69results about How to "Solve the high false positive rate" patented technology

Scanning of computer files for malware uses a classifying technique to classify an input file as a clean file or a dirty file. The parameters of the classifying technique are derived to train the classification on a corpus of reference files including clean files known to be free of malware and dirty files known to contain malware. The classification is performed using a representation of the files in a feature space defined by a set of predetermined features for respective file formats, the features being a predetermined value or range of values for one or more data fields of given meanings. The representation of a file is derived by determining the file format, parsing the file on the basis of the structure of data fields in the determined file format to identify the data fields and their meaning, and determining, on the basis of the identified data fields, which of the set of predetermined features are present.

A malicious behaviour detector (100) for detecting malicious behaviour on a network, comprises a processor unit (120) and associated system memory (130) containing computer program code. The computer program code provides a signature matching module (132) to perform malicious partial signature detection by reading the contents of packets of data passing through the network to look for partial signatures associated with malicious programs; a Domain Name Service, DNS, request and/or response detection module (134) to monitor the requests made by hosts connected to the network and/or responses thereto; and an evidence assessment module (138) to analyse the results of the partial signature detection and the DNS monitoring make a determination of the suspected presence of malicious behaviour on the network based upon the analysis of the results of both the partial signature detection and the DNS monitoring.

Disclosed is a method and system for detecting malicious entities and malicious behavior in a time evolving network via a graph framework by modeling activity in a network graph representing associations between entities. The system utilizes classification methods to give score predictions indicative of a degree of suspected maliciousness, and presents a unified graph inference method for surfacing previously undetected malicious entities that utilizes both the structure and behavioral features to detect malicious entities.

A hierarchical data structure of digested payload information (e.g., information within a payload, or information spanning two or more payloads) allows a payload excerpt to be attributed to earlier network flow information. These compact data structures permit data storage reduction, while permitting efficient query processing with a low level of false positives. One example of such a compact data structure is a hierarchical Bloom filter. Different layers of the hierarchy may correspond to different block sizes.

The invention relates to a decision model used for detecting malicious programs and a detecting method of the malicious programs. The detecting method includes rule generation and machine learning algorithm training, the rule generation is characterized in that a training sample set composed of a 'malicious program' sample set and a 'non-malicious program' sample set is generated into a decision rule set in an API (application programming interface) calling layer, a basic abstraction layer and a business abstraction layer; the decision rule set includes an API calling layer decision rule, a basic abstraction layer decision rule and a business abstraction layer decision rule; the machine learning algorithm training is characterized in that an OCSVM algorithm training model is utilized to obtain an OCSVM algorithm model. The malicious program related detecting method is characterized in that the programs to be detected are subjected to decision by the aid of the related decision model. According to the technical scheme, multilevel and systematic features are achieved, false detection can be effectively controlled in the steps, and feasibility of application of the malicious program behavior detection in actual computer security is improved.

Methods and systems for non-intrusively detecting the existence of fissile materials in a container via the measurement of energetic prompt neutrons are disclosed. The methods and systems use the unique nature of the prompt neutron energy spectrum from photo-fission arising from the emission of neutrons from almost fully accelerated fragments to unambiguously identify fissile material. The angular distribution of the prompt neutrons from photo-fission and the energy distribution correlated to neutron angle relative to the photon beam are used to distinguish odd-even from even-even nuclei undergoing photo-fission. The independence of the neutron yield curve (yield as a function of electron beam energy or photon energy) on neutron energy also is also used to distinguish photo-fission from other processes such as (γ, n). Different beam geometries are used to detect localized samples of fissile material and also fissile materials dispersed as small fragments or thin sheets over broad regions. These signals from photo-fission are unique and allow the detection of any material in the actinide region of the nuclear periodic table.

The invention discloses a patch security detecting method and system for software update.The method comprises the steps that by means of conducting preprocessing on patches, intermediate code analysis is conducted to obtain basic blocks affected by the patches, memory sensitive functions and variables thereof are obtained by means of the basic blocks affected by the patches, and a main execution path is constructed by starting from the memory sensitive functions; in the process of conducting symbolic execution on the path, the specific path condition is determined, and when the specific memory sensitive function is reached, the path condition at this time is combined with a corresponding security detector to judge whether a software update security problem exists or not.The patch security detecting method and system for the software update focus on detecting the security of different parts, relative to an older version, of the new version after software is updated, the problems that in an existing patch security detection scheme, the false positive rate and false negative rate are high, the efficiency is low due to duplicate detection, and overdependence on test set effectiveness exists are solved, the detection efficiency is significantly improved, and the accuracy of patch security detection is further improved.

The invention belongs to the technical field of human body distinguishing and discloses a human body heat source feature extracting and distinguishing method based on infrared pyroelectric information. The method aims to achieve 360-degree wide-range and remote detection and detection of static infrared heat sources and effectively solves the problems of confliction between comfort and energy saving of an intelligent air-conditioner and high false alarm rate of a safety supervision system. According to the technical scheme, the method comprises the steps that a stepping motor is adopted to drive a single pyroelectric infrared detector to rotate at a uniform speed, and then remote, 360-degree wide-range and static object detection is achieved; the detection range is a disk shape, wavelet packet analysis is conducted on collected human body heat source samples and non-human-body heat source samples, a signal wavelet entropy is taken as the features of the signal, 5-fold cross-validation is conducted by means of the BP neural network, and then human body heat resources are distinguished from non-human-body heat resources. The method is mainly applied to human body distinguishing.

The invention discloses a website security detection method, website security detection equipment and a website security detection system. The method includes the steps: according to a web address of a website, constructing a first detection web address relevant to the website; accessing to the website by the aid of the first detection web address, and detecting whether the website has vulnerabilities or not on the basis of return results of the website; under the circumstance that the detection results indicate that the website has the vulnerabilities, constructing a second detection web address corresponding to the first detection web address based on the first detection web address; and accessing to the website by the aid of the second detection web address, and verifying whether the website has the vulnerabilities or not on the basis of the return results of the website.

The invention discloses a software defined network security situation assessment method. The advantages of SDN (software defined network) control and direct and quick information collection are combined, an open SDN security situation assessment framework is provided on account of three kinds of attacks of an SDN forwarding surface, the security framework closely fits a framework of an SDN controller, an anomaly detection module extracts characteristic indexes according to the SDN and various attack characteristics, an SVM (support vector machine) classification algorithm is selected for identification, and attack pre-judgment is given. A security situation assessment module performs quantitative assessment on network security situations according to information collected by the anomaly detection module, and attack sensitivity and noise resistance of an assessment system are regulated through setting of a threshold value. Different weights are allocated to different attacks based on an AHP (analytic hierarchy process) so as to fit out a comprehensive security situation of the network. The method is flexible and simple, attach behaviors can be detected accurately, the quantitative assessment of the security situations of the network is given, and monitoring and assessment of the security status of the SDN forwarding surface are realized with lower cost.

A method of quickly detecting a handmade explosive in a bottle with a low erroneous alarm frequency is provided. A sample gas generated from a bottle placed on a bottle placement space is sucked-in, and ions of the sample gas are generated by an ion source and subjected to mass analysis. The presence/absence of a mass spectrum derived from the handmade explosive is determined from an obtained mass spectrum, and the result thereof is displayed on a monitor, thereby quickly detecting the handmade explosive in the bottle or the handmade explosive adhering to the surface of the bottle at a low erroneous alarm frequency.

The present invention discloses a combined detection method of electroencephalogram epilepsy sharp and spike wave discharge based on long short term memory network (LSTM) multi-channels. The combineddetection method comprises the following steps: step 1, filtering input original multi-lead electroencephalogram and eliminating artifacts caused by physiological activities of electrocatdiogram, chewing and swallowing; segmenting processed signals in time domain firstly according to characteristics of target waveform duration characteristics and transforming the signals into a recognition form ofthe subsequent steps; step 2, conducting feature extraction of data of each channel in the segmented signals through the long short term memory network and conducting feature fusion through an adaptive weighted fusion algorithm; and step 3, using results obtained by the feature fusion to classify multi-channel signal fragments through fully connected neural network and finally obtaining classification results of the entire signal at different periods, so as to achieve a purpose of sharp and spike wave discharge detection. The combined detection method can realize a sharp and spike detection effect with higher precision and stronger anti-interference ability under multi-channel signal input.

The invention discloses a method applied to recognition of a flame image generated by gas combustion associated in oil drilling, relating to the field of image processing technologies. According to the color characteristic and the motion characteristic of the flame image, and by combination with discrete fractal Brownian random field, the flame of gas combustion associated in oil drilling is recognized. The method disclosed by the invention can be applied to monitoring the flame of gas combustion gas combustion associated in onshore oil drilling platforms and offshore oil drilling platforms, and also can be used for monitoring the whole safety of the oil drilling platform by combination with an oil-well monitoring system.

The invention provides an antenna test method and system and a wireless terminal, and relates to the field of antennae. The antenna test method and system and the wireless terminal can reduce the test error rate. The method is applied to the wireless terminal. The wireless terminal comprises an antenna and a test circuit, wherein the antenna is electrically connected with one end of the test circuit, and test equipment outside the wireless terminal is electrically connected with the other end of the test circuit. The method comprises the steps that the test circuit receives test signals sent by the test equipment and transmits the test signals to the antenna, and the test circuit receives reflected signals which are reflected by the antenna and correspond to the test signals and transmits the reflected signals to the test equipment so that the test equipment can determine parameters of the antenna according to the reflected signals. The antenna test method and system and the wireless terminal are applied to the antenna test technology.

The invention discloses an encrypted traffic identification method and device and equipment, an AI model is used to identify encrypted traffic, and the training process of the AI model comprises the following steps: S100, establishing a plurality of primary AI training models related to different features of a data stream, and training the primary AI training models through the data stream; S200, fusing the primary AI training models by using a multi-model fusion set algorithm; S300, performing secondary AI training by using a supervised learning algorithm based on a fusion result of S200, and obtaining a secondary AI training model; and S400, when an output result obtained after target data flow flows through the secondary AI training model is out of a threshold range of a normal output result of the secondary AI training model, determining that the target data flow is malicious flow. The encrypted traffic is detected by using a multi-dimensional and multi-model method, and the problems of high false alarm rate and poor interpretability caused by dependence on a single AI model are well solved.

The invention is applicable to the field of communication and provides a secondary dialing method based on a wireless access system and wireless access equipment. The method comprises the steps that DTMF signals are produced through secondary dialing; early-stage control signals are generated according to the DTMF signals; high-frequency signals in the DTMF signals are filtered according to the early-stage control signals so that resolution of the DTMF signals can be blocked; an uplink audio frequency channel is controlled and closed according to the DTMF signals, and an AT instruction is generated; the AT instruction is subjected to signal processing and then transmitted to a network end. According to the secondary dialing method based on the wireless access system, a digital filter in the uplink audio frequency channel in a wireless communication unit is used for finishing blocking the DTMF signals of the uplink audio frequency channel so that a receiving terminal of an opposite side can not recognize the DTMF signals transmitted to the receiving terminal through the audio frequency channel, and the problems that the wireless access equipment is low in success rate, high in false alarm rate and prone to being disturbed by wireless radio frequency in the secondary dialing process are solved thoroughly; the method has the advantages of being economical, practical, simple and effective.

The invention relates to primers and probes for detection of various respiratory viruses and provides a primer-probe assembly for detecting cytomegalovirus, EB (epstein-barr) virus and adenovirus. Sequences of a probe and a primer pair for detecting the cytomegalovirus, the EB virus and the adenovirus are as shown in SEQ ID NO:1-9. The invention further provides a kit for simultaneous detection of the cytomegalovirus, the EB virus and the adenovirus. By combination of multiplex PCR (polymerase chain reaction) and Taqman fluorescent probe technique, simultaneous detection of various pathogens possibly causing acute respiratory infection can be realized in one-time detection, and quickness in diagnosis and treatment guide is realized. In addition, shortening of operation time and reduction of raw material consumption are realized, and high sensitivity and specificity, effectiveness in solving of PCR pollution, high automation degree and the like are achieved.

The invention relates to a liquid medium leakage automatic detection method and system based on thermal imaging. The system is characterized by comprising an image collection module which is used for obtaining a frame of thermal image of a to-be-detected region, and enabling the frame of thermal image to serve as a background image if the frame of thermal image is a first frame of infrared image; the background subtraction module is used for extracting a dynamic change region of the current frame of thermal image by adopting a background subtraction method; the processing stage setting module is used for setting a processing stage; the parameter calculation module is used for calculating parameters of the thermal image; the gas leakage judgment module is used for judging whether liquid medium leakage occurs in the to-be-detected area or not when the calculated parameters of the thermal image meet preset detection conditions; the alarm module is used for outputting an alarm signal when liquid medium leakage occurs in the to-be-detected area, and marking an extracted dynamic change area in the corresponding frame of thermal image; the hazardous chemical substance leakage detection system and the hazardous chemical substance leakage detection method can be widely applied to the field of hazardous chemical substance leakage detection.

The invention discloses magnet polarity detection equipment. The equipment comprises a feeding platform, a Hall detection platform, a stock bin platform, a manipulator and a control system; a Hall sensor for detecting the magnet polarity is arranged on the Hall detection platform; the manipulator is used for grabbing a to-be-detected product from the feeding platform, placing the to-be-detected product on the Hall detection platform, and grabbing the qualified product and placing the same on the stock bin platform after the to-be-detected product placed on the Hall detection platform is detected to be qualified; and the control system is used for controlling the manipulator to act, receiving an induction signal output by the Hall sensor after detection, judging the magnetic pole mounting direction of a magnet in the to-be-detected product according to the induction signal, and determining the to-be-detected product with the correct mounting direction as the qualified product. Accordingto the magnet polarity detection equipment, the polarity of the magnet arranged in the to-be-detected product can be automatically detected, and whether the magnetic pole of the magnet is reversely mounted in the to-be-detected product or not can be automatically recognized.

The invention discloses a fault decision-making method for a payload, and relates to the field of spacecrafts. The method comprises the steps that operation parameters of a payload are acquired; Establishing a decision-making model, and training the decision-making model according to the operation parameters and the Markov model, the decision-making model being a deep enhanced network model; And making a decision on the fault of the effective load according to the trained decision model. The invention provides a fault decision-making method, the depth enhancement network model is trained through the operation parameters and the Markov model; and fault diagnosis, Inference and decision are combined, According to the method, the complete cycle process from fault discovery to fault decision making of the spacecraft effective load is achieved, the problem that under the influence of multiple factors, the complex jump parameter relevance and the false detection rate are high is solved, probability fuzzy diagnosis and dynamic updating of the context logic relation are achieved, and the accuracy and timeliness of fault diagnosis and decision making are improved.

The invention discloses a detection method, a device and a detection device of a corn stalk. The method comprises the following steps: obtaining an image of the corn to be detected; Inputting the cornimage to be detected into a preset network model; Wherein, the preset network model is obtained by using a convolution neural network to train sample images of several corn stalks; The convolution neural network comprises at least one convolution layer, at least one activation layer, at least one pooling layer and at least one full connection layer. Detecting the corn stalk in the corn image to be detected based on the preset network model; Output Detection Effect Image. This method takes the sample image of corn tassels under the farmland as the object, The preset network model based on theimage feature detection of corn stalks is established by using the image features of corn stalks in the sample images. The preset network model can accurately detect the corn stalks in the corn imagesto be detected, which greatly reduces the false detection rate of corn stalks.

The invention discloses a comprehensive perimeter security and protection system and method with infrared thermal image verification. Compared with an existing security and protection system, the security and protection system in the invention has the advantages that an image acquisition device and an image recognition unit are additionally arranged; an alarm signal of the infrared correlation device is confirmed and verified, a controller controls an alarm to give an alarm only after successful verification is confirmed, and at the moment, the problem that the false alarm rate of the infraredcorrelation device is high due to the influence of fallen leaves or rain and snow and other external environments mentioned in the background technology can be effectively solved. The security and protection method is completed based on the security and protection system and has the advantages of being simple, high in operability, high in alarm accuracy and the like.

The embodiment of the invention discloses a detection method and system for micro bar materials. The method comprises the steps that a feeding system conveys the micro bar materials to be detected to a before-jumping carrying system; the before-jumping carrying system conveys the micro bar materials to a jumping detection system; the jumping detection system carries out jumping detection on the micro bar materials, and a jumping detection result is obtained; a cylindrical detection system carries out cylindrical detection on the micro bar materials obtained after jumping classifying, and a cylindrical detection result is obtained; and a defective-product sorting system carries out classified statistic on the detected micro bar materials according to the jumping detection result and the cylindrical detection result. According to the method and system, precision and repeat reliability of the detection result can be improved; in addition, the detection efficiency can be improved, different kinds of failed test samples can be subject to classified statistic, the problems that in the prior art, when manual detection is adopted, the false drop rate is high, the yield is low, and detection reliability is low are solved, and the problem that in the prior art, surface defective detection cannot be carried out on the micro bar materials can also be solved.