CN108765436A - Method for detecting image edge is piled up based on the irregular beverage bottle for improving Roberts operators - Google Patents
Method for detecting image edge is piled up based on the irregular beverage bottle for improving Roberts operators Download PDFInfo
- Publication number
- CN108765436A CN108765436A CN201810362962.XA CN201810362962A CN108765436A CN 108765436 A CN108765436 A CN 108765436A CN 201810362962 A CN201810362962 A CN 201810362962A CN 108765436 A CN108765436 A CN 108765436A
- Authority
- CN
- China
- Prior art keywords
- image
- operators
- roberts
- edge
- gradient
- 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
Links
- 230000001788 irregular Effects 0.000 title claims abstract description 29
- 238000000034 method Methods 0.000 title claims abstract description 26
- 235000013361 beverage Nutrition 0.000 title claims abstract description 25
- 238000003708 edge detection Methods 0.000 claims abstract description 55
- 230000000877 morphologic effect Effects 0.000 claims abstract description 45
- 238000004422 calculation algorithm Methods 0.000 claims abstract description 25
- 238000012545 processing Methods 0.000 claims description 8
- 238000001514 detection method Methods 0.000 claims description 7
- 239000000284 extract Substances 0.000 claims description 6
- 230000011218 segmentation Effects 0.000 claims description 5
- 230000004927 fusion Effects 0.000 claims description 4
- 230000033228 biological regulation Effects 0.000 claims description 3
- 238000004364 calculation method Methods 0.000 claims description 3
- 235000019504 cigarettes Nutrition 0.000 claims description 3
- 230000010339 dilation Effects 0.000 claims description 3
- 230000003628 erosive effect Effects 0.000 claims description 3
- 238000001914 filtration Methods 0.000 claims description 3
- 230000005540 biological transmission Effects 0.000 abstract 1
- 238000001228 spectrum Methods 0.000 abstract 1
- 238000005516 engineering process Methods 0.000 description 6
- 238000011160 research Methods 0.000 description 3
- 238000004458 analytical method Methods 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000009416 shuttering Methods 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 238000002474 experimental method Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000002068 genetic effect Effects 0.000 description 1
- 238000003709 image segmentation Methods 0.000 description 1
- 230000036039 immunity Effects 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 230000005764 inhibitory process Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000000717 retained effect Effects 0.000 description 1
- 238000005070 sampling Methods 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- XLYOFNOQVPJJNP-UHFFFAOYSA-N water Substances O XLYOFNOQVPJJNP-UHFFFAOYSA-N 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/13—Edge detection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T5/00—Image enhancement or restoration
- G06T5/70—Denoising; Smoothing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20024—Filtering details
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T2207/00—Indexing scheme for image analysis or image enhancement
- G06T2207/20—Special algorithmic details
- G06T2207/20036—Morphological image processing
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Image Analysis (AREA)
Abstract
A kind of irregular beverage bottle stacking method for detecting image edge based on improvement Roberts operators is claimed in the present invention, includes the following steps:S1, irregular beverage bottle stacking image is filtered using mathematic morphology smooth algorithm, retains marginal information and removes noise;S2, the Morphological Gradient for calculating single scale hypograph;According to the Morphological Gradient of the single scale hypograph, multiscale morphological gradient image is calculated;Determine the position of the maximum point in multiscale morphological gradient image;Corresponding Morphological Gradient edge is chosen using zero crossing situation template, and pseudo-edge is removed according to predetermined threshold value;S3, edge detection is carried out to multiscale morphological gradient image using improvement Roberts edge detection operators, obtains the marginal information that irregular beverage bottle piles up image, this communication base station can improve frequency spectrum apportionment ratio, improve efficiency of transmission.
Description
Technical field
The invention belongs to shop equipment control technology field more particularly to a kind of not advising based on improvement Roberts operators
Then beverage bottle piles up method for detecting image edge.
Background technology
Technique of image edge detection is mainly used in monitoring and entrance guard facilities, carries out control or the automobile of road traffic
The monitoring of gate inhibition.It is gradually used in the automatic control system of factory now, the stacking for lot cargo in factory is existing
Method be manually to pile up, take time and effort, need to pile up image carry out edge detection, can be only achieved automation pile up.
In addition, technique of image edge detection is equally peace with camera sampling technique, digital-to-analogue conversion technology, interfacing
One of the core technology of anti-system, there is deep Research foundation, and from nineteen sixty-five so far nearly 50 years, there are many learn both at home and abroad
Person studies it, and obtains many achievements in different field.The type of edge detection algorithm is more, can be divided into traditional algorithm
With emerging algorithm.Traditional algorithm has:Sobel operators, Prewitt operators, LOG operators, laplacian operators and Canny operators
Deng.Traditional algorithm is all based on greatly mathematical operation realization or the precision of noiseproof feature difference or edge detection is not high.Emerging
The research of edge detection algorithm is often intersected with Other subjects, is had in terms of engineering:Side based on wavelet analysis and wavelet packet
Edge detection method, the edge detection method based on fuzzy theory, the dividing method etc. based on neural network, in machine vision and people
The fields such as work intelligence have:Edge detection method, self-organizing clustering method, genetic algorithm based on mathematical morphology etc..With research
Go deep into, it is found that the accuracy of detection of Roberts operators is high, therefore have part using Roberts operators to carry out edge detection, than
Such as 201510172635.4 (title of Patent No.:Method for detecting image edge based on Roberts operators), this method compared to
The prior art is improved the precision and anti-noise ability of Image Edge-Detection, cannot under complicated noise but exist
Still there are preferable identification, and the problem that robustness is poor.Traditional Roberts edge detection operators edge positioning is accurate, but
It is to noise-sensitive.The apparent and less noise image segmentation suitable for edge.Roberts edge detection operators are a kind of utilizations
Local difference operator finds the operator at edge, and result edge is not very smooth after Robert operator image procossings.Through analyzing, due to
Robert operators would generally generate wider response in the region near image border, it is therefore desirable to make improvement.
Invention content
Present invention seek to address that the above problem of the prior art.Propose that a kind of automation control degree is high, improves and piles up essence
Exactness, the irregular beverage bottle stacking figure based on improvement Roberts operators that still can have higher robustness under complex environment
As edge detection method.
Technical scheme is as follows:
It is a kind of that method for detecting image edge is piled up based on the irregular beverage bottle for improving Roberts operators comprising following
Step:
S1, irregular beverage bottle stacking image is filtered, retains marginal information and gone using mathematic morphology smooth algorithm
Except noise, the specific steps are:Using the opening operation processing in morphology, if structural element is s1, defining opening operation operation is:
In formula, F indicates that irregular beverage bottle piles up the set of image,Indicate that opening operation operation, s indicate structural element, Θ
Indicate that image F is corroded by structural element s,Indicate expansion of the structural element to image F;At the closed operation in morphology
Reason, if structural element is s2, defining closed operation operation is:
In formula, F indicates the set of irregular cigarette packet image, indicates that closed operation operation, s indicate that structural element, Θ indicate
Image F is corroded by structural element s,Indicate expansion of the structural element to image F;
S2, the Morphological Gradient for calculating single scale hypograph;According to the Morphological Gradient of the single scale hypograph,
Multiscale morphological gradient image is calculated;Determine the position of the maximum point in multiscale morphological gradient image;It uses
Zero crossing situation template chooses corresponding Morphological Gradient edge, and removes pseudo-edge according to predetermined threshold value;
S3, edge is carried out to the multiscale morphological gradient image of step S2 using improvement Roberts edge detection operators
Detection, obtains the marginal information that irregular beverage bottle piles up image, and the improvement Roberts edge detection operators are mainly:It adopts
2 × 2 neighborhoods are replaced in Roberts algorithms with 3 × 3 neighborhoods to calculate gradient magnitude;Also calculated in improved Roberts edge detections
Improved Sobel operators are added on the basis of son, improved Roberts edge detection operators are mainly the improved Sobel
Operator mainly improves:In the template of traditional Sobel operators vertical and horizontal both direction, as unit of 45 degree will vertically with
Horizontal shuttering is divided into 8 direction templates.
Further, the improved Sobel operators of the step S3 mainly improve:Traditional Sobel operators vertically and water
In the template of flat both direction, vertical and horizontal template is divided into 8 direction template bodies as unit of 45 degree, is specifically included:
S31:Boundary is the graded of intensity level, and edge is the position of graded, the size with gradient vector and side
Always this variation is stated, the size and Orientation of edge gradient vector states this variation;
Gradient operator is first derivative operator, and image f (x, y) is defined as lower column vector in the gradient of position (i, j):
The range value of ▽ f is:
S32:The template of the horizontal and vertical both direction of Sobel operators does convolution algorithm with image f (x, y), close with this
Like the Grad at calculating (i, j), G can be solved by following equationxAnd GyValue:
F (x, y)=max | Gx|,|Gy|}
Gx=f (i+1, j-1)+2f (i+1, j)+f (i+1, j+1)-f (i-1, j-1) -2f (i-1, j)-f (i-1, j+1)
Gy=f (i-1, j+1)+2f (i, j+1)+f (i+1, j+1)-f (i-1, j-1) -2f (i, j-1)-f (i+1, j-1)
In formula, GxAnd GyGrad respectively horizontally and vertically;
S33:On the basis of traditional Sobel operators, increase the template of other six directions, respectively 45 °, 135 °,
180°,225°,270°,315°;
S34:According to above-mentioned 8 direction templates, the weights of different directions are calculated, calculation formula is as follows:
Lng (x, y)=- ln2 [d (x, y)2-u]
ω (x, y)=[g (x, y)]
In formula, d (x, y) indicates that the Euclidean distance between the element and central point of template, g (x, y) indicate the reality at (x, y)
Number weights, u indicate regulation coefficient, ω (x, y) are obtained to g (x, y) rounding;
S35:According to the weight of each point of preset template, then pixel corresponding with target image does convolution fortune
It calculates;
S36:The maximum value obtained in step s 25 is chosen, the corresponding target figure of template center's point is replaced with this maximum value
The pixel value of picture finally exports maximum gray value as the pixel output in all templates;
S37:Suitable threshold value T is set, if gradient magnitude ▽ f (i, j) >=T at (i, j), which is defined as edge
Point.
Further, the improved Roberts edge detection operators are replaced using 3 × 3 neighborhoods 2 in Roberts algorithms
× 2 neighborhoods calculate gradient magnitude, and using the matched denoising model of three-dimensional bits of similitude between image block, improve
The accuracy of detection and noiseproof feature of Roberts operators;It is obtained most instead of artificial specified threshold by optimal threshold alternative manner
Good segmentation threshold efficiently extracts objective contour in figure.
Further, the step S2 calculates the Morphological Gradient of single scale hypograph using following formula:
G (f)=(f ⊕ B)-(f Θ B);
Wherein, G (f) is the Morphological Gradient of single scale hypograph, and f is the spatial target images after filtering out noise, and B is
Structural elements, ⊕ and Θ indicate the dilation operation in Morphological scale-space and erosion operation respectively.
Further, multiscale morphological gradient image is calculated using formula as described below:
Wherein, MG (f) is multiscale morphological gradient;B i are i-th of structural elements, and size is (2i+1) × (2i+1);
N is scale parameter.
Further, Weighted Fusion between the improved Roberts edge detection operators and improved Sobel operators,
The weight coefficient of wherein improved Roberts edge detection operators is 0.5, and the weight coefficient of improved Sobel operators is 0.5.
It advantages of the present invention and has the beneficial effect that:
The present invention is carrying out edge extracting based on morphological gradient to spatial target images, in order to improve single structure
When member carries out Morphological scale-space, the contradiction between the accuracy and the noise immunity at edge of edge positioning utilizes mathematical morphology
Basic operation designs and has used to meet and detects abrupt local letter by the multiscale morphological gradient of processing image border characteristic
Breath to obtain the marginal information of image, used multiple and different scales structural elements carry out edge detection, used it is multiple not
Morphologic burn into expansive working is carried out to spatial target images with the morphological operator under scale, by between adjacent scale into
Row asks difference that approximate image border profile is calculated.By the processing on multiple scales, influence of noise can be effectively weakened,
The accurate of edge is kept simultaneously, the deficiency of single scale edge detection algorithm is effectively compensated for, substantially increases extraterrestrial target side
The accuracy and robustness of edge extraction, convenient for the processing of the subsequent images such as Target Segmentation, target identification, target following.
Another innovative point of the present invention is will to improve Roberts edge detection operators to calculate instead of Roberts using 3 × 3 neighborhoods
2 × 2 neighborhoods calculate gradient magnitude in method;And it using the matched denoising model of three-dimensional bits of similitude between image block, improves
The accuracy of detection and noiseproof feature of Roberts operators;It is obtained most instead of artificial specified threshold by optimal threshold alternative manner
Good segmentation threshold efficiently extracts objective contour in figure.The simulation experiment result shows that algorithm PSNR reaches 33dB or so, than
The edge detection algorithm noiseproof feature of anti-noise morphologic edge detection algorithm and a kind of improved Roberts and grey correlation analysis
It is good, while inhibiting noise jamming, marginal information, the overall profile of better extract target can be retained.Also improved
Improved Sobel operators are added on the basis of Roberts edge detection operators, improved Roberts edge detection operators are main
It is that the improved Sobel operators mainly improve:In the template of traditional Sobel operators vertical and horizontal both direction, with
Vertical and horizontal template is divided into 8 direction templates by 45 degree for unit.Increase by 45 °, 135 °, 180 °, 225 °, 270 °, 315 °
The template of six different directions increases the ability of the marginal information of Sobel operators processing all directions;Identical experimental situation
Under, still there are preferable identification, and stronger robustness under complicated noise, can further extract irregular component
The edge feature of picture largely improves the performance of irregular image identifying system.Of the invention innovative uses
Improved Roberts edge detection operators carry out edge detection with improved Sobel operators Weighted Fusion at a new operator,
Promote mutually in respective individually effect, realizes better edge detection.
Description of the drawings
Fig. 1 is that the present invention provides preferred embodiment based on the irregular beverage bottle stacking image side for improving Roberts operators
Edge detection method flow chart.
Specific implementation mode
Following will be combined with the drawings in the embodiments of the present invention, and technical solution in the embodiment of the present invention carries out clear, detailed
Carefully describe.Described embodiment is only a part of the embodiment of the present invention.
The present invention solve above-mentioned technical problem technical solution be:
It is a kind of irregular beverage bottle stacking method for detecting image edge based on improvement Roberts operators as shown in Figure 1,
It includes the following steps:
S1, irregular beverage bottle stacking image is filtered, retains marginal information and gone using mathematic morphology smooth algorithm
Except noise, the specific steps are:Using the opening operation processing in morphology, if structural element is s1, defining opening operation operation is:
In formula, F indicates that irregular beverage bottle piles up the set of image,Indicate that opening operation operation, s indicate structural element, Θ
Indicate that image F is corroded by structural element s,Indicate expansion of the structural element to image F;At the closed operation in morphology
Reason, if structural element is s2, defining closed operation operation is:
In formula, F indicates the set of irregular cigarette packet image, indicates that closed operation operation, s indicate that structural element, Θ indicate
Image F is corroded by structural element s,Indicate expansion of the structural element to image F;
S2, the Morphological Gradient for calculating single scale hypograph;According to the Morphological Gradient of the single scale hypograph,
Multiscale morphological gradient image is calculated;Determine the position of the maximum point in multiscale morphological gradient image;It uses
Zero crossing situation template chooses corresponding Morphological Gradient edge, and removes pseudo-edge according to predetermined threshold value;
S3, edge is carried out to the multiscale morphological gradient image of step S2 using improvement Roberts edge detection operators
Detection, obtains the marginal information that irregular beverage bottle piles up image, and the improvement Roberts edge detection operators are mainly:It adopts
2 × 2 neighborhoods are replaced in Roberts algorithms with 3 × 3 neighborhoods to calculate gradient magnitude;Also calculated in improved Roberts edge detections
Improved Sobel operators are added on the basis of son, improved Roberts edge detection operators are mainly the improved Sobel
Operator mainly improves:In the template of traditional Sobel operators vertical and horizontal both direction, as unit of 45 degree will vertically with
Horizontal shuttering is divided into 8 direction templates.
Preferably, the improved Sobel operators of the step S3 mainly improve:In traditional Sobel operators vertical and horizontal
In the template of both direction, vertical and horizontal template is divided into 8 direction template bodies as unit of 45 degree, is specifically included:
S31:Boundary is the graded of intensity level, and edge is the position of graded, the size with gradient vector and side
Always this variation is stated, the size and Orientation of edge gradient vector states this variation;
Gradient operator is first derivative operator, and image f (x, y) is defined as lower column vector in the gradient of position (i, j):
The range value of ▽ f is:
S32:The template of the horizontal and vertical both direction of Sobel operators does convolution algorithm with image f (x, y), close with this
Like the Grad at calculating (i, j), G can be solved by following equationxAnd GyValue:
F (x, y)=max | Gx|,|Gy|}
Gx=f (i+1, j-1)+2f (i+1, j)+f (i+1, j+1)-f (i-1, j-1) -2f (i-1, j)-f (i-1, j+1)
Gy=f (i-1, j+1)+2f (i, j+1)+f (i+1, j+1)-f (i-1, j-1) -2f (i, j-1)-f (i+1, j-1)
In formula, GxAnd GyGrad respectively horizontally and vertically;
S33:On the basis of traditional Sobel operators, increase the template of other six directions, respectively 45 °, 135 °,
180°,225°,270°,315°;
S34:According to above-mentioned 8 direction templates, the weights of different directions are calculated, calculation formula is as follows:
Lng (x, y)=- ln2 [d (x, y)2-u]
ω (x, y)=[g (x, y)]
In formula, d (x, y) indicates that the Euclidean distance between the element and central point of template, g (x, y) indicate the reality at (x, y)
Number weights, u indicate regulation coefficient, ω (x, y) are obtained to g (x, y) rounding;
S35:According to the weight of each point of preset template, then pixel corresponding with target image does convolution fortune
It calculates;
S36:The maximum value obtained in step s 25 is chosen, the corresponding target figure of template center's point is replaced with this maximum value
The pixel value of picture finally exports maximum gray value as the pixel output in all templates;
S37:Suitable threshold value T is set, if gradient magnitude ▽ f (i, j) >=T at (i, j), which is defined as edge
Point.
Preferably, the improved Roberts edge detection operators are replaced using 3 × 3 neighborhoods 2 × 2 in Roberts algorithms
Neighborhood calculates gradient magnitude, and using the matched denoising model of three-dimensional bits of similitude between image block, improves Roberts and calculate
The accuracy of detection and noiseproof feature of son;By optimal threshold alternative manner optimal segmentation threshold is obtained instead of artificial specified threshold
Value, efficiently extracts objective contour in figure.Between the improved Roberts edge detection operators and improved Sobel operators
Weighted Fusion, wherein the weight coefficient of improved Roberts edge detection operators is 0.5, the weight system of improved Sobel operators
Number is 0.5.
Preferably, the step S2 calculates the Morphological Gradient of single scale hypograph using following formula:
G (f)=(f ⊕ B)-(f Θ B);
Wherein, G (f) is the Morphological Gradient of single scale hypograph, and f is the spatial target images after filtering out noise, and B is
Structural elements, ⊕ and Θ indicate the dilation operation in Morphological scale-space and erosion operation respectively.Preferably, using public affairs as described below
Multiscale morphological gradient image is calculated in formula:
Wherein, MG (f) is multiscale morphological gradient;B i are i-th of structural elements, and size is (2i+1) × (2i+1);
N is scale parameter.
The preferred embodiment of the present invention has been described in detail above.It should be appreciated that those skilled in the art without
It needs creative work according to the present invention can conceive and makes many modifications and variations.Therefore, all technologies in the art
Personnel are available by logical analysis, reasoning, or a limited experiment on the basis of existing technology under this invention's idea
Technical solution, all should be in the protection domain being defined in the patent claims.
Claims (6)
1. a kind of piling up method for detecting image edge based on the irregular beverage bottle for improving Roberts operators, which is characterized in that packet
Include following steps:
S1, irregular beverage bottle stacking image is filtered using mathematic morphology smooth algorithm, retains marginal information and removes and make an uproar
Sound, the specific steps are:Using the opening operation processing in morphology, if structural element is s1, defining opening operation operation is:
In formula, F indicates that irregular beverage bottle piles up the set of image,Indicate that opening operation operation, s indicate that structural element, Θ indicate
Image F is corroded by structural element s,Indicate expansion of the structural element to image F;Using the closed operation processing in morphology, if
Structural element is s2, defining closed operation operation is:
In formula, F indicates the set of irregular cigarette packet image, indicates that closed operation operation, s indicate that structural element, Θ indicate image F
Corroded by structural element s,Indicate expansion of the structural element to image F;
S2, the Morphological Gradient for calculating single scale hypograph;According to the Morphological Gradient of the single scale hypograph, calculate
Obtain multiscale morphological gradient image;Determine the position of the maximum point in multiscale morphological gradient image;Use zero passage
Point situation template chooses corresponding Morphological Gradient edge, and removes pseudo-edge according to predetermined threshold value;
S3, edge detection is carried out to the multiscale morphological gradient image of step S2 using improvement Roberts edge detection operators,
The marginal information that irregular beverage bottle piles up image is obtained, the improvement Roberts edge detection operators are mainly:Using 3 ×
3 neighborhoods calculate gradient magnitude instead of 2 × 2 neighborhoods in Roberts algorithms;Also in improved Roberts edge detection operators base
Improved Sobel operators are added on plinth, improved Roberts edge detection operators are mainly the improved Sobel operators
It is main to improve:In the template of traditional Sobel operators vertical and horizontal both direction, by vertical and horizontal as unit of 45 degree
Template is divided into 8 direction templates.
2. according to claim 1 pile up Image Edge-Detection side based on the irregular beverage bottle for improving Roberts operators
Method, which is characterized in that the improved Sobel operators of step S3 mainly improve:In traditional Sobel operators vertical and horizontal two
In the template in a direction, vertical and horizontal template is divided into 8 direction template bodies as unit of 45 degree, is specifically included:
S31:Boundary is the graded of intensity level, and edge is the position of graded, with the size and Orientation of gradient vector come
State this variation, the size and Orientation of edge gradient vector states this variation;
Gradient operator is first derivative operator, and image f (x, y) is defined as lower column vector in the gradient of position (i, j):
Range value be:
S32:The template of the horizontal and vertical both direction of Sobel operators does convolution algorithm with image f (x, y), in terms of this approximation
The Grad at (i, j) is calculated, G can be solved by following equationxAnd GyValue:
F (x, y)=max | Gx|,|Gy|}
Gx=f (i+1, j-1)+2f (i+1, j)+f (i+1, j+1)-f (i-1, j-1) -2f (i-1, j)-f (i-1, j+1)
Gy=f (i-1, j+1)+2f (i, j+1)+f (i+1, j+1)-f (i-1, j-1) -2f (i, j-1)-f (i+1, j-1)
In formula, GxAnd GyGrad respectively horizontally and vertically;
S33:On the basis of traditional Sobel operators, increase the template of other six directions, respectively 45 °, 135 °, 180 °,
225°,270°,315°;
S34:According to above-mentioned 8 direction templates, the weights of different directions are calculated, calculation formula is as follows:
Lng (x, y)=- ln2 [d (x, y)2-u]
ω (x, y)=[g (x, y)]
In formula, d (x, y) indicates that the Euclidean distance between the element and central point of template, g (x, y) indicate the real number power at (x, y)
Value, u indicate regulation coefficient, ω (x, y) are obtained to g (x, y) rounding;
S35:According to the weight of each point of preset template, then pixel corresponding with target image does convolution algorithm;
S36:The maximum value obtained in step s 25 is chosen, the corresponding target image of template center's point is replaced with this maximum value
Pixel value finally exports maximum gray value as the pixel output in all templates;
S37:Suitable threshold value T is set, if gradient magnitude ▽ f (i, j) >=T at (i, j), which is defined as marginal point.
3. according to claim 1 pile up Image Edge-Detection side based on the irregular beverage bottle for improving Roberts operators
Method, which is characterized in that the improved Roberts edge detection operators are replaced using 3 × 3 neighborhoods 2 × 2 in Roberts algorithms
Neighborhood calculates gradient magnitude, and using the matched denoising model of three-dimensional bits of similitude between image block, improves Roberts and calculate
The accuracy of detection and noiseproof feature of son;By optimal threshold alternative manner optimal segmentation threshold is obtained instead of artificial specified threshold
Value, efficiently extracts objective contour in figure.
4. according to claim 1 pile up Image Edge-Detection side based on the irregular beverage bottle for improving Roberts operators
Method, which is characterized in that the step S2 calculates the Morphological Gradient of single scale hypograph using following formula:
Wherein, G (f) is the Morphological Gradient of single scale hypograph, and f is the spatial target images after filtering out noise, and B is structure
Member,WithThe dilation operation and erosion operation in Morphological scale-space are indicated respectively.
5. according to claim 4 pile up Image Edge-Detection side based on the irregular beverage bottle for improving Roberts operators
Method, which is characterized in that multiscale morphological gradient image is calculated using formula as described below:
Wherein, MG (f) is multiscale morphological gradient;B i are i-th of structural elements, and size is (2i+1) × (2i+1);N is
Scale parameter.
6. according to claim 2 pile up Image Edge-Detection side based on the irregular beverage bottle for improving Roberts operators
Method, which is characterized in that Weighted Fusion between the improved Roberts edge detection operators and improved Sobel operators, wherein
The weight coefficient of improved Roberts edge detection operators is 0.5, and the weight coefficient of improved Sobel operators is 0.5.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810362962.XA CN108765436A (en) | 2018-04-21 | 2018-04-21 | Method for detecting image edge is piled up based on the irregular beverage bottle for improving Roberts operators |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201810362962.XA CN108765436A (en) | 2018-04-21 | 2018-04-21 | Method for detecting image edge is piled up based on the irregular beverage bottle for improving Roberts operators |
Publications (1)
Publication Number | Publication Date |
---|---|
CN108765436A true CN108765436A (en) | 2018-11-06 |
Family
ID=64010999
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201810362962.XA Pending CN108765436A (en) | 2018-04-21 | 2018-04-21 | Method for detecting image edge is piled up based on the irregular beverage bottle for improving Roberts operators |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN108765436A (en) |
Cited By (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113344959A (en) * | 2021-08-06 | 2021-09-03 | 山东捷瑞数字科技股份有限公司 | Residual material analysis processing method and device and material conveying system |
CN113643290A (en) * | 2021-10-14 | 2021-11-12 | 昌亚新材料科技有限公司 | Straw counting method and device based on image processing and storage medium |
CN115330802A (en) * | 2022-10-17 | 2022-11-11 | 山东大学 | Carbon fiber composite material gas cylinder X-ray image debonding defect extraction method |
CN115689943A (en) * | 2022-11-18 | 2023-02-03 | 武汉保诚信科技有限公司 | Micro graph code motion blur detection method based on gradient symmetry |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104732530A (en) * | 2015-03-10 | 2015-06-24 | 北京环境特性研究所 | Image edge detection method |
CN106073698A (en) * | 2016-06-14 | 2016-11-09 | 江苏大学 | A kind of fundus imaging method based on Android |
CN107133962A (en) * | 2017-07-07 | 2017-09-05 | 哈尔滨理工大学 | A kind of diamond saw blade extracting thermal crack method based on rim detection |
-
2018
- 2018-04-21 CN CN201810362962.XA patent/CN108765436A/en active Pending
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104732530A (en) * | 2015-03-10 | 2015-06-24 | 北京环境特性研究所 | Image edge detection method |
CN106073698A (en) * | 2016-06-14 | 2016-11-09 | 江苏大学 | A kind of fundus imaging method based on Android |
CN107133962A (en) * | 2017-07-07 | 2017-09-05 | 哈尔滨理工大学 | A kind of diamond saw blade extracting thermal crack method based on rim detection |
Non-Patent Citations (1)
Title |
---|
YI ZHANG 等: "《Edge Detection Algorithm of Image Fusion Based on Improved Sobel Operator》", 《ITOEC》 * |
Cited By (6)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN113344959A (en) * | 2021-08-06 | 2021-09-03 | 山东捷瑞数字科技股份有限公司 | Residual material analysis processing method and device and material conveying system |
CN113643290A (en) * | 2021-10-14 | 2021-11-12 | 昌亚新材料科技有限公司 | Straw counting method and device based on image processing and storage medium |
CN115330802A (en) * | 2022-10-17 | 2022-11-11 | 山东大学 | Carbon fiber composite material gas cylinder X-ray image debonding defect extraction method |
CN115330802B (en) * | 2022-10-17 | 2024-01-19 | 山东大学 | Method for extracting debonding defect of X-ray image of carbon fiber composite gas cylinder |
CN115689943A (en) * | 2022-11-18 | 2023-02-03 | 武汉保诚信科技有限公司 | Micro graph code motion blur detection method based on gradient symmetry |
CN115689943B (en) * | 2022-11-18 | 2023-07-25 | 武汉保诚信科技有限公司 | Micro-image code motion blur detection method based on gradient symmetry |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN108734716A (en) | A kind of fire complex environment image detecting method based on improvement Prewitt operators | |
CN108765436A (en) | Method for detecting image edge is piled up based on the irregular beverage bottle for improving Roberts operators | |
CN103984961B (en) | A kind of image detecting method for being used to detect underbody foreign matter | |
CN102542289B (en) | Pedestrian volume statistical method based on plurality of Gaussian counting models | |
CN101398894B (en) | Automobile license plate automatic recognition method and implementing device thereof | |
CN104268872B (en) | Consistency-based edge detection method | |
CN104299008B (en) | Vehicle type classification method based on multi-feature fusion | |
CN103886589B (en) | Object-oriented automated high-precision edge extracting method | |
CN104134077B (en) | A kind of gait recognition method unrelated with visual angle based on the determination theories of learning | |
CN108764186A (en) | Personage based on rotation deep learning blocks profile testing method | |
CN107895376A (en) | Based on the solar panel recognition methods for improving Canny operators and contour area threshold value | |
CN103984946B (en) | High resolution remote sensing map road extraction method based on K-means | |
CN104794435B (en) | A kind of unmanned plane of view-based access control model moving target detecting method over the ground | |
CN107451999A (en) | foreign matter detecting method and device based on image recognition | |
CN109241871A (en) | A kind of public domain stream of people's tracking based on video data | |
CN107392885A (en) | A kind of method for detecting infrared puniness target of view-based access control model contrast mechanism | |
CN105184812B (en) | A kind of pedestrian based on target following hovers detection method | |
CN109918971B (en) | Method and device for detecting number of people in monitoring video | |
CN102163284A (en) | Chinese environment-oriented complex scene text positioning method | |
CN104061907A (en) | Viewing-angle greatly-variable gait recognition method based on gait three-dimensional contour matching synthesis | |
CN101527043B (en) | Video picture segmentation method based on moving target outline information | |
Karpagavalli et al. | Estimating the density of the people and counting the number of people in a crowd environment for human safety | |
CN109961416B (en) | Business license information extraction method based on morphological gradient multi-scale fusion | |
CN105741276A (en) | Ship waterline extraction method | |
CN106815583A (en) | A kind of vehicle at night license plate locating method being combined based on MSER and SWT |
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 | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20181106 |
|
RJ01 | Rejection of invention patent application after publication |