CN109359705A - A kind of dislocation books search system and its lookup method - Google Patents
A kind of dislocation books search system and its lookup method Download PDFInfo
- Publication number
- CN109359705A CN109359705A CN201810961866.7A CN201810961866A CN109359705A CN 109359705 A CN109359705 A CN 109359705A CN 201810961866 A CN201810961866 A CN 201810961866A CN 109359705 A CN109359705 A CN 109359705A
- Authority
- CN
- China
- Prior art keywords
- books
- dislocation
- image
- list
- call number
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06K—GRAPHICAL DATA READING; PRESENTATION OF DATA; RECORD CARRIERS; HANDLING RECORD CARRIERS
- G06K17/00—Methods or arrangements for effecting co-operative working between equipments covered by two or more of main groups G06K1/00 - G06K15/00, e.g. automatic card files incorporating conveying and reading operations
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a kind of dislocation books to search system and its lookup method, and it includes parameter setting module, bookshelf scanning device, image mosaic module, book label identification module, dislocation books searching module and data Reports module that the dislocation books, which search system,.The dislocation books lookup method specifically includes step: S1, shooting bookshelf full face are simultaneously spliced into a complete image P;S2, identification image P simultaneously extract the call number on each spine label;S3, books call number and books are successively stored in list A and list POS in the coordinate in image P;S4, scanning analysis list A data detect error bit call number;S5, the coordinate that dislocation call number is found in list POS, and with specific color mark in image P;S6, S1 to S5 is repeated, the dislocation books until completing all bookshelves are searched and label.The present invention has the advantages that method is simple, haves no need to change existing books equipment, is at low cost.
Description
Technical field
The invention belongs to library automation technical field, it is mainly concerned with a kind of dislocation books and searches system and its lookup
Method.
Background technique
The bibliography of modern society's library collection increases year by year, and book borrowing and reading activity is increasingly frequent.In reader's also book, by
Administrator scans books bar code with barcode scanning equipment, it is put in storage in book management system, then puts bookshelf specific bit again
It sets.However after reader voluntarily reads books in reading room, original position will not be often put back to, librarian needs periodically to arrange
Misplace books, but manual sorting mode inefficiency and easy mistakes and omissions, reader often occurs and looks into book management system
Asking certain books, obviously there are also collections to deposit this, but the case where can not find books on corresponding bookshelf.
In response to this, existing solution often carries out books positioning using RFID tag technology, but
It is that use cost and system integration aspect still have certain obstacle, it is difficult to extensive universal.In addition there are some other schemes
Also attempt to solve the problems, such as book bidding, such as: it is whole that Chinese invention patent file CN106504443A discloses a kind of automatic books
Device is managed, using walking vehicle body and arranges that elevating mechanism and movement mechanism on the car body carry out automatic arranging to bookshelf.China
Patent of invention file CN107862488A discloses a kind of crawler formula book bidding system, installs crawler scanning means on the shelf
The information of the books on bookshelf is scanned, then according to the information of books on bookshelf, figure is arranged by regulatory components, push and block component
Book.Chinese invention patent file CN107329469A discloses a kind of random frame book management system of autonomous robot, is bookshelf
Each layer encode and make two dimensional code, scanner module barcode scanning in robot obtains bookshelf information, scans the item on spine
Code obtains book information, and the books bar code on the entire bookshelf of Robot Scanning compares original position-information judges whether books are wrong
Frame realizes random frame taking care of books.These three above-mentioned existing schemes need to be transformed bookshelf or use automation equipment pair
Bookshelf is arranged, and if desired judges the whether wrong frame of books, it is also necessary to store the correct storage position of books in the database in advance
It sets, the higher cost of operation and maintenance, there are certain obstacles for popularization.
Summary of the invention
The main problem that books in libraries arranges is how to find dislocation books, once have found dislocation books, it is subsequent
Housekeeping by be accomplished manually it is not difficult;The dislocation books for still lacking a kind of relative low price currently on the market are searched
Method.The present invention proposes that a kind of dislocation books search system and its lookup method, is looked into using image recognition and data processing technique
Confuse position books.
In order to solve the above technical problems, present invention employs following technological means:
A kind of dislocation books search system, specifically include parameter setting module, bookshelf scanning device, image mosaic module,
Book label identification module, dislocation books searching module and data Reports module.
The parameter setting module is for being arranged indoor plane layout, bookshelf quantity, bookshelf position and shelf size;Institute
The picture of shooting for branch point shooting bookshelf picture and is uploaded to image mosaic module by the bookshelf scanning device stated;Described
Multiple bookshelf image mosaics that image mosaic module is used to that bookshelf scanning device to be shot and be uploaded are complete image;The book
Identification module spine label image for identification is marked, the call number of text formatting is converted into;The dislocation books are searched
Module finds out dislocation books therein for analyzing the call number data recognized;The data sheet module
The dislocation book information that storage dislocation books searching module is found, and dislocation book information is made into report and is looked into for staff
It sees.
Further, the bookshelf scanning device is using automation photographing device or artificial shooting.
Further, the image mosaic module carries out image using the Harris Corner Detection Algorithm based on standard deviation
Splicing.
A kind of dislocation books lookup method, specifically includes the following steps:
S1, multiple books images are shot in bookshelf front, picture is uploaded in system and is spliced into one in sequence
Complete image P;
S2, text identification is carried out to image P, extracts the call number in image P on each spine label;
S3, the coordinate of books call number and corresponding books label in image P is successively stored in list A and list POS
In;
S4, it is scanned analysis to the call number stored in list A, detects the call number wherein to misplace;
S5, the corresponding coordinate data of dislocation books call number detected in step S4 is found in list POS, according to seat
Mark data are made a mark in image P with specific color;
S6, step S1 to S5 is repeated, the dislocation books until completing all bookshelves are searched and label.
Further, the step S3 successively deposits the coordinate of books call number and corresponding books label in image P
It is put into list A and list POS, is realized especially by following steps:
One S31, construction list A construct a list POS, for storing for successively storing the call number of each books
Abscissa and ordinate of the books label in image P;
S32, it enables T for current book, is initially the first Book in the upper left corner image P;
S33, the call number of T is added in list A, the coordinate of T is added in list POS;
S34, T is enabled to be next book on the right side of current book, if right side is without books, enabling T is next layer in image P
Leftmost side books terminate process, otherwise go to step if current book has been last this book of the lower right corner in image P
S33。
Further, the S4 step is scanned analysis to the call number stored in list A, what detection wherein misplaced
Call number is realized especially by the following steps:
S41, creation and the identical array L of list A length, for each element in record list A oneself to be ending element
Longest be incremented by sub-sequence length, and enable L [1]=1;
S42, creation and the identical array P of list A length, for each element in record list A oneself to be ending member
The longest of element is incremented by the precursor element in subsequence, and enables P [1]=NULL;
S43, current lower label x is defined, enables x=2;
S44, the call number for detecting list A all elements before subscript x are found and are passed with the longest that A [x] is ending element
Increase subsequence and L [x] be calculated by following formula that is, to each i < x:
L [x]=max L (i)+1 | A [i] < A [x] ∧ i < x } (1)
And recording P [x] is the previous element subscript i that the longest obtained when calculating L [x] is incremented by subsequence;
S45, x=x+1 is enabled, if x≤| A |, S44 is gone to, S46 is otherwise gone to;
S46, traversal array L simultaneously find the subscript k at its maximum value, then start to recall forward out at P [k] with A [k]
Longest for the element that ends up is incremented by subsequence, is denoted as LIS;
S47, setting dislocation books list B are all books of the list A in addition to LIS.
The basic ideas of present invention dislocation books lookup method are: the spine portion of every book all posts mark in library
It signs, prints unique call number on label, books are always successively put by sequence according to call number in restocking.If on bookshelf
Books put sequence all it is correct, the call number of these books necessarily will form a monotone-increasing sequence;Conversely, if bookshelf
On there are dislocation books, the monotonic increase property of the sequence will be broken.Therefore, as long as can find out in call number sequence
Mistake number, the books of dislocation can be found.
Using following advantage can be obtained after the above technological means:
Originally a kind of dislocation books proposed by the present invention search system and its lookup method, pass through image recognition and data are analyzed
Technology realizes that dislocation books are searched, and does not need to carry out any transformation to the books or library equipment of script, cost is relatively low, no
By environmental restrictions, adaptive ability is stronger.The feature retrieval dislocation books being incremented by furthermore according to books call number, do not need rear
The correct storage position of books is stored in advance in platform database, greatly reduces Operating Complexity and operation difficulty, does not need
It is excessively safeguarded in subsequent use, use easy to spread.
Detailed description of the invention
Fig. 1 is that present invention dislocation books search system module schematic diagram.
Fig. 2 is the flow chart of present invention dislocation books lookup method.
Fig. 3 is present invention dislocation books lookup method books looked-up sequence schematic diagram.
Fig. 4 is the specific flow chart that the present invention carries out data analysis retrieval dislocation books.
Specific embodiment
Technical solution of the present invention is described further with reference to the accompanying drawing:
A kind of dislocation books lookup system, as shown in Figure 1, specifically including parameter setting module, bookshelf scanning device, image
Splicing module, book label identification module, dislocation books searching module and data Reports module.
Parameter setting module is for being arranged the parameters such as indoor plane layout, bookshelf quantity, bookshelf position and shelf size;Book
Frame scanning device uploads to image mosaic module, branch point shooting for branch point shooting bookshelf picture, and by the picture of shooting
It is for the clearer detailed information for recording each books, bookshelf scanning device can choose automatic camera equipment, can also be with
Shoot upload manually by staff;Image mosaic module is used for multiple bookshelf images that bookshelf scanning device is shot and uploaded
It is spliced into the image of complete display;Book label identification module spine label image for identification, is converted into the rope of text formatting
Book number;Dislocation books searching module is used to carry out finishing analysis to the call number data recognized, finds out dislocation book therein
Nationality;Data sheet module stores the dislocation book information found, and dislocation book information is made into report and is looked into for staff
It sees.
A kind of dislocation books lookup method, as shown in Fig. 2, specifically comprising the following steps:
S1, multiple books images are shot in bookshelf front, is then spliced into a complete image P.In this specific implementation
Feature detection is carried out using the Harris Corner Detection Algorithm based on standard deviation in example, and then carries out image mosaic.Believe in texture
Region abundant is ceased, Harris operator can extract a large amount of useful feature points, because its calculating process has pertained only to image
First derivative, so even if the extraction of angle steel joint is also more stable there are image rotation, grey scale change and viewpoint change.Figure
As 9 × 9 matrix expression of each angle point, the Euclidean distance of each corner description with corresponding corner description is calculated, to institute
There is angle point slightly to be matched, then reject pseudo- matching double points using the consistent RANSAC method of random sampling, calculate homography matrix,
It obtains accurately matching a little pair.The fusion of image uses smooth transition method, realizes the smooth transition between image, eliminates the trace of splicing
Mark.The complete image that splicing is completed finally is denoted as P.
S2, text identification is carried out to image P, extracts the call number on spine.Text on book labels is block letter text
Word is more standardized than hand-written script writing, and pixel distribution more regularly, is conducive to go to identify with the method for statistical nature.
Books tag recognition is carried out using the recognition methods based on statistical nature in this embodiment, using based on instruction
The statistical calculation for practicing sampled pixel obtains template, is matched with template and books label, obtains Text region as a result, specific step
Suddenly are as follows:
S21, image preprocessing.Image is pre-processed first, will be split after its gray processing;Then to segmentation after
Image be normalized, unified size is modified to each character using the method for linear interpolation;Next it is calculated using OTSU
Method calculates threshold value, by image binaryzation.
S22, image recognition.Template training is carried out by the distribution probability of statistics, letter image pixel, is then led to
The eigenmatrix for crossing calculating pictograph carries out Classification and Identification at a distance from each Character mother plate, and the image of each books label is known
Other result is stored in a character string.
S3, the coordinate of the call number of each books and the label in image P is divided according to the sequence of positions in bookshelf
It is not stored in list A and list POS, the specific steps are as follows:
One S31, construction list A, for storing the call number of each books;A list POS is constructed, for storing each mark
Sign abscissa and ordinate in image P;
S32, it enables T for current book, is initially the first Book in the upper left corner image P;
S33, the call number of T is added in list A, the coordinate of T is added in list POS;
S34, as shown in figure 3, enabling T is next book on the right side of current book, if it is image P that right side, which without books, enables T,
In next layer of leftmost side books;If current book has been last this book of the lower right corner in image P, terminate process, otherwise
Go to step S33.
S4, it is scanned analysis to the call number stored in list A, detects the call number wherein to misplace, as shown in figure 4,
Specific step is as follows:
S41, creation and the identical array L of list A length, for each element in record list A oneself to be ending element
Longest be incremented by sub-sequence length, and first element of L array is assigned a value of 1, i.e. L [1]=1;
S42, creation and the identical array P of list A length, for each element in record list A oneself to be ending member
The longest of element is incremented by the precursor element in subsequence, and enables P [1]=NULL;
S43, current lower label x is defined, enables x=2;
S44, the call number for detecting list A all elements before subscript x are found and are passed with the longest that A [x] is ending element
Increase subsequence and L [x] be calculated by following formula that is, to each i < x:
L [x]=max L (i)+1 | A [i] < A [x] ∧ i < x } (2)
And recording P [x] is the previous element subscript i that the longest obtained when calculating L [x] is incremented by subsequence;
S45, x=x+1 is enabled, if x≤| A |, S44 is gone to, S46 is otherwise gone to;
S46, traversal array L simultaneously find the subscript k at its maximum value, then start to recall forward out at P [k] with A [k]
Longest for the element that ends up is incremented by subsequence, is denoted as LIS;
S47, setting dislocation books list B are all books of the list A in addition to LIS.
S5, the dislocation books label that will be detected in S4 step in image P, the POS coordinate obtained according to S3 step are used
Specific color makes a mark.
S6, S1 to S5 step is repeated, until all bookshelves handle completion.
All dislocation books can be accurately found out by the above method, generate the image that dislocation books are marked
Group, this image group can be stored in a computer, can also print and check, the staff in library can be according to figure
As the dislocation books marked in group find corresponding bookshelf and books in library, the housekeeping of dislocation books is carried out.This hair
The dislocation books of bright proposition search system and lookup method, and not only cost is relatively low, but also not by library design limitation itself, only
It needs to analyze bookshelf image information the books that can find dislocation, early-stage preparations are simpler, subsequent also not need
Multidimensional shield, is suitable for each macro library.
Embodiments of the present invention are explained in detail above in conjunction with attached drawing, but the invention is not limited to above-mentioned
Embodiment within the knowledge of a person skilled in the art can also be before not departing from present inventive concept
It puts and makes a variety of changes.
Claims (6)
1. a kind of dislocation books search system, which is characterized in that including parameter setting module, bookshelf scanning device, image mosaic
Module, book label identification module, dislocation books searching module and data Reports module;Wherein,
The parameter setting module, for indoor plane layout, bookshelf quantity, bookshelf position and shelf size to be arranged;
The picture of shooting for branch point shooting bookshelf picture and is uploaded to image mosaic mould by the bookshelf scanning device
Block;
The image mosaic module, multiple bookshelf image mosaics for shooting and uploading bookshelf scanning device are complete graph
Picture;
The book label identification module, spine label image, is converted into the call number of text formatting for identification;
The dislocation books searching module finds out dislocation therein for analyzing the call number data recognized
Books;
The data sheet module, for storing the dislocation book information that finds of dislocation books searching module, and by the book that misplaces
Nationality information is made into report and checks for staff.
2. a kind of dislocation books according to claim 1 search system, which is characterized in that the bookshelf scanning device is adopted
With automation photographing device or artificial shooting.
3. a kind of dislocation books according to claim 1 search system, which is characterized in that the image mosaic module is adopted
Image mosaic is carried out with the Harris Corner Detection Algorithm based on standard deviation.
4. a kind of dislocation books lookup method, which comprises the following steps:
S1, multiple books images are shot in bookshelf front, picture is uploaded in system and be spliced into sequence one it is complete
Image P;
S2, text identification is carried out to image P, extracts the call number in image P on each spine label;
S3, books call number and corresponding books label are successively stored in list A and list POS in the coordinate in image P;
S4, it is scanned analysis to the call number stored in list A, detects the call number wherein to misplace;
S5, the corresponding coordinate data of dislocation books call number detected in step S4 is found in list POS, according to number of coordinates
It is made a mark according in image P with specific color;
S6, step S1 to S5 is repeated, the dislocation books until completing all bookshelves are searched and label.
5. a kind of dislocation books lookup method according to claim 4, which is characterized in that the step S3 is by books rope
Book number and corresponding books label are successively stored in list A and list POS in the coordinate in image P, especially by following steps
It realizes:
One S31, construction list A construct a list POS, for storing books for successively storing the call number of each books
Abscissa and ordinate of the label in image P;
S32, it enables T for current book, is initially the first Book in the upper left corner image P;
S33, the call number of T is added in list A, the coordinate of T is added in list POS;
S34, T is enabled to be next book on the right side of current book, if it is next layer most left in image P that right side, which without books, enables T,
Side books terminate process if current book has been last this book of the lower right corner in image P, otherwise go to step S33.
6. a kind of dislocation books lookup method according to claim 4, which is characterized in that the S4 step is to list A
The call number of middle storage is scanned analysis, detects the call number wherein to misplace, realizes especially by following steps:
S41, creation and the identical array L of list A length, for each element in record list A oneself to be to end up element most
Long incremental sub-sequence length, and enable L [1]=1;
S42, creation and the identical array P of list A length, for each element in record list A oneself to be ending element
Longest is incremented by the precursor element in subsequence, and enables P [1]=NULL;
S43, current lower label x is defined, enables x=2;
S44, the call number for detecting list A all elements before subscript x find and are incremented by son with the longest that A [x] is ending element
L [x] is calculated by following formula that is, to each i < x in sequence:
L [x]=max L (i)+1 | A [i] < A [x] ∧ i < x }
And recording P [x] is the previous element subscript i that the longest obtained when calculating L [x] is incremented by subsequence;
S45, x=x+1 is enabled, if x≤| A |, S44 is gone to, S46 is otherwise gone to;
S46, traversal array L simultaneously find the subscript k at its maximum value, then start to recall forward at P [k] out with A [k] for knot
The longest of tail element is incremented by subsequence, is denoted as LIS;
S47, setting dislocation books list B are all books of the list A in addition to LIS.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810961866.7A CN109359705B (en) | 2018-08-22 | 2018-08-22 | System and method for searching misplaced books |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810961866.7A CN109359705B (en) | 2018-08-22 | 2018-08-22 | System and method for searching misplaced books |
Publications (2)
Publication Number | Publication Date |
---|---|
CN109359705A true CN109359705A (en) | 2019-02-19 |
CN109359705B CN109359705B (en) | 2021-09-28 |
Family
ID=65349873
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810961866.7A Active CN109359705B (en) | 2018-08-22 | 2018-08-22 | System and method for searching misplaced books |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN109359705B (en) |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109949520A (en) * | 2019-03-01 | 2019-06-28 | 苏州洪昇新能源科技有限公司 | A kind of self-service book management system of enterprise-level |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101779864A (en) * | 2010-03-02 | 2010-07-21 | 山东大学 | Self-service error-detecting arrangement device for library |
KR101336124B1 (en) * | 2013-04-29 | 2013-12-03 | 주식회사 이도링크 | Library book position management system and library book position management method |
CN205210922U (en) * | 2015-12-16 | 2016-05-04 | 武汉科技大学 | Books position debugging system |
CN108000525A (en) * | 2017-10-23 | 2018-05-08 | 广东数相智能科技有限公司 | A kind of taking care of books robot and the misplaced recognition methods of books based on robot |
CN108304836A (en) * | 2018-02-01 | 2018-07-20 | 上海海事大学 | A kind of books Ordering Detection Algorithm based on image |
-
2018
- 2018-08-22 CN CN201810961866.7A patent/CN109359705B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN101779864A (en) * | 2010-03-02 | 2010-07-21 | 山东大学 | Self-service error-detecting arrangement device for library |
KR101336124B1 (en) * | 2013-04-29 | 2013-12-03 | 주식회사 이도링크 | Library book position management system and library book position management method |
CN205210922U (en) * | 2015-12-16 | 2016-05-04 | 武汉科技大学 | Books position debugging system |
CN108000525A (en) * | 2017-10-23 | 2018-05-08 | 广东数相智能科技有限公司 | A kind of taking care of books robot and the misplaced recognition methods of books based on robot |
CN108304836A (en) * | 2018-02-01 | 2018-07-20 | 上海海事大学 | A kind of books Ordering Detection Algorithm based on image |
Cited By (1)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109949520A (en) * | 2019-03-01 | 2019-06-28 | 苏州洪昇新能源科技有限公司 | A kind of self-service book management system of enterprise-level |
Also Published As
Publication number | Publication date |
---|---|
CN109359705B (en) | 2021-09-28 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
Luo et al. | Moran: A multi-object rectified attention network for scene text recognition | |
Shi et al. | Robust scene text recognition with automatic rectification | |
CN107067044B (en) | Financial reimbursement complete ticket intelligent auditing system | |
Lin et al. | STAN: A sequential transformation attention-based network for scene text recognition | |
Xu et al. | Automatic extraction of 1D barcodes from video scans for drone-assisted inventory management in warehousing applications | |
CN109829467A (en) | Image labeling method, electronic device and non-transient computer-readable storage medium | |
CN100511219C (en) | Electronic filing system searchable by a handwritten search query | |
EP1854051B1 (en) | Intelligent importation of information from foreign application user interface using artificial intelligence | |
CN110298291A (en) | Ox face and ox face critical point detection method based on Mask-RCNN | |
CN112101357A (en) | RPA robot intelligent element positioning and picking method and system | |
CN112651289B (en) | Value-added tax common invoice intelligent recognition and verification system and method thereof | |
Liu et al. | Data extraction from charts via single deep neural network | |
CN111476210B (en) | Image-based text recognition method, system, device and storage medium | |
CN105678322A (en) | Sample labeling method and apparatus | |
Schroth et al. | Exploiting text-related features for content-based image retrieval | |
US20210304422A1 (en) | Generation of non-occluded image based on fusion of multiple occulded images | |
CN111539417B (en) | Text recognition training optimization method based on deep neural network | |
Colter et al. | Tablext: A combined neural network and heuristic based table extractor | |
Xiao et al. | An extended attention mechanism for scene text recognition | |
Xiong et al. | Text detection in stores using a repetition prior | |
CN109359705A (en) | A kind of dislocation books search system and its lookup method | |
Ahmed et al. | A generic method for automatic ground truth generation of camera-captured documents | |
Tiwari et al. | Live attendance system via face recognition | |
Kataria et al. | CNN-bidirectional LSTM based optical character recognition of Sanskrit manuscripts: A comprehensive systematic literature review | |
Rahman et al. | UTRNet: High-Resolution Urdu Text Recognition in Printed Documents |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
CB02 | Change of applicant information | ||
CB02 | Change of applicant information |
Address after: 210032 No. 219 Ning six road, Jiangbei new district, Nanjing, Jiangsu Applicant after: Nanjing University of Information Science and Technology Address before: 211500 Yuting Square, 59 Wangqiao Road, Liuhe District, Nanjing City, Jiangsu Province Applicant before: Nanjing University of Information Science and Technology |
|
GR01 | Patent grant | ||
GR01 | Patent grant |