the complete flow cannot be obtained, non-complete flow detection can be conducted and when the non-complete flow detection fails or if necessary, complete flow detection is conducted, so that system occupation is reduced, the detection efficiency is improved, and the detection ratio is guaranteed.

The invention provides an industrial Internet of Things oriented abnormal traffic grading detection method, which comprises two times of anomaly detection: detecting frequency anomaly of network traffic of the first stage; firstly, calculating a differential value of flow frequency, and smoothing; secondly, performing data fitting by adopting an exponential weighted moving average model to enable the data to accord with a statistical law, and performing deviation correction on the model to reduce an error generated by an initial value; and finally, providing a double-layer threshold interval method, and dividing the traffic into three types of traffic, namely steady-state traffic, suspicious traffic and dangerous traffic. Second-level flow attribute detection is carried out on suspicious flow found in first-level detection, a clustering optimization anomaly detection algorithm oriented to complex data attribute features is adopted, weighting distances and safety coefficients of data are classified according to priorities of the flow attribute features, data with the high safety coefficient is rapidly selected as a clustering center, and matching the multi-feature data around the center into a cluster, and finally using the cluster for attribute anomaly detection.

The invention discloses an elevator control system based on infrared detectors, which comprises a signal recording module, a human body detection module, a PLC (Programmable Logic Controller) module and a main dragging system module, wherein the signal recording module is used for recording calling signals and floor sorting signals of all floors, and sending the information to the PLC module; the human body detection module comprises a group of infrared detectors of all the floors, and is used for detecting human bodies in front of the door of an elevator, and sending the detection results to the PLC module for processing; the PLC module is used for analyzing the information sent by the signal recording module and the human body detection module, and sending control signals to the main dragging system module; and the main dragging system module is used for controlling the elevating of the elevator according to the control signals sent by the PLC module.

The invention discloses preparation of a hydrophobic low-temperature co-melt and a method for extracting Congo red, acid golden yellow and red 2G from meat products and aquatic products. The inventionbelongs to the technical field of food detection. In order to solve the problems of tedious steps, high reagent consumption, high possibility of causing respiratory tract injury to personnel and thelike in the traditional method, the method comprises the steps of mixing a hydrogen bond receptor and a hydrogen bond donor, heating until the mixture is completely changed into transparent clear liquid, and cooling to room temperature. The detection method comprises the steps of taking a dry sample, adding water for ultrasonic treatment, centrifuging and taking a supernatant; mixing, whirling andstanding the hydrophobic low-temperature co-melt and all the supernatant to generate oily liquid drops, sucking all the oily liquid drops, dissolving with acetonitrile, fixing the volume, and filtering with a microporous organic filter membrane to obtain a filtrate; and measuring the filtrate by using a liquid chromatograph-mass spectrometer. Under the condition that the detection rate is guaranteed, the detection efficiency is greatly improved, the detection cost is reduced, the test steps are greatly simplified, and the detection method is good in applicability and stability.

The invention provides a detection method and a device for a specified travel tool. The method comprises the steps of acquiring a road monitoring image collected by a camera; detecting a pedestrian position in the road monitoring image; if yes, extracting a pedestrian foot area image from the road monitoring image according to the detected pedestrian position; and identifying whether a specified transportation tool exists in the pedestrian foot area image. By automatically detecting the unsafe travel tool from the road monitoring image, the problem of going to the road of the unsafe travel tool can be effectively monitored in the traffic monitoring system, the labor cost is reduced, and the renovation management efficiency is improved. Due to the fact that the pedestrian using the travel tool is far away from the camera and the target in the image is small, the detection rate can be guaranteed and the missing detection probability of the small target can be reduced in the mode that thepedestrian is detected firstly and then the foot area image of the pedestrian is extracted.