54results about How to "Good attack performance" patented technology

The invention discloses a ship whole elastic-plastic movement response predication method and system under the action of underwater explosion. The method comprises the following steps: S1, simplifying a ship into a free ship body beam with an equal section; S2, dividing load pressure of the free ship body beam into five time phases under the action of the underwater explosion; S3, solving movement displacement of the free ship body beam in the five time phases under the action of the underwater explosion; S4, solving a bending moment of the ship body beam; S5, solving forward plastic movement parameters of the ship body beam; S6, solving backward elastic unloading movement parameters of the ship body beam; S7, solving backward plastic movement parameters of the ship body beam; S8, solving forward plastic movement parameters of the ship body beam. The invention further provides a system for realizing the method. According to the method and the system disclosed by the invention, the ship plastic movement process is analyzed on the basis of comprehensively considering the impact effect on the whole ship body by impact waves and air bubble loads, and engineering predication of ship whole elastic-plastic movement deformation under the action of the underwater explosion can be accurately and conveniently realized.

The invention relates to a DWT-SVD geometric attack prevention blind watermark method based on SIFT. The method comprises a watermark embedding method and a watermark extraction method. The watermark embedding method includes the steps that discrete waterlet transformation is conducted on an original image, a low-frequency sub-band of the original image is divided into small blocks, singular value decomposition is conducted on each small block, chaotic encryption is conducted on watermarks to be embedded, and the watermarks are embedded into the maximum singular value of each small block through an optimal quantitative method, the SIFT feature points of a watermark image are stored as secret keys, the types of geometric attacks are judged according to the secret keys, and correction is conducted. The watermark extraction method is the inverse process of the watermark embedding method and includes the steps of correction of attacked images, watermark extraction and watermark decryption and restoration. According to the method, by the utilization of the rotation, zoom and translation invariance of the SIFT feature points and the combination of the advantages of DWT and SVD in the digital watermark aspect, robustness on geometric attacks is remarkably improved.

Embodiments of the invention provide an adversarial sample generation method and device. The method comprises the steps of obtaining a to-be-enhanced current adversarial sample in the current iteration; in the decreasing direction of the target loss function, performing preset geometric deformation on the current adversarial sample for a first time to obtain a deformed image; executing second-timepixel-by-pixel updating on the deformed image to obtain a first adversarial sample; performing third pixel-by-pixel updating on the current adversarial sample to obtain a second adversarial sample; determining an adversarial sample with a smaller corresponding loss value in the first adversarial sample and the second adversarial sample as an updated adversarial sample; when the iteration stoppingcondition is met, taking the updated adversarial sample as a final adversarial sample; and when the iteration stopping condition is not met, performing the next round of iteration based on the updated adversarial sample. The generated adversarial sample can have stronger aggressivity, so that targeted defense is realized.

The invention discloses a ship whipping motion response prediction method and system under a combined effect of underwater explosive shock waves and bubbles. The ship whipping motion response prediction method comprises the steps that step 1, a ship is equivalent to a uniform cross section free ship body girder, and therefore the prediction of ship motion responses are equivalent to the predication of free ship body girder motion responses; step 2, the load pressure process of the free ship body girder under the underwater explosion effect is divided into five time phases, and a shock wave pressure peak value, a bubble pulsation phase negative pressure peak value and a bubble first time pulsation pressure peak value are collected; step 3, motion displacements in the five time phases of underwater explosion of the free ship body girder are solved. The invention further provides a ship whipping motion response prediction system. The ship whipping motion response prediction method and system comprehensively consider the shock effect of the shock wave and the bubble load on the whole ship, the underwater explosion load pressure is simplified to be five phases, the motion displacements of a ship in different phases are calculated, and therefore project prediction of ship elastic motion or whipping motion responses under medium and distal field explosion effect is achieved accurately, simply and conveniently.

The invention discloses a black box attack method for a brain-computer interface system. Firstly, generating EEG samples with more information amount and diversity based on inquiry synthesis active learning to substitute model training samples, then training an alternative model based on the training sample, so that the alternative model can better approximate a target model, and finally, generating an adversarial sample on the trained alternative model, and performing black box attack on a brain-computer interface system of the EEG by using the adversarial sample, so that the brain-computer interface system can be wrongly classified on the target model. Compared with a traditional black box attack method based on a Jacobian matrix, the method has a better attack effect. The same or betterattack performance can be obtained under the condition of fewer inquiry times, the generated countermeasure sample has very low noise, is almost not different from the original EEG signal in time domain and frequency domain and cannot be easily found, and the attack efficiency of the black box attack in the brain-computer interface system is greatly improved.

The invention discloses a matt-white immersion-gold-resistant, anti-cracking and light-sensing solder mask material and a preparation method thereof. The matt-white immersion-gold-resistant, anti-cracking and light-sensing solder mask material mainly comprises a main agent and a curing agent, wherein the main agent comprises acrylic resin with a special function, acrylic resin, a photoinitiator, a dispersing agent, dibasic ester, fumed silica, barium sulfate, titanium dioxide and an antifoaming agent; the curing agent comprises thermocuring resin, a curing additive, melamine, a photocuring monomer, triglycidyl isocyanurate, titanium dioxide and dibasic ester. The prepared matt-white light-sensing solder mask material has high immersion gold resistance and yellowing resistance, and is more stable and reliable during normal technology use of a PCB (printed circuit board).

The invention provides a method for preventing a community structure in a complex network from being deeply mined. The method comprises the following steps: (1) for a given network, initializing a population according to a set coding mode, sequentially selecting an initial population, crossing, mutating and reserving an elite to obtain an optimal individual, namely, an optimal attack strategy of acomplex network clustering algorithm; (2) according to the optimal individuals, adding or deleting the corresponding edges in the network to obtain the new network after camouflage so as to prevent the community structure in the complex network from being accurately mined. The method for preventing the community structure in the complex network from being excavated in depth provided by the invention effectively reduces the performance of the complex network clustering algorithm and guarantees the privacy safety of the community in the network.

The invention relates to a martial art arena robot going onto a stage in a tail slope mode. The martial art arena robot comprises a base plate part, a slope stage going part, an upper layer structure,a shovel attacking device, a sensor module layout, a battery part and a control circuit part. The martial art arena robot goes onto an arena with a certain height by using the tail slope; the stage going speed is high; and the success rate is high. A plurality of laser sensors and grayscale sensors are installed, so that the arena robot accurately detects the edge of the arena, fast recognizes anopponent, and sensitively gives a response. In a control aspect, a serial bus connection mode is used. In a structure aspect, the martial art arena robot has the advantages that materials are saved;the processing is easy; the production period is short; and the transformation is easy, and the like.

The invention provides a method for attacking a pedestrian re-identification system. The method comprises the following steps: training the pedestrian re-identification system by using a training data set; testing a pedestrian re-recognition system by using the query image before the attack; based on the mobility between the substitution model and a target model and the inquirability of the target model, black box attack is adopted to generate an adversarial sample for the pedestrian re-recognition system, and the adversarial sample is generated by adding disturbance to all query images of an attacked person; when the attacked query image is used for testing the pedestrian re-recognition system, the matching degree of the pedestrian re-recognition system and the query library image is reduced. The same disturbance is added to all images of an attacked person, so that a corresponding adversarial sample is generated, a new black box attack mode is formed, and the black box attack mode is more practical; time and labor are saved, and the attack effect is good.

The invention relates to a novel method for preventing DoS attack of a wireless disassociation frame based on a rough set, and provides a disassociation frame validity decision rule based on various types of characteristic attributes of an 802.11 frame, wherein the characteristic attributes mainly comprises nine parts: frame number difference, the type of a previous frame, a time interval between a frame and the previous frame, field control byte configuration, frame receiver address setting, frame sender address setting, BSSID setting, Reason Code setting, and Duration ID setting. The method comprises the steps: building a decision table for discriminating whether the disassociation frame is legal or not according to the research of WiFi wireless communication features and the attribute characteristics of the disassociation frame; carrying out the attribute reduction of a conventional method which employs nine condition attributes for discriminating whether the disassociation frame is legal based on the rough set theory, wherein the novel method just needs three condition attributes for judging the authenticity of the disassociation frame; and building a novel '802.11w-rough set filtering ' method, wherein the method can enable a WiFi LAN to effectively prevent the DoS attack of the wireless disassociation frame at low and high attack rates.