CN105574077A - Battle plotting matching method and system - Google Patents

Battle plotting matching method and system Download PDF

Info

Publication number
CN105574077A
CN105574077A CN201510875672.1A CN201510875672A CN105574077A CN 105574077 A CN105574077 A CN 105574077A CN 201510875672 A CN201510875672 A CN 201510875672A CN 105574077 A CN105574077 A CN 105574077A
Authority
CN
China
Prior art keywords
image
eigenvector
database
matching
feature extraction
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.)
Pending
Application number
CN201510875672.1A
Other languages
Chinese (zh)
Inventor
程柏林
潘谊春
韩俊
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Individual
Original Assignee
Individual
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Individual filed Critical Individual
Priority to CN201510875672.1A priority Critical patent/CN105574077A/en
Publication of CN105574077A publication Critical patent/CN105574077A/en
Pending legal-status Critical Current

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/50—Information retrieval; Database structures therefor; File system structures therefor of still image data
    • G06F16/58—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually
    • G06F16/583—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content
    • G06F16/5838—Retrieval characterised by using metadata, e.g. metadata not derived from the content or metadata generated manually using metadata automatically derived from the content using colour

Abstract

The invention discloses a battle plotting matching method and system. The method comprises following steps: S1, obtaining basic image information; S2, taking a feature extraction algorithm as a basis, carrying out feature extraction treatment to every image in a database, meanwhile, arranging the obtained feature vectors into corresponding relation tables; S3, converting an image retrieval problem into matching treatment to the indexes of the feature vectors according to the fields of the feature vectors; S4, judging whether there are images in the database satisfying a threshold range, if so, outputting an image matching result to a user. According to the method, the retrieval speed can be improved and the user waiting time can be shortened; in adoption of a battle plotting comparison method based on a graph matching technique, relatively satisfied matching result can be provided; and military standard training intelligence and standardization can be realized.

Description

The matching process that a kind of operation is marked on a map and system
Technical field
The invention belongs to image technique field, more specifically, belong to matching process and system that a kind of operation marks on a map.
Background technology
Fighting marks on a map is an important content of military training, and the plotting situation that trainer marks on a map to fighting is the important evidence of reflection military training benefit.Although there is the software of marking on a map that some are special at present, but cannot assess automatically mark on a map the accuracy of result of trainer, it is one of difficult point faced at present that the army's mark automatically marked and drawed participant carries out correctness comparison.
Summary of the invention
The technical matters that the present invention is directed to solution is: provide a kind of matching process that can effectively mark on a map to the operation of measurement of army's mark training effect and system.
Its concrete technical scheme is as follows:
The matching process that operation is marked on a map, comprises the following steps:
S1, acquisition image essential information;
S2, based on feature extraction algorithm, each image in database is done feature extraction process, the eigenvector obtained is placed in mapping table simultaneously;
S3, foundation eigenvector field, transfer the process of mating the index of eigenvector to by image retrieval problem;
S4, meet the image of threshold range as having in database, to the result of user's output image coupling.
Compared with prior art, the present invention compared with prior art has the following advantages and beneficial effect:
The method effectively can improve retrieval rate, shorten period of reservation of number, and the operation of graphic based matching technique is marked on a map the intellectuality of the matching result that Comparison Method can provide comparatively satisfied, the army's of achieving mark training, standardization.
Accompanying drawing explanation
Fig. 1 is dividing method process flow diagram of the present invention.
Fig. 2 is anode segmentation effect figure under the present invention's 400 times of optical microscopes.
Embodiment
In order to make object of the present invention, technical scheme and advantage clearly understand, below in conjunction with drawings and Examples, the present invention is further elaborated.Should be appreciated that specific embodiment described herein only in order to explain the present invention, be not intended to limit the present invention.In addition, if below in described each embodiment of the present invention involved technical characteristic do not form conflict each other and just can mutually combine.
Schematic illustration:
Information base unit in routine data storehouse all has and defines accurately, therefore its semantic and meaning be all clear definitely; But the information base unit in image data base is formed by the gray-scale value set of each pixel, even if the numerical value of so each pixel is fixing, be invalid with the object that this numerical value is corresponding, with regard to level, can not show a candle to semanteme and meaning that image comprises.Therefore, following two kinds of search methods can be taked to adapt to this species diversity:
(1) artificial semantic identification process is done to each image, after the semanteme obtaining each image and meaning, utilize traditional database index method to carry out matching treatment.But the cost of this method is very large, and for the system that view data is large, its feasibility is very poor.
(2) employing take content as the coupling of sample, relates to the important technology such as image procossing and pattern-recognition, utilizes the content of image to carry out token image, carrys out retrieving images according to content simultaneously.The method has advantage flexible and efficient, with low cost, and its application prospect is large.
Content-based image matching method, need given the mated image of user again according to image realize coupling, focus on the feature excavating image, image pixel matrix is replaced with eigenvector, realize the search problem of image to be reduced to retrieve same Given Graph as immediate characteristics of image vector in space of feature vectors, complete the function of images match.
Specific embodiment
Operation is marked on a map in comparison, have employed content-based image matching method, and the army's of realization mark Graphic Pattern Matching need set up two different tables, comprising: image data table, deposits image related data; Characteristic Vectors scale, deposits the characteristics of image vector of extraction.Concrete function is realized by view data initialization module, characteristics of image vector extraction module, content indexing matching module, result for retrieval output module, and its operation logic as shown in Figure 1.
As shown in Figure 2, provide the matching process that a kind of operation is marked on a map, comprise the following steps:
Step S1, acquisition image essential information;
Concrete, this step comprises in relation table corresponding for the eigenvector of all images (as color, shape and texture etc.) input database, and these eigenvectors are done clustering processing, find the center of its cluster successively, again cluster centre is treated as key images, finally in mapping table, label is stamped to these images.
Step S2, based on feature extraction algorithm, each image in database is done feature extraction process, the eigenvector obtained is placed in mapping table simultaneously;
Concrete, image I in each storehouse of this step precomputation i, i=1,2,3 ... with each key images K s, s=1,2 ... the distance d between the color of m, shape, textural characteristics vector c(I i, K s), d s(I i, K s), d t(I i, K s).
Then, read the inquiry example images Q that user provides, extract its color, shape, textural characteristics vector, and calculate Q and each key images K s, s=1,2 ... the distance d between each eigenvector between m c(I i, K s), d s(I i, K s), d t(I i, K s).
Calculate image I in each storehouse i, i=1,2,3 ... and the color of inquiry example images Q, shape, distance d between textural characteristics vector c(I i, K s), d s(I i, K s), d t(I i, K s) lower bound
l c ( I i , Q ) = m a x l ≤ s ≤ m | d c ( I i , K s ) - d c ( Q , K s ) | ,
l s ( I i , Q ) = m a x l ≤ s ≤ m | d s ( I i , K s ) - d s ( Q , K s ) | ,
l t ( I i , Q ) = m a x l ≤ s ≤ m | d t ( I i , K s ) - d t ( Q , K s ) | .
Calculate image I in each storehouse i, i=1,2,3 ... and the color of inquiry example images Q, shape, Gaussian normalization distance d' between textural characteristics vector c(I i, K s), d ' s(I i, K s), d ' t(I i, K s), determine its lower bound l' c(I i, Q), l ' s(I i, Q), l ' t(I i, K s), concrete grammar is as follows: calculate inquiry example images Q and each key images K respectively at pretreatment stage s, s=1,2 ... the distance d between each eigenvector of m c(I i, K s), d s(I i, K s), d t(I i, K s) average m c, m s, m tand variances sigma c, σ s, σ t; Then linear transformation the lower bound of Gaussian normalization distance respectively.
Step S3, foundation eigenvector field, transfer the process of mating the index of eigenvector to by image retrieval problem;
Concrete, this step calculates image I in each storehouse i, i=1,2,3 ... and the total eigenvector distance d'(I between the vector of inquiry example images Q i, Q) and=W cd' c(I i, Q) and+W sd ' s(I i, Q) and+W td ' t(I i, Q) lower bound l'(I i, Q) and=W cl' c(I i, Q) and+W sl ' s(I i, Q) and+W tl ' t(I i, Q), W c, W s, W tfor the weights preset, for l'(I i, Q) and the image of > T (T is given threshold value), can determine that it is not similar to retrieval example images, and in corresponding image table, do selection markers screened.
S4, meet the image of threshold range as having in database, to the result of user's output image coupling;
Concrete, this step directly calculates Q and each storehouse image I do not screened j, j=1,2,3 ..., total eigenvector distance D of u c(Q ,i j), meet D'(Q, I j) < T 0(T 0for given threshold value) storehouse image be that input operation is marked on a map the similar army mark of image.The minimum image of threshold value finally exports to user as matching result.。
Based on said method, provide the system realizing the method, this system comprises as lower module simultaneously:
(1) view data initialization module
This module mainly completes the input of database images, the function such as foundation of relation table in the acquisition of image essential information and database.
(2) characteristics of image vector extraction module
Based on feature extraction algorithm, each image in database is done feature extraction process, the eigenvector obtained is placed in mapping table simultaneously.
(3) content indexing matching module
According to eigenvector field, image retrieval problem is transferred to the process that the index of eigenvector is mated.
(4) result for retrieval output module
The image of threshold range is met, to the result of user's output image coupling as having in database.
An operation is marked on a map, is generally made up of multiple fundamental figure (straight line, rectangle, curve etc.), and the base attribute such as size, color of each fundamental figure has requirement, and mutual alignment between them and angular relationship have regulation.So the comparison of marking on a map of fighting, not only will have the comparison of the single fundamental figure in local, also will have the Inspection of mutual relationship between each overall figure.Military symbols element is various, needs elements of comparison a lot, and this difficult point place just.
Realizing in army's symbol comparison process, because each variant between each label, so can only, for specific army mark label, provide special comparison function, and the comparison principle that each operation is marked on a map be substantially identical.One is analyze the key element composition of fighting and marking on a map, as total how many key element composition, respectively by several straight lines, and several curve compositions; Two is that element is entirely true, analyzes the characteristic of single key element successively, as length, width, color, line weight, with horizontal direction angle, and the relative distance etc. of range image center.
Mark on a map in comparison in operation, color characteristic adopts color histogram eigenvector, and shape facility adopts edge orientation histogram eigenvector, and textural characteristics adopts co-occurrence matrix textural characteristics vector.After feature extraction, carry out characteristic distance calculating according to corresponding criterion, the more similar then characteristic distance of two width images is less.Can set a threshold value, be less than army's logo image of threshold value if exist, can think the matching symbols of image of marking on a map containing this operation, symbol figure minimum for threshold value can be exported as matching result, its Graphic Pattern Matching process flow diagram as shown in Figure 2.
Last it is noted that above embodiment is only in order to illustrate technical scheme of the present invention, be not intended to limit; Although with reference to previous embodiment to invention has been detailed description, those of ordinary skill in the art is to be understood that: it still can be modified to the technical scheme described in foregoing embodiments, or carries out equivalent replacement to wherein portion of techniques feature; And these amendments or replacement, do not make the essence of appropriate technical solution depart from the scope of various embodiments of the present invention technical scheme.

Claims (2)

1. an operation matching process of marking on a map, is characterized in that, comprise the following steps:
S1, acquisition image essential information;
S2, based on feature extraction algorithm, each image in database is done feature extraction process, the eigenvector obtained is placed in mapping table simultaneously;
S3, foundation eigenvector field, transfer the process of mating the index of eigenvector to by image retrieval problem;
S4, judge in database, whether there is the image meeting threshold range, if any, to the result of user's output image coupling.
2. an operation matching system of marking on a map, is characterized in that, comprise the following steps:
View data initialization module, this module is for obtaining image essential information;
Characteristics of image vector extraction module, for based on feature extraction algorithm, does feature extraction process by each image in database, the eigenvector obtained is placed in mapping table simultaneously;
Content indexing matching module, for according to eigenvector field, transfers the process of mating the index of eigenvector to by image retrieval problem;
Whether result for retrieval output module, have for judging the image meeting threshold range in database, if any, to the result of user's output image coupling.
CN201510875672.1A 2015-12-03 2015-12-03 Battle plotting matching method and system Pending CN105574077A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510875672.1A CN105574077A (en) 2015-12-03 2015-12-03 Battle plotting matching method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510875672.1A CN105574077A (en) 2015-12-03 2015-12-03 Battle plotting matching method and system

Publications (1)

Publication Number Publication Date
CN105574077A true CN105574077A (en) 2016-05-11

Family

ID=55884208

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510875672.1A Pending CN105574077A (en) 2015-12-03 2015-12-03 Battle plotting matching method and system

Country Status (1)

Country Link
CN (1) CN105574077A (en)

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109241226A (en) * 2018-08-29 2019-01-18 武汉市星盟科技有限公司 A kind of synthesis optimal in structure Map Analysis System
CN113052945A (en) * 2021-03-09 2021-06-29 中国人民解放军陆军防化学院 Computer automatic judging method for key map

Cited By (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109241226A (en) * 2018-08-29 2019-01-18 武汉市星盟科技有限公司 A kind of synthesis optimal in structure Map Analysis System
CN113052945A (en) * 2021-03-09 2021-06-29 中国人民解放军陆军防化学院 Computer automatic judging method for key map

Similar Documents

Publication Publication Date Title
CN104599275B (en) The RGB-D scene understanding methods of imparametrization based on probability graph model
CN102609681B (en) Face recognition method based on dictionary learning models
CN105069481B (en) Natural scene multiple labeling sorting technique based on spatial pyramid sparse coding
CN111079847B (en) Remote sensing image automatic labeling method based on deep learning
CN105931253A (en) Image segmentation method combined with semi-supervised learning
CN105138998A (en) Method and system for re-identifying pedestrian based on view angle self-adaptive subspace learning algorithm
CN105930873B (en) A kind of walking across mode matching method certainly based on subspace
CN103345760B (en) A kind of automatic generation method of medical image object shapes template mark point
CN105005760A (en) Pedestrian re-identification method based on finite mixture model
Zhu et al. Text recognition in images based on transformer with hierarchical attention
CN102385592A (en) Image concept detection method and device
CN104268552A (en) Fine category classification method based on component polygons
CN110458132A (en) One kind is based on random length text recognition method end to end
CN105183857A (en) Automatic picture training sample extracting method and system
CN107657276B (en) Weak supervision semantic segmentation method based on searching semantic class clusters
CN109753962A (en) Text filed processing method in natural scene image based on hybrid network
Wang et al. Fast vanishing point detection method based on road border region estimation
CN103295026B (en) Based on the image classification method of space partial polymerization description vectors
CN102722717B (en) Cell division identification method
CN105574077A (en) Battle plotting matching method and system
CN106250818B (en) A kind of total order keeps the face age estimation method of projection
CN109685103A (en) A kind of text Multi-label learning method based on broad sense K mean algorithm
CN111144469B (en) End-to-end multi-sequence text recognition method based on multi-dimensional associated time sequence classification neural network
CN105844299B (en) A kind of image classification method based on bag of words
Hu et al. Building extraction using mask scoring R-Cnn network

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
WD01 Invention patent application deemed withdrawn after publication

Application publication date: 20160511

WD01 Invention patent application deemed withdrawn after publication