A kind of Educational toy external member and its recognition methodss of component and electric wire
Technical field
The present invention relates to Computer Vision Detection processing technology field, particularly to a kind of Educational toy external member and its circuit
Element and the recognition methodss of electric wire.
Background technology
Now many interesting children education game application or children's game are had on panel computer, but often simply allow little
Friend points at picture, interactive shortcoming on screen, look at screen for a long time and easily eyes is damaged, and lacks physics
The education of knowledge, thinking of logic culture shortcoming;And when some interactive strong traditional game toys have had been detached from instantly
The development in generation, the demand in form cannot meet child's study, playing, also it is not easy to the interactive communication of child and the head of a family.
In order to solve the above problems, computer vision and technical field of image processing successfully develop a kind of Educational toy
External member, comprising: support, helmet detector and base plate, installs games, by the photographic head of panel computer in panel computer
Collection is positioned over the image of the base plate in plane.Although above-mentioned Educational toy external member solves the interaction of game in panel computer
Property shortcoming problem, but form is still single, simply simple interaction, physical knowledge embedded be not it is impossible to child
From little culture electricity knowledge, also cannot accomplish from little culture child to the understanding of electricity and interest.
Therefore, Computer Vision Detection processing technology field is badly in need of a kind of Educational toy external member and its component and electric wire
Recognition methodss, be by circuit components and electric wire be positioned over game base plate on, in panel computer install games, pass through
The photographic head collection of panel computer is positioned over circuit components and electric wire image on base plate, then is based on predefined color, wheel
Wide information, color-coded information, identification circuit element and electric wire, circuit components are connected by child with electric wire, judge to connect
Whether good circuit is correct, strengthens the imaginative power of child, and increase game is interesting, makes child may learn tandem circuit and knows
Know, the interest of culture child.
Content of the invention
The present invention is in order to solve the above problems, there is provided the identification of a kind of Educational toy external member and its component and electric wire
Method, technical scheme is as follows:
A kind of Educational toy external member, comprising: base plate, circuit components and electric wire base plate are positioned in plane, circuit elements device
Part and electric wire are positioned on base plate.
Preferably, in a kind of above-mentioned Educational toy external member, base plate is the rectangle with fillet, at 4 angles of rectangle
On be provided with collimation angle.
Preferably, in a kind of above-mentioned Educational toy external member, collimation angle is red circular arc line.
In a kind of Educational toy external member, the recognition methodss of component and electric wire, comprise the steps:
Step one, installs games in panel computer, then base plate is positioned in plane it is ensured that the one side of collimation angle
Upward
Step 2, completes the connection of circuit components and electric wire on base plate, real by the post-positioned pick-up head of panel computer
When gather coloured image, mobile panel computer is it is ensured that at least contain 3 collimation angle in the coloured image of post-positioned pick-up head collection;
Step 3, extracts effective identification region from the coloured image of step 2;
Step 4, the circuit components that detection is located in the effective identification region of coloured image;
Step 5, the electric wire that detection is located in the effective identification region of coloured image;
Step 6, judges whether circuit components are accurate with the connection of electric wire.
Preferably, in the recognition methodss of component and electric wire in a kind of above-mentioned Educational toy external member, in step 2
The coloured image of post-positioned pick-up head collection is ixy, ixy=f (x, y)=(rxy, gxy, bxy), wherein, (x, y) represents coloured image
The position coordinateses of pixel, f (x, y) represents pixel value at pixel coordinate position for the image, rxyRepresent that image slices vegetarian refreshments exists
The color-values of red channel, gxyRepresent the color-values in green channel for the image slices vegetarian refreshments, bxyRepresent image slices vegetarian refreshments blue logical
The color-values in road.
Preferably, in the recognition methodss of component and electric wire in a kind of above-mentioned Educational toy external member, in step 3
Concretely comprising the following steps of effective identification region is extracted from coloured image:
A) according to priori, the coloured image of step 2 is partitioned into 4 pieces of calibration angular zones, according in hsv space
Priori threshold value, 4 pieces of collimation angle area images are carried out binary conversion treatment, obtain 4 pieces of collimation angle binary map;
B) the 4 pieces of collimation angle binary map obtaining in scanning step a), obtain corresponding edge contour figure, further according to edge wheel
Wide eccentricity and the priori of size, filter out irrational profile;
C) the remaining edge contour being obtained according to step b), calculates the boundary rectangle of 4 collimation angle, in identification process
In, when having qualified collimation angle at least three footmarks, effective cog region that its boundary rectangle as calculates
Domain.
Preferably, in the recognition methodss of component and electric wire in a kind of above-mentioned Educational toy external member, in step 4
The tool step of the circuit components that detection is located in color image recognition region is:
1, because the color of each circuit components shell is different, therefore passes through color distortion, be partitioned into each circuit elements
Device, and extract the Internal periphery of each circuit components shell;
2, according to the Internal periphery of the circuit components shell extracting in step 1, calculate the position of each circuit components
Put and deflection angle;
3, the deflection angle rotation circuit components and parts calculating according to step 2, then partitioning circuitry components and parts, compiled by color
Code identifies the classification of circuit components.
Preferably, in the recognition methodss of component and electric wire in a kind of above-mentioned Educational toy external member, carry in step 1
Take the concretely comprising the following steps of Internal periphery of each circuit components shell:
A) because the color of circuit components is unfavorable for separated in rgb color space, illumination variation is also compared
Sensitivity, so, by the region of interest extracting area image by rgb color space conversion to the hsv face laying particular emphasis on color representation
The colour space, concrete conversion formula is:
V=max { c (r '), c (g '), c (b ') };
Wherein, h represents tone value, and s represents intensity value, and v represents brightness value, and max { c (r '), c (g '), c (b ') } represents
In original image, pixel is in the pixel maximum of three passages of red, green, blue, min { c (r '), c (g '), c (b ') } table
Show in original image the pixel minimum in three passages of red, green, blue for the pixel, and the span of h is located at 0-
Between 360;
B) in hsv color space, color according to involved by the circuit components priori threshold value in hsv space,
Coloured image is carried out binary conversion treatment, concrete formula is as follows:
In binary picture, (x, y) &b_s (during x, y) &b_v (x, y), as generates binary system figure to b (x, y)=b_h
Picture;
Wherein, b (x, y) represents the binary pixel values of image slices vegetarian refreshments (x, y), and h (x, y), s (x, y), v (x, y) are respectively
Represent tone value in hsv color space for image slices vegetarian refreshments (x, y), intensity value, brightness value;B_h (x, y), b_s (x, y),
B_v (x, y) represents image slices vegetarian refreshments (x, y) respectively whether respectively in specified h, s, v region, if it is, value is 1,
Otherwise, value is 0;hmin、hmaxRepresent that the priori of color tone in hsv color space of certain components and parts shell is minimum respectively
And maximum;smin、smaxRepresent respectively color saturation in hsv color space of certain components and parts shell priori minimum and
Maximum;vmin、vmaxRepresent the priori minimum and maximum of color brightness in hsv color space of certain components and parts shell respectively
Value.
C) scan binary image, find out all edge contours;
Binary image can be regarded as the gray level image of only two values, and the edge of image refers to gray scale in gray level image
Change more violent part, the intensity of variation of gray value using neighbor between graded carry out quantificational expression, gradient is
The two dimensional equivalent formula of single order two dimension derivative, concrete calculating process is:
First, calculate the difference of neighbor, concrete formula is:
gx=f [i, j+1]-f [i, j]
gy=f [i, j]-f [i+1, j]
Wherein, gxRepresent neighbor difference in the x direction, gyRepresent neighbor difference in y-direction, f [i, j
+ 1] represent the pixel value that image arranges in the i-th row jth+1, f [i, j] represents the pixel value that image arranges in the i-th row jth;F [i+1, j]
Represent the pixel value that image arranges in i+1 row jth;
Further, calculate the gradient between neighbor, concrete formula is:
Wherein, g (x, y) represents the Grad on (x, y) point of image,Represent pixel value derivation in the x direction,Represent pixel value derivation in y-direction;
Further, calculate the gradient magnitude of marginal point, the gradient magnitude set of all marginal points is the edge extracting
Profile;
Further, calculating nonstatic electric fan and the method for remaining circuit components and parts ear is according to circuit components ear
The priori of piece contour shape, size and eccentricity, filters irrational profile, thus obtaining 2 ears of circuit components
Piece position;
Because the ear of static electric fan is possible to be covered some or all of by flabellum it is therefore desirable to carry out face to step b)
Color is split, and is partitioned into blue binary map and green binary map;
Further, blue binary map is scanned, is known by the priori of the contour shape of flabellum, eccentricity, size
Knowing, determining whether there is the presence of flabellum, if it has, then judging this circuit components as fan;Otherwise, leap to step 2;
Further, after judging this circuit components as fan, the computational methods for the ear of static fan are as follows: right
Step 1 Green binary map carries out clustering processing, green point close together is polymerized to a class, by of a sort point set number
Amount, the size of point set minimum outsourcing rectangle, eccentricity, the priori of point set position, filter to the point set being polymerized to a class,
If being only left two class point sets after filtering then it is assumed that the position of current flabellum has covered the ear of electric components with having greater probability
Piece, then the central point position of circuit components ear is calculated by the center position of two class point sets and the profile information of components and parts
Put.
Preferably, in the recognition methodss of component and electric wire in a kind of above-mentioned Educational toy external member, step 2 calculates
The concretely comprising the following steps of the position of each circuit components and deflection angle:
According to ear location, the center position of the circuit components calculating, existed with the ear ensureing circuit components
Horizontal direction is standard, calculates the anglec of rotation of circuit components.
Preferably, in the recognition methodss of component and electric wire in a kind of above-mentioned Educational toy external member, know in step 3
Do not go out concretely comprising the following steps of circuit components classification:
Firstly, it is necessary to be that all circuit components arrange a kind of coding rule in advance, each circuit components is made to have only
One coding;Because circuit components quantity to be identified is limited, therefore chooses red, yellow, blue, the green four kinds colors easily distinguished and make
For coding characteristic color;During the printing without permission of circuit components is concentrated mainly on, in, lower in three places, when these three places
A certain colored pixels exceed colored pixels threshold value then it is assumed that this color is the color in this region;
According to circuit components color of shell, upper in, neutralization lower in print color without permission, red be designated as 1, Huang is designated as 2, Lan Ji
For 3, green be designated as 4, be no designated as 0, ignore current color and be also denoted as 0,11 circuit components can be encoded, according to circuit
Components and parts encode, you can unique type determining current detection circuit components;
Filter out the circuit components not meeting coding, by remaining circuit components type, circuit components central point,
The anglec of rotation passes to upper layer software (applications) together;
Preferably, in the recognition methodss of component and electric wire in a kind of above-mentioned Educational toy external member, colored pixels
Threshold value is 30% of each palace lattice area in nine grids.
Preferably, in the recognition methodss of component and electric wire in a kind of above-mentioned Educational toy external member, in step 5
Detection is located at the concretely comprising the following steps of electric wire in the effective identification region of coloured image:
Step 1), it is partitioned into electric wire, extracts skeleton by color distortion;
Step 2), by the skeleton extracting, calculate end points and the bifurcation point of electric wire.
Preferably, in the recognition methodss of component and electric wire in a kind of above-mentioned Educational toy external member, step 1) in
It is partitioned into electric wire, extracts concretely comprising the following steps of skeleton by color distortion:
First, in effective identification region hsv space of step 3, color according to the electric wire priori in hsv space
Threshold value, image binaryzation is processed, and obtains electric wire binary image;
Further, electric wire binary image is scanned, by the priori filter of the shape of electric wire profile, size
Except the profile not meeting electric wire feature, remaining profile is filled with, obtains the binary picture of a new live wire;
Further, skeleton refinement is carried out to the binary picture of above-mentioned live wire, obtain the skeleton of electric wire.
Preferably, in the recognition methodss of component and electric wire in a kind of above-mentioned Educational toy external member, step 2) in
By the skeleton extracting, calculate the end points of electric wire and concretely comprising the following steps of bifurcation point: by step 1) in the electric wire bone that obtains
Bone, and the end points of electric wire, the priori of bifurcation point feature, find the end points in electric wire skeleton and bifurcation point, if there are point
Trouble point, by electric wire skeleton from bifurcation point separately, is divided into multiple line segments, without bifurcation point, origin wire skeleton is exactly a line
Section, all of line segment and line segment end points type in skeleton is supplied to upper layer software (applications).
Preferably, in the recognition methodss of component and electric wire in a kind of above-mentioned Educational toy external member, in step 6
Judge whether circuit components and the connection of electric wire accurately concretely comprise the following steps: the circuit that step 4 is identified by upper layer software (applications)
The electric wire that components and parts and step 5 identify links together, and the circuit diagram then realizing storage with upper layer software (applications) is compared,
If consistent with circuit diagram then it is assumed that circuit connects accurately.
Beneficial effects of the present invention:
1st, the present invention is cleverly by Appliance computer vision pattern recognition technique and hsv color space, binary conversion treatment, figure
As cutting technique combines use, can interpolate that out type and the electric wire of circuit components, and can interpolate that out circuit even
Whether accurately to connect, there is fast operation, accurate positioning, hardware is united well with software engineering, game interaction sets
Meter is ingenious;Attractive in appearance simple, judge more quick, strengthen the imaginative power of child, increase game is interesting, so that child is may learn
Tandem circuit knowledge, the interest of culture child.
2nd, detection algorithm of the present invention more science, maturation, by the color conversion of image, image convolution, image cutting, skeleton
Refinement scheduling algorithm combines use, can quickly judge type and the electric wire of circuit components.
3rd, calculating speed of the present invention is fast;Every time detection and localization takes in 200ms, provides smooth use body for player
Test.
4th, stable performance of the present invention, is installed on the situation in Educational toy external member in different illumination, to different panel computers
Under, carry out collecting test for 3,000 width pictures, false recognition rate and loss are below 0.2%.
Brief description
To describe the present invention with reference to the accompanying drawings and detailed description in detail:
Fig. 1 is a kind of structural representation of present invention Educational toy external member.
Fig. 2 is the flow chart of the recognition methodss of component and electric wire in a kind of present invention Educational toy external member.
Fig. 3 is the form of the physical circuit components and parts coding of step 3 in step 4 in a kind of present invention Educational toy external member
Figure.
Wherein, the corresponding relation between the reference in Fig. 1-3 and component names is:
Base plate 1, circuit components 2, electric wire 3.
Specific embodiment
In order that measure, creation characteristic, reached purpose and effect that the technology of the present invention is realized are easy to understand, below will
In conjunction with the accompanying drawing in the embodiment of the present invention, the technical scheme in the embodiment of the present invention is clearly and completely described it is clear that
Described embodiment is only a part of embodiment of the present invention, rather than whole embodiments.Based on the enforcement in the present invention
Example, the every other embodiment that those of ordinary skill in the art are obtained under the premise of not making creative work, broadly fall into
The scope of protection of the invention.
Fig. 1 is a kind of structural representation of present invention Educational toy external member.
As shown in figure 1, the recognition methodss of a kind of toy external member and its component and electric wire, base plate 1, circuit components 2
With electric wire 3, base plate 1 is positioned in plane, and circuit components 2 and electric wire 3 are positioned on base plate 3.
In the present embodiment, base plate is the rectangle with fillet, is provided with collimation angle on 4 angles of rectangle;Preferably, school
Quasi- angle is red circular arc line.
Fig. 2 is the flow chart of the recognition methodss of component and electric wire in a kind of present invention Educational toy external member.
As shown in Fig. 2 in a kind of Educational toy external member component and electric wire recognition methodss, comprise the steps:
Step one, installs games in panel computer, then base plate is positioned in plane it is ensured that the one side of collimation angle
Upward;
Step 2, completes the connection of circuit components and electric wire on base plate, real by the post-positioned pick-up head of panel computer
When gather coloured image, mobile panel computer it is ensured that at least containing 3 collimation angle in the coloured image of post-positioned pick-up head collection,
Concretely comprise the following steps:
The coloured image of post-positioned pick-up head collection is ixy, ixy=f (x, y)=(rxy,gxy,bxy), wherein, (x, y) represents
The position coordinateses of color image pixel point, f (x, y) represents pixel value at pixel coordinate position for the image, rxyRepresent image
Pixel is in the color-values of red channel, gxyRepresent the color-values in green channel for the image slices vegetarian refreshments, bxyRepresent image slices vegetarian refreshments
Color-values in blue channel;
Step 3, extracts effective identification region from the coloured image of step 2, concretely comprises the following steps:
A) according to priori, the coloured image of step 2 is partitioned into 4 pieces of calibration angular zones, according in hsv space
Priori threshold value, 4 pieces of collimation angle area images are carried out binary conversion treatment, obtain 4 pieces of collimation angle binary map;
B) the 4 pieces of collimation angle binary map obtaining in scanning step a), obtain corresponding edge contour figure, further according to edge wheel
Wide eccentricity and the priori of size, filter out irrational profile;
C) the remaining edge contour being obtained according to step b), calculates the boundary rectangle of 4 collimation angle, in identification process
In, when having qualified collimation angle at least three footmarks, effective cog region that its boundary rectangle as calculates
Domain;
Step 4, the circuit components that detection is located in the effective identification region of coloured image, concretely comprise the following steps:
1, because the color of each circuit components shell is different, therefore passes through color distortion, be partitioned into each circuit elements
Device, and extract the Internal periphery of each circuit components shell, concretely comprise the following steps:
A) because the color of circuit components is unfavorable for separated in rgb color space, illumination variation is also compared
Sensitivity, so, by the region of interest extracting area image by rgb color space conversion to the hsv face laying particular emphasis on color representation
The colour space, concrete conversion formula is:
V=max { c (r '), c (g '), c (b ') };
Wherein, h represents tone value, and s represents intensity value, and v represents brightness value, and max { c (r '), c (g '), c (b ') } represents
In original image, pixel is in the pixel maximum of three passages of red, green, blue, min { c (r '), c (g '), c (b ') } table
Show in original image the pixel minimum in three passages of red, green, blue for the pixel, and the span of h is located at 0-
Between 360;
B) in hsv color space, color according to involved by the shell of the circuit components priori in hsv space
Threshold value, coloured image is carried out binary conversion treatment, and concrete formula is as follows:
In binary picture, (x, y) &b_s (during x, y) &b_v (x, y), as generates binary system figure to b (x, y)=b_h
Picture;
Wherein, b (x, y) represents the binary pixel values of image slices vegetarian refreshments (x, y), and h (x, y), s (x, y), v (x, y) are respectively
Represent tone value in hsv color space for image slices vegetarian refreshments (x, y), intensity value, brightness value;B_h (x, y), b_s (x, y),
B_v (x, y) represents image slices vegetarian refreshments (x, y) respectively whether respectively in specified h, s, v region, if it is, value is 1,
Otherwise, value is 0;hmin、hmaxRepresent that the priori of color tone in hsv color space of certain components and parts shell is minimum respectively
And maximum;smin、smaxRepresent respectively color saturation in hsv color space of certain components and parts shell priori minimum and
Maximum;vmin、vmaxRepresent the priori minimum and maximum of color brightness in hsv color space of certain components and parts shell respectively
Value;
C) scan binary image, find out all edge contours;
Binary image can be regarded as only two worth gray level images, and the edge of image refers to gray scale in gray level image
Change more violent part, the intensity of variation of gray value using neighbor between graded carry out quantificational expression, gradient is
The two dimensional equivalent formula of single order two dimension derivative, concrete calculating process is:
First, calculate the difference of neighbor, concrete formula is:
gx=f [i, j+1]-f [i, j]
gy=f [i, j]-f [i+1, j]
Wherein, gxRepresent neighbor difference in the x direction, gyRepresent neighbor difference in y-direction, f [i, j
+ 1] represent the pixel value that image arranges in the i-th row jth+1, f [i, j] represents the pixel value that image arranges in the i-th row jth;F [i+1, j]
Represent the pixel value that image arranges in i+1 row jth,
Further, calculate the gradient between neighbor, concrete formula is:
Wherein, g (x, y) represents the Grad on (x, y) point of image,Represent pixel value derivation in the x direction,Represent pixel value derivation in y-direction;
Further, calculate the gradient magnitude of marginal point, the gradient magnitude set of all marginal points is the edge extracting
Profile;
Further, calculating nonstatic electric fan and the method for remaining circuit components and parts ear is according to circuit components ear
The priori of piece contour shape, size and eccentricity, filters irrational profile, thus obtaining 2 ears of circuit components
Piece position;
Because the ear of static electric fan is possible to be covered some or all of by flabellum it is therefore desirable to carry out face to step b)
Color is split, and is partitioned into blue binary map and green binary map;
Further, blue binary map is scanned, is known by the priori of the contour shape of flabellum, eccentricity, size
Knowing, determining whether there is the presence of flabellum, if it has, then judging this circuit components as fan;Otherwise, leap to step 2;
Further, after judging this circuit components as fan, the computational methods for the ear of static fan are as follows: right
Step 1 Green binary map carries out clustering processing, green point close together is polymerized to a class, by of a sort point set number
Amount, the size of point set minimum outsourcing rectangle, eccentricity, the priori of point set position, filter to the point set being polymerized to a class,
If being only left two class point sets after filtering then it is assumed that the position of current flabellum has covered the ear of electric components with having greater probability
Piece, then the central point position of circuit components ear is calculated by the center position of two class point sets and the profile information of components and parts
Put;
2, according to the profile of the circuit components extracting in step 1, calculate the position and partially of each circuit components
Gyration, concretely comprises the following steps:
According to ear location, the center position of the circuit components calculating, existed with the ear ensureing circuit components
Horizontal direction is standard, calculates the anglec of rotation of circuit components.According to the ear location of the circuit components calculating,
Center position, with ensure circuit components ear in the horizontal direction as standard, calculate the anglec of rotation of circuit components
Degree;
3, the deflection angle rotation circuit components and parts calculating according to step 2, then partitioning circuitry components and parts, compiled by color
Code identifies the classification of circuit components, concretely comprises the following steps:
Firstly, it is necessary to be that all circuit components arrange a kind of coding rule in advance, each circuit components is made to have only
One coding;Because circuit components quantity to be identified is limited, therefore chooses red, yellow, blue, the green four kinds colors easily distinguished and make
For coding characteristic color;During the printing without permission of circuit components is concentrated mainly on, in, lower in three places, when these three places
A certain color pixel values exceed priori threshold value then it is assumed that this color is the color in this region;
According to circuit components color of shell, upper in, neutralization lower in print color without permission, red be designated as 1, Huang is designated as 2, Lan Ji
For 3, green be designated as 4, be no designated as 0, ignore current color and be also denoted as 0,11 circuit components can be encoded, according to circuit
Components and parts encode, you can unique type determining current detection circuit components;
Filter out the circuit components not meeting coding, by remaining circuit components type, circuit components central point,
The anglec of rotation passes to upper layer software (applications) together, and physical circuit components and parts coding is as shown in Figure 3;
Step 5, the electric wire that detection is located in the effective identification region of coloured image, concretely comprise the following steps:
Step 1), it is partitioned into electric wire, extracts skeleton by color distortion, concretely comprise the following steps:
First, in effective identification region hsv space of step 3, color according to the electric wire priori in hsv space
Threshold value, image binaryzation is processed, and obtains electric wire binary image;
Further, electric wire binary image is scanned, by the priori filter of the shape of electric wire profile, size
Except the profile not meeting electric wire feature, remaining profile is filled with, obtains the binary picture of a new live wire;
Further, skeleton refinement is carried out to the binary picture of above-mentioned live wire, obtain the skeleton of electric wire;
Step 2), by the skeleton extracting, calculate end points and the bifurcation point of electric wire, concretely comprise the following steps: by step 1)
In the electric wire skeleton that obtains, and the end points of electric wire, the priori of bifurcation point feature, find end points in electric wire skeleton and point
Trouble point, if there are bifurcation point, by electric wire skeleton from bifurcation point separately, is divided into multiple line segments, without bifurcation point, origin wire bone
Bone is exactly a line segment, and all of line segment and line segment end points type in skeleton is supplied to upper layer software (applications);
Step 6, judges that whether circuit components are accurate with the connection of electric wire, concretely comprise the following steps:
The step 4 circuit components identifying and the electric wire that step 5 identifies are linked together by upper layer software (applications), then
The circuit diagram realizing storage with upper layer software (applications) is compared, if consistent with circuit diagram then it is assumed that circuit connect accurate.
In the present embodiment, colored pixels threshold value is 30% of each palace lattice area in nine grids.
The present invention is cleverly by Appliance computer vision pattern recognition technique and hsv color space, binary conversion treatment, image
Cutting technique combines use, can interpolate that out type and the electric wire of circuit components, and can interpolate that out that circuit connects
Whether accurately, there is fast operation, accurate positioning, hardware is united well with software engineering, game interaction designs
Ingenious;Attractive in appearance simple, judge more quick, strengthen the imaginative power of child, increase game is interesting, makes child may learn base
Plinth electric circuit knowledge, the interest of culture child.
Detection algorithm of the present invention more science, maturation, will be thin to the color conversion of image, image convolution, image cutting, skeleton
Change scheduling algorithm to combine use, can quickly judge type and the electric wire of circuit components.
Calculating speed of the present invention is fast;Every time detection and localization takes in 200ms, provides smooth use body for player
Test.
Stable performance of the present invention, in the case of different illumination, different panel computers being installed in Educational toy external member,
Carry out collecting test for 3,000 width pictures, false recognition rate and loss are below 0.2%.
Ultimate principle, principal character and the advantages of the present invention of the present invention have been shown and described above.The technology of the industry
, it should be appreciated that the present invention is not restricted to the described embodiments, the simply explanation described in above-described embodiment and description is originally for personnel
Invention principle, without departing from the spirit and scope of the present invention the present invention also have various changes and modifications, these change
Change and improvement both falls within scope of the claimed invention.Claimed scope by appending claims and its
Equivalent defines.