WO2021031445A1 - 一种基于三维动态特征的离线笔迹个体识别系统及方法 - Google Patents

一种基于三维动态特征的离线笔迹个体识别系统及方法 Download PDF

Info

Publication number
WO2021031445A1
WO2021031445A1 PCT/CN2019/122176 CN2019122176W WO2021031445A1 WO 2021031445 A1 WO2021031445 A1 WO 2021031445A1 CN 2019122176 W CN2019122176 W CN 2019122176W WO 2021031445 A1 WO2021031445 A1 WO 2021031445A1
Authority
WO
WIPO (PCT)
Prior art keywords
handwriting
feature
suspicious
image
sample
Prior art date
Application number
PCT/CN2019/122176
Other languages
English (en)
French (fr)
Inventor
陈晓红
杨旭
王雅晨
王楠
卢启萌
Original Assignee
司法鉴定科学研究院
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 司法鉴定科学研究院 filed Critical 司法鉴定科学研究院
Priority to US16/968,833 priority Critical patent/US20230245501A1/en
Publication of WO2021031445A1 publication Critical patent/WO2021031445A1/zh

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/24—Classification techniques
    • G06F18/241—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches
    • G06F18/2411—Classification techniques relating to the classification model, e.g. parametric or non-parametric approaches based on the proximity to a decision surface, e.g. support vector machines
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/30—Writer recognition; Reading and verifying signatures
    • G06V40/37—Writer recognition; Reading and verifying signatures based only on signature signals such as velocity or pressure, e.g. dynamic signature recognition
    • G06V40/376—Acquisition
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T7/00—Image analysis
    • G06T7/20—Analysis of motion
    • G06T7/246—Analysis of motion using feature-based methods, e.g. the tracking of corners or segments
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/20—Image preprocessing
    • G06V10/34—Smoothing or thinning of the pattern; Morphological operations; Skeletonisation
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/30—Writer recognition; Reading and verifying signatures
    • G06V40/37—Writer recognition; Reading and verifying signatures based only on signature signals such as velocity or pressure, e.g. dynamic signature recognition
    • G06V40/382—Preprocessing; Feature extraction
    • G06V40/388—Sampling; Contour coding; Stroke extraction
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
    • G06V40/30—Writer recognition; Reading and verifying signatures
    • G06V40/37—Writer recognition; Reading and verifying signatures based only on signature signals such as velocity or pressure, e.g. dynamic signature recognition
    • G06V40/394—Matching; Classification
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
    • G06T2207/00—Indexing scheme for image analysis or image enhancement
    • G06T2207/30—Subject of image; Context of image processing
    • G06T2207/30241—Trajectory

Definitions

  • the invention relates to the technical field of identity recognition based on behavior characteristics, in particular to an offline handwriting individual recognition system and method based on three-dimensional dynamic characteristics.
  • the existing handwriting recognition technology is divided into online handwriting recognition and offline handwriting recognition.
  • online handwriting recognition can obtain more information about writing sequence, speed, pressure, angle and other characteristics that are beneficial to identity recognition in real time, but this information needs to be obtained by special equipment, thus limiting its application scope and development prospects;
  • Offline handwriting is the trajectory of writing motion, that is, displayed in front of people in the form of a two-dimensional static image, so offline handwriting recognition needs to mine and extract feature information from it.
  • the existing offline handwriting individual recognition technology is mainly for the extraction and analysis of two-dimensional static features of handwriting, or the extraction and analysis of single three-dimensional topographic features, but the two-dimensional static features cannot effectively identify imitated handwriting, and there are serious problems in application.
  • the single three-dimensional topography feature has limited identification ability due to too little information.
  • the present invention provides an offline handwriting individual recognition system based on three-dimensional dynamic features, which specifically includes:
  • the data acquisition module is used to scan the acquired suspicious handwriting to obtain the first white light image and the first three-dimensional image corresponding to the suspicious handwriting, and to scan the acquired sample handwritings to obtain the corresponding sample handwriting.
  • the data preprocessing module connected to the data acquisition module, is used to perform image preprocessing on the first white light image to obtain the first preprocessed image corresponding to the suspicious handwriting, and to perform all the second white light images separately The image preprocessing to obtain a second preprocessed image corresponding to each of the sample handwriting;
  • the skeleton extraction module connected to the data preprocessing module, is used to perform skeleton extraction on the first preprocessed image to obtain the first skeleton image corresponding to the suspicious handwriting, and perform the respective processing on each of the second preprocessed images. Extracting the skeleton to obtain a second skeleton image corresponding to the sample handwriting;
  • the line width of the first skeleton image and each of the second skeleton images is a single pixel
  • the stroke order recognition module connected to the skeleton extraction module, is configured to perform handwriting tracking according to the first skeleton image to obtain the first writing trajectory corresponding to the suspicious handwriting, and to perform the handwriting tracking respectively according to each of the second skeleton images Obtaining a second writing track corresponding to each of the sample handwriting;
  • the first feature extraction module is respectively connected to the data acquisition module and the stroke order recognition module, and is used to perform feature extraction on the corresponding first white light image according to the first writing trajectory to obtain each of the first writing trajectories
  • the first dynamic feature of each pixel point, and the second dynamic feature of each pixel point in each of the second writing trajectories is extracted from each corresponding second white light image according to each of the second writing trajectories;
  • the second feature extraction module is respectively connected to the data acquisition module and the stroke order recognition module, and is configured to extract each of the first writing trajectories from the corresponding first three-dimensional image according to the first writing trajectory A first three-dimensional feature of a pixel, and extracting a second three-dimensional feature of each pixel in each of the second writing trajectories from each corresponding second three-dimensional image according to each of the second writing trajectories;
  • the first data processing module is respectively connected to the first feature extraction module and the second feature extraction module, and is configured to use the first dynamic feature and the first three-dimensional feature corresponding to the suspicious handwriting, and each Processing the second dynamic feature and the second three-dimensional feature corresponding to the sample handwriting to obtain a correlation coefficient between the suspicious handwriting and each of the sample handwriting;
  • the second data processing module connected to the first data processing module, is used to process the data of the suspicious handwriting based on the first probability density distribution data and the second probability density distribution data obtained by statistics in advance, and each of the correlation coefficients. Individual identification results.
  • the first data processing module specifically includes:
  • a feature vector generating unit configured to add the first dynamic feature and the first three-dimensional feature corresponding to the suspicious handwriting to a first feature vector
  • the feature matrix generating unit is configured to respectively add the second dynamic feature and the second three-dimensional feature corresponding to each of the sample handwriting to a second feature matrix, and the second feature matrix includes multiple rows and columns arranged in rows and columns. Second feature vectors, each of the second feature vectors corresponds to the second dynamic feature and the second three-dimensional feature of each of the sample handwriting;
  • a vector pair generating unit respectively connected to the feature vector generating unit and the feature matrix generating unit, and configured to combine the first feature vector and each of the second feature vectors in pairs to generate multiple vector pairs;
  • the dynamic time adjustment unit is connected to the vector pair generation unit, and is configured to perform dynamic time warping between the first feature vector and the second feature vector in each vector pair, so that the first feature vector and The strokes corresponding to the second feature vector and the same parts of the strokes are matched one by one;
  • the data calculation unit is connected to the dynamic time warping unit, and is used to respectively calculate the correlation coefficient between the first feature vector and the second feature vector in each of the vector pairs after the dynamic time warping is performed.
  • the second data processing module specifically includes:
  • a first data processing unit configured to perform averaging processing on all the correlation coefficients corresponding to the suspicious handwriting to obtain the mean correlation coefficient corresponding to the suspicious handwriting;
  • the second data processing unit connected to the first data processing unit, is used to calculate the first probability corresponding to the mean correlation coefficient in the first probability density distribution data, and to calculate the mean correlation coefficient in the first probability density distribution data. 2.
  • the first probability is the probability that the suspicious handwriting and the sample handwriting corresponding to the mean correlation coefficient are written by the same person
  • the second probability is the probability that the suspicious handwriting and the sample handwriting corresponding to the mean correlation coefficient are Probability that the sample handwriting is not written by the same person
  • a third data processing unit connected to the second data processing unit, and configured to calculate the ratio of the first probability to the second probability to obtain the likelihood ratio of the suspicious handwriting;
  • the result judging module is connected to the third data processing unit, and is used for matching with a preset likelihood ratio scale according to the likelihood ratio to obtain the individual recognition result of the suspicious handwriting.
  • the first probability density distribution data is the probability density distribution of the correlation coefficient between the handwritings of several real samples of the person written by the same person, and
  • the second probability density distribution data is the probability density distribution of the correlation coefficients between the handwriting of several other people and the handwriting of the real sample obtained by different people according to the handwriting of the real sample of the person.
  • the handwriting of the other person includes handwriting for signature, and/or copying handwriting, and/or copying handwriting.
  • the image preprocessing method includes removing the background, and/or filling the whiteness in the strokes, and/or smoothing the strokes, and/or binarization.
  • the first dynamic feature includes width, and/or gray scale, and/or radian, and the first three-dimensional feature is height.
  • the second dynamic feature includes width, and/or gray scale, and/or radian, and the second three-dimensional feature is height.
  • An offline handwriting individual recognition method based on three-dimensional dynamic features which is applied to the offline handwriting individual recognition system described in any one of the above, specifically includes the following steps:
  • Step S1 the offline handwriting individual recognition system scans the acquired suspicious handwriting to obtain the first white light image and the first three-dimensional image corresponding to the suspicious handwriting, and scans the acquired sample handwritings to obtain each The second white light image and the second three-dimensional image corresponding to the sample handwriting;
  • Step S2 the offline handwriting individual recognition system performs image preprocessing on the first white light image to obtain a first preprocessed image corresponding to the suspicious handwriting, and performs the image preprocessing on each of the second white light images respectively Obtaining a second preprocessed image corresponding to each of the sample handwriting;
  • Step S3 the offline handwriting individual recognition system performs skeleton extraction on the first preprocessed image to obtain a first skeleton image corresponding to the suspicious handwriting, and performs the skeleton extraction on each of the second preprocessed images respectively The second skeleton image corresponding to the sample handwriting;
  • Step S4 the offline handwriting individual recognition system separately performs handwriting tracking according to the first skeleton image to obtain the first writing trajectory corresponding to the suspicious handwriting, and performs the handwriting tracking separately according to each of the second skeleton images to obtain each The second writing track corresponding to the sample handwriting;
  • Step S5 the offline handwriting recognition system performs feature extraction on the corresponding first white light image according to the first writing trajectory to obtain the first dynamic feature of each pixel in the first writing trajectory, and Extracting from each corresponding second white light image according to each second writing track to obtain a second dynamic feature of each pixel in each second writing track;
  • Step S6 the offline handwriting individual recognition system extracts the first three-dimensional feature of each pixel in the first writing trajectory on the corresponding first three-dimensional image according to the first writing trajectory, and Extracting the second writing trajectory from each corresponding second three-dimensional image to obtain the second three-dimensional feature of each pixel in each second writing trajectory;
  • Step S7 the offline handwriting individual recognition system according to the first dynamic feature and the first three-dimensional feature corresponding to the suspicious handwriting, and the second dynamic feature and the second dynamic feature corresponding to each sample handwriting Three-dimensional features, processed to obtain the correlation coefficient between each of the suspicious handwriting and each of the sample handwriting;
  • Step S8 the offline handwriting individual recognition system processes the individual recognition results of each suspicious handwriting according to the first probability density distribution data and the second probability density distribution data obtained by statistics in advance, and the respective correlation coefficients.
  • the step S7 specifically includes:
  • Step S71 the offline handwriting individual recognition system adds the first dynamic feature and the first three-dimensional feature corresponding to the suspicious handwriting to a first feature vector;
  • Step S72 the offline handwriting individual recognition system respectively adds the second dynamic feature and the second three-dimensional feature corresponding to each of the sample handwriting to a second feature matrix, the second feature matrix including rows and columns A plurality of second feature vectors of the cloth, each of the second feature vectors corresponds to the second dynamic feature and the second three-dimensional feature of each of the sample handwriting;
  • Step S73 The offline handwriting individual recognition system combines the first feature vector and each of the second feature vectors to generate multiple vector pairs;
  • Step S74 the offline handwriting individual recognition system performs dynamic time warping between the first feature vector and the second feature vector in each vector pair, so that the first feature vector and the second feature vector The strokes corresponding to the vector and the corresponding parts of the strokes are matched one by one;
  • Step S75 the offline handwriting individual recognition system separately calculates the correlation coefficient between the first feature vector and the second feature vector in each of the vector pairs after the dynamic time warping.
  • the step S8 specifically includes:
  • Step S81 the offline handwriting individual recognition system performs average processing on all the correlation coefficients corresponding to the suspicious handwriting to obtain the mean correlation coefficient corresponding to the suspicious handwriting;
  • Step S82 the offline handwriting individual recognition system calculates the first probability corresponding to the mean correlation coefficient in the first probability density distribution data, and calculates the mean correlation coefficient corresponding to the second probability density distribution data Second probability
  • the first probability is the probability that the suspicious handwriting and the sample handwriting corresponding to the mean correlation coefficient are written by the same person
  • the second probability is the probability that the suspicious handwriting and the sample handwriting corresponding to the mean correlation coefficient are Probability that the sample handwriting is not written by the same person
  • Step S83 the offline handwriting individual recognition system calculates the ratio of the first probability to the second probability to obtain the likelihood ratio of the suspicious handwriting
  • step S84 the offline handwriting individual recognition system performs matching with a preset likelihood ratio scale according to the likelihood ratio to obtain an individual recognition result of the suspicious handwriting.
  • Image preprocessing technology effectively removes the interference of factors such as paper background, adapts to the writing results of different thickness writing tools, and enhances the compatibility of different detection objects;
  • the correlation coefficient between suspicious handwriting and sample handwriting is further processed through the probability density distribution obtained by pre-statistics, which further improves the recognition accuracy.
  • FIG. 1 is a schematic structural diagram of an offline handwriting individual recognition system based on three-dimensional dynamic features in a preferred embodiment of the present invention
  • Figure 2 is a schematic diagram of a likelihood ratio scale in a preferred embodiment of the present invention.
  • FIG. 3 is a schematic flow chart of an offline handwriting individual recognition method based on three-dimensional dynamic features in a preferred embodiment of the present invention
  • FIG. 4 is a schematic diagram of a sub-process of an offline handwriting individual recognition method based on three-dimensional dynamic features in a preferred embodiment of the present invention
  • Fig. 5 is a schematic diagram of a sub-process of an offline handwriting individual recognition method based on three-dimensional dynamic features in a preferred embodiment of the present invention.
  • FIG. 1 specifically includes:
  • the data acquisition module 1 is used to scan the obtained suspicious handwriting to obtain the first white light image and the first three-dimensional image corresponding to the suspicious handwriting, and to scan several sample handwritings respectively to obtain the second white light corresponding to each sample handwriting.
  • the data preprocessing module 2 connected to the data acquisition module 1, is used to perform image preprocessing on the first white light image to obtain the first preprocessed image corresponding to the suspicious handwriting, and perform image preprocessing on each second white light image to obtain each sample handwriting.
  • the skeleton extraction module 3 connected to the data preprocessing module 2, is used to perform skeleton extraction on the first preprocessed image to obtain the first skeleton image corresponding to the suspicious handwriting, and perform skeleton extraction on each second preprocessed image to obtain the corresponding sample handwriting.
  • Second skeleton image is used to perform skeleton extraction on the first preprocessed image to obtain the first skeleton image corresponding to the suspicious handwriting, and perform skeleton extraction on each second preprocessed image to obtain the corresponding sample handwriting.
  • the line width of the first skeleton image and each second skeleton image is a single pixel
  • the stroke order recognition module 4 connected to the skeleton extraction module 3, is used to track the handwriting according to the first skeleton image to obtain the first writing trajectory corresponding to the suspicious handwriting, and to separately track the handwriting according to each second skeleton image to obtain the second handwriting corresponding to each sample Writing track
  • the first feature extraction module 5 is respectively connected to the data acquisition module 1 and the stroke order recognition module 4, and is used to perform feature extraction on the corresponding first white light image according to the first writing trajectory to obtain the first writing trajectory of each pixel A dynamic feature, and extracting the second dynamic feature of each pixel in each second writing track from each corresponding second white light image according to each second writing track;
  • the second feature extraction module 6, respectively connected to the data acquisition module 1 and the stroke order recognition module 4, is used to extract the first three-dimensional feature of each pixel in the first writing trajectory on the corresponding first three-dimensional image according to the first writing trajectory , And extract the second three-dimensional feature of each pixel in each second writing trajectory from each corresponding second three-dimensional image respectively according to each second writing trajectory;
  • the first data processing module 7 is respectively connected to the first feature extraction module 5 and the second feature extraction module 6, and is used for the first dynamic feature and the first three-dimensional feature corresponding to the suspicious handwriting, and the second dynamic feature corresponding to each sample handwriting And the second three-dimensional feature, processed to obtain the correlation coefficient between each suspicious handwriting and each sample handwriting;
  • the second data processing module 8 is connected to the first data processing module 7, and is used to process the data of the suspicious handwriting according to the first probability density distribution data and the second probability density distribution data obtained by statistics in advance, and each of the correlation coefficients. Individual identification results.
  • the scanning operation is first performed by using a large-area optical three-dimensional topography scanning instrument, specifically placing each suspicious handwriting and sample handwriting to be detected on the carrier of the above-mentioned large-area optical three-dimensional topography scanning instrument. On the stage, and decide whether to turn on the vacuum suction device according to the condition of the paper.
  • the computer connected to the above-mentioned large-area optical three-dimensional profile scanning instrument simultaneously obtains the first white light image and the first three-dimensional image corresponding to the suspicious handwriting, and At the same time, the second white light image and the second three-dimensional image corresponding to each sample handwriting are acquired.
  • the first white light image and the second white light image are respectively subjected to image preprocessing to obtain a first preprocessed image corresponding to the first white light image and a second preprocessed image corresponding to the second white light image.
  • image preprocessing includes removing Operations such as background, filling in whiteness in strokes, smoothing strokes and binarization.
  • the processing procedures for suspicious handwriting and sample handwriting are similar. Therefore, only the processing procedure of suspicious handwriting is taken as an example for detailed description, and the processing procedure for sample handwriting is not repeated here.
  • the process of handling suspicious handwriting is as follows:
  • the skeleton of the corresponding suspicious handwriting is extracted from the first preprocessed image to obtain the first skeleton image corresponding to the suspicious handwriting, and the first writing track corresponding to the suspicious handwriting is obtained by tracking the handwriting on the first skeleton image.
  • the above-mentioned handwriting tracking is completed in the form of human-computer interaction, which specifically includes:
  • the acquired first writing trajectory acquiring the first dynamic feature of each pixel on the first writing trajectory on the corresponding first white light image, that is, features such as stroke width, stroke gray scale, and stroke arc;
  • the first writing track acquires the first three-dimensional feature of each pixel on the first writing track on the corresponding first three-dimensional image, that is, the height data of each pixel.
  • the above-mentioned first dynamic feature and the first three-dimensional feature together constitute the three-dimensional dynamic feature of the suspicious handwriting.
  • the sample handwriting is processed according to the same processing procedure described above to obtain the three-dimensional dynamic characteristics of the sample handwriting. Then the correlation coefficient between the three-dimensional dynamic characteristics of the suspicious handwriting and the three-dimensional dynamic characteristics of the sample handwriting is calculated, and the similarity between the suspicious handwriting and the sample handwriting is characterized by the correlation coefficient. Specifically, because the formation of handwriting is a process, even if the handwriting is written by the same person, because the writing speed is not exactly the same each time, the length of each corresponding stroke is different, resulting in different sequence lengths corresponding to the handwriting, and the same stroke occurs in the sequence. dislocation.
  • the sequence length of each stroke when the three-dimensional dynamic feature of the suspicious handwriting is projected on the above-mentioned coordinate system, and the three-dimensional dynamic feature of the sample handwriting when each stroke is projected on the above-mentioned coordinate system The sequence length of is not exactly the same, resulting in the inability to calculate the effective correlation coefficient between the three-dimensional dynamic features of suspicious handwriting and the three-dimensional dynamic features of sample handwriting. Therefore, before calculating the correlation coefficient, it is necessary to perform dynamic time adjustment on the three-dimensional dynamic characteristics of a set of suspicious handwriting and the three-dimensional dynamic characteristics of the sample handwriting that need to be calculated for the correlation coefficient.
  • DTW technology is used for dynamic time adjustment, so that The three-dimensional dynamic features of the suspicious handwriting and the three-dimensional dynamic features of the sample handwriting correspond one-to-one in the same stroke, and the same stroke is the same in sequence length.
  • the above-mentioned similarity that is, the correlation coefficient
  • the correlation coefficient is the mean correlation coefficient according to the present invention. It should be noted that during handwriting recognition, several suspicious handwritings and several sample handwritings are usually obtained. For example, two suspicious handwritings and three sample handwritings are obtained by combining the three-dimensional dynamic characteristics of the two suspicious handwritings with the three sample handwritings. For each suspicious handwriting, the corresponding three correlation coefficient calculation results will be obtained.
  • the above-mentioned first probability density distribution data obtained by statistics in advance is the probability density distribution of the correlation coefficients between several real sample handwritings of the same person written by the same person
  • the second probability density distribution data is the probability density distribution of the correlation coefficient between different people according to the actual The probability density distribution of the correlation coefficients between the handwritings of some other people and the real sample handwritings obtained by the sample handwriting.
  • hundreds of thousands of handwriting data are obtained in advance, and a corresponding handwriting database is established based on the handwriting data.
  • the handwriting database includes a number of real sample handwritings written by the same person, that is, the handwriting of the person, and different people.
  • a number of other people's handwritings obtained based on the actual sample handwriting of the person that is, the handwriting for signatures obtained from the above-mentioned personal handwriting, and/or copying handwriting, and/or copying the handwriting of others.
  • the probability density distribution of the own handwriting is obtained; by separately calculating the correlation coefficients between each other’s handwriting and the own handwriting, and according to the obtained Several correlation coefficients get the probability density distribution of others' handwriting.
  • the likelihood ratio of suspicious handwriting is further calculated by the following formula:
  • SLR means the likelihood ratio of suspicious handwriting
  • H p means that the suspicious handwriting and the sample handwriting are written by the same person
  • H p means that the suspicious handwriting and the sample handwriting are not written by the same person
  • E U means the suspicious handwriting
  • E S means the sample handwriting ;
  • a likelihood ratio scale is set in advance, and the calculation result of the likelihood ratio is matched with the likelihood ratio scale. If the likelihood ratio is greater than 10,000, the individual The recognition result strongly supports that the suspicious handwriting and the sample handwriting are written by the same person; if the likelihood ratio is less than 0.001, the individual recognition result given strongly supports that the suspicious handwriting and the sample handwriting are not written by the same person; other matches The result can be deduced by analogy.
  • the first data processing module 7 specifically includes:
  • the feature vector generating unit 71 is configured to add the first dynamic feature and the first three-dimensional feature corresponding to the suspicious handwriting to a first feature vector;
  • the feature matrix generating unit 72 is configured to respectively add the second dynamic feature and the second three-dimensional feature corresponding to each sample handwriting to a second feature matrix.
  • the second feature matrix includes a plurality of second feature vectors arranged in rows and columns.
  • the second feature vector corresponds to the second dynamic feature and the second three-dimensional feature of each sample handwriting;
  • the vector pair generating unit 73 respectively connects the feature vector generating unit 71 and the feature matrix generating unit 72, and is configured to combine the first feature vector and each second feature vector in pairs to generate multiple vector pairs;
  • the dynamic time adjustment unit 74 the connection vector pair generation unit 73, is used to perform dynamic time warping between the first feature vector and the second feature vector in each vector pair, so that the first feature vector corresponds to the second feature vector
  • the strokes and the corresponding parts of the strokes are matched one by one;
  • the data calculation unit 75 is connected to the dynamic time warping unit 74 and is used to calculate the correlation coefficients between the first feature vector and the second feature vector in each vector pair after dynamic time warping.
  • the second data processing module 8 specifically includes:
  • the first data processing unit 81 is configured to perform average processing on all correlation coefficients corresponding to the suspicious handwriting to obtain the mean correlation coefficient corresponding to the suspicious handwriting;
  • the second data processing unit 82 connected to the first data processing unit 81, is used to calculate the first probability of the mean correlation coefficient in the first probability density distribution data, and calculate the mean correlation coefficient in the second probability density distribution data. Second probability
  • the first probability is the probability that the suspicious handwriting corresponding to the mean correlation coefficient and the sample handwriting are written by the same person
  • the second probability is the probability that the suspicious handwriting corresponding to the mean correlation coefficient and the sample handwriting are not written by the same person
  • the third data processing unit 83 connected to the second data processing unit 82, is used to calculate the ratio of the first probability to the second probability to obtain the likelihood ratio of the suspicious handwriting;
  • the result determination module 84 connected to the third data processing unit 83, is used for matching in a preset likelihood ratio scale according to the likelihood ratio to obtain an individual identification result of the suspicious handwriting.
  • the first probability density distribution data is the probability density distribution of the correlation coefficients between several real sample handwritings written by the same person
  • the second probability density distribution data is based on different people The probability density distribution of the correlation coefficients between the handwritings of several others obtained by the handwriting of the real sample and the handwriting of the real sample.
  • the handwriting of another person includes handwriting for signature, and/or copying handwriting, and/or copying handwriting.
  • the method of image preprocessing includes removing the background, and/or filling the whiteness in the strokes, and/or smoothing the strokes, and/or binarization.
  • the first dynamic feature includes width, and/or gray scale, and/or radian, and the first three-dimensional feature is height.
  • the second dynamic feature includes width, and/or gray scale, and/or radian, and the second three-dimensional feature is height.
  • An offline handwriting individual recognition method based on three-dimensional dynamic features applied to any of the above offline handwriting individual recognition systems, as shown in Figure 3, specifically includes the following steps:
  • Step S1 the offline handwriting individual recognition system scans the acquired suspicious handwriting to obtain the first white light image and the first three-dimensional image corresponding to the suspicious handwriting, and scans the acquired sample handwritings to obtain the second corresponding to each sample handwriting.
  • Step S2 the offline handwriting individual recognition system performs image preprocessing on the first white light image to obtain a first preprocessed image corresponding to the suspicious handwriting, and performs image preprocessing on each second white light image to obtain a second preprocessing corresponding to each sample handwriting. image;
  • Step S3 the offline handwriting recognition system performs skeleton extraction on the first preprocessed image to obtain a first skeleton image corresponding to the suspicious handwriting, and performs skeleton extraction on each second preprocessed image to obtain a second skeleton image corresponding to the sample handwriting;
  • Step S4 the offline handwriting recognition system performs handwriting tracking according to the first skeleton image to obtain a first writing trajectory corresponding to the suspicious handwriting, and performs handwriting tracking according to each second skeleton image to obtain a second writing trajectory corresponding to each sample handwriting;
  • Step S5 The offline handwriting recognition system performs feature extraction on the corresponding first white light image according to the first writing trajectory to obtain the first dynamic feature of each pixel in the first writing trajectory, and according to each second writing trajectory, The second dynamic feature of each pixel in each second writing track is extracted from each corresponding second white light image;
  • Step S6 the offline handwriting individual recognition system extracts the first three-dimensional feature of each pixel in the first writing trajectory on the corresponding first three-dimensional image according to the first writing trajectory, and according to each second writing trajectory, respectively Extracting the second three-dimensional feature of each pixel in each second writing track from the second three-dimensional image;
  • Step S7 the offline handwriting recognition system processes the first dynamic feature and the first three-dimensional feature corresponding to the suspicious handwriting, as well as the second dynamic feature and the second three-dimensional feature corresponding to each sample handwriting, to obtain the difference between each suspicious handwriting and each sample handwriting.
  • step S8 the offline handwriting individual recognition system processes and obtains the individual recognition result of the suspicious handwriting according to the first probability density distribution data and the second probability density distribution data obtained by statistics in advance, and various correlation coefficients.
  • step S7 specifically includes:
  • Step S71 The offline handwriting individual recognition system separately adds the first dynamic feature and the first three-dimensional feature corresponding to the suspicious handwriting to a first feature vector;
  • Step S72 the offline handwriting individual recognition system separately adds the second dynamic feature and the second three-dimensional feature corresponding to each sample handwriting to a second feature matrix.
  • the second feature matrix includes a plurality of second feature vectors arranged in rows and columns.
  • the second feature vector corresponds to the second dynamic feature and the second three-dimensional feature of each sample handwriting;
  • Step S73 the offline handwriting individual recognition system combines the first feature vector and each second feature vector in pairs to generate multiple vector pairs;
  • step S74 the offline handwriting individual recognition system performs dynamic time warping between the first feature vector and the second feature vector in each vector pair, so that the strokes and corresponding parts of the strokes corresponding to the first feature vector and the second feature vector are one One match
  • Step S75 The offline handwriting individual recognition system respectively calculates the correlation coefficient between the first feature vector and the second feature vector in each vector pair after dynamic time warping.
  • step S8 specifically includes:
  • Step S81 the offline handwriting individual recognition system averages all the correlation coefficients corresponding to the suspicious handwriting to obtain the mean correlation coefficient corresponding to the suspicious handwriting;
  • Step S82 the offline handwriting individual recognition system calculates the first probability corresponding to the mean correlation coefficient in the first probability density distribution data, and calculates the second probability corresponding to the mean correlation coefficient in the second probability density distribution;
  • the first probability is the probability that the suspicious handwriting corresponding to the mean correlation coefficient and the sample handwriting are written by the same person
  • the second probability is the probability that the suspicious handwriting corresponding to the mean correlation coefficient and the sample handwriting are not written by the same person
  • Step S83 the offline handwriting individual recognition system calculates the ratio of the first probability to the second probability to obtain the likelihood ratio of the suspicious handwriting
  • step S84 the offline handwriting individual recognition system performs matching with a preset likelihood ratio scale according to the likelihood ratio to obtain an individual recognition result of the suspicious handwriting.

Landscapes

  • Engineering & Computer Science (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Multimedia (AREA)
  • Human Computer Interaction (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Data Mining & Analysis (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Evolutionary Computation (AREA)
  • General Engineering & Computer Science (AREA)
  • Collating Specific Patterns (AREA)

Abstract

一种基于三维动态特征的离线笔迹个体识别系统及方法，涉及身份识别技术领域，包括扫描可疑笔迹得到第一白光图像和第一三维图像，扫描样本笔迹得到第二白光图像和第二三维图像；将第一白光图像和第二白光图像预处理得到第一预处理图像和第二预处理图像；将第一预处理图像和第二预处理图像提取第一骨架图像和第二骨架图像；根据第一骨架图像和第二骨架图像得到第一书写轨迹和第二书写轨迹；根据第一书写轨迹和第二书写轨迹提取第一动态特征、第一三维特征和第二动态特征、第二三维特征；处理得到可疑笔迹与样本笔迹之间的相关系数，并根据相关系数处理得到个体识别结果。该方法包含的信息量更大，特征种类更丰富，有效提高识别准确率。

Description

一种基于三维动态特征的离线笔迹个体识别系统及方法 技术领域
本发明涉及基于行为特征的身份识别技术领域，尤其涉及一种基于三维动态特征的离线笔迹个体识别系统及方法。
背景技术
随着科学技术的进步，基于生物特征的身份识别技术也不断发展。每个个体都有唯一的可以测量或可自动识别、验证的生理特征或行为特征，其中生理特征包括眼虹膜、视网膜、指纹及面貌等，行为特征包括步态、声音及笔迹等。离线笔迹个体识别技术是根据人的书写技能习惯特性在纸上书写的自己、符号、绘画中的反映，通过存疑笔迹与样本笔迹的比较、鉴别，从而确定笔迹书写人的一项个体识别技术。笔迹是一个人独特的行为特征，不同人的笔迹存在很大差别，每个人书写习惯不同，一般情况下，笔迹模仿者只能模仿字形，却无法准确还原原作者的书写习惯，模仿的笔迹与原笔迹在细节上会存在差异，因此可以利用该部分的差异性和独特性，通过测量书写者的字形及笔画的速度、顺序和压力等特征进行身份识别。
根据笔迹获取途径的不同，现有笔迹识别技术分为在线笔迹识别和离线笔迹识别两种。其中，在线笔迹识别能够实时获取更多关于书写顺序、速度、压力、角度等有益于身份识别的特征信息，但这些信息需要运用特殊的设备来获取，因此限定了其应用范围及发展前景； 而离线笔迹则是以书写运动的轨迹，即二维静态图像的形式展现在人们面前，因此离线笔迹识别需要从中挖掘和提取特征信息。现有的离线笔迹个体识别技术主要是笔迹二维静态特征的提取和分析上，或者单一的三维形貌特征的提取和分析上，但二维静态特征无法有效识别摹仿笔迹，在应用上存在严重的瓶颈问题，而单一的三维形貌特征由于信息量过少，鉴别能力有限。
发明内容
针对现有技术中存在的问题，本发明提供一种基于三维动态特征的离线笔迹个体识别系统，具体包括：
数据采集模块，用于将获取到的可疑笔迹进行扫描得到所述可疑笔迹对应的第一白光图像和第一三维图像，以及将获取到的若干样本笔迹分别进行扫描得到各所述样本笔迹对应的第二白光图像和第二三维图像；
数据预处理模块，连接所述数据采集模块，用于将所述第一白光图像进行图像预处理得到所述可疑笔迹对应的第一预处理图像，以及将各所述第二白光图像分别进行所述图像预处理得到各所述样本笔迹对应的第二预处理图像；
骨架提取模块，连接所述数据预处理模块，用于将所述第一预处理图像进行骨架提取得到所述可疑笔迹对应的第一骨架图像，以及将各所述第二预处理图像分别进行所述骨架提取得到所述样本笔迹对应的第二骨架图像；
所述第一骨架图像和各所述第二骨架图像的线幅均为单个像素；
笔顺识别模块，连接所述骨架提取模块，用于根据所述第一骨架图像进行笔迹追踪得到所述可疑笔迹对应的第一书写轨迹，以及根据各所述第二骨架图像分别进行所述笔迹追踪得到各所述样本笔迹对应的第二书写轨迹；
第一特征提取模块，分别连接数据采集模块和所述笔顺识别模块，用于根据所述第一书写轨迹在对应的所述第一白光图像上进行特征提取，得到所述第一书写轨迹中每个像素点的第一动态特征，以及根据各所述第二书写轨迹分别在对应的各所述第二白光图像上提取得到各所述第二书写轨迹中每个像素点的第二动态特征；
第二特征提取模块，分别连接所述数据采集模块和所述笔顺识别模块，用于根据所述第一书写轨迹在对应的所述第一三维图像上提取得到所述第一书写轨迹中每个像素点的第一三维特征，以及根据各所述第二书写轨迹分别在对应的各所述第二三维图像上提取得到各所述第二书写轨迹中每个像素点的第二三维特征；
第一数据处理模块，分别连接所述第一特征提取模块和所述第二特征提取模块，用于根据所述可疑笔迹对应的所述第一动态特征和所述第一三维特征，以及各所述样本笔迹对应的所述第二动态特征和所述第二三维特征，处理得到所述可疑笔迹与各所述样本笔迹之间的相关系数；
第二数据处理模块，连接所述第一数据处理模块，用于根据预先统计得到的第一概率密度分布数据和第二概率密度分布数据，以及各 所述相关系数，处理得到所述可疑笔迹的个体识别结果。
优选的，所述第一数据处理模块具体包括：
特征向量生成单元，用于将所述可疑笔迹对应的所述第一动态特征和所述第一三维特征加入一第一特征向量；
特征矩阵生成单元，用于分别将各所述样本笔迹对应的所述第二动态特征和所述第二三维特征加入一第二特征矩阵，所述第二特征矩阵包括以行列形式排布的多个第二特征向量，每个所述第二特征向量对应每个所述样本笔迹的所述第二动态特征和所述第二三维特征；
向量对生成单元，分别连接所述特征向量生成单元和所述特征矩阵生成单元，用于将所述第一特征向量与各所述第二特征向量之间进行两两组合生成多个向量对；
动态时间调整单元，连接所述向量对生成单元，用于分别将各向量对中的所述第一特征向量和所述第二特征向量之间进行动态时间规整，使得所述第一特征向量与所述第二特征向量对应的笔画及所述笔画的相同部位一一相匹配；
数据计算单元，连接所述动态时间规整单元，用于分别计算进行所述动态时间规整后的各所述向量对中的所述第一特征向量和所述第二特征向量之间的相关系数。
优选的，所述第二数据处理模块具体包括：
第一数据处理单元，用于将所述可疑笔迹对应的所有所述相关系数进行均值处理得到所述可疑笔迹对应的均值相关系数；
第二数据处理单元，连接所述第一数据处理单元，用于计算所述 均值相关系数在所述第一概率密度分布数据中对应的第一概率，以及计算所述均值相关系数在所述第二概率密度分布数据中对应的第二概率；
所述第一概率为所述均值相关系数对应的所述可疑笔迹与所述样本笔迹是同一人所写的概率，所述第二概率为所述均值相关系数对应的所述可疑笔迹与所述样本笔迹不是同一人所写的概率；
第三数据处理单元，连接所述第二数据处理单元，用于计算所述第一概率与所述第二概率的比值得到所述可疑笔迹的似然比；
结果判定模块，连接所述第三数据处理单元，用于根据所述似然比于预先设置的似然比量表中进行匹配，以得到所述可疑笔迹的个体识别结果。
优选的，所述第一概率密度分布数据为同一人书写得到的若干本人真实样本笔迹两两之间的相关系数的概率密度分布，以及
所述第二概率密度分布数据为不同人根据所述本人真实样本笔迹书写得到的若干他人笔迹与本人真实样本笔迹两两之间的相关系数的概率密度分布。
优选的，所述他人笔迹包括代签笔迹，和/或临摹笔迹，和/或套摹笔迹。
优选的，所述图像预处理的方式包括去除背景，和/或填补笔画中的露白，和/或平滑笔画，和/或二值化。
优选的，所述第一动态特征包括宽度，和/或灰度，和/或弧度，所述第一三维特征为高度。
优选的，所述第二动态特征包括宽度，和/或灰度，和/或弧度，所述第二三维特征为高度。
一种基于三维动态特征的离线笔迹个体识别方法，应用于以上任意一项所述的离线笔迹个体识别系统，具体包括以下步骤：
步骤S1，所述离线笔迹个体识别系统将获取到的可疑笔迹进行扫描得到所述可疑笔迹对应的第一白光图像和第一三维图像，以及将获取到的若干样本笔迹分别进行扫描得到各所述样本笔迹对应的第二白光图像和第二三维图像；
步骤S2，所述离线笔迹个体识别系统将所述第一白光图像进行图像预处理得到所述可疑笔迹对应的第一预处理图像，以及将各所述第二白光图像分别进行所述图像预处理得到各所述样本笔迹对应的第二预处理图像；
步骤S3，所述离线笔迹个体识别系统将所述第一预处理图像进行骨架提取得到所述可疑笔迹对应的第一骨架图像，以及将各所述第二预处理图像分别进行所述骨架提取得到所述样本笔迹对应的第二骨架图像；
步骤S4，所述离线笔迹个体识别系统根据所述第一骨架图像分别进行笔迹追踪得到所述可疑笔迹对应的第一书写轨迹，以及根据各所述第二骨架图像分别进行所述笔迹追踪得到各所述样本笔迹对应的第二书写轨迹；
步骤S5，所述离线笔迹个体识别系统根据所述第一书写轨迹在对应的所述第一白光图像上进行特征提取，得到所述第一书写轨迹中 每个像素点的第一动态特征，以及根据各所述第二书写轨迹分别在对应的各所述第二白光图像上提取得到各所述第二书写轨迹中每个像素点的第二动态特征；
步骤S6，所述离线笔迹个体识别系统根据所述第一书写轨迹在对应的所述第一三维图像上提取得到所述第一书写轨迹中每个像素点的第一三维特征，以及根据各所述第二书写轨迹分别在对应的各所述第二三维图像上提取得到各所述第二书写轨迹中每个像素点的第二三维特征；
步骤S7，所述离线笔迹个体识别系统根据所述可疑笔迹对应的所述第一动态特征和所述第一三维特征，以及各所述样本笔迹对应的所述第二动态特征和所述第二三维特征，处理得到各所述可疑笔迹与各所述样本笔迹之间的相关系数；
步骤S8，所述离线笔迹个体识别系统根据预先统计得到的第一概率密度分布数据和第二概率密度分布数据，以及各所述相关系数，处理得到各所述可疑笔迹的个体识别结果。
优选的，所述步骤S7具体包括：
步骤S71，所述离线笔迹个体识别系统将所述可疑笔迹对应的所述第一动态特征和所述第一三维特征加入一第一特征向量；
步骤S72，所述离线笔迹个体识别系统分别将各所述样本笔迹对应的所述第二动态特征和所述第二三维特征加入一第二特征矩阵，所述第二特征矩阵包括以行列形式排布的多个第二特征向量，每个所述第二特征向量对应每个所述样本笔迹的所述第二动态特征和所述第 二三维特征；
步骤S73，所述离线笔迹个体识别系统将所述第一特征向量与各所述第二特征向量之间进行两两组合生成多个向量对；
步骤S74，所述离线笔迹个体识别系统分别将各向量对中的所述第一特征向量和所述第二特征向量之间进行动态时间规整，使得所述第一特征向量与所述第二特征向量对应的笔画及所述笔画的相应部位一一相匹配；
步骤S75，所述离线笔迹个体识别系统分别计算进行所述动态时间规整后的各所述向量对中的所述第一特征向量和所述第二特征向量之间的相关系数。
优选的，所述步骤S8具体包括：
步骤S81，所述离线笔迹个体识别系统将所述可疑笔迹对应的所有所述相关系数进行均值处理得到所述可疑笔迹对应的均值相关系数；
步骤S82，所述离线笔迹个体识别系统计算所述均值相关系数在所述第一概率密度分布数据中对应的第一概率，以及计算所述均值相关系数在所述第二概率密度分布数据中对应的第二概率；
所述第一概率为所述均值相关系数对应的所述可疑笔迹与所述样本笔迹是同一人所写的概率，所述第二概率为所述均值相关系数对应的所述可疑笔迹与所述样本笔迹不是同一人所写的概率；
步骤S83，所述离线笔迹个体识别系统计算所述第一概率与所述第二概率的比值得到所述可疑笔迹的似然比；
步骤S84，所述离线笔迹个体识别系统根据所述似然比于预先设置的似然比量表中进行匹配，以得到所述可疑笔迹的个体识别结果。
上述技术方案具有如下优点或有益效果：
1)与传统的二维图像特征或单一的三维形貌特征相比，包含的信息量更大，特征种类更丰富；
2)图像预处理技术有效去除纸张背景等因素的干扰，适应不同粗细的书写工具的书写结果，增强了对不同检测对象的兼容性；
3)特征的自动提取，最大限度的排除了人为主观因素的参与度，从源头上避免了人为因素的干扰，有效提高识别准确率；
4)通过预先统计得到的概率密度分布对可疑笔迹与样本笔迹之间的相关系数做进一步处理，进一步提高了识别准确率。
附图说明
图1为本发明的较佳的实施例中，一种基于三维动态特征的离线笔迹个体识别系统的结构示意图；
图2为本发明的较佳的实施例中，似然比量表的示意图；
图3为本发明的较佳的实施例中，一种基于三维动态特征的离线笔迹个体识别方法的流程示意图；
图4为本发明的较佳的实施例中，一种基于三维动态特征的离线笔迹个体识别方法的子流程示意图；
图5为本发明的较佳的实施例中，一种基于三维动态特征的离线笔迹个体识别方法的子流程示意图。
具体实施方式
下面结合附图和具体实施例对本发明进行详细说明。本发明并不限定于该实施方式，只要符合本发明的主旨，则其他实施方式也可以属于本发明的范畴。
本发明的较佳的实施例中，基于现有技术中存在的上述问题，现提供一种基于三维动态特征的离线笔迹个体识别系统，如图1所示，具体包括：
数据采集模块1，用于将获取到的可疑笔迹进行扫描得到可疑笔迹对应的第一白光图像和第一三维图像，以及将获取到的若干样本笔迹分别进行扫描得到各样本笔迹对应的第二白光图像和第二三维图像；
数据预处理模块2，连接数据采集模块1，用于将第一白光图像进行图像预处理得到可疑笔迹对应的第一预处理图像，以及将各第二白光图像分别进行图像预处理得到各样本笔迹对应的第二预处理图像；
骨架提取模块3，连接数据预处理模块2，用于将第一预处理图像进行骨架提取得到可疑笔迹对应的第一骨架图像，以及将各第二预处理图像分别进行骨架提取得到样本笔迹对应的第二骨架图像；
第一骨架图像和各第二骨架图像的线幅均为单个像素；
笔顺识别模块4，连接骨架提取模块3，用于根据第一骨架图像进行笔迹追踪得到可疑笔迹对应的第一书写轨迹，以及根据各第二骨 架图像分别进行笔迹追踪得到各样本笔迹对应的第二书写轨迹；
第一特征提取模块5，分别连接数据采集模块1和笔顺识别模块4，用于根据第一书写轨迹在对应的第一白光图像上进行特征提取，得到第一书写轨迹中每个像素点的第一动态特征，以及根据各第二书写轨迹分别在对应的各第二白光图像上提取得到各第二书写轨迹中每个像素点的第二动态特征；
第二特征提取模块6，分别连接数据采集模块1和笔顺识别模块4，用于根据第一书写轨迹在对应的第一三维图像上提取得到第一书写轨迹中每个像素点的第一三维特征，以及根据各第二书写轨迹分别在对应的各第二三维图像上提取得到各第二书写轨迹中每个像素点的第二三维特征；
第一数据处理模块7，分别连接第一特征提取模块5和第二特征提取模块6，用于根据可疑笔迹对应的第一动态特征和第一三维特征，以及各样本笔迹对应的第二动态特征和第二三维特征，处理得到各可疑笔迹与各样本笔迹之间的相关系数；
第二数据处理模块8，连接第一数据处理模块7，用于根据预先统计得到的第一概率密度分布数据和第二概率密度分布数据，以及各所述相关系数，处理得到所述可疑笔迹的个体识别结果。
具体地，本实施例中，首先通过采用大面积光学三维形貌扫描仪器进行扫描操作，具体为依次将各待检测的可疑笔迹和样本笔迹置于上述大面积光学三维形貌扫描仪器的载物台上，并根据纸张的情况决定是否开启真空吸附装置，在扫描完成后，与上述大面积光学三维形 貌扫描仪器连接的计算机同时获取可疑笔迹对应的第一白光图像和第一三维图像，以及同时获取每个样本笔迹对应的第二白光图像和第二三维图像。随后将第一白光图像和第二白光图像分别进行图像预处理得到对应于第一白光图像的第一预处理图像，以及对应于第二白光图像的第二预处理图像，上述图像预处理包括去除背景、填补笔画中的露白、平滑笔画和二值化等操作。
本实施例中，对于可疑笔迹和样本笔迹的处理过程类似，因此此处仅以可疑笔迹的处理过程为例进行详细描述，对于样本笔迹的处理过程此处不再赘述。可疑笔迹的处理过程如下：
在第一预处理图像上提取对应的可疑笔迹的骨架，以得到对应于可疑笔迹的第一骨架图像，通过在第一骨架图像上进行笔迹追踪即得到对应于可疑笔迹的第一书写轨迹。上述笔迹追踪通过人机交互的形式完成，具体包括：
首先使用鼠标、电子笔或触摸屏的方式在第一骨架图像上的笔画起始位置附近点击，以自动获取笔画起点，随后沿笔画运行方向进行追踪；若当前追踪笔画有误或由于人为导致误操作，可以通过撤销当前追踪笔画，以返回上一个笔画并继续进行笔迹追踪，直至完成对整个第一骨架图像上的每个笔画的笔迹追踪，最终获得上述第一书写轨迹。
进一步地，根据获取的第一书写轨迹在对应的第一白光图上获取第一书写轨迹上每个像素点的第一动态特征，即笔画宽度、笔画灰度和笔画弧度等特征；还包括根据第一书写轨迹在对应的第一三维图上 获取第一书写轨迹上每个像素点的第一三维特征，即每个像素点的高度数据。上述第一动态特征和第一三维特征共同构成可疑笔迹的三维动态特征。
本实施例中，按照上述相同的处理过程，对样本笔迹进行处理得到样本笔迹的三维动态特征。随后计算可疑笔迹的三维动态特征与样本笔迹的三维动态特征之间的相关系数，通过相关系数表征可疑笔迹与样本笔迹的相似度。具体地，由于笔迹的形成是一个过程，即使是同一人书写的笔迹，由于每次的书写速度不完全相同，各个对应笔画的长度不同，造成笔迹对应的序列长度不同，相同笔画在序列上发生错位。从而在以序列长度为横轴的坐标系上，可疑笔迹的三维动态特征投射在上述坐标系上时每个笔画的序列长度，与样本笔迹的三维动态特征投射在上述坐标系上时每个笔画的序列长度是不完全相同的，造成无法计算出有效的可疑笔迹的三维动态特征与样本笔迹的三维动态特征之间的相关系数。因此，在计算相关系数之前，需要对需要进行相关系数计算的一组可疑笔迹的三维动态特征和样本笔迹的三维动态特征进行动态时间调整，本实施例中，采用DTW技术进行动态时间调整，使得可疑笔迹的三维动态特征和样本笔迹的三维动态特征在相同笔画一一对应，且相同笔画在序列长度上相同。
本实施例中，在获取可疑笔迹与样本笔迹的相似度后，为进一步提高识别准确率，还需要将上述相似度，即相关系数代入预先统计得到的概率密度分布中进行进一步处理。在进行笔迹识别时，仅获取一个可疑笔迹和一个样本笔迹时，对比的结果仅有一个相关系数，该相 关系数即为本发明所述的均值相关系数。需要说明的是，通常进行笔迹识别时，会获取到若干可疑笔迹和若干样本笔迹，如获取两个可疑笔迹和三个样本笔迹，通过将两个可疑笔迹的三维动态特征分别与三个样本笔迹的三维动态特征进行相关系数的计算，对于每个可疑笔迹，将得到对应的三个相关系数的计算结果。
本实施例中，对于每个可疑笔迹，无需将上述计算结果中的三个相关系数均代入预先统计得到的概率密度分布中进行进一步处理，而是先将三个相关系数取均值后得到相应的均值相关系数，再代入预先统计得到的概率密度分布中进行进一步处理。
具体地，上述预先统计得到的第一概率密度分布数据为同一人书写得到的若干本人真实样本笔迹两两之间的相关系数的概率密度分布，以及第二概率密度分布数据为不同人根据本人真实样本笔迹书写得到的若干他人笔迹与本人真实样本笔迹两两之间的相关系数的概率密度分布。本实施例中，预先获取数十万数量的笔迹数据，并根据该笔迹数据建立相应的笔迹数据库，该笔迹数据库中包括同一人书写得到的若干本人真实样本笔迹，即本人笔迹，还包括不同人根据本人真实样本笔迹书写得到的若干他人笔迹，即根据上述本人笔迹得到的代签笔迹，和/或临摹笔迹，和/或套摹笔迹等他人笔迹。通过分别计算两两本人笔迹之间的相关系数，并根据得到的若干相关系数得到本人笔迹的概率密度分布；通过分别计算每个他人笔迹与本人笔迹两两之间的相关系数，并根据得到的若干相关系数得到他人笔迹的概率密度分布。随后通过将上述得到的每个可疑笔迹的均值相关系数分别代 入到本人笔迹的概率密度分布和他人笔迹的概率密度分布中，以得到表征该可疑笔迹与样本笔迹是同一人所写的第一概率，以及表征该可疑笔迹与样本笔迹不是同一人所写的第二概率。
本实施例中，进一步通过下述公式计算可疑笔迹的似然比：
Figure PCTCN2019122176-appb-000001
其中，SLR表示可疑笔迹的似然比；H p表示可疑笔迹与样本笔迹是同一人所写；H p表示可疑笔迹与样本笔迹不是同一人所写；E U表示可疑笔迹；E S表示样本笔迹；
本实施例中，如图2所示，预先设有似然比量表，将上述似然比的计算结果与该似然比量表进行匹配，若似然比大于10000，则给出的个体识别结果为极强支持该可疑笔迹与样本笔迹是同一人所写；若似然比小于0.001，则给出的个体识别结果为极强支持该可疑笔迹与样本笔迹不是同一人所写；其他匹配结果以此类推。
本发明的较佳的实施例中，第一数据处理模块7具体包括：
特征向量生成单元71，用于将可疑笔迹对应的第一动态特征和第一三维特征加入一第一特征向量；
特征矩阵生成单元72，用于分别将各样本笔迹对应的第二动态特征和第二三维特征加入一第二特征矩阵，第二特征矩阵包括以行列形式排布的多个第二特征向量，每个第二特征向量对应每个样本笔迹的第二动态特征和第二三维特征；
向量对生成单元73，分别连接特征向量生成单元71和特征矩阵生成单元72，用于将第一特征向量与各第二特征向量之间进行两两 组合生成多个向量对；
动态时间调整单元74，连接向量对生成单元73，用于分别将各向量对中的第一特征向量和第二特征向量之间进行动态时间规整，使得第一特征向量与第二特征向量对应的笔画及笔画的相应部位一一相匹配；
数据计算单元75，连接动态时间规整单元74，用于分别计算进行动态时间规整后的各向量对中的第一特征向量和第二特征向量之间的相关系数。
本发明的较佳的实施例中，第二数据处理模块8具体包括：
第一数据处理单元81，用于将可疑笔迹对应的所有相关系数进行均值处理得到可疑笔迹对应的均值相关系数；
第二数据处理单元82，连接第一数据处理单元81，用于计算均值相关系数在第一概率密度分布数据中对应的第一概率，以及计算均值相关系数在第二概率密度分布数据中对应的第二概率；
第一概率为均值相关系数对应的可疑笔迹与样本笔迹是同一人所写的概率，第二概率为均值相关系数对应的可疑笔迹与样本笔迹不是同一人所写的概率；
第三数据处理单元83，连接第二数据处理单元82，用于计算第一概率与第二概率的比值得到可疑笔迹的似然比；
结果判定模块84，连接第三数据处理单元83，用于根据似然比于预先设置的似然比量表中进行匹配，以得到可疑笔迹的个体识别结果。
本发明的较佳的实施例中，第一概率密度分布数据为同一人书写得到的若干本人真实样本笔迹两两之间的相关系数的概率密度分布，以及第二概率密度分布数据为不同人根据本人真实样本笔迹书写得到的若干他人笔迹与本人真实样本笔迹两两之间的相关系数的概率密度分布。
本发明的较佳的实施例中，他人笔迹包括代签笔迹，和/或临摹笔迹，和/或套摹笔迹。
本发明的较佳的实施例中，图像预处理的方式包括去除背景，和/或填补笔画中的露白，和/或平滑笔画，和/或二值化。
本发明的较佳的实施例中，第一动态特征包括宽度，和/或灰度，和/或弧度，第一三维特征为高度。
本发明的较佳的实施例中，第二动态特征包括宽度，和/或灰度，和/或弧度，第二三维特征为高度。
一种基于三维动态特征的离线笔迹个体识别方法，应用于以上任意一项的离线笔迹个体识别系统，如图3所示，具体包括以下步骤：
步骤S1，离线笔迹个体识别系统将获取到的可疑笔迹进行扫描得到可疑笔迹对应的第一白光图像和第一三维图像，以及将获取到的若干样本笔迹分别进行扫描得到各样本笔迹对应的第二白光图像和第二三维图像；
步骤S2，离线笔迹个体识别系统将第一白光图像进行图像预处理得到可疑笔迹对应的第一预处理图像，以及将各第二白光图像分别进行图像预处理得到各样本笔迹对应的第二预处理图像；
步骤S3，离线笔迹个体识别系统将第一预处理图像进行骨架提取得到可疑笔迹对应的第一骨架图像，以及将各第二预处理图像分别进行骨架提取得到样本笔迹对应的第二骨架图像；
步骤S4，离线笔迹个体识别系统根据第一骨架图像进行笔迹追踪得到可疑笔迹对应的第一书写轨迹，以及根据各第二骨架图像分别进行笔迹追踪得到各样本笔迹对应的第二书写轨迹；
步骤S5，离线笔迹个体识别系统根据第一书写轨迹在对应的第一白光图像上进行特征提取，得到第一书写轨迹中每个像素点的第一动态特征，以及根据各第二书写轨迹分别在对应的各第二白光图像上提取得到各第二书写轨迹中每个像素点的第二动态特征；
步骤S6，离线笔迹个体识别系统根据第一书写轨迹在对应的第一三维图像上提取得到第一书写轨迹中每个像素点的第一三维特征，以及根据各第二书写轨迹分别在对应的各第二三维图像上提取得到各第二书写轨迹中每个像素点的第二三维特征；
步骤S7，离线笔迹个体识别系统根据可疑笔迹对应的第一动态特征和第一三维特征，以及各样本笔迹对应的第二动态特征和第二三维特征，处理得到各可疑笔迹与各样本笔迹之间的相关系数；
步骤S8，离线笔迹个体识别系统根据预先统计得到的第一概率密度分布数据和第二概率密度分布数据，以及各相关系数，处理得到可疑笔迹的个体识别结果。
本发明的较佳的实施例中，如图4所示，步骤S7具体包括：
步骤S71，离线笔迹个体识别系统分别将可疑笔迹对应的第一动 态特征和第一三维特征加入一第一特征向量；
步骤S72，离线笔迹个体识别系统分别将各样本笔迹对应的第二动态特征和第二三维特征加入一第二特征矩阵，第二特征矩阵包括以行列形式排布的多个第二特征向量，每个第二特征向量对应每个样本笔迹的第二动态特征和第二三维特征；
步骤S73，离线笔迹个体识别系统将第一特征向量与各第二特征向量之间进行两两组合生成多个向量对；
步骤S74，离线笔迹个体识别系统分别将各向量对中的第一特征向量和第二特征向量之间进行动态时间规整，使得第一特征向量与第二特征向量对应的笔画及笔画的相应部位一一相匹配；
步骤S75，离线笔迹个体识别系统分别计算进行动态时间规整后的各向量对中的第一特征向量和第二特征向量之间的相关系数。
本发明的较佳的实施例中，如图5所示，步骤S8具体包括：
步骤S81，离线笔迹个体识别系统将可疑笔迹对应的所有相关系数进行均值处理得到可疑笔迹对应的均值相关系数；
步骤S82，离线笔迹个体识别系统计算均值相关系数在第一概率密度分布数据中对应的第一概率，以及计算均值相关系数在第二概率密度分布中对应的第二概率；
第一概率为均值相关系数对应的可疑笔迹与样本笔迹是同一人所写的概率，第二概率为均值相关系数对应的可疑笔迹与样本笔迹不是同一人所写的概率；
步骤S83，离线笔迹个体识别系统计算第一概率与第二概率的比 值得到可疑笔迹的似然比；
步骤S84，离线笔迹个体识别系统根据似然比于预先设置的似然比量表中进行匹配，以得到可疑笔迹的个体识别结果。
以上所述仅为本发明较佳的实施例，并非因此限制本发明的实施方式及保护范围，对于本领域技术人员而言，应当能够意识到凡运用本说明书及图示内容所作出的等同替换和显而易见的变化所得到的方案，均应当包含在本发明的保护范围内。

Claims (11)

  1. 一种基于三维动态特征的离线笔迹个体识别系统，其特征在于，具体包括：
    数据采集模块，用于将获取到的可疑笔迹进行扫描得到所述可疑笔迹对应的第一白光图像和第一三维图像，以及将获取到的若干样本笔迹分别进行扫描得到各所述样本笔迹对应的第二白光图像和第二三维图像；
    数据预处理模块，连接所述数据采集模块，用于将所述第一白光图像进行图像预处理得到所述可疑笔迹对应的第一预处理图像，以及将各所述第二白光图像分别进行所述图像预处理得到各所述样本笔迹对应的第二预处理图像；
    骨架提取模块，连接所述数据预处理模块，用于将所述第一预处理图像进行骨架提取得到所述可疑笔迹对应的第一骨架图像，以及将各所述第二预处理图像分别进行所述骨架提取得到所述样本笔迹对应的第二骨架图像；
    所述第一骨架图像和各所述第二骨架图像的线幅均为单个像素；
    笔顺识别模块，连接所述骨架提取模块，用于根据所述第一骨架图像进行笔迹追踪得到所述可疑笔迹对应的第一书写轨迹，以及根据各所述第二骨架图像分别进行所述笔迹追踪得到各所述样本笔迹对应的第二书写轨迹；
    第一特征提取模块，分别连接数据采集模块和所述笔顺识别模块，用于根据所述第一书写轨迹在对应的所述第一白光图像上进行特征提取，得到所述第一书写轨迹中每个像素点的第一动态特征，以及 根据各所述第二书写轨迹分别在对应的各所述第二白光图像上提取得到各所述第二书写轨迹中每个像素点的第二动态特征；
    第二特征提取模块，分别连接所述数据采集模块和所述笔顺识别模块，用于根据所述第一书写轨迹在对应的所述第一三维图像上提取得到所述第一书写轨迹中每个像素点的第一三维特征，以及根据各所述第二书写轨迹分别在对应的各所述第二三维图像上提取得到各所述第二书写轨迹中每个像素点的第二三维特征；
    第一数据处理模块，分别连接所述第一特征提取模块和所述第二特征提取模块，用于根据所述可疑笔迹对应的所述第一动态特征和所述第一三维特征，以及各所述样本笔迹对应的所述第二动态特征和所述第二三维特征，处理得到所述可疑笔迹与各所述样本笔迹之间的相关系数；
    第二数据处理模块，连接所述第一数据处理模块，用于根据预先统计得到的第一概率密度分布数据和第二概率密度分布数据，以及各所述相关系数，处理得到所述可疑笔迹的个体识别结果。
  2. 根据权利要求1所述的离线笔迹个体识别系统，其特征在于，所述第一数据处理模块具体包括：
    特征向量生成单元，用于将所述可疑笔迹对应的所述第一动态特征和所述第一三维特征加入一第一特征向量；
    特征矩阵生成单元，用于分别将各所述样本笔迹对应的所述第二动态特征和所述第二三维特征加入一第二特征矩阵，所述第二特征矩阵包括以行列形式排布的多个第二特征向量，每个所述第二特征向量 对应每个所述样本笔迹的所述第二动态特征和所述第二三维特征；
    向量对生成单元，分别连接所述特征向量生成单元和所述特征矩阵生成单元，用于将所述第一特征向量与各所述第二特征向量之间进行两两组合生成多个向量对；
    动态时间规整单元，连接所述向量对生成单元，用于分别将各向量对中的所述第一特征向量和所述第二特征向量之间进行动态时间规整，使得所述第一特征向量与所述第二特征向量对应的笔画及笔画的相应部位一一对应；
    数据计算单元，连接所述动态时间规整单元，用于分别计算进行所述动态时间规整后的各所述向量对中的所述第一特征向量和所述第二特征向量之间的相关系数。
  3. 根据权利要求1所述的离线笔迹个体识别系统，其特征在于，所述第二数据处理模块具体包括：
    第一数据处理单元，用于将所述可疑笔迹对应的所有所述相关系数，进行均值处理得到所述可疑笔迹对应的均值相关系数；
    第二数据处理单元，连接所述第一数据处理单元，用于计算所述均值相关系数在所述第一概率密度分布数据中对应的第一概率，以及计算所述均值相关系数在所述第二概率密度分布数据中对应的第二概率；
    所述第一概率为所述均值相关系数对应的所述可疑笔迹与所述样本笔迹是同一人所写的概率，所述第二概率为所述均值相关系数对应的所述可疑笔迹与所述样本笔迹不是同一人所写的概率；
    第三数据处理单元，连接所述第二数据处理单元，用于计算所述第一概率与所述第二概率的比值得到所述可疑笔迹的似然比；
    结果判定模块，连接所述第三数据处理单元，用于根据所述似然比于预先设置的似然比量表中进行匹配，以得到所述可疑笔迹的个体识别结果。
  4. 根据权利要求3所述的离线笔迹个体识别系统，其特征在于，所述第一概率密度分布数据为同一人书写得到的若干本人真实样本笔迹两两之间的相关系数的概率密度分布，以及
    所述第二概率密度分布数据为不同人根据所述本人真实样本笔迹书写得到的若干他人笔迹与所述本人真实样本笔迹两两之间的相关系数的概率密度分布。
  5. 根据权利要求4所述的离线笔迹个体识别系统，其特征在于，所述他人笔迹包括代签笔迹，和/或临摹笔迹，和/或套摹笔迹。
  6. 根据权利要求1所述的离线笔迹个体识别系统，其特征在于，所述图像预处理的方式包括去除背景，和/或填补笔画中的露白，和/或平滑笔画，和/或二值化。
  7. 根据权利要求1所述的离线笔迹个体识别系统，其特征在于，所述第一动态特征包括宽度，和/或灰度，和/或弧度，所述第一三维特征为高度。
  8. 根据权利要求1所述的离线笔迹个体识别系统，其特征在于，所述第二动态特征包括宽度，和/或灰度，和/或弧度，所述第二三维特征为高度。
  9. 一种基于三维动态特征的离线笔迹个体识别方法，其特征在于，应用于如权利要求1-8中任意一项所述的离线笔迹个体识别系统，具体包括以下步骤：
    步骤S1，所述离线笔迹个体识别系统将获取到的可疑笔迹进行扫描得到所述可疑笔迹对应的第一白光图像和第一三维图像，以及将获取到的若干样本笔迹分别进行扫描得到各所述样本笔迹对应的第二白光图像和第二三维图像；
    步骤S2，所述离线笔迹个体识别系统将所述第一白光图像进行图像预处理得到所述可疑笔迹对应的第一预处理图像，以及将各所述第二白光图像分别进行所述图像预处理得到各所述样本笔迹对应的第二预处理图像；
    步骤S3，所述离线笔迹个体识别系统将所述第一预处理图像进行骨架提取得到所述可疑笔迹对应的第一骨架图像，以及将各所述第二预处理图像分别进行所述骨架提取得到所述样本笔迹对应的第二骨架图像；
    步骤S4，所述离线笔迹个体识别系统根据所述第一骨架图像分别进行笔迹追踪得到所述可疑笔迹对应的第一书写轨迹，以及根据各所述第二骨架图像分别进行所述笔迹追踪得到各所述样本笔迹对应的第二书写轨迹；
    步骤S5，所述离线笔迹个体识别系统根据所述第一书写轨迹在对应的所述第一白光图像上进行特征提取，得到所述第一书写轨迹中每个像素点的第一动态特征，以及根据各所述第二书写轨迹分别在对 应的各所述第二白光图像上提取得到各所述第二书写轨迹中每个像素点的第二动态特征；
    步骤S6，所述离线笔迹个体识别系统根据所述第一书写轨迹在对应的所述第一三维图像上提取得到所述第一书写轨迹中每个像素点的第一三维特征，以及根据各所述第二书写轨迹分别在对应的各所述第二三维图像上提取得到各所述第二书写轨迹中每个像素点的第二三维特征；
    步骤S7，所述离线笔迹个体识别系统根据所述可疑笔迹对应的所述第一动态特征和所述第一三维特征，以及各所述样本笔迹对应的所述第二动态特征和所述第二三维特征，处理得到所述可疑笔迹与各所述样本笔迹之间的相关系数；
    步骤S8，所述离线笔迹个体识别系统根据预先统计得到的第一概率密度分布数据和第二概率密度分布数据，以及各所述相关系数，处理得到所述可疑笔迹的个体识别结果。
  10. 根据权利要求9所述的离线笔迹个体识别方法，其特征在于，所述步骤S7具体包括：
    步骤S71，所述离线笔迹个体识别系统将所述可疑笔迹对应的所述第一动态特征和所述第一三维特征加入一第一特征向量；
    步骤S72，所述离线笔迹个体识别系统分别将各所述样本笔迹对应的所述第二动态特征和所述第二三维特征加入一第二特征矩阵，所述第二特征矩阵包括以行列形式排布的多个第二特征向量，每个所述第二特征向量对应每个所述样本笔迹的所述第二动态特征和所述第 二三维特征；
    步骤S73，所述离线笔迹个体识别系统将所述第一特征向量与各所述第二特征向量之间进行两两组合生成多个向量对；
    步骤S74，所述离线笔迹个体识别系统分别将各所述向量对中的所述第一特征向量和所述第二特征向量之间进行动态时间规整，使得所述第一特征向量与所述第二特征向量对应的笔画及笔画的对应部位一一相匹配；
    步骤S75，所述离线笔迹个体识别系统分别计算进行所述动态时间规整后的各所述向量对中的所述第一特征向量和所述第二特征向量之间的相关系数。
  11. 根据权利要求9所述的离线笔迹个体识别方法，其特征在于，所述步骤S8具体包括：
    步骤S81，所述离线笔迹个体识别系统将所述可疑笔迹对应的所有所述相关系数进行均值处理得到所述可疑笔迹对应的均值相关系数；
    步骤S82，所述离线笔迹个体识别系统计算所述均值相关系数在所述第一概率密度分布数据中对应的第一概率，以及计算所述均值相关系数在所述第二概率密度分布数据中对应的第二概率；
    所述第一概率为所述均值相关系数对应的所述可疑笔迹与所述样本笔迹是同一人所写的概率，所述第二概率为所述均值相关系数对应的所述可疑笔迹与所述样本笔迹不是同一人所写的概率；
    步骤S83，所述离线笔迹个体识别系统计算所述第一概率与所述 第二概率的比值得到所述可疑笔迹的似然比；
    步骤S84，所述离线笔迹个体识别系统根据所述似然比于预先设置的似然比量表中进行匹配，以得到所述可疑笔迹的个体识别结果。
PCT/CN2019/122176 2019-08-22 2019-11-29 一种基于三维动态特征的离线笔迹个体识别系统及方法 WO2021031445A1 (zh)

Priority Applications (1)

Application Number Priority Date Filing Date Title
US16/968,833 US20230245501A1 (en) 2019-08-22 2019-11-29 Offline handwriting individual recognition system and method based on three-dimensional dynamic features

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201910780141.2 2019-08-22
CN201910780141.2A CN110490153B (zh) 2019-08-22 2019-08-22 一种基于三维动态特征的离线笔迹个体识别系统及方法

Publications (1)

Publication Number Publication Date
WO2021031445A1 true WO2021031445A1 (zh) 2021-02-25

Family

ID=68552970

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2019/122176 WO2021031445A1 (zh) 2019-08-22 2019-11-29 一种基于三维动态特征的离线笔迹个体识别系统及方法

Country Status (3)

Country Link
US (1) US20230245501A1 (zh)
CN (1) CN110490153B (zh)
WO (1) WO2021031445A1 (zh)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110490153B (zh) * 2019-08-22 2021-10-19 司法鉴定科学研究院 一种基于三维动态特征的离线笔迹个体识别系统及方法
CN113468987B (zh) * 2021-06-17 2023-04-18 重庆傲雄在线信息技术有限公司 一种电子笔迹鉴定方法、系统、电子设备及存储介质
CN113407921B (zh) * 2021-06-30 2023-04-07 重庆建安仪器有限责任公司 一种笔迹识别的登录方法

Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110311141A1 (en) * 2008-12-30 2011-12-22 Guangdong Guobi Technology Co., Ltd. Method and system for recognizing a handwritten character
CN104200239A (zh) * 2014-09-09 2014-12-10 河海大学常州校区 一种基于图像特征融合识别的签名鉴伪系统及方法
CN104299000A (zh) * 2014-10-09 2015-01-21 南通大学 基于局部片段分布特征的笔迹识别方法
CN107239753A (zh) * 2017-05-23 2017-10-10 刘青 一种笔迹鉴定的系统
CN109387501A (zh) * 2018-11-29 2019-02-26 中国政法大学 利用快速三维拉曼光谱重构技术鉴定交叉笔画顺序的方法
CN110490153A (zh) * 2019-08-22 2019-11-22 司法鉴定科学研究院 一种基于三维动态特征的离线笔迹个体识别系统及方法

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103049207B (zh) * 2012-12-27 2016-01-13 富泰华工业（深圳）有限公司 手写输入识别系统及方法
CN107977155B (zh) * 2017-11-23 2020-12-29 广州视睿电子科技有限公司 一种手写识别方法、装置、设备和存储介质

Patent Citations (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20110311141A1 (en) * 2008-12-30 2011-12-22 Guangdong Guobi Technology Co., Ltd. Method and system for recognizing a handwritten character
CN104200239A (zh) * 2014-09-09 2014-12-10 河海大学常州校区 一种基于图像特征融合识别的签名鉴伪系统及方法
CN104299000A (zh) * 2014-10-09 2015-01-21 南通大学 基于局部片段分布特征的笔迹识别方法
CN107239753A (zh) * 2017-05-23 2017-10-10 刘青 一种笔迹鉴定的系统
CN109387501A (zh) * 2018-11-29 2019-02-26 中国政法大学 利用快速三维拉曼光谱重构技术鉴定交叉笔画顺序的方法
CN110490153A (zh) * 2019-08-22 2019-11-22 司法鉴定科学研究院 一种基于三维动态特征的离线笔迹个体识别系统及方法

Also Published As

Publication number Publication date
CN110490153A (zh) 2019-11-22
US20230245501A1 (en) 2023-08-03
CN110490153B (zh) 2021-10-19

Similar Documents

Publication Publication Date Title
WO2021031446A1 (zh) 一种基于二维动态特征的离线笔迹个体识别系统及方法
Ross et al. Fingerprint warping using ridge curve correspondences
AU2004271639B2 (en) Systems and methods for biometric identification using handwriting recognition
WO2021031445A1 (zh) 一种基于三维动态特征的离线笔迹个体识别系统及方法
Feng et al. Robust and efficient algorithms for separating latent overlapped fingerprints
CN106778586A (zh) 离线手写签名鉴别方法及系统
CN104392229B (zh) 基于笔画片段方向分布特征的笔迹识别方法
Chansri et al. Hand gesture recognition for Thai sign language in complex background using fusion of depth and color video
US20110227923A1 (en) Image synthesis method
CN110796101A (zh) 一种嵌入式平台的人脸识别方法及系统
CN109858433B (zh) 一种基于三维人脸模型识别二维人脸图片的方法及装置
Rahman et al. Human ear recognition using geometric features
Ter Haar et al. A 3D face matching framework for facial curves
CN112185515A (zh) 一种基于动作识别的病患辅助系统
CN108090476A (zh) 一种针对具有外部遮挡的3d人脸识别方法
CN110222660B (zh) 一种基于动态与静态特征融合的签名鉴伪方法及系统
Segundo et al. Real-time scale-invariant face detection on range images
Wang et al. 3D face recognition by local shape difference boosting
Latha et al. A novel method for person authentication using retinal images
Bharadi et al. Multi-modal biometric recognition using human iris and dynamic pressure variation of handwritten signatures
Ravidas et al. Deep learning for pose-invariant face detection in unconstrained environment
Vezzetti et al. Application of geometry to rgb images for facial landmark localisation-a preliminary approach
Boukamcha et al. 3D face landmark auto detection
Lin et al. 3D face authentication by mutual coupled 3D and 2D feature extraction
CN112270287A (zh) 一种基于旋转不变的手掌静脉识别方法

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 19942501

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 19942501

Country of ref document: EP

Kind code of ref document: A1