50results about How to "Does not affect recognition" patented technology

The invention belongs to the technical field of confrontation sample defense. The invention discloses an adversarial sample defense method and system based on VAE-GAN. A variational automatic encoderVAE and a generative adversarial network GAN are used to denoise an adversarial sample, the VAE is used as a preprocessing model of a classifier to denoise the adversarial sample, and the GAN is usedto assist training of the VAE, so that an image result output by the VAE is closer to an original noiseless image. The adversarial sample defense method based on VAE-GAN provided by the invention belongs to input preprocessing, and learning migration can be carried out among different classification models; the original classification network does not need to be retrained, and the training cost islow; the classification precision of the original noiseless sample is hardly influenced; an adversarial sample is not needed, so that an external training adversarial sample is not needed; the defense effect on adversarial samples with low noise is also very good; the preprocessing speed is high, and the output image quality is close to that of an original noiseless image.

The invention provides a stamp managing machine, which is provided with a casing, wherein the interior of the casing is provided with a support assembly and a controller, the support assembly is provided with more than two stamp assemblies, the support assembly is provided with a rotary disc, a first driving assembly, a code disc and a photoelectric sensor, the first driving assembly is used for driving the rotary disc to rotate, the stamp assemblies are arranged on the rotary disc, the code disc is positioned above the rotary disc and is synchronously rotated with the rotary disc, the periphery of the code disc is provided with more than two notches, and the photoelectric sensor is positioned at the outer side of the periphery of the code disc and is used for detecting the notches. The stamp managing machine has the advantages that the stamps can be effectively managed, the efficient and accurate stamping operation is realized, and the working process of selecting the stamps and stamping is smooth.

The invention provides a pressure touch sensor, a display device and a drive method for the same and relates to the technical field of display, wherein wrong recognition of pressure sensing can be prevented when the display device is bent. The pressure touch sensor comprises a drive electrode, a sensing electrode as well as a piezoelectric material layer and a piezo-resistance material layers disposed between the drive electrode and the sensing electrode.

The invention provides a password safety system (3000). The password safety system comprises a server (300) and a terminal (500), wherein the terminal (500) is used for receiving password information and images input by a user, processing the images, integrating the password information into the data of the images in a mode that the use of the images is not influenced, and returning the processed images; and the server (300) is used for receiving the processed images input by the user, processing the processed images input by the user, extracting the password information which is integrated into the processed images input by the user in the mode that the use of the images is not influenced, and returning the extracted password information. The invention also provides a password safety method.

The present invention discloses a novel mitochondrial genome editing tool, and belongs to the field of genome engineering. According to the prevent invention, the constructed mtCRISPR/Cas9 system mainly comprises two parts such as gRNA entering mitochondria and Cas9 nuclease localized in mitochondria, wherein the constructed mt-gRNA has two forms, the one mt-gRNA comprises a RNA mitochondrial localizing guide sequence, a targeting sequence and a gRNA skeleton sequence, the other mt-gRNA comprises a RNA mitochondrial localizing guide sequence, any one tRNA sequence encoded by mitochondrial or other additional spacer sequences, a targeting sequence and a gRNA skeleton sequence, the obtained combined material acts on the mitochondrial genome to break the target sequence in a targeted manner after the mt-gRNA and the mtCas9 nuclease are combined, and the action efficiency of the second mt-gRNA action is high than the action efficiency of the mt-gRNA; and the results verify that the constructed mtCRISPR/Cas9 system has characteristics of high efficiency and strong specificity.

The invention discloses a high-molecular material containing cholic acid and a drug-loaded nanoparticle modified by the same. The material has a structural formula as represented by formula III, and n in the formula is equal to 25 to 40. Monomers of the material are obtained by reacting hydroxy groups at C3 position of cholic acid methylester with 4-vinylbenzyl chloride and are initiated by an initiator so as to obtain a polymer precursor; ester groups at C24 position of the polymer precursor are hydrolyzed and turn into carboxyl groups so as to obtain the material in the invention. The material has amphipathy and is capable of assembling and of modifying the surface of the nanoparticles in the process of preparing the nanoparticles by using the method of emulsification-solvent evaporation. The molecular structure of cholic acid in the material can specifically bind to bile acid transporters on the surface of parenchymal hepatic cells; under the mediation of the bile acid transporters, the nanoparticles are absorbed by the parenchymal hepatic cells, and therefore, a drug is targetedly delivered to the parenchymal hepatic cells. Thus, the material provided in the invention can be used as an ideal surface modification material for development of a liver-targeting nanometer drug loading system.

The invention provides a polymer material, which is represented by the formula I and contains cholic acid, and a preparation method thereof. The invention further provides a liposome, which takes phosphatide and cholesterol as the carriers and is modified by the polymer material represented by the formula I. Drugs are coated in the phosphatide bi-molecule layer. The provided liposome has double functions of active targeting and passive targeting; and furthermore, the preparation method is simple and practical and can be applied to massive production. The in-vivo experiment results show that the liposome has a strong targeting performance on livers and thus has a good application prospect.

The invention discloses an intermediate for synthesizing benzothiadiazole malononitrile, a synthesis method thereof and a method for detecting CN<->. The intermediate for synthesizing the benzothiadiazole malononitrile is shown as a formula (III): FORMULA. The synthesis method of the intermediate shown as the formula (III) comprises the following steps: (1), synthesizing an intermediate TPA-BTD-Br; (2), synthesizing the intermediate TPA-BTD-CHO shown as the formula (III). The method for detecting the CN<-> comprises the following steps: (1), with tetrahydrofuran THF as a solvent, adding benzothiadiazole malononitrile TPA-BTD-BT to obtain a tetrahydrofuran THF solution of the benzothiadiazole malononitrile TPA-BTD-BT; (2), adding a sample to be detected into the tetrahydrofuran THF solutionof the benzothiadiazole malononitrile TPA-BTD-BT; (3) identifying whether the CN<-> exists in the sample to be detected through visual observation, UV-vis absorption spectra or fluorescence spectra.The intermediate shown as the formula (III) can be applied to construction of a fluorescent sensor with turn-on-type high selectivity and strong anti-interference ability to the CN<->.

The invention discloses a video frame transmission method, a video frame transmission system and a server, and the method comprises the steps: enabling a sending end to obtain original video data andreference network information fed back by a receiving end; determining a first code rate and a second code rate according to the reference network information; encoding a to-be-transmitted video framein the original video data according to the first code rate and the second code rate respectively to obtain a first video frame corresponding to the first code rate and a second video frame corresponding to the second code rate; obtaining current network information, and matching the first video frame, the second video frame and the current network information to obtain a matching result; and sending the first video frame or sending the second video frame or not sending the video frame to a receiving end according to the matching result. According to the technical scheme provided by the invention, the video fluency and stability are improved.

The invention relates to the field of artificial intelligence and automatic control and discloses a two-dimensional matrix dot matrix based on a BCD (Binary-Coded Decimal) code for tracing of a wheelhub. The two-dimensional matrix dot matrix based on the code for tracing of the wheel hub comprises a direction code, a check code and a plurality of code text units, wherein the direction code is located at the upper left corner of the two-dimensional matrix dot matrix and is used for representing an identification direction of the code text units; the check code is located at the last bit of thetwo-dimensional matrix dot matrix and is used for detecting the correctness of the code text units or correcting error-reading code text units; the plurality of code text units are sequentially located on remaining positions of the two-dimensional matrix dot matrix and is used for representing tracing information of the wheel hub; and the code text units, the direction code and the check code arein a 2*2 dot matrix structure. The two-dimensional matrix dot matrix for tracing of the wheel hub is simple in structure and high in identifiability; the two-dimensional matrix dot matrix can be increased and decreased, adjusted and deformed according to actual demands and is suitable for engraving and detection demands of the code texts under various complicated conditions, the check code is provided, thus the two-dimensional matrix dot matrix has a certain error prevention and error correction capabilities.

The variable-speed printing control algorithm of CIJ inkjet printer disclosed by the present invention includes: first setting the initial printing speed for a piece of printing data, and the initial printing speed can make the printing data clearly printed; If the data with different horizontal resolutions is required, the printing data is divided into the modified speed area and the constant speed area. For the printing data in the modified speed area, by inserting the corresponding printing ink drop between the printing ink drop and the printing ink drop Increase the spacing between printed ink dots by increasing the spacing between printed ink dots by more non-printing ink drops than in the speed constant area to reduce the printing speed in the modified speed area. The invention can effectively reduce the interference between ink dots in the high-resolution printing information and reduce the scattered ink around the information, improve the neatness and aesthetics of the printing content, and realize the best printing effect.

The protective film applied to the curved-screen mobile phone at least comprises a protective film layer, a functional coating, a composite film layer and a pressure-sensitive adhesive layer, whereinthe protective film layer is the outermost layer of the protective film; the functional coating layer is formed on the inner side of the protective film layer; the composite film layer is positioned on the inner side of the functional coating and is used for providing fitness suitable for the curved screen of the mobile phone; and the pressure-sensitive adhesive layer is positioned on the inner side of the composite film and is used for providing adhesiveness. The protective film disclosed by the invention not only has good fitting performance on a curved screen, but also is beneficial to fingerprint identification under the screen, and the use performance of the mobile phone protective film is greatly improved.

The invention discloses a packing bag two-dimensional code printing method. The packing bag two-dimensional code printing method comprises the following steps in sequence: (1) a decoration pattern is printed on the frontal surface of a packing material for manufacturing a bag body; and a blank position for printing a two-dimensional code is reserved in the decoration pattern; (2) the two-dimensional code is printed in the blank position; and meanwhile, a number same with the content of the two-dimensional code and a hidden fluorescent digital code are printed on the back surface of the packing material; (3) the frontal surface of the packing material is coated with a transparent white film; and the back surface of the packing material is coated with a hot-sealed inner film; (4) the packing material is fed in a curing chamber for heating and curing; and (5) the hot-sealed inner film is inwards formed to a packing bag or slit to a winding film.

The invention belongs to the technical field of document recognition, and particularly relates to a table or text line processing method, a bill processing method and device and a storage medium. The method mainly comprises the following steps: acquiring points on two opposite boundaries of a to-be-processed object, and respectively counting the points into a first boundary point set and a second boundary point set; obtaining a first boundary curve according to the first boundary point set, and obtaining a second boundary curve according to the second boundary point set; source control points with the corresponding number are extracted from the first boundary curve and the second boundary curve respectively; aligning the source control points to obtain corresponding target control points; obtaining an interpolation function according to each source control point, the target control point and the minimum energy function; and processing the text line region corresponding to the to-be-processed object by applying the interpolation function. The thin-plate spline interpolation technology is applied to processing of curved text lines or curved tables, and the problem that the correction effect is limited by training samples when a machine learning method is used for correcting curved texts in the prior art is effectively solved.

The invention discloses a durable paper money rolling belt which comprises an upper paper money rolling belt and a lower paper money rolling belt, and comprises photic areas and lightproof areas, wherein one end of the upper paper money rolling belt is twisted on an upper rolling wheel; the other end of the upper paper money rolling belt bypasses an upper reverse wheel, and is fixed on a paper money rolling drum; the photic areas I with certain length are distributed at the two ends of the upper paper money rolling belt, and the lightproof area I is arranged between the two sections of photic areas I; one end of the lower paper money rolling belt is twisted on a lower rolling wheel; the other end of the lower paper money rolling belt bypasses the lower rolling wheel, and is fixed on the paper money rolling drum; the lower paper money rolling belt comprises one section of lightproof area II and one section of photic area II, the length of the section of the lightproof area II is greater than the length of the section of the photic area II, and the lightproof area II is arranged at the end connected with the lower rolling wheel. According to the paper money rolling belt realized through the technological mean, whether the paper money rolling belt is attached with stains or not, sensors can accurately induct the starting position of paper money rolling and the stopping position of paper money rolling, ATM machine failures caused by the paper money rolling belt are obviously reduced, and when the frequently used end of the paper money rolling belt is abraded, the paper money rolling belt can also be used continuously after the two ends are exchanged, so that the service time of the paper money rolling belt is greatly prolonged, and the waste of the paper money rolling belt is also reduced.

The invention discloses a graphic identification code generation method for man-machine identification, belongs to the field of network security testing, and solves the problems that an existing verification code is easily and automatically identified and skipped by a robot trained based on a machine learning technology, and an existing picture confrontation sample generation mode is long in processing time and large in calculation scale. The method comprises the following steps: presetting a picture size, selecting a picture set based on the picture size, inputting the picture set and a classification result of the picture set into an adversarial sample generation system to generate disturbance, and adding the disturbance to a picture frame of the picture set to obtain an initialized picture adversarial sample set; based on the initialized picture confrontation sample set, optimizing interference by using a white-box training model, and generating a final picture confrontation sample, namely a picture interference frame; and adding the picture interference frame added to the to-be-interfered picture, and forming a picture identification code, namely a picture verification code, wherein the size of the to-be-interfered picture is the same as the size of the picture in the picture set. The method is used for generating the interference frame.

The invention relates to an express post address distinguishing method, computer equipment and a storage medium. The express post address distinguishing method comprises the steps: pre-storing at least one express post address; inputting an express address; converting the express address into a standardized address; calculating the similarity between each express post address and an optimized address; and sorting the at least one express post address according to the size sequence of the similarity to generate a predicted target address list. By standardizing and optimizing the express addressfilled by a consumer, the identification of the express address is not influenced even if the express address is not filled completely or is not standard, so that the correct delivery is ensured, thedelivery efficiency is improved, the delivery time is shortened, and the economic loss caused by delivery errors is avoided.

The invention discloses a one-time forming different-color double-line processing method. The method comprises the following steps: designing a double-core double-hole inner mold and a double-core double-hole outer mold; selecting an inner wire of a metal material and an outer wire of high-density PE and foaming PE; a single-color integrated double-spliced line is machined and formed through the inner mold and the outer mold; after the double-spliced lines are longitudinally distributed according to the double lines, glue is injected to the two transverse sides of one branch line through a vertical glue injection machine, and marking lines of different colors are achieved; according to the invention, twisted pairs of different colors are formed by twisting and twisting the processed double-spliced lines, and the integrated forming process of the same-color double-spliced lines is realized by adopting integrated forming processing, so that during construction and wiring, sub-lines cannot generate gaps and intervals due to bending, signal interference and signal attenuation cannot be caused, and the quality of the double-spliced lines is improved. The double-spliced cable is effectively used in the fields of medical treatment, network cables, aviation lines and the like, and after the double-spliced cable is twisted, the sub-cable of one of the double-spliced cables is displayed in a different-color identification manner, so that the identification is not influenced.