CN105809100B - Person's handwriting and the signature quantization total area method of test sensitivity feature - Google Patents
Person's handwriting and the signature quantization total area method of test sensitivity feature Download PDFInfo
- Publication number
- CN105809100B CN105809100B CN201410856199.8A CN201410856199A CN105809100B CN 105809100 B CN105809100 B CN 105809100B CN 201410856199 A CN201410856199 A CN 201410856199A CN 105809100 B CN105809100 B CN 105809100B
- Authority
- CN
- China
- Prior art keywords
- feature
- value
- handwriting
- level
- conformity
- Prior art date
- Legal status (The legal status 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 status listed.)
- Expired - Fee Related
Links
Landscapes
- Collating Specific Patterns (AREA)
- Character Discrimination (AREA)
Abstract
Person's handwriting and the signature quantization total area method of test sensitivity feature belong to a kind of method of the inspection of document and identification technology field more particularly to person's handwriting and signature quantization test sensitivity in forensic science.The present invention provides a kind of objective, standard handwriting Signature quantization test sensitivity method.1) present invention is the following steps are included: construct handwriting characteristic hierarchical structure and value assessment analysis model；2) weighted value of each feature hierarchy is determined；3) the probability interval value of feature degree of conformity Yu feature occurrence rate is established；4) production grade feature deck watch calculates person's handwriting quantization test sensitivity degree of conformity and always accumulates with feature；5) corresponding relationship of quantization evaluation and test total value and test sensitivity opinion is established.
Description
Technical field
The invention belongs to the inspection of document in forensic science and identification technology field more particularly to a kind of person's handwritings and signature to quantify
The method of test sensitivity.
Background technique
Person's handwriting is that the writing system of symbols with personal touch is formed by writing activity.Verification of handwriting be surveyor in order to
Whether the hand-written writing being determined as on the material evidence of evidence is by a kind of a certain particular person is write and carries out Special Survey.Signature
Verification of handwriting occupies larger proportion in verification of handwriting work, is always forensic science since handwriting signature number of words is few, stroke is few
Difficulties in field.
Identified and sentenced mainly by the special knowledge and experience of handwriting expert currently, carrying out person's handwriting (signature) test sensitivity
It is disconnected, sometimes due to lacking enough, compellent objective basis, it is difficult to make expert's conclusion, or expert's conclusion is occurring
When disagreement, none objective standard and solid foundation are unified the understanding, and are caused difficulty to investigation and the administration of justice, are also made
The technology development of handwriting Signature test sensitivity is limited to, and science is under suspicion.
Summary of the invention
The present invention addresses the above problem, provides a kind of objective, standard handwriting Signature quantization test sensitivity method.
To achieve the above object, the present invention adopts the following technical scheme that, the present invention the following steps are included:
1) building handwriting characteristic hierarchical structure and value assessment analysis model；
2) weighted value of each feature hierarchy is determined；
3) the probability interval value of feature degree of conformity Yu feature occurrence rate is established；
4) production grade feature deck watch calculates person's handwriting quantization test sensitivity degree of conformity and always accumulates with feature；
5) corresponding relationship of quantization evaluation and test total value and test sensitivity opinion is established.
As a preferred embodiment, 1) building handwriting characteristic hierarchical structure and value assessment analysis model packet of the present invention
It includes: setting destination layer --- handwriting characteristic hierarchical structure and value；Determining middle layer --- macro-level feature, middle level are special
It seeks peace microcosmic level characteristics；Feature that refinement scheme layer --- feature that macro-level feature includes, middle level feature include,
The feature that microcosmic level characteristics include.
As another preferred embodiment, macro-level feature of the present invention is individual character feature, and intermediate level feature is outer
Shape, content, space, temporal characteristics, microcosmic level characteristics are two dimension, three-dimensional feature.
As another preferred embodiment, the macro-level feature in solution layer of the present invention includes: h1- feature is whole
Body layout, word line direction and ruling relationship；h2- whole style, writing level；h3Relationship between-entire combination and word；
Middle level feature in solution layer includes: z1- body and size；z2- structure content；z3- stroke and
Radical writes successive time sequencing；z4- stroke and radical blend proportion space；
Microcosmic level characteristics in solution layer include: w1- point, horizontal, vertical, left, flick, folding, curved, hook；w2It is shone between-stroke
Answer connection relationship and trend, connection form；w3The speed and the regularity of distribution of-writing speed；w4- write the weight of pressure and divide
Cloth rule.
As another preferred embodiment, 2) the weighted value calculating of the present invention for determining each feature hierarchy includes: in pen
When mark feature weight value determines, according to hierarchy Model, layer-by-layer Judgement Matricies from top to bottom；Each layer of element is all with phase
Adjacent upper level each element is criterion, compares Judgement Matricies two-by-two by 1 to 9 grade of scale, reuses method for root and calculates feature
Weighted value.
As another preferred embodiment, 3) the probability interval value of the present invention for establishing feature degree of conformity Yu feature occurrence rate
Including following division:
Section degree of conformity is divided into: complying fully with 100%, pole and meets 90%, meet very much 80%, relatively meet 70%, have centainly
Meet 60%；
Section diversity factor is divided into: complete difference -100%, diversity factor greatly -90%, diversity factor very big -80%, diversity factor
Larger -70%, there is different -60%；
The occurrence rate of each feature is divided into: only a few person writing 1%；Very few person writing 10%；Fewer person writing
30%；Respectively account for half person writing 50%；The writing 70% of majority.
As another preferred embodiment, 4) production grade feature deck watch of the present invention calculates person's handwriting quantization and examines mirror
Determining degree of conformity and feature, always product uses following formula:
The calculation formula of section degree of conformity and probabilistic relation:
viIt represents: (totally 11 grades) the section degrees of conformity of feature and the value of probabilistic relation at different levels, CiIt represents: each feature
Section degree of conformity；PiIt represents: the occurrence rate of single feature.
Secondly, the quantization evaluation and test total value and the corresponding relationship of test sensitivity opinion of the present invention of 5) establishing includes following step
It is rapid:
The total value V of case section degree of conformity and probabilistic relation:
V=VH+VZ+VW
Macro-level feature value: VH=(vh1×h1+vh2×h2+vh3×h3+vh4×h4)×H
Middle level feature value: VZ=(vz1×z1+vz2×z2+vz3×z3+vz4×z4)×Z
Microcosmic level feature value: VW=(vw1×w1+vw2×w2+vw3×w3+vw4×w4)×W。
It is closed in addition, case section of the present invention degree of conformity and the total value V of probabilistic relation are corresponding with test sensitivity opinion
System: 100 > v >=70 affirmative is same, and 70 >=v >=40 are likely to same, and 40 >=v >=0 may be same, and 0 >=v >=-40 may be non-same
One, -40 >=v >=-70 are likely to non-same, and -70 >=v > -100 negative is same.
The invention has the advantages that: the present invention to provide a kind of combination of qualitative and quantitative analysis, has checkout procedure modelling, thinks
The person's handwriting (signature) that the mathematicization of dimension process, check conclusion objectify quantifies test sensitivity analysis method.
Detailed description of the invention
The present invention will be further described with reference to the accompanying drawings and detailed description.The scope of the present invention not only limits to
In the statement of the following contents.(since black and white attached drawing cannot clearly show that shown content, retaining color)
Fig. 1 is invention's principle block diagram.
Fig. 2 is that macro-level aspect ratio of the invention is intended to expression.
Fig. 3 is that middle-level aspect ratio of the invention is intended to expression.
Fig. 4 is that microcosmic level aspect ratio of the invention is intended to expression.
Fig. 5 is that the present invention examines opinion and feature degree of conformity and occurrence rate relationship three-dimensional model diagram.
Fig. 6 is the corresponding relationship of person's handwriting of the present invention (signature) quantization test sensitivity assignment measure and calculate table and expert's conclusion
Specific embodiment
The present invention includes building person's handwriting (signature) handwriting characteristic hierarchical structure and value assessment analysis model, determines person's handwriting
The weighted value of (signature) each feature hierarchy, the probability interval value for establishing feature degree of conformity and feature occurrence rate, building person's handwriting (label
Name) quantization test sensitivity degree of conformity and feature always accumulate calculation method, establish quantization evaluation and test total value and test sensitivity opinion
Corresponding relationship.
Person's handwriting (signature) the handwriting characteristic hierarchical structure and value assessment analysis model are to utilize the level in operational research
Analytic approach building.Wherein, destination layer --- handwriting characteristic hierarchical structure and value assessment are set；Determine middle layer --- macroscopic view
Level characteristics, middle level feature and microcosmic level characteristics；Each category feature that refinement scheme layer --- macro-level feature includes,
Each category feature that middle level feature includes, each category feature that microcosmic level characteristics include.
Handwriting characteristic hierarchical structure and value assessment analysis model are as shown in table 1 below:
Table 1: handwriting characteristic hierarchical structure and value assessment analysis model
On the basis of constructing handwriting characteristic hierarchical structure and value assessment analysis model, four person's handwritings (label are constructed respectively
Name) feature value judgment matrix, and 1 to 9 grade of scaling law and method for root is utilized to complete the calculating of person's handwriting (signature) feature weight value,
Concrete outcome is shown in Table 2.
Table 2: person's handwriting (signature) feature weight value
Weighted value in table 2 can be calculated using root method and by consistency checking.
The feature degree of conformity and the probability interval value of feature occurrence rate include following division: the section of each feature meets
Spend (Ci): section degree of conformity be divided into comply fully with (100%), pole meets (90%), meets very much (80%), relatively meets (70%) and have
Centainly meet (60%)；Section diversity factor be divided into complete difference (- 100%) diversity factor greatly (- 90%), diversity factor it is very big (-
80%), diversity factor larger (- 70%), have different (- 60%)；Occurrence rate (the P of each featureiRepresent): only a few people's book
The feature occurrence rate value write is 1%；The feature occurrence rate value of very few person writing is 10%；The feature of fewer person writing occurs
Rate value is 30%；The feature occurrence rate value for respectively accounting for half person writing is 50%；The writing feature occurrence rate value of majority is 70%.
In specific test sensitivity implementation process, in order to ensure features at different levels are (see handwriting characteristic hierarchical structure and value
Analysis and assessment model) analysis accurately relatively comprehensively, three sight level characteristics deck watch should be made (see Fig. 2 to 4).
The calculation method that person's handwriting (signature) the quantization test sensitivity degree of conformity and feature are always accumulated is using following formula: section
The calculation formula of degree of conformity and probabilistic relation:
viIt represents: (totally 11 grades) the section degrees of conformity of feature and the value of probabilistic relation at different levels, all features referred to as at different levels
Value.
CiIt represents: the section degree of conformity of each feature；PiIt represents: the occurrence rate of single feature,
The corresponding relationship of quantization evaluation and test total value and test sensitivity opinion of the present invention: difference is respectively corresponded by quantization score value
Expert opinion statement, specific corresponding relationship is shown in Table 3:
Table 3: the corresponding relationship of quantization evaluation and test total value and test sensitivity opinion
Wherein, specific case quantization evaluation and test value is represented with V: the total value V of case section degree of conformity and probabilistic relation, it with
There are corresponding relationships for the case test sensitivity opinion.The calculation method of V are as follows:
V=VH+VZ+VW
Macro-level feature value: VH=(vh1×h1+vh2×h2+vh3×h3+vh4×h4)×H
Middle level feature value: Vz=(vz1×z1+vz2×z2+vz3×z3+vz4×z4)×Z
Microcosmic level feature value: VW=(vw1×w1+vw2×w2+vw3×w3+vw4×w4)×W
Reviewer can correlation model method and calculation formula through the invention, and according to specific case sample and sample
Concrete condition production three see deck watch, and according to each level characteristics be respectively compared respectively indicate feature similarities and differences, then
Statistical related law is combined according to the experience of test sensitivity personnel, according to probability interval, brings calculation formula into, calculating is provided
The feature degree of conformity and occurrence rate binary regression value of body case, specific case calculated value is contrasted with test sensitivity opinion, is given
Corresponding test sensitivity opinion out completes the quantization test sensitivity of person's handwriting (signature).
It is to be understood that being merely to illustrate the present invention above with respect to specific descriptions of the invention and being not limited to this
Technical solution described in inventive embodiments, those skilled in the art should understand that, still the present invention can be carried out
Modification or equivalent replacement, to reach identical technical effect；As long as meet use needs, all protection scope of the present invention it
It is interior.
Claims (1)
1. person's handwriting and signature quantization test sensitivity feature always accumulate method, it is characterised in that the following steps are included:
1) building handwriting characteristic hierarchical structure and value assessment analysis model；
2) weighted value of each feature hierarchy is determined；
3) the probability stepping value of feature degree of conformity Yu feature occurrence rate is established；
4) production grade feature deck watch calculates person's handwriting quantization test sensitivity degree of conformity and always accumulates with feature；
5) corresponding relationship of quantization evaluation and test total value and test sensitivity opinion is established；
1) building handwriting characteristic hierarchical structure and the value assessment analysis model include: setting destination layer --- handwriting characteristic layer
Secondary structure and value；Determine middle layer --- macro-level feature, middle level feature and microcosmic level characteristics；Refinement scheme
The feature that feature that layer --- feature that macro-level feature includes, middle level feature include, microcosmic level characteristics include；
The macro-level feature is individual character feature, and intermediate level feature is shape, content, space, temporal characteristics, microcosmic level
Feature is two dimension, three-dimensional feature；
Macro-level feature in the solution layer includes: h1- feature is integral layout, word line direction and ruling relationship；h2—
Whole style, writing level；h3Relationship between-entire combination and word；
Middle level feature in solution layer includes: z1- body and size；z2- structure content；z3- stroke and radical
Write successive time sequencing；z4- stroke and radical blend proportion space；
Microcosmic level characteristics in solution layer include: w1- point, horizontal, vertical, left, flick, folding, curved, hook；w2Correlate company between-stroke
Connect relationship and trend, connection form；w3The speed and the regularity of distribution of-writing speed；w4- write the weight of pressure and be distributed rule
Rule；
2) the weighted value calculating for determining each feature hierarchy includes: when handwriting characteristic weighted value determines, according to level knot
Structure model, from top to bottom layer-by-layer Judgement Matricies；Each layer of element is all using adjacent upper level each element as criterion, by 1 to 9
Grade scale compares Judgement Matricies two-by-two, reuses method for root and calculates feature weight value；
3) the feature degree of conformity and the probability stepping value of feature occurrence rate established includes following division:
Stepping degree of conformity is divided into: complying fully with 100%, pole and meets 90%, meet very much 80%, relatively meet 70%, having and centainly meet
60%；
Stepping diversity factor is divided into: complete difference -100%, diversity factor greatly -90%, diversity factor are very big -80%, diversity factor it is larger -
70%, there is different -60%；
The occurrence rate of each feature is divided into: only a few person writing 1%；Very few person writing 10%；Fewer person writing 30%；Respectively
Account for half person writing 50%；The writing 70% of majority；
4) the production grade feature deck watch, calculating person's handwriting quantization test sensitivity degree of conformity and feature, always product uses following public affairs
Formula:
The calculation formula of stepping degree of conformity and probabilistic relation:
ViIt represents: the stepping degree of conformity of features at different levels and the value of probabilistic relation, CiIt represents: the stepping degree of conformity of each feature；Pi
Represent: the occurrence rate of single feature, n represent feature quantity；
It is described 5) establish quantization evaluation and test total value and test sensitivity opinion corresponding relationship the following steps are included:
The total value V of case stepping degree of conformity and probabilistic relation:
V=VH+VZ+VW
Macro-level feature value: VH=(vh1×h1+vh2×h2+vh3×h3)×H
Middle level feature value: Vz=(vz1×z1+vz2×z2+vz3×z3+vz4×z4)×Z
Microcosmic level feature value:
VW=(vw1×w1+vw2×w2+vw3×w3+vw4×w4)×W；
vh1~vh3Respectively h1~h3Feature value quantized value；
vz1~vz4Respectively z1~z4Feature value quantized value；
vw1~vw4Respectively w1~w4Feature value quantized value；
H is the weight of macro-level feature；
Z is the weight of middle level feature；
W is the weight of microcosmic level characteristics.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410856199.8A CN105809100B (en) | 2014-12-31 | 2014-12-31 | Person's handwriting and the signature quantization total area method of test sensitivity feature |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410856199.8A CN105809100B (en) | 2014-12-31 | 2014-12-31 | Person's handwriting and the signature quantization total area method of test sensitivity feature |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN105809100A CN105809100A (en) | 2016-07-27 |
| CN105809100B true CN105809100B (en) | 2019-04-19 |
Family
ID=56465140
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410856199.8A Expired - Fee Related CN105809100B (en) | 2014-12-31 | 2014-12-31 | Person's handwriting and the signature quantization total area method of test sensitivity feature |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN105809100B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112990000A (en) * | 2021-03-11 | 2021-06-18 | 中国刑事警察学院 | Electronic signature handwriting characteristic typing and characteristic chain comparison method |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1701323A (en) * | 2001-10-15 | 2005-11-23 | 西尔弗布鲁克研究有限公司 | Digital ink database searching using handwriting feature synthesis |
| CN102693420A (en) * | 2012-05-25 | 2012-09-26 | 深圳市亚略特生物识别科技有限公司 | Automatic updating method for fingerprint template |
| CN103823880A (en) * | 2014-03-03 | 2014-05-28 | 国家认证认可监督管理委员会信息中心 | Attribute weight-based method for calculating similarity between detection mechanisms |
| CN104050468A (en) * | 2013-03-11 | 2014-09-17 | 日电（中国）有限公司 | Handwriting identification method, device and system |
-
2014
- 2014-12-31 CN CN201410856199.8A patent/CN105809100B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1701323A (en) * | 2001-10-15 | 2005-11-23 | 西尔弗布鲁克研究有限公司 | Digital ink database searching using handwriting feature synthesis |
| CN102693420A (en) * | 2012-05-25 | 2012-09-26 | 深圳市亚略特生物识别科技有限公司 | Automatic updating method for fingerprint template |
| CN104050468A (en) * | 2013-03-11 | 2014-09-17 | 日电（中国）有限公司 | Handwriting identification method, device and system |
| CN103823880A (en) * | 2014-03-03 | 2014-05-28 | 国家认证认可监督管理委员会信息中心 | Attribute weight-based method for calculating similarity between detection mechanisms |
Non-Patent Citations (2)
| Title |
|---|
| 刍议物证与物证特征的系统分类;李彪等;《中国司法鉴定》;20120630(第3期);第74-78页 |
| 签名笔迹量化检验研究用实验样本的设计与收集;卫燕茹等;《证据科学》;20141031(第5期);第632-641页 |
Also Published As
| Publication number | Publication date |
|---|---|
| CN105809100A (en) | 2016-07-27 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Sousa et al. | Technical efficiency of the Brazilian municipalities: correcting nonparametric frontier measurements for outliers | |
| Kelly et al. | The economic returns to field of study and competencies among higher education graduates in Ireland | |
| Agasisti et al. | Evaluating the efficiency of Italian public universities (2008–2011) in presence of (unobserved) heterogeneity | |
| Cai et al. | Asymmetric persistence in convergence for carbon dioxide emissions based on quantile unit root test with Fourier function | |
| Aparicio et al. | Economic crisis and public education. A productivity analysis using a Hicks-Moorsteen index | |
| Li et al. | Cluster analysis of the relationship between carbon dioxide emissions and economic growth | |
| CN105787522A (en) | Writing attitude evaluation method and writing attitude evaluation system based on handwriting | |
| Chen et al. | Quantifying circular urban expansion patterns of compact Chinese cities: the case of Yangtze River Delta, China | |
| CN109840466A (en) | Based on the comprehensive multiple measurement learning method of cluster and the overall situation/local distance | |
| CN105069268A (en) | Landscape visual impact comprehensive evaluation method based on GIS | |
| Yuan et al. | The educational efficiency evaluation framework: by using DEA model and CA method | |
| Lin et al. | The network structure of occupations: fragmentation, differentiation, and contagion | |
| CN105809100B (en) | Person's handwriting and the signature quantization total area method of test sensitivity feature | |
| Mazziotta et al. | A composite index for measuring Italian regions’ development over time | |
| CN104573726B (en) | Facial image recognition method based on the quartering and each ingredient reconstructed error optimum combination | |
| CN103955676B (en) | Human face identification method and system | |
| Bich | The cycle of transparency, accountability, corruption, and administrative performance: evidence from Vietnam | |
| CN112836137A (en) | Person network support degree calculation system and method, terminal, device, and storage medium | |
| Bramley | Multivariate representations of subject difficulty | |
| CN105893746B (en) | Material evidence quantifies test sensitivity method | |
| Seguino et al. | Contractionary monetary policy and the dynamics of US race and gender stratification | |
| Qu et al. | The risk factor analysis for software project based on the interpretative structural modelling method | |
| Yi et al. | Using DEA and CA Approachs for Educational Production Efficiency: Evidence from ShangHai PISA Measurement. | |
| CN107292213B (en) | Handwriting quantitative inspection and identification method | |
| Ganjour et al. | Gender inequality regarding retirement benefits in Switzerland |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| C10 | Entry into substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20190419 Termination date: 20211231 |