76 results about "Fingerprint alignment" patented technology

The invention relates to an embedded single safety chip biological fingerprint recognize system and relative method, wherein said safety chip contains chip operation system, internal micro processor, volatile memory and non-volatile memory, safety module and other manage module; the fingerprint collecting, processing and comparing modules are arranged inside the safety chip; the output of fingerprint sensor is connected to the fingerprint collecting module, the fingerprint processing module, and the fingerprint comparing module; and the inventive method comprises: collecting the original image information of fingerprint; extracting fingerprint character information to generate recognize code, to judge if initializing fingerprint register setting, and store recognize code into non-volatile memory of safety chip based on judged result, or extract the recognize code stored in the safety chip to be compared. The invention has high dual-factor safety and reliability, high stability and wider application.

The present invention belongs to the computer biometrics technology and relates to a method and a system which adopt certainty coding to accurately match fingerprints during the process of fingerprint comparison. The present invention has the accurate matching by using characteristic of the certainty coding that a characteristic area can be selected from the fingerprint image and the characteristic value corresponding to the characteristic area, such as the total length of fingerprint ridge line, the entropy of the main direction of the direction field of the ridge line and the direction field of the pixel point, to conduct matching work. The certainty coding method and the fingerprint identification system built by the method overcome the defects, such as low accuracy rate and potential safety problems which are caused by adopting fuzzy matching algorithm in the traditional point fingerprint identification system.

The invention provides a method for fingerprint alignment, a fingerprint acquisition device and a fingerprint alignment device, wherein the method for the fingerprint alignment comprises the following steps of: extracting a knuckle line from a two-dimensional fingerprint image; calculating a position relation between the extracted knuckle line and a prestored knuckle line; and aligning the extracted knuckle line with the prestored knuckle line according to the position relation so that a fingerprint containing knuckle texture aligns with a prestored fingerprint. The method for the fingerprint alignment performs contrast calculation on the extracted knuckle line and the prestored knuckle line by extracting the knuckle line, and acquires an angle between the extracted knuckle line and the prestored knuckle line, thereby achieving the aim of aligning fingerprints by rotating the angle. The method only needs to realize the coincidence of the extracted knuckle line and the prestored knuckle line, avoids the complexity of the process that the prior method for the fingerprint alignment needs to match minutiaes, shortens the time of the fingerprint alignment, and improves the efficiency of the fingerprint alignment.

The invention relates to a fingerprint eigenvalue encryption-decryption system which comprises a fingerprint sensor, a nonvolatile store and a fingerprint encryption-decryption chip. The fingerprint encryption-decryption chip comprises a special interface of the fingerprint sensor, a fingerprint matching module, a fingerprint eigenvalue extracting module, a special channel of the fingerprint matching and the fingerprint eigenvalue extracting, a bus interface of the fingerprint eigenvalue extracting module, a fingerprint comparing module, a bus interface of the fingerprint comparing module, an interface of the nonvolatile store, a controller, a bus interface of the controller, a communication port, an internal bus. The fingerprint encryption-decryption chip also comprises an encryption-decryption module, a one time programmable program store, a bus interface of the encryption-decryption module and a bus interface of the one time programmable program store. The invention can improve the safety of the fingerprint eigenvalue and enhancing the safety level. The integration of the fingerprint eigenvalue extracting, the fingerprint comparing and the encryption-decryption reduces the cost of the system; the system solves the safety problem of remote transmission of the fingerprint eigenvalue, thus leading a fingerprint database to be easily centralized and conveniently managed and searched.

To enable extraction of a specific area (pattern area) that includes a characteristic part of the inherent pattern of individual fingerprint from an inputted fingerprint image of a human being or other creature. The pattern area is extracted as the minimum area that is surrounded by a right pattern area slope, a left pattern area slope, and pattern area bases. The right (left) pattern area slope is a fingerprint ridge which runs towards the outside on the right (left) side from the start point that is located on the upper side of the center point of the fingerprint image, and satisfies a prescribed condition. The pattern area base is a fingerprint ridge which runs on the lower side of the center and satisfies a prescribed condition.

Provided is a motor fingerprint identification device and a method, wherein a fingerprint identification module and a controller which are connected to a firing circuit of the motor are arranged on a car body of the motor, and main control fingerprints are presupposed and stored in the module. When a user starts the fingerprint identification module and executes locking action, the next time when starting the motor, the user is required to perform fingerprint alignment identification, if the alignment result matches with the fingerprint used for locking originally, the motor is allowed to start, otherwise start is inhibited, thereby preventing the motor from thieving.

The invention discloses a multi-path depth encoded information fingerprint extraction method based on the bit space. The method comprises a first step of constructing a bit window; a second step of constructing a bit plane; a third step of reducing dimensions and constructing a bit chain; a fourth step of encoding the bit chain; and a fifth step of comparing fingerprints. The fourth step includes conducting transversal and statistics for the constructed bit chain with reduced dimensions, acquiring the number of continuous 0 and the number of continuous 1 in sequence, and constructing a new decimal sequence of number; and binarizing the obtained decimal sequence of number to obtain a new binary bit chain with the first bit not being 0, repeating the transversal and statistics operation for the obtained bit chain, carrying out loop iteration in sequence until the newly obtained decimal sequence of number has only one element, and recording the value of the element obtained at last and the number of times of loop iteration as a feature value of an information fingerprint feature space The invention has the advantages of improving the efficiency in the same information detection and the like.

The invention relates to a small-area fingerprint image feature extraction method and system, a terminal and a storage medium, belongs to the technical field of fingerprint identification, is appliedto small-area fingerprint feature information extraction, and comprises the following steps: S1, acquiring a fingerprint image; s2, preprocessing the fingerprint image, separating a foreground regionand a background region of the fingerprint image, obtaining data information of the foreground region, and performing classification filtering processing on the fingerprint image according to the datainformation of the foreground region; s3, extracting key feature points, performing block division on the fingerprint image, selecting the key feature points according to data information of a fingerprint image foreground area in a corresponding block, wherein the data information of the fingerprint image foreground area comprises curvature information of pixel points; and step S4, screening of key feature points: screening out the key feature points with good stability through comparative analysis. The method is used for solving the problem of insufficient extraction of small-area fingerprint feature information, is effectively suitable for small-area fingerprint comparison and has good reliability.

The invention relates to the technical field of chemical detection of cigarettes, and especially provides an identification method for oil stained cigarettes. The GC/MS identification method for oil stained cigarettes comprises the following steps: 1) collecting a plurality of types of oil directly related to production of cigarettes and respectively dissolving and extracting the types of oil with ethyl acetate; 2) adjusting GC/MS parameters, carrying out gas phase/mass spectrometry detection on the extracted solutions of oil, establishing different chromatograms to form a chromatogram database, and determining fingerprints; 3) taking cigarette papers having oil stains, immersing the cigarette papers with ethyl acetate, carrying out extracting, concentrating the extract with a distillation apparatus, and carrying out gas phase/mass spectrometry on the concentrate to obtain a chromatographic fingerprint; 4) comparing fingerprints of the chromatographic fingerprint of oil stained cigarettes with fingerprints in the established chromatogram database to identify which oil stain is on the oil stained cigarettes. The invention enables fast and effective identification of oil stains generated in the production of cigarettes, provides a basis for identifying a problematic process in the production of cigarettes, and allows problematic cigarettes to be traceable.

The invention provides an application system fingerprint login method and system based on a domestic CPU and an operating system, belonging to the field of fingerprint authentication technology, the method is based on domestic CPU and operating system environment; fingerprint collection and fingerprint comparison services which can be operated independently are provided; fingerprint registration and fingerprint authentication login of an application system user are realized; a complete rich service interface required by a service process from fingerprint collection, fingerprint registration to initialization, addition, deletion and fingerprint comparison verification of a 'fingerprint template library 'is provided for the outside; encrypted transmission is carried out on the fingerprint template in the transmission process, and a simple, convenient, safe and stable application system login solution based on a domestic CPU and an operating system environment is provided.

The invention discloses an image big data similarity comparison method, and the method comprises the steps: capturing the image data to be compared through a crawler capture technology and big data search, analyzing and processing the captured image data and storing the image data in a structured mode, and performing classification comparison according to attributes of images, wherein the comparison process is that the features contained in the images are calculated a set of fingerprints are generated and the fingerprints of the images are compared to judge the similarity of the images; therefore, the image recognition efficiency can be improved, the method can be combined with big data, and the comparison of the similarity of the images in the big data range is achieved.

An embodiment of the application discloses a fingerprint storage method, and an authentication method, device and system, and the fingerprint storage method comprises the following steps of a smart door lock monitors an ID card information input event, and obtains a user's ID card information from the ID card information input event, wherein the ID card information includes the user's ID number and fingerprint information; the smart door lock obtains an authentication result of the ID number; and if the authentication result indicates that the ID number has pass the authentication, the smart door lock stores the fingerprint information, wherein the fingerprint information is used to trigger the smart door lock to be unlocked when the comparison with the collected request unlock fingerprintindicates a success. The above scheme is used to not only realize the user self-entering of the fingerprint and reduce the management cost, but also reduce the risk of remote transmission of the fingerprint. Compared with the direct fingerprint entry method, the invention is simpler and can guarantee the integration of person and witnesses.

The invention relates to an interference source individual identification method based on envelope fingerprint features and belongs to the technical field of target identification. The method comprises the following steps of 1) establishing a radar interference source signal model; 2) establishing a simulation model of a wireless channel; 3) extracting a leading edge feature enveloped by a radar interference source signal as a fingerprint feature identified by a radiation source individual; 4) repeating the step 3), extracting the envelope leading edge feature of each radar interference source signal, and establishing an envelope fingerprint feature library; 5) processing the radar echoes to obtain envelope fingerprint features of the standard radar echoes; 6) identifying active interference signals, and classifying the active interference signals into suppressing interference and deceptive interference; and 7) receiving and analyzing the echo envelope fingerprints of the deceptive interference identified in the step 6), and comparing the echo envelope fingerprints with the envelope fingerprints in the envelope fingerprint feature library in the step 4). The method is simple and convenient; scientific basis is provided for electronic system design, and identification efficiency is improved.

The invention discloses an efficient data deletion method based on source end deduplication. In the backup process, a data stream of a source end is segmented into data blocks, fingerprints are calculated, the fingerprints are compared, if the fingerprints do not exist, it is indicated that the data stream is a new block, the corresponding data blocks are transmitted to a container of a server tobe stored, the corresponding container is marked as 1, the container is written into a data file after the container is full, and then a new container is created; automatically clearing up the expiredbackup set, and clearing up the guid object record; and performing data block and fingerprint cleaning on the container marked as 0 by utilizing a preset cyclic deletion logic in idle time beyond thenormal service window period, wherein the container marked as 0 indicates that the data block and the fingerprint in the container can be cleaned when not being referenced. The method has the advantages that a marking mode is adopted, the statistical logic is simpler, the cleaning logic is not influenced by the size of the deduplication library, and the method is more efficient.

The invention discloses an integrated computer based on an Shenwei 421 processor and relates to the technical field of computers. The computer comprises a case, wherein a mainboard, a hard disk, a display screen, a biological recognition module and a gravity sensing module are arranged in the case, an SW 421 processor, a CPLD control chip, a memory, domestic UEFI firmware and a display card are arranged on the mainboard, the display screen is embedded in a front panel of the case, and the display screen is connected with the display card and used for displaying a user interface; the biologicalrecognition module is used for carrying out biological recognition to determine whether to log in the operating system or not and comprises an image collector and a fingerprint recognition module, the image collector is used for shooting an image of a user for face recognition, and the fingerprint recognition module is used for collecting fingerprints for fingerprint comparison and recognition; the gravity sensing module is used for sensing rotation of the case and sending out a screen rotation instruction to adjust the display direction of the display screen.

The invention discloses a web system data transplantation prevention method based on an asymmetric encryption algorithm. The method comprises the following steps: S1, a user logs in a system through a browser 1, reads a unique data calculator machine code and calculator feature information of a client through a client fingerprint generator to generate a client fingerprint, and stores and binds the client fingerprint with the user and the browser; S2, the client fingerprint is combined with the current time and the public key to encrypt and generate a string of character strings, and the character strings are client signatures; S3, a server interface is visited, and when a normal interface accesses a server, a client signature and a server private key are carried for encrypting to obtain a fingerprint and time; S4, the parameter is copied to access the server interface through the browser 2, and if the verification of the client fingerprint verifier is not passed, the service response fails. The method has the characteristics of preventing pretending, authenticating identity and preventing parameter tampering, and a timestamp comparison and fingerprint comparison technology exists, so that replay attacks can be prevented when the data of the previous timestamp is invalid.

The invention provides a picture fingerprint comparison method and a device based on a large integer type, equipment and a medium, and the method comprises the steps: setting a large integer type with a user-defined byte length to store a picture id of M bytes and a picture fingerprint of N bytes, and forming storage data of at least M + N bytes; arranging the storage data according to a sequence from small to large, and sequentially storing the storage data into corresponding files with different ranges and sizes; obtaining to-be-compared storage data stored in the same way, and positioning a target file matched with the range size of the to-be-compared storage data according to the high X bits of the N bytes of the fingerprint of the to-be-compared storage data; if target storage data with the same high X bits are stored in the file, counting the M-bit picture id corresponding to the target storage data to be hit by the fingerprint for one time. According to the method, a large integer type with an adjustable length is set, so that storage and comparison are facilitated, and the efficiency of picture storage and picture fingerprint comparison is greatly improved.

The embodiment of the invention discloses an unlocking method, which is used for realizing a double insurance mechanism of a service library so as to improve the use safety of the service library. The method provided by the embodiment of the invention comprises the following steps: when the central control board receives a fingerprint comparison passing notification sent by a bank background, the central control board receives random number information from a target lock body; the central control board sends the random number information to the password host to generate unlocking code information of the target lock body; and the central control board sends the unlocking code information to the bank background, so that the bank background carries out unlocking processing on the target lock body.

The invention belongs to the technical field of fingerprint comparison, and relates to a type I diabetes risk assessment system based on a deep neural network, which comprises a data input module, a site acquisition module, a site screening module, a model training module and a model inspection module, wherein the data input module is used for inputting or collecting sample data; the locus acquisition module is used for acquiring genotype information of whole genome loci of the sample data; the site screening module is used for carrying out GWAS statistics on the genotype information, carrying out quality control on a GWAS statistical result and screening sites according to an analysis result after the quality control; the model training module is used for training a neural network model according to the screened sites to obtain an optimal model; and the model inspection module is used for inspecting the optimal model. Under the condition that no experiment cost is increased, screening of high-genetic-risk patients is achieved as early as possible on a large scale, cost is low, accuracy is high, and operation is easy and convenient.

The invention discloses an elevator taking person recognition and elevator control method and device based on a neural network. The method comprises the steps of collecting an elevator scene image and a fingerprint image of an elevator calling person; respectively performing face detection and fingerprint feature detection on the scene image and the fingerprint image; decomposing the detected face image into a low-frequency sub-band image, a horizontal edge sub-band image and a vertical edge sub-band image; identifying the identity characteristics of the elevator taking personnel through the decomposed images, and performing fingerprint comparison on the detected fingerprint characteristics; and when identity feature recognition and/or fingerprint feature recognition fails, elevator taking personnel and/or elevator calling personnel are judged as specific personnel, saving face images and/or fingerprint images of the specific personnel in a database, and when fingerprint comparison succeeds, activating an elevator lifting key to change the locked state of the elevator into the usable state. The use safety of the elevator is improved.