Journal of Forensic Science and Medicine

: 2018  |  Volume : 4  |  Issue : 3  |  Page : 167--173

Handwriting identification: Challenges and solutions

Liu Xiaohong1, Lian Yuanyuan2,  
1 Beijing Huaxia Evidence Identification Centre, China University of Political Science and Law, Beijing, China
2 Key Laboratory of Evidence Science, Ministry of Education, China University of Political Science and Law, Beijing; Collaborative Innovation Center of Judicial Civilization, China

Correspondence Address:
Lian Yuanyuan
No 25, Xitucheng Road, Haidian District, Beijing
Liu Xiaohong
9/F, Beike Building, No 27, Xisanhuan North Road, Haidian District, Beijing


As a historic technology, handwriting identification, especially that of Chinese characters, plays an important role in judicial trials both in domestic and abroad. However, in recent decades, handwriting identification has faced a challenging situation; the scientificity of handwriting identification has been questioned. This article analyzes the reasons for questioning handwriting identification, focusing on the aspects of standards, methods, quantitative analysis of handwriting identification, and qualification, training, and assessment of document examiners. Besides, corresponding solutions for the above aspects and some other characteristics, such as systematic identification, cross-examination of expert opinions, and rule of corroboration, are proposed.

How to cite this article:
Xiaohong L, Yuanyuan L. Handwriting identification: Challenges and solutions.J Forensic Sci Med 2018;4:167-173

How to cite this URL:
Xiaohong L, Yuanyuan L. Handwriting identification: Challenges and solutions. J Forensic Sci Med [serial online] 2018 [cited 2019 Jun 18 ];4:167-173
Available from:

Full Text


The importance of handwriting identification was emphasized by the century-heritage dispute case between “the first rich woman” Nina Wang and her father-in-law Wang Din-shin in Hong Kong, which took 8 years and costed hundreds of millions. The focus of the case was on the authenticity of Teddy Wang's signature in the will before and after. Nina Wang finally won the case, but the judgment was not based on handwriting identification opinions. The court held that the authenticity of Teddy Wang's signature was difficult to identify and handwriting experts on both the sides could not make any overwhelming conclusion. Finally, the judge made a decision based on the burden of proof and environmental testimony in common law.[1]

Recently, Taiwan Central News Agency reported that after Ye Chongde, the founder of the Germany organic company died, a dispute about his heritage broke out between his first and second wives. The lawsuit lasted for many years, and had been sent back by the Supreme Court several times. The judge convicted his wives and their children in the first, second and first retrial, based on the evidence that the will was forged, because professor Henry Lee, the forensic expert, and professor Chen Husheng of Taiwan's Central Police University, suggested that a seriously ill person cannot write such a will. However, in the second retrial, the collegial panel acquitted them based on the expert opinion of investigation bureau and other units that they were unable to confirm whether the will was forged because copy would affect the appraisal result. The final judgment was in favor of his first wife, who was proven not guilty, disregarding Dr. Henry Lee's expert opinion.[2]

From the above cases, it can be observed that handwriting identification has been faced with an embarrassing situation; on the one hand, experts think handwriting identification is scientific; on the other hand, handwriting identification has been questioned and criticized.

 The History and Principle of Handwriting Identification

Handwriting identification, used as evidence, has appeared and developed with the development of character and the social legal system. In China, using handwriting identification into trial has been recorded as early as in the Qin dynasty, and related records were more detailed after the Han dynasty, such as some of the records in Shih Chi and Three Kingdoms. However, the development of domestic handwriting identification into a more mature modern identification technology happened after the founding of new China, especially after the Chinese economic reform (simply termed as “reform and opening-up”). Internationally, historical materials have shown that the forged documents appeared almost simultaneously with words. Historically, the Romans developed the law about the expert testimony, and it was generally adopted by the court of common law of many countries until a few centuries ago.[3] After the establishment of the theory of writing habit in the 1930s, a mature theoretical system of handwriting identification gradually developed.

The particularity, stability, and reflective nature of handwriting are the scientific foundation of handwriting identification. Writing habit is formed after repeated practice, and it would be hard to § because of the general change in conditions, so everyone's writing habit is generally stable. Writing habits can be reflected through writing activities, and this kind of habits varies between different people. Beyond this, theory of individualism, principle of systematology, and the principle of Pavlov's theory of higher nervous activity constitute the theoretical basis of handwriting identification.[4] However, handwriting identification uses the more subjective, induction technique as its dominating research method.[5] Although the handwriting identification has been accepted widely as scientific evidence, in criminal, civil, and administrative cases, it authenticity has always been questioned.

 Challenges to Handwriting Identification in China and Abroad

In the United States, a strong challenge against handwriting identification started in the 1990s after the Patrick case. Three law professors, D. Michael Rising, Mark P. Denbeaux, and Michael J. Saks, criticized that handwriting identification was not professional knowledge. They argued that there was no proof for the existence of the profession of handwriting identification based on experience through empirical research. Their views grabbed the attention of forensic experts, juristical scientists, and evidence law scholars, but could not change the attitude of judges, who have been accepting handwriting identification opinions for a long term. The Daubert rule presented a really huge impact on the opinions of handwriting identification after the case Daubert v. Merrell Dow Pharmaceuticals in 1993, after which, some judges began to question the reliability of the handwriting expert opinion. Then, after the case Kumho Tire Co. V. Carmichael, some judges still admitted handwriting identification expert's testimony; some ruled it out. Some judges did not admit the handwriting experts' appraisal conclusion, but still allowed handwriting comparison test.[6] In short, in the United States, the handwriting identification opinion, once treated as solid scientific evidence, has received serious challenges, and are faced with an unprecedented predicament. In China, although questions about its authenticity are not as strong as in the United States, there are cases where parties are not satisfied with the expert opinion and express strong dissatisfaction through various approaches.

There are basically three main causes for the challenges that are faced: (i) the handwriting identification mainly relies on subjective experience, lacking the support of objective statistical method;[8] (ii) it lacks unified and comprehensive appraisal standards; (iii) the quality and ability of experts need to improve. The later two aspects are more prominent in China.

Question of the scientificity of handwriting identification

Both domestic and foreign scholars have raised questions about whether handwriting identification has a reliable scientific basis. Foreign scientists have argued that such a plausible hypothesis based on intuition is not scientific. To investigate whether each person's handwriting is stable, a lot of people need to be randomly chosen, classified and measurement on the changes of their handwriting, and some other researches need to be performed. Currently, there is no standardized measurement method and test database that can be used for this kind of scientific experiment.[7] Some domestic scholars point out that the principles of handwriting identification are yet to be confirmed, and the statement of its principle is a kind of metaphysical statement, lacking statistical research and calculation. The choice of handwriting characteristics lacks an objective standard, and handwriting identification experts choose handwriting characteristics according to their own subjective criteria.[8] However, some scholars point out that the subjective observation does not necessarily lead to the subjective capriciousness. Experience science, such as handwriting identification, cannot be negated just because some regular content may not be explained by current sciences. The professionals, engaged in handwriting identification, hold that handwriting identification has its scientific basis, and provide evidence for the same. American Professor Sargar organized a handwriting computer test collecting handwriting samples from 1568 people, and used computer algorithms to extract features, which is highly credible. The study of handwriting provides a strong support to scientific foundation of handwriting identification.[8] However, there is no denying that subjective observation method of handwriting identification cannot effectively control subjective arbitrariness in the process of identification, leading to the questions on the scientificity of handwriting identification.

Lack of unified and comprehensive evaluation standard

Unified and comprehensive standard is the fundamental guarantee of scientific conclusion of handwriting identification and a measurement of the credibility of appraisal conclusions.[9] Unified appraisal standard is more advantageous to unified appraisal methods and procedures, and more conducive to the surveyor to appear in court for cross-examination based on the unified standard. In brief, unified and comprehensive standards can make the appraisal opinions more convincing. However, there are no systematic and specified standards of handwriting identification in China. Comparatively speaking, handwriting identification standards are more detailed and comprehensive abroad. Current standards used by the nongovernmental judicial expertise units are the judicial authentication technical specifications issued by the Administration Agency of Judicial Authentication of the Ministry of Justice in 2010. According to handwriting identification code, expert witnesses can use professional knowledge and practical experience, to give a comprehensive opinion on handwriting features. There are nine kinds of handwriting identification opinions as well as the corresponding meaning in different grades and different levels, such as “most likely,” “likely” (often expressed as “tendency” in practice).[10] Compared with the standard of the Ministry of Justice, the standard made by the Ministry of Public Security is more detailed. The standard of the Ministry of Public Security divides handwriting identification into normal handwriting inspection, disguised handwriting inspection, and condition-change handwriting inspection.[11] In the United States, The Scientific Working Group for Forensic Document Examination (SWGDOC) issued guidelines to forensic document examination as early as in 2000, which is the first technical standard about forensic document examination.[12] Although it did not become a formal industry standard, the file provides reference and guidance to internal document inspection industry. Document inspection guidelines issued by SWGDOC in 2012 have provided more detailed provisions about handwriting inspection. Compared to the United States, the rule of thumb is still a dominant method in the domestic process of identification. Since there is no unified standard of handwriting identification, the appraisal conclusion issued by appraisal institutions cannot be compared objectively.[9] There is no distinguishing standard about handwriting features because the handwriting of different people has different features even for the same Chinese characters. Therefore, the appraisal personnel need to make a concrete analysis of each case.

Question the expert witnesses' professional quality

Some scientific research institutions in the United States have performed several experiments, to study the effect of appraisers' professional skills on handwriting identification results. Statistics show that the accuracy of appraisal result, made by professionals and laymen, varies a lot, but it also highlights that even professional appraisal is prone to certain errors on appraisal conclusion. Forensic Science Organization in the United States took handwriting comparison test experiments in the national crime labs from 1984 to 1989. The experimental results showed that when the handwriting samples of real cases were tested, the result accuracy was 70% for general cases, and for some more difficult cases, the result accuracy was only 40%. Related research institutions in the United States investigated the handwriting inspection accuracy through experiments comparing professionals with college students' ability in handwriting inspection. Experimental results showed that the accuracy of trained professionals is far higher than that of the laymen, but there was nearly 7% error rate. In 2002, research data organized by the United States Department of Justice about identifying true signature handwriting and imitation signature handwriting revealed that the error rate of well-trained professionals was 3.4%, and that of nonprofessional personnel was as high as 19.3%.[13] Statistical data from another research organized in 2010 for testing normal handwriting and forged handwriting pointed out that 23.1% of well-trained professionals gave the opinion that they could not ensure whether the tested handwriting was normal or forged in comparison to 8.4% by nonprofessional personnel. The data also pointed out that only 4.3% of professionals gave misleading opinions, while 12.2% of nonprofessionals gave misleading opinions on the case.[14] Hence, not only the professional ability of handwriting identification personnel would affect the appraisal result, but also the appraiser's subjective bias would impact it. Therefore, handwriting identification still needs effective program design to limit the impact of the subjective bias of expert witnesses on the appraisal result.[15] At present, the main problem faced by appraisers domestically is that the professional quality guarantee system is not perfect. There are many appraisers who do not have solid professional knowledge and rich practical experience. Furthermore, the administration departments grant appraisal rights only to organizations, paying no enough attention to the quality of the appraisers, which has reduced the quality of the appraisers to a certain degree.[16]

Handwriting identification is unique and different from technologies which can rely on physical and chemical experiments, equipment, and data analysis, because the role of handwriting identification personnel cannot be entirely replaced even if the appraisal standard is more complete and the equipment is more advanced. Therefore, the authenticity of handwriting identification depends on the appraiser's professional skills as well as is affected by the poor technical experience or deliberate false identification of expert witness.

 Suggestions to Improve the Scientificity of Handwriting Identification

Establish uniform and comprehensive identification standard

The lack of identification standard is especially prominent in China. Handwriting identification opinions lack credibility. The fundamental reason is that such identification technique lacks systematic and comprehensive scientific technical standards, in which experimental elements occupy a large part, scientific identification methods, and a complete and uniform standard system.[17] Therefore, we should refer to international identification standard and adopt suggestions from domestic experts on handwriting identification, trying to establish an examination document for handwriting identification as soon as possible.

The guidelines to forensic document examination issued by SWGDOC in 2000 are a standard worthy of reference. In the guidelines to forensic document examination, the contents about handwriting identification are composed of four correlative parts. Among them, the guidelines to written project examination is the subject and basis, and the other three guidelines focus on doubtful handwriting, nonoriginal handwriting, and disguised handwriting. The guidelines to written project examination stipulate specific procedures of handwriting identification: step one is the comparison between the check-material and samples and it is comparison of unknown handwritings between two or more writers; step two is to determine whether the sample is original; and step three is to make sure whether the examined sample is disguised using the following procedures: assessing the check-material, assessing the samples, comparing the check-material and the samples, and proposing conclusions. Based on the results, the next set of operation that needs to be adopted has also been pointed out. These seven identification procedures encompass the complete process of handwriting identification, so there are standards to be referred to for each step of handwriting identification.[12] Though there are many differences between identification of Chinese and English characters, there are still many commonalities in identification procedures, so they can be referred to reasonably. In addition, foreign handwriting identification experts have also proposed suggestions about handwriting identification. For example, Professor Zou Mingli from Southwest University of Political Science and Law proposed the following suggestions concerning handwriting identification standards: there should be minimum condition standards and degrees according to normal handwriting, conditional changing handwriting, disguised handwriting, imitated handwriting, signature handwriting, and number handwriting. The limits should have basic methods and recommended special methods, and there should be stipulations on the effectiveness and technical grades for each method. Irrespective of whether it is a basic method or special method, it should conform to the standards, i.e. no arguments; stable identification effects, high accuracy degree of identification practice proof, having passed expert review, and agreed to be used in identification. As for establishing, the identity document verification like handwriting identification, five basic technical indexes should be considered: characteristic cardinals of the check-material document; proportion of quantity and quality of the same characteristics between the check-material and sample document; proportion of quantity and quality of the different characteristics between the check-material and sample document; the general situation and nature of the same and different characteristics between the check-material and sample document (essentially agreeable and nonessentially different or nonessentially agreeable and essentially different); and in the scientific explanation of the general nature of the same and different characteristics between the check-material and sample documents, these five indexes are indispensable.[17]

If handwriting identification is to develop into a mature scientific technology, there should be certain standards and basis. On the basis of guaranteeing correct identification conclusions, the more detailed identification standards, the more possible it is to control the identifiers arbitrary, and the more persuasive the identification results. Furthermore, it can be applied in the court, and there is basis for both precourt investigation and on-court cross-examination, which can conform to requirements of “scientific evidence.”

Combining computer technology with handwriting identification

The combination of computer technology with handwriting identification is a development trend for handwriting identification and there have been a large number of studies both in China and abroad. As early as in 1966, the American IBM Company began researches on automatic character identification technology. In the following decades, computer automatic identification technology developed rapidly. In China, Tsinghua University cooperated with the Second Research Institute of The Ministry of Public Security of PRC and developed computer handwriting identification system in 2002, which won the technical recognition from the Ministry of Public Security in 2005. This is the first handwriting identification method that utilizes single character statistics in China and it solves computer automatic identification of normal Chinese character handwriting. According to relative statistics, the identification accuracy of the system can reach 92.69%, which is the world-leading standard.[18] In addition, other researchers and agencies have conducted researches on other aspects, such as wavelet transform [19] and texture analysis technology [20] for computer automatic identification. However, these technologies are still in the preliminary stage and have not been applied for handwriting identification dependably.

Some researchers applied computer technology into the quantitative research of signature handwriting identification and proposed to conduct quantitative analysis from three layers, which are macro, meso, and microscopic. The macroscopic layer focuses on the distribution and the direction of the signature, relationship with the ruling, the integral syntagmatic relation with the signature, and the value of the relationship between the two characters and their characteristics; the meso layer emphasizes on the shape, size, and specific appearance of the characters, specific content of the elements and the structure, the specific spatial relationship of the collocation and proportion, the writing order of the strokes and character component, and the value of the characteristics; the microscopic layer focuses on the basic stroke form, analysis and grading of joint part and form, writing speed and writing strength; after finishing the handwriting (signature), verifying the grades of all second-level indexes of quantitative evaluation standards, then all quantitative evaluation systems will calculate the grades of corresponding first-level index characteristic automatically, and finally confirm results of handwriting quantitative evaluation.[21] However, it is noteworthy to reflect that if quantitative analysis is emphasized excessively in handwriting identification, it might lead to the wrong path, so quantitative analysis of handwriting identification should be conditional and relative rather than absolute. Therefore, some scholars propose that in the field of handwriting identification, reasonable combination of quantitative analysis and qualitative analysis is the development trend of handwriting identification; the two methods should be conducted alternatively, and identifier's logical analysis, occupational skills, and judgment experience are still the main parts in integration, supplemented by quantitative statistics and analysis explanation, so that qualitative analysis and quantitative analysis can exert their separate advantages.[22]

Irrespective of whether it is a quantitative analysis or computer automatic handwriting identification technology, they cannot replace the role of professional identification staff in handwriting identification, since handwriting in its entirety is composed of multiple elements. Therefore, to verify handwriting, an integral analysis of various subjective and objective elements that form the handwriting and adherence to the principles of system analysis is necessary. Though quantitative analysis is based on the handwriting itself, issues concerning the writer's psychology and the environmental influence on the handwriting, which needs analysis from identification staff. Therefore, a combination of computer automatic identification technology such as quantitative analysis with identifier's identification is a more scientific and reasonable developmental trend for handwriting identification.

Enhance qualification, training, and evaluation of appraiser

Handwriting identification is queried at present, partially because some problems in professional skills of the appraiser exist. Particularly in China, the mixed professional skills of appraisers (good and bad) cannot be neglected. In practice, the professional ethics of a few appraisers are not high and can be reflected as follows: (i) facing complicated interpersonal relationship and tempted by material benefits; (ii) failure to stand firm; (iii) being content with the status quo and declining to be updated, resulting in aging and insufficiency of professional knowledge. This further leads to difficultly in satisfying requirements of practical work; (iv) short of responsibility awareness; (v) work carelessly; and (vi) being prejudiced by first impressions, and subjective assumptions which adversely influence their work.[23] Therefore, in order to improve credibility of appraisal, it is important to enhance qualification, training, and evaluation of appraisers.

The related regulations in terms of qualification, training, and evaluation of appraisers are still incomplete in China. By contrast, they are more detailed and comprehensive abroad and should be adapted. In the application for employment, there are no specific regulations on the number of years the applicant has worked abroad, but is focused on the applicant's basic professional skill education and training. In China, applicants (except applicants with senior professional and technical titles) are required to have at least 5 years of work experience (above 10 years of work experience for special major) related to the major with premise that they have related academic background. Therefore, in terms of talent acquisition, China pays more attention to employment experience instead of professional skill, whereas it works the other way around abroad. Of the systems related to training and evaluation of appraisers are more comprehensive abroad. For example, in the United States, prior to official employment, handwriting identification specialists shall receive 2–3 years of professional skill training and can only achieve appraiser qualification after passing qualification certification. In Australia, unified national and internal training, together with evaluation from the accrediting body, is essential for certification.[24] In the Netherlands, experts with qualification for appraiser shall still take regular training, which has become a systematic regulation. Similar to the Netherlands, Finland is also strict with training and requires appraisers to receive lifelong training after getting license. Besides, it also trains appraiser's ability to witness at court, specific study of related laws and regulations, code of conduct when appearing in court as a witness, and simulation of appearance in court.[11] Presently, in China, more attention is emphasized on common regulation and internal document for quality control of training of appraisers. For example, internal document in evidence identification center of the Ministry of Public Security, and regulation on judicial expertise training openly issued by Ministry of Justice, have neither made specific regulations on training and evaluation of appraisers. Compared with English handwriting identification, Chinese one is far more complicated, hence more attention should be focused on the formulation of specific regulations on qualification, training, and regular evaluation of appraisers.

Perfect the examination of the opinion, the evidence, and the rule of corroboration

The handwriting identification opinion is only a kind of legal evidence, so the identification opinion should be strictly examined and cross-examined, such as other material evidences, documentary evidences, and witness evidences. Regarding the identification opinion as “apotheosis” is not only a misconception of identification opinion, but also an irresponsible behavior for an objective and fair trial.

Under the Western adversarial system of lawsuit, whether the evidence could be adopted depends on the judges. However, whether the evidence has probative force for the case and the strength of the probative force is dependent on the jury. In China, since there is no jury to further examine the expert opinions, the judge should be more responsible for strictly evaluating the scientific evidence such as handwriting identification. As some scholars have said, “Scientific evidence cannot take the place of the jurisdiction of the court, because scientific evidence itself is uncertain.” Whether guilty or innocent depends on the judge and the jury.[8] As a consequence, the establishment of a strict pretrial review system, setting the expert opinion open, is beneficial not only for both parties to facilitate the preparation for the trial, but also for the judge to check the professional knowledge involved in the expert opinion.

It is more essential for the appraiser to make cross-examination on the court and to leverage its role for the opinion of the handwriting identification. Full cross-examination by the expert assistant appraiser, the parties in the trial hired by both sides, and one of the judges, for their expert opinion on technical investigation, is more conducive not only for the identification of exposed problems in the process, but also for the realization of justice of judiciary. Some people think that “Similar to handwriting identification, the expert opinion based on experience and knowledge requires substantive investigation for the specific identification process and the supporting reasons in the adoption of special needs.” Hence, the scientific expert conclusion should not be given unified treatment, and cross-examination should be considered to be particularly vital.[25]

Admissibility of expert opinion should give full play to the role of the rule of corroboration. Since scientific evidence is mostly indirect evidence, it requires other evidences to confirm each other, which cannot simply rely on an expert opinion for the final judgment. As for the handwriting identification opinion evidence, as an exception of the judges' free judgment of evidence to the principle of its establishment, there are two main reasons for the reinforcing rules: on the one hand, some people still doubt the scientificity of handwriting identification. If there is no reinforcement for the evidence with doubts, the suspect will spread to justice itself; on the other hand, the handwriting identification is quite professional, especially for the judge in our country. Without professional knowledge and due to excessive reliance on the handwriting expert opinion, the handwriting identification opinion might substitute the judgment of judges. The rule of corroboration could play the role of reviewing and correcting the opinion of handwriting identification.[4] Therefore, for the handwriting appraisal opinion, it especially needs the court to leverage the role of examination and cross-examination of evidence.


The questioning about handwriting identification is not completely correct, but it is reasonable. Handwriting identification does have a few defects, whether on scientific principle or standard system. However, handwriting identification is scientific to a certain extent because it has existed for thousands of years and has been developing until now. We should be objective and tolerant for handwriting identification. After all, scientific evidence such as handwriting identification cannot solve all the judicial issues. As the American scholar John Hord said, “no matter how significant progress the science will make, we cannot solve all the judicial issues, because science is a relative term and its accuracy could never reach 100%.” On the other hand, the question about handwriting identification will promote its development toward a more scientific direction, such as making more comprehensive and scientific appraisal standard, improve appraisal personnel training assessment mechanism, and promote research on automatic identification. Questioning these techniques will bring about a forced development of handwriting identification, which is bound to have a more broad space for development in the future.


This study was supported by Program for school-level Research in China University of Political Science and Law (16ZFQ82008).

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1Yanan W. On evidentiary effectiveness of handwriting identification – In the view of handwriting identification in the case of fighting for the century heritage. Chin J Forensic Sci 2009;6:49-53.
2Available from: 03004599-260402.
3Xiaonan Li. A comparative study of Chinese and English handwriting identification methods [D]. Southwest University of Political Science and Law, 2008.
4Zhichun D, Yuansheng S. The principle and legal interpretation of handwriting identification evidence. ECUPL J 2009;2:34-40.
5Jing L. The scientific analysis of handwriting identification. J Liaoning Police Coll 2008;5:89-92.
6Fumei Z. Scientific Appraisal and Criminal Investigation; 2006.
7Min Y, Lee HC. The application and development of scientific evidence in the United States. J Pol Sci Law 2013;30:96-106.
8Baoguo F. Distortion and prevent of scientific evidence. J Lanzhou Univ 2012;40:107-13.
9Lin WL. Discuss about the standard of handwriting identification conclusions. Technol Manage 2011;10:24-6.
10Yanru W, Xiangchen W, Bin Y. Design and collection of experimental samples for signature quantitative test. Evid Sci 2014;22:632-41.
11Bing L. Comparative studies on quality control of documentary forensic professionals. Evid Sci 2015;23:350-8.
12Zebo S, Na L. Evaluate and analyse of guidelines for forensic document examination. Police Science and Technology 2005,4:63-5.
13Sita J, Found B, Rogers DK. Forensic handwriting examiners' expertise for signature comparison. J Forensic Sci 2002;47:1117-24.
14Bird C, Found B, Rogers D. Forensic document examiners' skill in distinguishing between natural and disguised handwriting behaviors. J Forensic Sci 2010;55:1291-5.
15Stoel RD, Dror LE, Miller LS. Bias among forensic document examination: Still a need for procedural changes. Aust J Forensic Sci 2014;46:91-7.
16Jin L. Analysis of differences reason of handwriting identification conclusions. Chin J Forensic Sci 2006;4:25-7.
17Mingli Z. The difficult road of the standardization of forensic verification of document should take advantage of opportunity. Chin Forensic Verification 2009;3:39-45.
18Yong W. Analysis of the application of automatic handwriting verification in handwriting verification. Silicon Valley 2011;22:127.
19Huidang Z, Yuyao H. Handwriting verification methods based on wavelet transform and nerve grid integration. Res Comput Appl 2008;25:741-3.
20Lin FC, Mijaljica D, Sarkar S, Zoladek T. Autophagy researchers: Fu-Cheng lin, Dalibor Mijaljica, Sovan Sarkar, Teresa Zoladek. Autophagy 2014;10:552-5.
21Bin Y. Handwriting (Signature) research and implementation of quantitative inspection and appraisal analysis evaluation system. J China Crim Police Coll Instit 2013;3:44-6.
22Xiaoming C. On quantitative inspection in handwriting verification. Rule Law Soc 2014;11:167-8.
23Yi C. Realistic problem and future prospects of document examination. Sichuan Police Coll J 2005;6:53-6.
24Vbelaker DH, editor. The Global Practice of Forensic Science. Vol. 1. Hoboken, NJ: John Wiley and Sons Inc.; 2015. p. 23.
25Guixue Z. To Determine the Position and Qualification of Expert Opinion – Thinking of Identification of the Selection System in the Lawsuit of Seeking Truth from Facts; 2012. p. 76-8.