CN106203135B - A kind of passive measuring method for RSID hiding information - Google Patents
A kind of passive measuring method for RSID hiding information Download PDFInfo
- Publication number
- CN106203135B CN106203135B CN201610517202.2A CN201610517202A CN106203135B CN 106203135 B CN106203135 B CN 106203135B CN 201610517202 A CN201610517202 A CN 201610517202A CN 106203135 B CN106203135 B CN 106203135B
- Authority
- CN
- China
- Prior art keywords
- document
- value
- container
- rsid
- noise
- 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
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F21/00—Security arrangements for protecting computers, components thereof, programs or data against unauthorised activity
- G06F21/60—Protecting data
- G06F21/602—Providing cryptographic facilities or services
Abstract
The present invention relates to a kind of passive measuring methods for RSID hiding information.This method comprises: extracting settings.xml and document.xml file data from DOCX document；The attribute value that w:rsid is extracted from settings.xml file, is stored in the first container；The attribute value that prefix is w:rsid is extracted from document.xml file, is stored in second container；It is sorted from small to large to the element in second container according to numerical values recited, and removes repeat element；According to the element in the first container and second container, the noise figure of DOCX document to be detected is calculated；According to detection pattern set by user, detection threshold value is determined；Compare the noise figure of document and the size relation of detection threshold value, determines whether document hides classified information.The present invention is suitable for detecting whether some DOCX document using RSID mark hides classified information.
Description
Technical field
The present invention relates to the Stego-detection technical field of Information hiding, more particularly to a kind of for RSID hiding information
Passive measuring method.
Background technique
In order to more efficiently operate and management document, Microsoft utilizes a kind of new Office from after 2007 version of Office
Open XML (OOXML) format stores Office document.It, can be with since OOXML format uses ZIP file compress technique
Disk space needed for reducing storage file, and reduce through bandwidth needed for Email, network and Web transmission file.
In addition, OOXML format compared with binary format, also has the advantages such as interoperability, reliability, high efficiency and safety.It repairs
Change a kind of important element that identifier (Revision Identifier, RSID) is OOXML format, it can uniquely be identified
Every time to the editor of document.The DOCX file benefit of Microsoft's Office Word document (including 2007/2010/2013 version of Word)
Modification of the user to document is tracked with RSID mark, and can merge more parts of different revisions according to RSID.But
The RSID of DOCX file is not necessary to resolution file.Therefore, many users using the RSID of DOCX file by being identified
Redundancy be embedded in classified information, be snugly transferred to message receiver.
As the update of Office Word software upgrades, Most current Word document is carried out using DOCX format
Storage.In terms of the Information hiding research based on RSID mark redundant content, there is special steganography tool software, such as
Brain etc..Brain software be it is a it is special for RSID mark steganography software, it support DOCX format file carrier and
Plaintext message insertion.Brain tool is hiding to be embedded in by 3, the end byte data for directly replacing RSID identity property value
Message has very strong concealment because of the content without DOCX document after change modification.Due to each RSID identity property value
3 byte informations can be hidden, and are identified usually in DOCX document comprising a large amount of RSID, therefore the steganography of brain software
Capacity is very big.
Currently, some research achievements are had been achieved in the research for identifying hiding information and context of detection based on RSID.
Liberation army Polytechnics Xu Min in 2009 et al. proposes a kind of new method of Word2007 document hiding information, and this method utilizes
The characteristics of Word XML, proposes the Information hiding scheme (reference: what 2007 document information of Word was hidden of replacement " modified logo "
New method, " Journal of Computer Research and Development ", 46:112-116,2009), Zhengzhou University Dong Yan in 2015 et al. utilizes pseudorandom two
System sequence extracts specific " modified logo " attribute value of bearer documents, and proposing a kind of innovatory algorithm, (reference: one kind is based on
The Information Hiding of Word XML, " computer technology and development ", the 2nd phase, 2015).Italian Naples in 2011
Second university Castiglione et al. proposes a kind of Information Hiding Algorithms (reference: New for replacing " modified logo " attribute value
Steganographic Techniques for the OOXML File Format, ARES 2011, LNCS 6908,
Pp.344-358,2011).Chinese invention patent 201210424754.0 in 2012 proposes a kind of based on the OOX for deleting label
Watermark information is embedded into the symbol w:rsidDel (reference: patent of the deletion modified logo in OOX document by document digital watermark method
CN102968596A is a kind of based on the OOX document digital watermark method for deleting label, application number 201210424754.0).But
It is found by literature survey analysis, currently yet there are no open text in terms of the detection method research for modified logo hiding information
It offers.
Summary of the invention
The technical problem to be solved by the present invention is to the passive detections of RSID mark hiding information, provide a kind of for RSID
The passive measuring method of hiding information is not needing original load suitable for detecting whether some text document includes hiding information
Detect that hiding information whether there is under the premise of body file.Particularly, whether the present invention is suitable for detecting some DOCX document sharp
Classified information is hidden with RSID mark.
A kind of the technical solution of the invention is as follows passive measuring method for RSID hiding information, passes through utilization
The characteristics of OOXML format, it is only necessary to according to current detection document it is determined that whether DOCX document contains hiding information, mainly
Include the following steps:
(1) DOCX document to be detected is parsed, settings.xml and document.xml file data is extracted；
(2) settings.xml file is parsed, the attribute value of w:rsid is extracted from settings.xml file, and will knot
Fruit is stored sequentially in the first container；
(3) document.xml file is parsed, mode is traversed according to directory tree preamble, is mentioned from document.xml file
Taking prefix is the attribute value of w:rsid, and result is stored sequentially in second container；
(4) it sorted from small to large to the element in second container according to numerical values recited, remove repeat element；
(5) according to the element in the first container and second container, the noise figure of DOCX document to be detected is calculated；
(6) according to detection pattern set by user, detection threshold value is determined；
(7) by comparing the size relation of document noise figure and detection threshold value, determine whether document hides classified information.
Further, step 1) parses DOCX document to be detected using ZIP file format mode.
Further, it is 4 table of bytes that step 2) sequence, which successively extracts the attribute value of w:rsid in settings.xml file,
The character string is converted into corresponding decimal system integer, and result is stored sequentially in by the hexadecimal values character string shown
In the first container.
Further, step 3) traverses mode according to directory tree preamble, and sequence is successively extracted in document.xml file
Prefix is that the attribute value of w:rsid is the hexadecimal values character string of 4 byte representations, which is converted into corresponding ten
System integer, and result is stored sequentially in the second container.
Further, the step 6) detection pattern include: the low mode of priority of false dismissed rate, the low mode of priority of false alarm rate, at
Power mode of priority.
" container " of the present invention can be using sequential core-pullings such as vector, deque, list in C++ container class.
The beneficial effect of the present invention compared with prior art is:
(1) in the present invention, a kind of algorithm for calculating DOCX document noise figure is given, calculation method is suitable for any
.docx file obtains a kind of intrinsic essential characteristic that document noise figure characterizes document.
(2) in the present invention, whether classified information is hidden based on a kind of detection DOCX document of document noise figure feature extraction
Detection method.
(3) in the present invention, detection method is a kind of passive measuring method, it requires no knowledge about initial carrier document data,
It can be realized as whether including that secret information differentiates to document only by the feature using document to be detected itself.
(4) in the present invention, user can freely select three kinds of different detection patterns with actual demand.Detection algorithm can
The analysis detection that suitable algorithm threshold value is hidden message is arranged, to meet difference in the detection pattern according to set by user
The practical application of user needs.
(5) in the present invention, the training algorithm of general threshold value T a kind of is given, arbitrary user's application scenarios are suitable for,
Using given training algorithm threshold value T, the false dismissed rate and false alarm rate of detection algorithm under set scene can reduce.
Detailed description of the invention
Fig. 1 is the implementation flow chart of embodiment of the present invention method；
Fig. 2 is the logical construction schematic diagram of embodiment of the present invention method DOCX document；
Fig. 3 is the schematic diagram of settings.xml file<w:rsids>element in the method for the present invention；
Fig. 4 is the directory tree structure schematic diagram of document.xml file in the method for the present invention；
Fig. 5 is the flow diagram that DOCX document noise figure is calculated in the method for the present invention.
Specific embodiment
In order to make the foregoing objectives, features and advantages of the present invention clearer and more comprehensible, below by specific embodiment and
Attached drawing, the present invention will be further described.
As shown in Figure 1, be the present invention realize passive detection flow diagram, it is of the invention the specific implementation process is as follows:
(S1) DOCX document to be detected is parsed, settings.xml and document.xml file data is extracted.
DOCX document to be detected is inputted for user, is parsed using ZIP file format mode, available document
Logical construction be a bibliographic structure.By taking Example.docx document as an example, directory tree structure is as shown in Fig. 2, can send out
It include settings.xml and document.xml file under present ./word/ catalogue.For the detection in subsequent step, need
Extract ./word/settings.xml and ./word/document.xml file data.
(S2) w:rsid attribute value is extracted from settings.xml file, and stores the result into container v0。
The settings.xml file extracted in processing step S1, in settings.xml root element < w:
Settings>in include daughter element<w:rsids>, and element<w:rsids>is by 1 w:rsidRoot element and several w:
Rsid element composition.Successively sequence extract w:rsid in attribute w:val attribute value, it be 4 byte representations 16 into
Numeric character string processed namely the character string include 8 characters and each character is ' 0 '~' 9 ' or ' A '~' F '.By above-mentioned word
Symbol string is converted into corresponding decimal integer according to hexadecimal number, and the integer is successively stored according to sequence of extraction to container
vector<int>v0。
By taking Example.docx document as an example, content such as Fig. 3 institute of element<w:rsids>in settings.xml file
Show.The attribute value for successively extracting attribute w:val in w:rsid is character string " 00300F3E " and " 005579D9 ", is converted it into
Decimal system integer is respectively equal to 3,149,630 and 5,601,753.Then successively sequential storage to container vector<int>v0
In obtain, v0[0]=3,149,630, v0[1]=5,601,753.
(S3) mode is traversed according to directory tree preamble, the attribute that prefix is w:rsid is extracted from document.xml file
Value, and store the result into container v1。
The document.xml file extracted in processing step S1, the root element<w:document>of the xml document
Comprising daughter element<w:body>, and element<w:body>is made of daughter elements such as several<w:p>,<w:r>and<w:sectPr>,
And it can nested multiple daughter elements in each daughter element.Xml formatted file is a directory tree structure, is traversed using the preamble of tree
Algorithm may search for each node (element) of document.Then matching element property name prefix in each node is " w:rsid "
Attribute only exists 4 attribute in document.xml file, is w:rsidP, w:rsidR, w:rsidRPr and w respectively:
rsidRDefault.The attribute value of these attribute-names can be obtained, it is the hexadecimal values character of 4 byte representations
String namely the character string include 8 characters and each character is ' 0 '~' 9 ' or ' A '~' F '.By above-mentioned character string according to ten
Senary number is converted into corresponding decimal integer, and the integer is successively stored according to sequence of extraction to container vector < int
>v1。
By taking Example.docx document as an example, the content of document.xml file is as shown in Figure 4.According to node preamble
Traversal mode is searched for, successively extract attribute-name prefix be " w:rsid " attribute value be character string " 005579D9 ",
" 00300F3E " and " 005579D9 " converts it into decimal system integer and is respectively equal to 5,601,753,3,149,630 and 5,
601,753.Then successively sequential storage to container vector<int>v1In obtain, v1[0]=5,601,753, v1[1]=3,
149,630、v1[2]=5,601,753.
(S4) to v1In element sorted from small to large according to numerical values recited, remove repeat element.
Container v in processing step S31In element, due to v1In each element be integer, therefore can use sequence
Algorithm is to it according to being ranked up from small to large.Then container v is deleted1In repeat element, generate one orderly and without weight
The v of complex element1。
With the container v obtained in step S31For, the v that generates after the duplicate removal that sorts1Are as follows: v1[0]=3,149,630, v1[1]
=5,601,753.
(S5) v is utilized0And v1Calculate the characteristic noise value N of DOCX document.
Pass through the container v obtained using step S20The container v generated with step S41, calculate making an uproar for DOCX document to be detected
Sound value N, as shown in figure 5, circular is as follows:
(S5-1) successively sequence reads v1Each element x until all elements reading finish；
(S5-2) sequential search v0In element, if discovery v0It is middle to then follow the steps S5-1 in the presence of the element equal with x value,
It is no to then follow the steps S5-3；
(S5-3) decimal system integer x is expressed as to the binary number of 32 bits, an insufficient high position is filled with ' 0 ',
And sequence<1,2 ... is used, 32>according to the order from high significance bit to low order mark this 32 bits, record n value etc.
In there is the minimum ordinal number of bit ' 1 '；
(S5-4) n value is added to document noise figure N (initial value N=0), executes step S5-1.
According to step S5-1 to step S5-4 is calculated, the calculation expression of the characteristic noise value N of DOCX document to be detected is,
Wherein nxIt indicates to start counting from most significant bit to first appearing ratio in the integer x that 32 bit-binaries characterize
The ordinal number of spy ' 1 ',Indicate x in container v1In but not in container v0In.
(S6) the threshold value T of detection algorithm is determined.
According to detection pattern set by user, including the low mode of priority of false dismissed rate, the low mode of priority of false alarm rate and success rate
Mode of priority, determines the threshold value T of detection algorithm, and the corresponding relationship of detection pattern and threshold value T are as shown in table 1.I.e. when selection false dismissal
When the low mode of priority of rate, threshold value T=16；When the low mode of priority of selection false alarm rate, threshold value T=128；It is preferential when being chosen to power
When mode, threshold value T=48.
The mapping table of table 1 detection pattern and threshold value T
It is assumed that event A indicates that document containing hiding information is determined as that, without hiding information, event B is indicated without hiding information text
Shelves are determined as containing hiding information, then the false dismissed rate P of detection algorithmMD, false alarm rate PFA, success rate PSUCCIt can be expressed as,
It is worth noting that, threshold value T is not limited to the value in table 1, user can determine according to actual needs, table 1
The value of middle threshold value T is the empirical value obtained according to usual model training, has generality.
(S7) by comparing the size of N value and T value, determine whether document hides classified information.
By comparing the size for setting T value in N value and step S6 is calculated in step S5, it is secret to determine whether document is hidden
Close message, specific determination method are as shown in table 2.I.e. as N > T, DOCX document to be detected contains hiding information；As N≤T
When, DOCX document to be detected does not contain hiding information.
Table 2 determines whether document contains steganography information
| Fiducial value | Determine result |
| N > T | Contain |
| N≤T | It does not contain |
The training algorithm specific implementation step of heretofore described threshold value T is as follows:
Step 1: selecting K to positive, negative sample in the set scene of user, remember positive sample set
Negative sample setUsually require that sample size K >=500；
Step 2: calculating the characteristic noise value of sample using step S1 in detection algorithm to step S5, obtain result as the positive
The noise value set of sampleThe noise value set of negative sampleWhereinTable
Show certain single sampleCharacteristic noise value；
Step 3: noteIndicate positive sample noise value set N+In minimal noise value,Indicate that negative sample is made an uproar
Sound value set N-In maximum noise value, i.e.,Then basis
Training set data calculates threshold value T, and the results are shown in Table 3, specific as follows,
(1) if the low mode of priority of selection false dismissed rate, threshold value
(2) if the low mode of priority of selection false alarm rate, threshold value
(3) if being chosen to power mode of priority, the calculation expression of threshold value T is,
(3-a) whenWhen,
(3-b) whenWhen,
Table 3 calculates threshold value T according to training set
The above embodiments are merely illustrative of the technical solutions of the present invention rather than is limited, the ordinary skill of this field
Personnel can be with modification or equivalent replacement of the technical solution of the present invention are made, without departing from the spirit and scope of the present invention, this
The protection scope of invention should be subject to described in claims.
Claims (6)
1. a kind of passive measuring method for RSID hiding information, which comprises the following steps:
1) DOCX document to be detected is parsed using ZIP file format, extracts settings.xml and document.xml text
Number of packages evidence；
2) settings.xml file is parsed, the attribute value of w:rsid is extracted from settings.xml file, and result is deposited
Storage is in the first container；
3) document.xml file is parsed, mode is traversed according to directory tree preamble, before extracting in document.xml file
Sew be w:rsid attribute value, and store the result into second container；
4) it is sorted from small to large to the element in second container according to numerical values recited, and removes repeat element；
5) according to the element in the first container and second container, the noise figure of DOCX document to be detected is calculated, is included the following steps,
Wherein v0Indicate the first container, v1Indicate the second container:
5-1) successively sequence reads v1Each element x until all elements reading finish；
5-2) sequential search v0In element, if discovery v0It is middle to then follow the steps 5-1 in the presence of the element equal with x value), otherwise hold
Row step 5-3)；
The decimal system integer x 5-3) is expressed as to the binary number of 32 bits, insufficient high-order ' 0 ' filling, and is adopted
With sequence<1,2 ..., 32>according to the order from high significance bit to low order this 32 bits being marked, record n value, which is equal to, to be occurred
The minimum ordinal number of bit ' 1 '；
N value 5-4) is added to document noise figure N, initial value N=0, then executes step 5-1)；
According to step 5-1) arrive step 5-4), the calculation expression of the characteristic noise value N of DOCX document to be detected is,
Wherein nxIt indicates to start counting from most significant bit to first appearing bit ' 1 ' in the integer x that 32 bit-binaries characterize
Ordinal number；
6) according to detection pattern set by user, detection threshold value is determined；
7) by comparing the noise figure of DOCX document to be detected and the size relation of detection threshold value, determine whether document hides secret
Message.
2. the method as described in claim 1, it is characterised in that: step 2) sequence successively extracts w in settings.xml file:
The attribute value of rsid is the hexadecimal values character string of 4 byte representations, which is converted into corresponding decimal system integer
Number, and result is stored sequentially in the first container.
3. the method as described in claim 1, it is characterised in that: step 3) traverses mode according to directory tree preamble, and sequence is successively
Extracting the attribute value that prefix in document.xml file is w:rsid is the hexadecimal values character string of 4 byte representations, will
The character string is converted into corresponding decimal system integer, and result is stored sequentially in the second container.
4. the method as described in claim 1, which is characterized in that the step 6) detection pattern includes: the low preferential mould of false dismissed rate
The low mode of priority of formula, false alarm rate, success rate mode of priority.
5. method as claimed in claim 4, which is characterized in that false dismissed rate P in step 6)MD, false alarm rate PFA, success rate PSUCC's
Calculation method is:
Wherein, event A indicates that document containing hiding information is determined as that, without hiding information, event B indicates to be free of hiding information document
It is determined as containing hiding information.
6. method as claimed in claim 4, which is characterized in that the training method of the step 6) threshold value is as follows:
It 6-1) selects K to positive, negative sample in the set scene of user, remembers positive sample setNegative sample
This setSample size K >=500；
The characteristic noise value for 6-2) calculating sample obtains the noise value set that result is positive sampleIt is negative
The noise value set of sampleWhereinIndicate certain single sampleCharacteristic noise value；
6-3) rememberIndicate positive sample noise value set N+In minimal noise value,Indicate negative sample noise figure collection
Close N-In maximum noise value, i.e.,Then according to training set
Data calculate threshold value T's as a result, specific as follows:
(1) if the low mode of priority of selection false dismissed rate, threshold value
(2) if the low mode of priority of selection false alarm rate, threshold value
(3) if being chosen to power mode of priority, the calculation expression of threshold value T is,
(3-a) whenWhen,
(3-b) whenWhen,
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610517202.2A CN106203135B (en) | 2016-07-04 | 2016-07-04 | A kind of passive measuring method for RSID hiding information |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201610517202.2A CN106203135B (en) | 2016-07-04 | 2016-07-04 | A kind of passive measuring method for RSID hiding information |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN106203135A CN106203135A (en) | 2016-12-07 |
| CN106203135B true CN106203135B (en) | 2019-07-16 |
Family
ID=57465935
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201610517202.2A Expired - Fee Related CN106203135B (en) | 2016-07-04 | 2016-07-04 | A kind of passive measuring method for RSID hiding information |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106203135B (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1920877A (en) * | 2006-09-19 | 2007-02-28 | 北京邮电大学 | Statistic supervision and structure supervision based hidden messages analysis system |
| CN102968596A (en) * | 2012-10-30 | 2013-03-13 | 南京信息工程大学 | Delete marker-based office open xml (OOX) document digital watermarking method |
| CN104699661A (en) * | 2015-01-29 | 2015-06-10 | 中国科学院信息工程研究所 | Selecting method and system of privacy code words facing Unicode coded documents |
| CN105046159A (en) * | 2015-06-18 | 2015-11-11 | 中国科学院信息工程研究所 | Modification identifier based OOX text document privacy information detection method |
-
2016
- 2016-07-04 CN CN201610517202.2A patent/CN106203135B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN1920877A (en) * | 2006-09-19 | 2007-02-28 | 北京邮电大学 | Statistic supervision and structure supervision based hidden messages analysis system |
| CN102968596A (en) * | 2012-10-30 | 2013-03-13 | 南京信息工程大学 | Delete marker-based office open xml (OOX) document digital watermarking method |
| CN104699661A (en) * | 2015-01-29 | 2015-06-10 | 中国科学院信息工程研究所 | Selecting method and system of privacy code words facing Unicode coded documents |
| CN105046159A (en) * | 2015-06-18 | 2015-11-11 | 中国科学院信息工程研究所 | Modification identifier based OOX text document privacy information detection method |
Also Published As
| Publication number | Publication date |
|---|---|
| CN106203135A (en) | 2016-12-07 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109445834B (en) | Program code similarity rapid comparison method based on abstract syntax tree | |
| US8938461B2 (en) | Method for organizing large numbers of documents | |
| Xiang et al. | Linguistic steganalysis using the features derived from synonym frequency | |
| US20170279840A1 (en) | Automated event id field analysis on heterogeneous logs | |
| CA2743273C (en) | Method and device for intercepting junk mail | |
| CN108182370B (en) | Client webpage tampering detection method based on dynamic and static separation template | |
| US8316041B1 (en) | Generation and processing of numerical identifiers | |
| CN107688748A (en) | Fragility Code Clones detection method and its device based on leak fingerprint | |
| CN104778158A (en) | Method and device for representing text | |
| KR20120124581A (en) | Method, device and computer readable recording medium for improvded detection of similar documents | |
| US20070027882A1 (en) | Record boundary identification and extraction through pattern mining | |
| CN111159387A (en) | Recommendation method based on multi-dimensional alarm information text similarity analysis | |
| CN104765882B (en) | A kind of internet site statistical method based on web page characteristics character string | |
| CN106960058A (en) | A kind of structure of web page alteration detection method and system | |
| CN110007906B (en) | Script file processing method and device and server | |
| US20230418578A1 (en) | Systems and methods for detection of code clones | |
| CN105653949A (en) | Malicious program detection method and device | |
| Shi et al. | An approach to text steganography based on search in internet | |
| Fu et al. | Digital forensics of Microsoft Office 2007–2013 documents to prevent covert communication | |
| CN106203135B (en) | A kind of passive measuring method for RSID hiding information | |
| CN106780280B (en) | Digital watermarking encryption method and device | |
| US8984028B2 (en) | Systems and methods for storing data and eliminating redundancy | |
| CN110399485B (en) | Data tracing method and system based on word vector and machine learning | |
| CN115344538A (en) | Log processing method, device and equipment and readable storage medium | |
| CN112925755A (en) | Intelligent storage method and device for ultra-long path of file system |
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 |
Granted publication date: 20190716 Termination date: 20210704 |
|
| CF01 | Termination of patent right due to non-payment of annual fee |