CN108876711A - A kind of sketch generation method, server and system based on image characteristic point - Google Patents
A kind of sketch generation method, server and system based on image characteristic point Download PDFInfo
- Publication number
- CN108876711A CN108876711A CN201810635109.0A CN201810635109A CN108876711A CN 108876711 A CN108876711 A CN 108876711A CN 201810635109 A CN201810635109 A CN 201810635109A CN 108876711 A CN108876711 A CN 108876711A
- Authority
- CN
- China
- Prior art keywords
- characteristic point
- image
- solid images
- sketch
- point
- 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
- 238000000034 method Methods 0.000 title claims abstract description 76
- 239000007787 solid Substances 0.000 claims abstract description 52
- 238000000605 extraction Methods 0.000 claims abstract description 16
- 230000003044 adaptive effect Effects 0.000 claims abstract description 10
- 239000011159 matrix material Substances 0.000 claims description 16
- 230000010339 dilation Effects 0.000 claims description 14
- 238000012545 processing Methods 0.000 claims description 13
- 230000004044 response Effects 0.000 claims description 12
- 230000008569 process Effects 0.000 claims description 10
- 238000007781 pre-processing Methods 0.000 claims description 7
- 230000009467 reduction Effects 0.000 claims description 7
- 238000011946 reduction process Methods 0.000 claims description 7
- 238000010276 construction Methods 0.000 claims description 6
- 238000001914 filtration Methods 0.000 claims description 6
- 238000013461 design Methods 0.000 abstract description 11
- 238000004422 calculation algorithm Methods 0.000 description 14
- 238000001514 detection method Methods 0.000 description 14
- 230000006870 function Effects 0.000 description 14
- 238000010586 diagram Methods 0.000 description 10
- 241000282693 Cercopithecidae Species 0.000 description 9
- 238000004590 computer program Methods 0.000 description 8
- 238000003860 storage Methods 0.000 description 4
- 230000000694 effects Effects 0.000 description 3
- 239000000284 extract Substances 0.000 description 3
- 230000004048 modification Effects 0.000 description 3
- 238000012986 modification Methods 0.000 description 3
- 238000012360 testing method Methods 0.000 description 3
- 244000025254 Cannabis sativa Species 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 2
- 230000005540 biological transmission Effects 0.000 description 2
- 238000012938 design process Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 101100136092 Drosophila melanogaster peng gene Proteins 0.000 description 1
- 241000196324 Embryophyta Species 0.000 description 1
- 238000012550 audit Methods 0.000 description 1
- 238000005452 bending Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 235000013399 edible fruits Nutrition 0.000 description 1
- 230000008030 elimination Effects 0.000 description 1
- 238000003379 elimination reaction Methods 0.000 description 1
- 238000004519 manufacturing process Methods 0.000 description 1
- 230000007246 mechanism Effects 0.000 description 1
- 239000000203 mixture Substances 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
Classifications
-
- G06T3/04—
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T11/00—2D [Two Dimensional] image generation
- G06T11/20—Drawing from basic elements, e.g. lines or circles
- G06T11/203—Drawing of straight lines or curves
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
Abstract
The invention discloses a kind of sketch generation method, server and system based on image characteristic point.Wherein, the sketch generation method based on image characteristic point includes:Pretreatment operation is carried out to the solid images of acquisition;The characteristic point at pretreated solid images edge is extracted based on small echo adaptive H arris detective operators;The fitting for being carried out profile to the characteristic point at the solid images edge of extraction using Piecewise Spline Interpolation Method method, is obtained smooth closed curved profile, generates corresponding sketch.This method can be improved design efficiency, meet the use of conventional design personnel.
Description
Technical field
The invention belongs to threedimensional model shape-designing field more particularly to a kind of sketch generation sides based on image characteristic point
Method, server and system.
Background technique
With being constantly progressive for scientific and technological level, people step up the interest-degree of threedimensional model, and three-dimensional modeling has become
For a hot topic in computer research field.Whether it in 3D game design, machine-building, smart home or is curing
Different degrees of threedimensional model can be all related to during treatment etc. is multi-field.However, as customer demand is continuously improved, the design of model
Efficiency is also constantly accelerated, and the drafting of sketch is a wherein important link in threedimensional model design process, with life step
The quickening cut down, it is desirable that model also become relative complex, thus need the sketch drawn more efficient.
Research for this respect is existing very much, such as:Peng etc. passes through the Analysis of Entropy and constructs the descriptor of sketch,
Cartographical sketching model is analyzed, Liu Kai, Sun Zhengxing et al. propose a kind of exchange method for generating 3 D complex curved surface, support
User constructs the shape of threedimensional model by modification cartographical sketching and related personnel passes through based on Hough transform method and random circle
The tied mechanism of the profile geometrical characteristic recognition methods of detection method (RCD) completes the drafting etc. of sketch.
But still have following problem:
1) by the Analysis of Entropy skeletonizing descriptor, structure is more single, during layout design, lines and its
Profile drawing modification larger workload, in key point selection, the bending degree of lines, the processing of smoothness and complex figure
Aspect is drawn, the efficiency that grass draws mode is lower;
Although 2) can be identified to geometrical characteristic profile by Hough transform method and random loop truss method, identification
Be all straight line and circular arc essential characteristic.And it is directed to for curved wire-frame model and does not have applicability.Such as:To having
It is bent for the figure and rough figure of inflection point, Hough transform method and random loop truss method are only able to detect in figure
Straight line and random lines, testing result is not highly desirable.And the present invention is directly handled on the image, with based on small echo from
It adapts to Harris detective operators directly further to extract inflection point, characteristic point in image zooming-out image border point, detection knot
Fruit is more clear.
3) tool that the sketch drafting software of mainstream is related to is more, and ordinary user is difficult to carry out correlation model in a short time
Drafting work.
In conclusion, as the demand of client steps up, how to accelerate the drafting effect of sketch in actual design process
Rate, the design work for improving model is that designer needs the Important Problems that solve.
Summary of the invention
In order to solve the deficiencies in the prior art, there is provided a kind of grass based on image characteristic point for the first object of the present invention
Drawing generating method, which reduce time complexities, improve drafting efficiency.
A kind of sketch generation method based on image characteristic point of the invention, including:
Pretreatment operation is carried out to the solid images of acquisition;
The characteristic point at pretreated solid images edge is extracted based on small echo adaptive H arris detective operators;
The fitting for being carried out profile to the characteristic point at the solid images edge of extraction using Piecewise Spline Interpolation Method method, is obtained smooth
Closed curved profile generates corresponding sketch.
Further, the boundary that pretreatment operation includes noise reduction process and solid images is carried out to the solid images of acquisition
Tracking.
Due to image in transmission process it is possible that the phenomenon that noise pollution makes image quality decrease, primarily
Work be to image carry out noise reduction process.
Further, successively use median filtering denoising method and dilation operation method to the solid images noise reduction of acquisition.
It can guarantee the clarity of image while removing the noise of image in this way.
Further, the spy at pretreated solid images edge is extracted based on small echo adaptive H arris detective operators
Levying the detailed process put includes:
Bianry image is converted by pretreated solid images;
Calculate both horizontally and vertically component of the bianry image on default scale, the modulus value and width of wavelet structure coefficient
Angle;
The modulus value and argument of wavelet coefficient according to construction calculate bianry image in both horizontally and vertically gradient;
According to bianry image in both horizontally and vertically gradient, characteristic point Gauss is weighted using Gaussian function, is added simultaneously
Enter Euclidean distance, calculate difference between points, generates a matrix element value sequence;
In the matrix element value sequence, further determined that by calculating the Harris response of each image pixel
Angle point, by the Harris detective operators based on small wave self-adaption to the characteristic point for extracting image border.
The second object of the present invention is to provide a kind of sketch generation server based on image characteristic point, and which reduce the times
Complexity improves drafting efficiency.
A kind of sketch based on image characteristic point of the invention generates server, including:
Preprocessing module is configured as:Pretreatment operation is carried out to the solid images of acquisition;
Edge Feature Points extraction module, is configured as:Pre- place is extracted based on small echo adaptive H arris detective operators
The characteristic point at the solid images edge after reason;
Edge Feature Points fitting module, is configured as:Using Piecewise Spline Interpolation Method method to the solid images edge of extraction
Characteristic point carry out profile fitting, obtain smooth closed curved profile, generate corresponding sketch.
Further, in the preprocessing module, carrying out pretreatment operation to the solid images of acquisition includes at noise reduction
The frontier tracing of reason and solid images.
Further, in the preprocessing module, median filtering denoising method and dilation operation method pair are successively used
The solid images noise reduction of acquisition.
Further, the Edge Feature Points extraction module, including:
Bianry image converts submodule, is configured as:Bianry image is converted by pretreated solid images;
The modulus value and argument of wavelet coefficient construct submodule, are configured as:Bianry image is calculated on default scale
Both horizontally and vertically component, the modulus value and argument of wavelet structure coefficient;
Both horizontally and vertically gradient computational submodule is configured as:The modulus value and width of wavelet coefficient according to construction
Angle calculates bianry image in both horizontally and vertically gradient;
Matrix element value sequence generates submodule, is configured as:According to bianry image in both horizontally and vertically gradient,
Characteristic point Gauss is weighted using Gaussian function, while Euclidean distance is added, difference between points is calculated, generates a square
Array element element value sequence;
Feature point extraction submodule, is configured as:In the matrix element value sequence, by calculating each image slices
The Harris response of element further determines that angle point, by Harris detective operators based on small wave self-adaption to extracting image
The characteristic point at edge.
The third object of the present invention is to provide a kind of sketch generation system based on image characteristic point, multiple which reduce the time
Miscellaneous degree improves drafting efficiency.
A kind of sketch based on image characteristic point of the invention generates system, including described above based on image characteristic point
Sketch generate server.
Compared with prior art, the beneficial effects of the invention are as follows:
(1) present invention extracts characteristic point according to solid images, controls Harris Corner Detection by using Euclidean distance difference
Operator sketches the contours of contour edge in conjunction with Piecewise Spline Interpolation Method algorithm, and then completes the design of sketch, reduces time complexity,
Improve drafting efficiency.
(2) present invention is segmented according to Corner Detection Algorithm to image key points detection processing from the angle of image
Spline interpolation fitting completes sketch and generates work, has good sketch drafting effect, can be improved design efficiency, meets common
The use of designer.
Detailed description of the invention
The accompanying drawings constituting a part of this application is used to provide further understanding of the present application, and the application's shows
Meaning property embodiment and its explanation are not constituted an undue limitation on the present application for explaining the application.
Fig. 1 is a kind of sketch generation method flow chart based on image characteristic point of the invention.
Fig. 2 (a) is structural element A.
Fig. 2 (b) is processed object X.
Fig. 2 (c) is the result after processed object X dilation operation.
Fig. 3 (a) is dilation operation original graph.
Fig. 3 (b) is 1 dilation operation result figure.
Fig. 3 (c) is 2 dilation operation result figures.
Fig. 3 (d) is 3 dilation operation result figures.
Fig. 4 (a) is colored initial profile.
Fig. 4 (b) is the initial profile of monkey.
Fig. 5 (a) is colored initial profile characteristic point.
Fig. 5 (b) is the initial profile characteristic point of monkey.
Fig. 6 (a) is B-spline fitting schematic diagram.
Fig. 6 (b) is least square method fitting schematic diagram.
Fig. 6 (c) is that Piecewise Spline Interpolation Method is fitted schematic diagram three times.
Fig. 7 (a) is that Piecewise Spline Interpolation Method is fitted to obtain colored sketch three times.
Fig. 7 (b) is that Piecewise Spline Interpolation Method is fitted to obtain the sketch of monkey three times.
Fig. 8 is that a kind of sketch based on image characteristic point of the invention generates server architecture schematic diagram.
Specific embodiment
It is noted that following detailed description is all illustrative, it is intended to provide further instruction to the application.Unless another
It indicates, all technical and scientific terms used herein has usual with the application person of an ordinary skill in the technical field
The identical meanings of understanding.
It should be noted that term used herein above is merely to describe specific embodiment, and be not intended to restricted root
According to the illustrative embodiments of the application.As used herein, unless the context clearly indicates otherwise, otherwise singular
Also it is intended to include plural form, additionally, it should be understood that, when in the present specification using term "comprising" and/or " packet
Include " when, indicate existing characteristics, step, operation, device, component and/or their combination.
Fig. 1 is a kind of sketch generation method flow chart based on image characteristic point of the invention.
As shown in Figure 1, a kind of sketch generation method based on image characteristic point of the invention, including:
S1:Pretreatment operation is carried out to the solid images of acquisition.
The frontier tracing that pretreatment operation includes noise reduction process and solid images is carried out to the solid images of acquisition.
Due to image in transmission process it is possible that the phenomenon that noise pollution makes image quality decrease, primarily
Work be to image carry out noise reduction process.
Successively use median filtering denoising method and dilation operation method to the solid images noise reduction of acquisition.
It can guarantee the clarity of image while removing the noise of image in this way.
As shown in Fig. 2 (a)-Fig. 2 (c), dilation operation method includes:
Aa is obtained after the structural element A as shown in Fig. 2 (a) is translated a, if Aa hits X;
Such as Fig. 2 (b), point a is recorded, point a is the point of dash area in figure.
The collection for meeting the point a composition of above-mentioned condition is collectively referred to as the result that X is expanded by A.It is formulated as:D (X)=a | Aa ↑
X }=X A, as shown in Fig. 2 (c), wherein X is processed object, and A is structural element, for any one in dash area
Point a, Aa hits X, and dash area is the result that X is expanded by A.
By the comparative result figure of dilation operation, as shown in Fig. 3 (a)-Fig. 3 (d), the above-mentioned boundary point traced into is carried out
Median filter process carries out boundary expansion by dilation operation, and elimination includes the subtle noise in image target area.
S2:The characteristic point at pretreated solid images edge is extracted based on small echo adaptive H arris detective operators.
The tool of the characteristic point at pretreated solid images edge is extracted based on small echo adaptive H arris detective operators
Body process includes:
Step S21:Need wavelet structure function.If Φ (x, y) is a two-dimentional smooth function, then along the direction x and y
First derivative be defined as two wavelets
Step S22:Using the convolution of wavelet and image in both direction as small echo both horizontally and vertically on
Component, then there are two wavelet transform components, the component along x horizontal direction is random two-dimensional image function f (x, y):
It is along y vertically-oriented component:
Wherein wt1And wt2It is image along the gradient value on the direction x and y, b is constant here.
Step S23:Image modulus value (formula 1) and argument (formula 2) are calculated according to small echo direction scale j, along gradient-norm
The local maximum direction of amount carries out Corner Detection.
Wherein,WithRefer respectively to the modulus value in different directions gradient.
Specifically, the feature at pretreated solid images edge is extracted based on small echo adaptive H arris detective operators
Point detailed process include:
Bianry image is converted by pretreated solid images;
Calculate both horizontally and vertically component of the bianry image in scale i, the modulus value and argument of wavelet structure coefficient;
The modulus value and argument of wavelet coefficient according to construction calculate bianry image in both horizontally and vertically gradient;
Wherein, in formula (3), V is definedm,nIt is coefficient of the Gauss window at (m, n), A and B are respectively along m and n
The horizontal vertical component in direction;The horizontal and vertical component coordinates coefficient of (m, n) expression Gauss window.
According to horizontal vertical component is determined, characteristic point Gauss is weighted using Gaussian function, while Euclidean distance is added, into
Row between points difference (indicated with d, whereinGenerator matrix X element value sequence;
In matrix X, angle point is further determined that by calculating the Harris response K of each image pixel, passes through base
In small wave self-adaption Harris detective operators to image detection.
K (m, n)=det [X (m, n)]-Ytr2[X(m,n)] (4)
In formula (4), what det [X (m, n)] was indicated is the value of the determinant of X matrix, and tr2[X (m, n)] is expressed as square
The mark of battle array X;For the characteristic value of matrix X;Y takes recommendation 0.05;
Wherein, picture point is divided by the size of X characteristic value, whenWithAll than it is more prominent when, i.e., value compares
It is determined as angle point when big;And work asWithWhen value is less than preset threshold, it is not considered as angle point;When whereinEitherWhen, it is determined as edge.
S3:The fitting for being carried out profile to the characteristic point at the solid images edge of extraction using Piecewise Spline Interpolation Method method, is obtained
Smooth closed curved profile, generates corresponding sketch.
Such as:It is handled using cubic spline interpolation:
Step S31:Through it is above-mentioned detect the contour feature point of solid images after, it is special to pass through Cubic Spline Functions Fitting profile
Sign point.
Step S32:The above-mentioned data point for being detected and being extracted is imported into segment processing in origin, according to data
The coordinate position of point is ranked up x, its result is put into array c, successively iterative processing, until all key points are orderly.
Step S33:According to the length n of array c, the matrix of a n*1 is generated.
According to cubic spline interpolation formula,Interpolation fitting will make h0=0
Then j=1;Therefore, y and x are changed into function interpolation point related with h;Wherein, hjIndicate the characteristic point in spline interpolation;hj-1
Indicate hjPreceding interpolation point;xj-1,yj-1It is illustrated respectively in hj-1The transverse and longitudinal coordinate of interpolation point;xj,yjRespectively indicate hjInterpolation point
Transverse and longitudinal coordinate.
Step S34:According to key point and smoothness, the data point in array c is successively called, is inserted by being segmented batten three times
Value method is fitted operation, fits smooth spline curve, finally obtains complete sketch outline, meets at designer's audit
Reason.
A practical application example is given below:
Utilize the process and superiority of the invention for illustrating sketch drafting for colored and monkey photo.
Step 1:It is handled on two images that client gives, image is denoised and enhanced for convenience
The processing of function converts gray level image for two width instance graphs respectively.Using edge following algorithm, found in gray level image with
The starting point of track, and frontier tracing is carried out in eight neighborhood by the point, after expanded operation and coordinatograph, record its position
Coordinate.
Step 2:To be compared and analyze, carry out equidistant dot interlace sampling pretreatment, and record position coordinates a little with
The number of sampled point, as shown in Fig. 4 (a), wherein colored initial profile point number is 3804, response time 2.632207s;
And the initial profile point number of the monkey as shown in Fig. 4 (b) is 2295, response time 1.229416s.
Step 3:By in traditional Harris algorithm be added wavelet function and Euclidean distance it is poor, the present invention can guarantee
In the complete situation of image outline local feature, obtained shown in image characteristic point such as Fig. 5 (a)-Fig. 5 (b) in the short time.
Step 4:Piecewise Spline Interpolation Method is handled three times.
Be utilized respectively B-spline fitting, least square method fitting and three times Piecewise Spline Interpolation Method fitting as a result, such as Fig. 6
(a) shown in-Fig. 6 (c).
With the fitting of Piecewise Spline Interpolation Method three times, ultimately constructed flowers and plants figure is shown in Fig. 7 (a), the sketch of monkey such as Fig. 7 (b) institute
Show.
In this example, by taking colored and monkey picture as an example, several Corner Detection Algorithms are had chosen, and compared it
Compared with detailed comparison result is as shown in Tables 1 and 2, the experimental results showed that the present invention has certain superiority.
Table 1 spends testing result
Original image | Moravec algorithm | Fast algorithm | This paper algorithm | |
Initial samples point | 3804 | 17 | 524 | 270 |
Response time | 2.632207s | 7.430317s | 2.915905s | 1.028065s |
2 monkey testing result of table
Original image | Moravec algorithm | Fast algorithm | This paper algorithm | |
Initial samples point | 2295 | 40 | 762 | 351 |
Response time | 1.229416s | 2.225066s | 1.291240s | 1.404667s |
As can be seen from the table, the instance graph key point of the Harris Corner Detection flower proposed by the present invention based on small echo
Number is 270, response time 1.028065s;Harris Corner Detection example monkey graph key point number based on small echo is
351, response time 1.404667s.On the used time, accurate characteristic point can be obtained within the less time, for setting
For meter personnel's manual mode skeletonizing, the time is saved, efficiency is improved.
The sketch that the present invention also provides a kind of based on image characteristic point generates server, which reduce time complexity,
Improve drafting efficiency.
As shown in figure 8, a kind of sketch based on image characteristic point of the invention generates server, including:
(1) preprocessing module is configured as:Pretreatment operation is carried out to the solid images of acquisition.
Specifically, in the preprocessing module, carrying out pretreatment operation to the solid images of acquisition includes noise reduction process
And the frontier tracing of solid images.
Specifically, in the preprocessing module, successively using median filtering denoising method and dilation operation method to obtaining
The solid images noise reduction taken.
(2) Edge Feature Points extraction module is configured as:It is pre- to extract based on small echo adaptive H arris detective operators
The characteristic point at treated solid images edge.
Specifically, the Edge Feature Points extraction module, including:
Bianry image converts submodule, is configured as:Bianry image is converted by pretreated solid images;
The modulus value and argument of wavelet coefficient construct submodule, are configured as:Bianry image is calculated on default scale
Both horizontally and vertically component, the modulus value and argument of wavelet structure coefficient;
Both horizontally and vertically gradient computational submodule is configured as:The modulus value and width of wavelet coefficient according to construction
Angle calculates bianry image in both horizontally and vertically gradient;
Matrix element value sequence generates submodule, is configured as:According to bianry image in both horizontally and vertically gradient,
Characteristic point Gauss is weighted using Gaussian function, while Euclidean distance is added, difference between points is calculated, generates a square
Array element element value sequence;
Feature point extraction submodule, is configured as:In the matrix element value sequence, by calculating each image slices
The Harris response of element further determines that angle point, by Harris detective operators based on small wave self-adaption to extracting image
The characteristic point at edge.
(3) Edge Feature Points fitting module is configured as:Using Piecewise Spline Interpolation Method method to the solid images side of extraction
The characteristic point of edge carries out the fitting of profile, obtains smooth closed curved profile, generates corresponding sketch.
The present invention extracts characteristic point according to solid images, calculates by using Euclidean distance difference control Harris Corner Detection
Son sketches the contours of contour edge in conjunction with Piecewise Spline Interpolation Method algorithm, and then completes the design of sketch, reduces time complexity, mentions
High drafting efficiency.
The present invention is from the angle of image, according to Corner Detection Algorithm to image key points detection processing, segmented sample
Interpolation fitting completes sketch and generates work, has good sketch drafting effect, can be improved design efficiency, and satisfaction is commonly set
The use of meter personnel.
The present invention also provides a kind of, and the sketch based on image characteristic point generates system, and which reduce time complexities, mention
High drafting efficiency.
A kind of sketch based on image characteristic point of the invention generates system, including as shown in Figure 8 based on characteristics of image
The sketch of point generates server.
The present invention extracts characteristic point according to solid images, calculates by using Euclidean distance difference control Harris Corner Detection
Son sketches the contours of contour edge in conjunction with Piecewise Spline Interpolation Method algorithm, and then completes the design of sketch, reduces time complexity, mentions
High drafting efficiency.
It should be understood by those skilled in the art that, the embodiment of the present invention can provide as method, system or computer program
Product.Therefore, the shape of hardware embodiment, software implementation or embodiment combining software and hardware aspects can be used in the present invention
Formula.Moreover, the present invention, which can be used, can use storage in the computer that one or more wherein includes computer usable program code
The form for the computer program product implemented on medium (including but not limited to magnetic disk storage and optical memory etc.).
The present invention be referring to according to the method for the embodiment of the present invention, the process of equipment (system) and computer program product
Figure and/or block diagram describe.It should be understood that every one stream in flowchart and/or the block diagram can be realized by computer program instructions
The combination of process and/or box in journey and/or box and flowchart and/or the block diagram.It can provide these computer programs
Instruct the processor of general purpose computer, special purpose computer, Embedded Processor or other programmable data processing devices to produce
A raw machine, so that being generated by the instruction that computer or the processor of other programmable data processing devices execute for real
The device for the function of being specified in present one or more flows of the flowchart and/or one or more blocks of the block diagram.
These computer program instructions, which may also be stored in, is able to guide computer or other programmable data processing devices with spy
Determine in the computer-readable memory that mode works, so that it includes referring to that instruction stored in the computer readable memory, which generates,
Enable the manufacture of device, the command device realize in one box of one or more flows of the flowchart and/or block diagram or
The function of being specified in multiple boxes.
These computer program instructions also can be loaded onto a computer or other programmable data processing device, so that counting
Series of operation steps are executed on calculation machine or other programmable devices to generate computer implemented processing, thus in computer or
The instruction executed on other programmable devices is provided for realizing in one or more flows of the flowchart and/or block diagram one
The step of function of being specified in a box or multiple boxes.
Those of ordinary skill in the art will appreciate that realizing all or part of the process in above-described embodiment method, being can be with
Relevant hardware is instructed to complete by computer program, the program can be stored in a computer-readable storage medium
In, the program is when being executed, it may include such as the process of the embodiment of above-mentioned each method.Wherein, the storage medium can be magnetic
Dish, CD, read-only memory (Read-Only Memory, ROM) or random access memory (Random
AccessMemory, RAM) etc..
Above-mentioned, although the foregoing specific embodiments of the present invention is described with reference to the accompanying drawings, not protects model to the present invention
The limitation enclosed, those skilled in the art should understand that, based on the technical solutions of the present invention, those skilled in the art are not
Need to make the creative labor the various modifications or changes that can be made still within protection scope of the present invention.
Claims (9)
1. a kind of sketch generation method based on image characteristic point, which is characterized in that including:
Pretreatment operation is carried out to the solid images of acquisition;
The characteristic point at pretreated solid images edge is extracted based on small echo adaptive H arris detective operators;
The fitting for carrying out profile to the characteristic point at the solid images edge of extraction using Piecewise Spline Interpolation Method method, obtains smooth closing
Curved profile, generate corresponding sketch.
2. a kind of sketch generation method based on image characteristic point as described in claim 1, which is characterized in that the reality of acquisition
Body image carries out the frontier tracing that pretreatment operation includes noise reduction process and solid images.
3. a kind of sketch generation method based on image characteristic point as claimed in claim 2, which is characterized in that in successively using
The solid images noise reduction of value filtering denoising method and dilation operation method to acquisition.
4. a kind of sketch generation method based on image characteristic point as described in claim 1, which is characterized in that certainly based on small echo
Harris detective operators, which are adapted to, come the detailed process for extracting the characteristic point at pretreated solid images edge includes:
Bianry image is converted by pretreated solid images;
Calculate both horizontally and vertically component of the bianry image on default scale, the modulus value and argument of wavelet structure coefficient;
The modulus value and argument of wavelet coefficient according to construction calculate bianry image in both horizontally and vertically gradient;
According to bianry image in both horizontally and vertically gradient, characteristic point Gauss is weighted using Gaussian function, while Europe is added
Family name's distance calculates difference between points, generates a matrix element value sequence;
In the matrix element value sequence, angle is further determined that by calculating the Harris response of each image pixel
Point, by the Harris detective operators based on small wave self-adaption to the characteristic point for extracting image border.
5. a kind of sketch based on image characteristic point generates server, which is characterized in that including:
Preprocessing module is configured as:Pretreatment operation is carried out to the solid images of acquisition;
Edge Feature Points extraction module, is configured as:After extracting pretreatment based on small echo adaptive H arris detective operators
Solid images edge characteristic point;
Edge Feature Points fitting module, is configured as:Using Piecewise Spline Interpolation Method method to the spy at the solid images edge of extraction
Sign point carries out the fitting of profile, obtains smooth closed curved profile, generates corresponding sketch.
6. a kind of sketch based on image characteristic point as claimed in claim 5 generates server, which is characterized in that described pre-
In processing module, the frontier tracing that pretreatment operation includes noise reduction process and solid images is carried out to the solid images of acquisition.
7. a kind of sketch based on image characteristic point as claimed in claim 6 generates server, which is characterized in that described pre-
In processing module, successively use median filtering denoising method and dilation operation method to the solid images noise reduction of acquisition.
8. a kind of sketch based on image characteristic point as claimed in claim 5 generates server, which is characterized in that the edge
Feature point extraction module, including:
Bianry image converts submodule, is configured as:Bianry image is converted by pretreated solid images;
The modulus value and argument of wavelet coefficient construct submodule, are configured as:Calculate level of the bianry image on default scale
With vertical direction component, the modulus value and argument of wavelet structure coefficient;
Both horizontally and vertically gradient computational submodule is configured as:The modulus value and argument of wavelet coefficient according to construction, meter
Bianry image is calculated in both horizontally and vertically gradient;
Matrix element value sequence generates submodule, is configured as:According to bianry image in both horizontally and vertically gradient, utilize
Gaussian function weights characteristic point Gauss, while Euclidean distance is added, and calculates difference between points, generates a matrix element
Plain value sequence;
Feature point extraction submodule, is configured as:In the matrix element value sequence, by calculating each image pixel
Harris response further determines that angle point, by Harris detective operators based on small wave self-adaption to extracting image border
Characteristic point.
9. a kind of sketch based on image characteristic point generates system, which is characterized in that including any one of such as claim 5-8 institute
The sketch based on image characteristic point stated generates server.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810635109.0A CN108876711B (en) | 2018-06-20 | 2018-06-20 | Sketch generation method, server and system based on image feature points |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810635109.0A CN108876711B (en) | 2018-06-20 | 2018-06-20 | Sketch generation method, server and system based on image feature points |
Publications (2)
Publication Number | Publication Date |
---|---|
CN108876711A true CN108876711A (en) | 2018-11-23 |
CN108876711B CN108876711B (en) | 2023-01-31 |
Family
ID=64339627
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810635109.0A Active CN108876711B (en) | 2018-06-20 | 2018-06-20 | Sketch generation method, server and system based on image feature points |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108876711B (en) |
Cited By (2)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112529016A (en) * | 2020-12-21 | 2021-03-19 | 浙江欣奕华智能科技有限公司 | Method and device for extracting feature points in image |
CN112862851A (en) * | 2021-01-18 | 2021-05-28 | 网娱互动科技(北京)股份有限公司 | Automatic image matting method and system based on image recognition technology |
Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140169683A1 (en) * | 2012-12-18 | 2014-06-19 | Samsung Electronics Co., Ltd. | Image retrieval method, real-time drawing prompting method, and devices thereof |
CN105718552A (en) * | 2016-01-19 | 2016-06-29 | 北京服装学院 | Clothing freehand sketch based clothing image retrieval method |
CN106846340A (en) * | 2017-02-17 | 2017-06-13 | 大连理工大学 | A kind of striation boundary extraction method based on on-fixed characteristic point |
CN107368792A (en) * | 2017-06-29 | 2017-11-21 | 广州智慧城市发展研究院 | A kind of finger vein identification method and system based on wave filter and Bone Edge |
CN108009550A (en) * | 2017-11-09 | 2018-05-08 | 深圳大学 | High spectrum image feature detection method and device based on curve of spectrum fitting |
-
2018
- 2018-06-20 CN CN201810635109.0A patent/CN108876711B/en active Active
Patent Citations (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20140169683A1 (en) * | 2012-12-18 | 2014-06-19 | Samsung Electronics Co., Ltd. | Image retrieval method, real-time drawing prompting method, and devices thereof |
CN105718552A (en) * | 2016-01-19 | 2016-06-29 | 北京服装学院 | Clothing freehand sketch based clothing image retrieval method |
CN106846340A (en) * | 2017-02-17 | 2017-06-13 | 大连理工大学 | A kind of striation boundary extraction method based on on-fixed characteristic point |
CN107368792A (en) * | 2017-06-29 | 2017-11-21 | 广州智慧城市发展研究院 | A kind of finger vein identification method and system based on wave filter and Bone Edge |
CN108009550A (en) * | 2017-11-09 | 2018-05-08 | 深圳大学 | High spectrum image feature detection method and device based on curve of spectrum fitting |
Non-Patent Citations (1)
Title |
---|
徐真: "基于图像的3D造型交互设计方法研究", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑 2019 年 第01期》 * |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN112529016A (en) * | 2020-12-21 | 2021-03-19 | 浙江欣奕华智能科技有限公司 | Method and device for extracting feature points in image |
CN112862851A (en) * | 2021-01-18 | 2021-05-28 | 网娱互动科技(北京)股份有限公司 | Automatic image matting method and system based on image recognition technology |
CN112862851B (en) * | 2021-01-18 | 2021-10-15 | 网娱互动科技(北京)股份有限公司 | Automatic image matting method and system based on image recognition technology |
Also Published As
Publication number | Publication date |
---|---|
CN108876711B (en) | 2023-01-31 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN102592136B (en) | Three-dimensional human face recognition method based on intermediate frequency information in geometry image | |
CN103870845B (en) | Novel K value optimization method in point cloud clustering denoising process | |
CN111462175A (en) | Space-time convolution twin matching network target tracking method, device, medium and equipment | |
CN110688947B (en) | Method for synchronously realizing human face three-dimensional point cloud feature point positioning and human face segmentation | |
CN104217406B (en) | SAR image noise reduction method based on shear wave coefficient processing | |
CN109766866B (en) | Face characteristic point real-time detection method and detection system based on three-dimensional reconstruction | |
CN107341844A (en) | A kind of real-time three-dimensional people's object plotting method based on more Kinect | |
CN109034131A (en) | A kind of semi-automatic face key point mask method and storage medium | |
CN108765317A (en) | A kind of combined optimization method that space-time consistency is stablized with eigencenter EMD adaptive videos | |
CN104992403A (en) | Hybrid operator image redirection method based on visual similarity measurement | |
Mebatsion et al. | A novel profile based model for virtual representation of quasi-symmetric plant organs | |
CN110060260A (en) | A kind of image processing method and system | |
CN111858997B (en) | Cross-domain matching-based clothing template generation method | |
CN108876711A (en) | A kind of sketch generation method, server and system based on image characteristic point | |
CN105069766B (en) | A kind of an inscription on a tablet restorative procedure based on the description of Chinese character image contour feature | |
CN108447038A (en) | A kind of mesh denoising method based on non local full variation operator | |
CN101430789A (en) | Image edge detection method based on Fast Slant Stack transformation | |
CN113628170A (en) | Laser line extraction method and system based on deep learning | |
Barreiro et al. | Automatic Reconstruction of Semantic 3D Models from 2D Floor Plans | |
CN109242854A (en) | A kind of image significance detection method based on FLIC super-pixel segmentation | |
CN115601301B (en) | Fish phenotype characteristic measurement method, system, electronic equipment and storage medium | |
Luo et al. | Improved orb-slam2 algorithm based on information entropy and image sharpening adjustment | |
CN110136086A (en) | Interval threshold image de-noising method based on BEMD | |
Lyu et al. | Laplacian-based 3D mesh simplification with feature preservation | |
Chen et al. | DBSC‐based animation enhanced with feature and motion |
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 | ||
GR01 | Patent grant | ||
GR01 | Patent grant |