CN107767387B - Contour detection method based on variable receptive field scale global modulation - Google Patents
Contour detection method based on variable receptive field scale global modulation Download PDFInfo
- Publication number
- CN107767387B CN107767387B CN201711098829.XA CN201711098829A CN107767387B CN 107767387 B CN107767387 B CN 107767387B CN 201711098829 A CN201711098829 A CN 201711098829A CN 107767387 B CN107767387 B CN 107767387B
- Authority
- CN
- China
- Prior art keywords
- pixel point
- value
- scale
- response
- function
- 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.)
- Active
Links
- 238000001514 detection method Methods 0.000 title claims abstract description 23
- 230000004044 response Effects 0.000 claims abstract description 46
- 238000001914 filtration Methods 0.000 claims abstract description 38
- 230000001629 suppression Effects 0.000 claims abstract description 19
- 238000012545 processing Methods 0.000 claims abstract description 14
- 230000005764 inhibitory process Effects 0.000 claims abstract description 11
- 230000000638 stimulation Effects 0.000 claims abstract description 10
- 238000010606 normalization Methods 0.000 claims description 3
- 239000000126 substance Substances 0.000 claims description 3
- 238000000034 method Methods 0.000 abstract description 13
- 230000000694 effects Effects 0.000 abstract description 5
- 238000004088 simulation Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 31
- 230000009286 beneficial effect Effects 0.000 description 2
- 238000011156 evaluation Methods 0.000 description 2
- 208000003098 Ganglion Cysts Diseases 0.000 description 1
- 208000005400 Synovial Cyst Diseases 0.000 description 1
- 210000004027 cell Anatomy 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 210000002569 neuron Anatomy 0.000 description 1
- 230000009897 systematic effect Effects 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 230000000007 visual effect Effects 0.000 description 1
Images
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
- G06T7/00—Image analysis
- G06T7/10—Segmentation; Edge detection
- G06T7/136—Segmentation; Edge detection involving thresholding
-
- 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
Landscapes
- Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Image Analysis (AREA)
- Image Processing (AREA)
Abstract
The invention aims to provide a contour detection method based on variable receptive field scale global modulation, which comprises the following steps: A. inputting the image to be detected after gray processing, and calculating to obtain a normalized Gaussian difference filtering value of each pixel point; B. presetting a high scale value, a low scale value and a threshold value of the scale function, respectively comparing the normalized Gaussian difference filter value of each pixel point with the threshold value, and determining the scale function value of each pixel point; C. presetting a plurality of direction parameters of the suppression strength and the equipartition circumference; carrying out Gabor filtering on each pixel point in the image to be detected according to each direction parameter, and calculating the classical receptive field stimulus response of each pixel point; D. calculating the inhibition response of each pixel point; E. and calculating the classical receptive field stimulation response and the inhibition response of each pixel point to obtain the contour response of the pixel point, and processing to obtain the final contour value of each pixel point so as to obtain a final contour map. The method has the characteristics of good simulation effect and high outline identification rate.
Description
Technical Field
The invention relates to the field of computer image processing, in particular to a contour detection method based on variable receptive field scale global modulation.
Background
Contour detection is a fundamental task in the field of computer vision, and unlike edges, which are defined as strong intensity variations, contours usually represent the boundary of one object to another. The basic method for improving the contour detection performance is to fuse global information, and in order to improve the performance of a contour detection model, many researchers try to improve an original detection operator and a suppression model; based on the scale space theory, each scale corresponds to the size of a group of neuron receptive fields, and the different receptive field sizes of ganglion cells have the characteristics under different scales; therefore, considering the dimension of the receptive field model in the model can be taken as the development direction of the field.
Disclosure of Invention
The invention aims to provide a contour detection method based on variable receptive field scale global modulation, which has the characteristics of good simulation effect and high contour recognition rate.
The technical scheme of the invention is as follows:
A. inputting a to-be-detected image subjected to gray processing, performing Gaussian difference filtering on the gray value of each pixel point by using a Gaussian difference function to obtain a Gaussian difference filtering value of each pixel point, and respectively performing normalization processing on a positive value and a negative value in the Gaussian difference filtering values of each pixel point to obtain a normalized Gaussian difference filtering function of each pixel point; performing convolution on the normalized Gaussian difference filter function of each pixel point and the gray value of the corresponding pixel point respectively to obtain a normalized Gaussian difference filter value of each pixel point;
B. presetting a high scale value, a low scale value and a threshold value of a scale function, comparing the normalized Gaussian difference filter value of each pixel point with the threshold value respectively, if the normalized Gaussian difference filter value of the pixel point is greater than or equal to the threshold value, the scale function value corresponding to the pixel point is the high scale value, and if the normalized Gaussian difference filter value of the pixel point is less than the threshold value, the scale function value corresponding to the pixel point is the low scale value;
C. presetting inhibition intensity, and presetting a plurality of direction parameters for equally dividing the circumference; b, performing Gabor filtering on each pixel point in the image to be detected according to each direction parameter to obtain a response value of each pixel point in each direction, wherein a scale function value adopted in the Gabor filtering is a high scale value or a low scale value determined by each pixel point in the step B; for each pixel point, selecting the maximum value in the response values of the pixel point in each direction as the classical receptive field stimulation response of the pixel point;
D. convolving the classical receptive field stimulation response of each pixel point with the distance weight function of the pixel point, and multiplying the convolved response by the corresponding scale of the pixel point to obtain the inhibition response of each pixel point;
E. and subtracting the product of the suppression response and the suppression strength of each pixel point from the classical receptive field stimulation response of each pixel point to obtain the contour response of the pixel point, and performing non-maximum suppression and double-threshold processing on the contour response to obtain the final contour value of each pixel point so as to obtain a final contour map.
Preferably, the step a specifically comprises:
the Gaussian difference function DoGσ(x, y) is:
wherein σ is an initialization scale;
the normalized Gaussian difference filter function wσ(x, y) is:
the normalized Gaussian difference filtering value woutσ(x, y) is:
woutσ(x,y)=I(x,y)*wσ(x,y) (3)。
preferably, the step B specifically comprises:
the scale function σ (x, y) is:
wherein sigmaHAt a high scale value, σLAt a low scale value, σH=σL+ s, t is a threshold approaching 0, s is the step size of the scale.
Preferably, the step C specifically includes:
the two-dimensional Gabor function expression of the Gabor filter bank is as follows:
whereinGamma is a constant representing the ratio of the long axis to the short axis of the elliptical field, the parameter lambda is the wavelength, sigma (x, y) is the scale function, 1/lambda is the spatial frequency of the cosine function,is a phase angle parameter, theta is a direction parameter of Gabor filtering;
i (x, y) is the gray value of each pixel point of the image to be detected, and is a convolution operator;
the Gabor energy value was calculated as follows:
wherein theta isiFor a certain direction of Gabor filtering, NθThe number of the Gabor filtering directions;
the classical receptive field stimulus response Ec (x, y; sigma (x, y)) is:
preferably, the step D specifically includes:
the distance weight function wσ(x, y) is:
wherein | · | purple sweet1Is L1Norm, h (x) max (0, x), DoG (x, y) is an expression corresponding to DoG template;
the suppression response Inh (x, y) of each pixel point is as follows:
Inh(x,y)=Ec(x,y;σ(x,y))*wσ(x,y)σ(x,y) (11)。
preferably, the step E specifically comprises:
the profile response R (x, y) is:
R(x,y)=H(Ec(x,y;σ(x,y))-αInh(x,y)) (12);
where h (x) max (0, x), α is the inhibition intensity.
The invention distinguishes and selects the high and low scale values based on the principle that the high scale value is beneficial to extracting the contour and the low scale value is beneficial to inhibiting the texture, so that the judgment is carried out according to the normalized Gaussian difference filtering value of each pixel point, if the filtering value is obviously greater than 0, the pixel point is probably positioned at the position of the contour, and the high scale value is selected to extract the contour; if the filtering value is close to 0, the pixel point is probably positioned at the position of the texture, so that the texture is restrained by selecting a low-scale value; normalizing the Gaussian difference operator to ensure that the weights of the positive and negative filtering values are consistent when filtering is carried out, and reducing the error identification rate; therefore, the filtering value is combined to judge whether the pixel point corresponds to the contour or the texture, so that the corresponding high-scale value or low-scale value is selected to calculate the suppression response, the suppression of the texture is also considered while the contour is not missed, and the influence on the contour identification effect caused by too much useless texture is avoided; in addition, the suppression response is calculated by combining the distance weight function, the suppression characteristic of the non-classical receptive field can be better reflected, and the outline detection rate is improved.
In summary, the contour detection method of the present invention not only maintains the integrity of the contour, but also greatly removes the redundant texture background, and better conforms to the spatial frequency characteristics of the visual receptive field.
Drawings
FIG. 1 is a comparison graph of the profile test of the method of example 1 of the present invention and the method of reference 1.
Detailed Description
The present invention will be described in detail with reference to the accompanying drawings and examples.
Example 1
The contour detection method based on variable receptive field scale global modulation provided by the embodiment comprises the following steps:
A. inputting a to-be-detected image subjected to gray processing, performing Gaussian difference filtering on the gray value of each pixel point by using a Gaussian difference function to obtain a Gaussian difference filtering value of each pixel point, and respectively performing normalization processing on a positive value and a negative value in the Gaussian difference filtering values of each pixel point to obtain a normalized Gaussian difference filtering function of each pixel point; performing convolution on the normalized Gaussian difference filter function of each pixel point and the gray value of the corresponding pixel point respectively to obtain a normalized Gaussian difference filter value of each pixel point;
the Gaussian difference function DoGσ(x, y) is:
wherein σ is an initialization scale;
the normalized Gaussian difference filter function wσ(x, y) is:
the normalized Gaussian difference filtering value woutσ(x, y) is:
woutσ(x,y)=I(x,y)*wσ(x,y) (3);
B. presetting a high scale value, a low scale value and a threshold value of a scale function, comparing the normalized Gaussian difference filter value of each pixel point with the threshold value respectively, if the normalized Gaussian difference filter value of the pixel point is greater than or equal to the threshold value, the scale function value corresponding to the pixel point is the high scale value, and if the normalized Gaussian difference filter value of the pixel point is less than the threshold value, the scale function value corresponding to the pixel point is the low scale value;
the step B is specifically as follows:
the scale function σ (x, y) is:
wherein sigmaHAt a high scale value, σLAt a low scale value, σH=σL+ s, t is a threshold approaching 0, s is the step length of the scale;
C. presetting inhibition intensity, and presetting a plurality of direction parameters for equally dividing the circumference; b, performing Gabor filtering on each pixel point in the image to be detected according to each direction parameter to obtain a response value of each pixel point in each direction, wherein a scale function value adopted in the Gabor filtering is a high scale value or a low scale value determined by each pixel point in the step B; for each pixel point, selecting the maximum value in the response values of the pixel point in each direction as the classical receptive field stimulation response of the pixel point;
the two-dimensional Gabor function expression of the Gabor filter bank is as follows:
whereinGamma is a constant representing the ratio of the long axis to the short axis of the elliptical field, the parameter lambda is the wavelength, sigma (x, y) is the scale function, 1/lambda is the spatial frequency of the cosine function,is a phase angle parameter, theta is a direction parameter of Gabor filtering;
i (x, y) is the gray value of each pixel point of the image to be detected, and is a convolution operator;
the Gabor energy value was calculated as follows:
wherein theta isiFor a certain direction of Gabor filtering,NθThe number of the Gabor filtering directions;
the classical receptive field stimulus response Ec (x, y; sigma (x, y)) is:
D. convolving the classical receptive field stimulation response of each pixel point with the distance weight function of the pixel point, and multiplying the convolved response by the corresponding scale of the pixel point to obtain the inhibition response of each pixel point;
the step D is specifically as follows:
the distance weight function wσ(x, y) is:
wherein | · | purple sweet1Is L1Norm, h (x) max (0, x), DoG (x, y) is an expression corresponding to DoG template;
the suppression response Inh (x, y) of each pixel point is as follows:
Inh(x,y)=Ec(x,y;σ(x,y))*wσ(x,y)σ(x,y) (11);
E. subtracting the product of the suppression response and the suppression strength of each pixel point from the classical receptive field stimulation response of each pixel point to obtain the contour response of the pixel point, and performing non-maximum suppression and double-threshold processing on the contour response to obtain the final contour value of each pixel point so as to obtain a final contour map;
the step E is specifically as follows:
the profile response R (x, y) is:
R(x,y)=H(Ec(x,y;σ(x,y))-αInh(x,y)) (12);
where h (x) max (0, x), α is the inhibition intensity.
The effectiveness of the contour detection method of the present embodiment is compared with the contour detection isotropic model provided in document 1, where document 1 is as follows:
document 1: cosmin Grigorescu, Nicolai Petkov, and MichelA. Westenberg. Contourdetection Based on non-systematic receiving Fieldingdestination [ J ]. IEEE Transactions on image processing, vol.12, No.7, july 2003729-;
to ensure the validity of the comparison, the same non-maximum suppression and double-threshold processing as in document 1 are employed for the present embodiment, including two thresholds th,tlIs set to tl=0.5thCalculated from a threshold quantile p;
the performance evaluation index F of this time adopts the following criteria given in document 2:
reference 2 is "d.r. martin, c.c. fowles, and j.malik," Learning to detected natural image bounding data using local brightness, color, and texture documents, "ieee transactions on pattern analysis and machine interpretation, vol.26, pp.530-549,2004";
in the formula, P represents the accuracy, R represents the coverage rate, and the evaluation index P is between [0 and 1], wherein the closer to 1, the better the contour detection effect is;
selecting 3 classical images shown in FIG. 1 for effectiveness comparison, and performing contour detection on the 3 images by using the method in document 1 and the method in example 1, wherein the parameter set selected by the method in example 1 is shown in Table 1;
table 1 example 1 parameter set table
The method in document 1 uses 80 sets of parameters, α ═ {1.0, 1.2}, σ ═ 1.4, 1.6, 1.8, 2.0, 2.2, 2.4, 2.6, 2.8}, p ═ 0.5, 0.4, 0.3, 0.2, 0.1 };
as shown in fig. 1, the optimal contour detected by the original image, the actual contour image and the document 1 method of 3 pairs of classical images, and the optimal contour detected by the embodiment 1 method; as shown in table 2, the optimal F value detected by the method of document 1 for the 3 images is compared with the optimal F value detected by the method of example 1;
TABLE 2F-value comparison table
As can be seen from the above results, the method of example 1 is superior to the contour detection method in document 1 both in terms of the effect of contour extraction and in terms of the performance index parameter.
Claims (6)
1. A contour detection method based on variable receptive field scale global modulation is characterized by comprising the following steps:
A. inputting a to-be-detected image subjected to gray processing, performing Gaussian difference filtering on the gray value of each pixel point by using a Gaussian difference function to obtain a Gaussian difference filtering value of each pixel point, and respectively performing normalization processing on a positive value and a negative value in the Gaussian difference filtering values of each pixel point to obtain a normalized Gaussian difference filtering function of each pixel point; performing convolution on the normalized Gaussian difference filter function of each pixel point and the gray value of the corresponding pixel point respectively to obtain a normalized Gaussian difference filter value of each pixel point;
B. presetting a high scale value, a low scale value and a threshold value of a scale function, comparing the normalized Gaussian difference filter value of each pixel point with the threshold value respectively, if the normalized Gaussian difference filter value of the pixel point is greater than or equal to the threshold value, the scale function value corresponding to the pixel point is the high scale value, and if the normalized Gaussian difference filter value of the pixel point is less than the threshold value, the scale function value corresponding to the pixel point is the low scale value;
C. presetting inhibition intensity, and presetting a plurality of direction parameters for equally dividing the circumference; b, performing Gabor filtering on each pixel point in the image to be detected according to each direction parameter to obtain a response value of each pixel point in each direction, wherein a scale function value adopted in the Gabor filtering is a high scale value or a low scale value determined by each pixel point in the step B; for each pixel point, selecting the maximum value in the response values of the pixel point in each direction as the classical receptive field stimulation response of the pixel point;
D. convolving the classical receptive field stimulation response of each pixel point with the distance weight function of the pixel point, and multiplying the convolved response by the corresponding scale of the pixel point to obtain the inhibition response of each pixel point;
E. and subtracting the product of the suppression response and the suppression strength of each pixel point from the classical receptive field stimulation response of each pixel point to obtain the contour response of the pixel point, and performing non-maximum suppression and double-threshold processing on the contour response to obtain the final contour value of each pixel point so as to obtain a final contour map.
2. The profile detection method based on variable receptive field scale global modulation according to claim 1, characterized in that:
the step A is specifically as follows:
the Gaussian difference function DoGσ(x, y) is:
wherein σ is an initialization scale;
the normalized Gaussian difference filter function wσ(x, y) is:
the normalized Gaussian difference filtering value woutσ(x, y) is:
woutσ(x,y)=I(x,y)*wσ(x,y) (3)。
3. the profile detection method based on variable receptive field scale global modulation according to claim 2, characterized in that:
the step B is specifically as follows:
the scale function σ (x, y) is:
wherein sigmaHAt a high scale value, σLAt a low scale value, σH=σL+ s, t is a threshold approaching 0, s is the step size of the scale.
4. The profile detection method based on variable receptive field scale global modulation according to claim 3, characterized in that:
the step C is specifically as follows:
the two-dimensional Gabor function expression of the Gabor filter bank is as follows:
whereinGamma is a constant representing the ratio of the long axis to the short axis of the elliptical field, the parameter lambda is the wavelength, sigma (x, y) is the scale function, 1/lambda is the spatial frequency of the cosine function,is a phase angle parameter, theta is a direction parameter of Gabor filtering;
i (x, y) is the gray value of each pixel point of the image to be detected, and is a convolution operator;
the Gabor energy value was calculated as follows:
wherein theta isiFor a certain direction of Gabor filtering, NθThe number of the Gabor filtering directions;
the classical receptive field stimulus response Ec (x, y; sigma (x, y)) is:
5. the profile detection method based on variable receptive field scale global modulation according to claim 4, characterized in that:
the step D is specifically as follows:
the distance weight function wσ(x, y) is:
wherein | · | purple sweet1Is L1Norm, h (x) max (0, x), DoG (x, y) is an expression corresponding to DoG template;
the suppression response Inh (x, y) of each pixel point is as follows:
Inh(x,y)=Ec(x,y;σ(x,y))*wσ(x,y)σ(x,y) (11)。
6. the profile detection method based on variable receptive field scale global modulation according to claim 5, characterized in that:
the step E is specifically as follows:
the profile response R (x, y) is:
R(x,y)=H(Ec(x,y;σ(x,y))-αInh(x,y)) (12);
where h (x) max (0, x), α is the inhibition intensity.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711098829.XA CN107767387B (en) | 2017-11-09 | 2017-11-09 | Contour detection method based on variable receptive field scale global modulation |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201711098829.XA CN107767387B (en) | 2017-11-09 | 2017-11-09 | Contour detection method based on variable receptive field scale global modulation |
Publications (2)
Publication Number | Publication Date |
---|---|
CN107767387A CN107767387A (en) | 2018-03-06 |
CN107767387B true CN107767387B (en) | 2020-05-05 |
Family
ID=61272282
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201711098829.XA Active CN107767387B (en) | 2017-11-09 | 2017-11-09 | Contour detection method based on variable receptive field scale global modulation |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN107767387B (en) |
Families Citing this family (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN109766919B (en) * | 2018-12-18 | 2020-11-10 | 通号通信信息集团有限公司 | Gradual change type classification loss calculation method and system in cascade target detection system |
CN109919945B (en) * | 2019-02-01 | 2022-03-25 | 广西科技大学 | Contour detection method based on non-classical receptive field non-linear two-side subunit response |
CN109949324B (en) * | 2019-02-01 | 2022-04-22 | 广西科技大学 | Contour detection method based on non-classical receptive field nonlinear subunit response |
CN109978898B (en) * | 2019-02-01 | 2023-07-18 | 广西科技大学 | Contour detection method based on vector field energy calculation |
CN111062957B (en) * | 2019-10-28 | 2024-02-09 | 广西科技大学鹿山学院 | Non-classical receptive field contour detection method |
CN111179293B (en) * | 2019-12-30 | 2020-10-02 | 广西科技大学 | Bionic contour detection method based on color and gray level feature fusion |
CN111080663B (en) * | 2019-12-30 | 2020-09-22 | 广西科技大学 | Bionic contour detection method based on dynamic receptive field |
CN111539969B (en) * | 2020-04-23 | 2023-06-09 | 武汉铁路职业技术学院 | Image edge detection method, device, computer equipment and storage medium |
CN111968140B (en) * | 2020-06-24 | 2023-07-14 | 广西科技大学 | Contour detection method based on classical receptive field vision-strengthening micro-motion mechanism |
Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5422962A (en) * | 1992-03-19 | 1995-06-06 | Fujitsu Limited | Method and apparatus for extracting line segments from an image of an object |
US7130484B2 (en) * | 2001-10-15 | 2006-10-31 | Jonas August | Biased curve indicator random field filters for enhancement of contours in images |
CN101673345A (en) * | 2009-07-01 | 2010-03-17 | 北京交通大学 | Method for extracting target closed contour based on shape prior |
CN102034105A (en) * | 2010-12-16 | 2011-04-27 | 电子科技大学 | Object contour detection method for complex scene |
CN103473759A (en) * | 2013-06-24 | 2013-12-25 | 南京理工大学 | Low-light-level image significant contour extraction method of WKPCA homogeneity degree correction nCRF inhibition |
US8971614B2 (en) * | 2012-05-14 | 2015-03-03 | University Of Southern California | Extracting object edges from images |
CN104484667A (en) * | 2014-12-30 | 2015-04-01 | 华中科技大学 | Contour extraction method based on brightness characteristic and contour integrity |
CN106033609A (en) * | 2015-07-24 | 2016-10-19 | 广西科技大学 | Target contour detection method of biomimetic jumping eye movement information processing mechanism |
CN106485724A (en) * | 2016-09-20 | 2017-03-08 | 华中科技大学 | A kind of profile testing method that modulates based on combination receptive field and towards feature |
Family Cites Families (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20020154833A1 (en) * | 2001-03-08 | 2002-10-24 | Christof Koch | Computation of intrinsic perceptual saliency in visual environments, and applications |
JP4208485B2 (en) * | 2001-05-31 | 2009-01-14 | キヤノン株式会社 | Pulse signal processing circuit, parallel processing circuit, pattern recognition device, and image input device |
US20140254922A1 (en) * | 2013-03-11 | 2014-09-11 | Microsoft Corporation | Salient Object Detection in Images via Saliency |
-
2017
- 2017-11-09 CN CN201711098829.XA patent/CN107767387B/en active Active
Patent Citations (9)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US5422962A (en) * | 1992-03-19 | 1995-06-06 | Fujitsu Limited | Method and apparatus for extracting line segments from an image of an object |
US7130484B2 (en) * | 2001-10-15 | 2006-10-31 | Jonas August | Biased curve indicator random field filters for enhancement of contours in images |
CN101673345A (en) * | 2009-07-01 | 2010-03-17 | 北京交通大学 | Method for extracting target closed contour based on shape prior |
CN102034105A (en) * | 2010-12-16 | 2011-04-27 | 电子科技大学 | Object contour detection method for complex scene |
US8971614B2 (en) * | 2012-05-14 | 2015-03-03 | University Of Southern California | Extracting object edges from images |
CN103473759A (en) * | 2013-06-24 | 2013-12-25 | 南京理工大学 | Low-light-level image significant contour extraction method of WKPCA homogeneity degree correction nCRF inhibition |
CN104484667A (en) * | 2014-12-30 | 2015-04-01 | 华中科技大学 | Contour extraction method based on brightness characteristic and contour integrity |
CN106033609A (en) * | 2015-07-24 | 2016-10-19 | 广西科技大学 | Target contour detection method of biomimetic jumping eye movement information processing mechanism |
CN106485724A (en) * | 2016-09-20 | 2017-03-08 | 华中科技大学 | A kind of profile testing method that modulates based on combination receptive field and towards feature |
Non-Patent Citations (6)
Title |
---|
Contour detection operators based on surround inhibition;Cosmin Grigorescu 等;《Proceedings 2003 International Conference on Image Processing》;20031231;第3卷;第III-437-III440页 * |
Improved Contour Detection by Non-classical Receptive Field Inhibition;Cosmin Grigorescu 等;《BMCV 2002》;20021231;第50-59页 * |
Improved contour detection model with spatial summation properties based on nonclassical receptive field;Chuan Lin 等;《Journal of Electronic Imaging》;20160831;第25卷(第4期);第1-10页 * |
Orientation Histogram-Based Center-Surround Interaction: An Integration Approach for Contour Detection;Rongchang Zhao 等;《Neural Computation》;20161130;第29卷(第4期);第1-23页 * |
基于多通道Gabor滤波的指纹图像二值化方法;林川 等;《科学技术与工程》;20130831;第13卷(第22期);第6487-6491页 * |
基于非经典感受野多尺度机制的图像分析方法;许跃颖 等;《信息技术》;20170725(第7期);第5-8页 * |
Also Published As
Publication number | Publication date |
---|---|
CN107767387A (en) | 2018-03-06 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN107767387B (en) | Contour detection method based on variable receptive field scale global modulation | |
CN107657279B (en) | Remote sensing target detection method based on small amount of samples | |
CN105956582B (en) | A kind of face identification system based on three-dimensional data | |
CN109753914B (en) | License plate character recognition method based on deep learning | |
CN109117826B (en) | Multi-feature fusion vehicle identification method | |
CN108022233A (en) | A kind of edge of work extracting method based on modified Canny operators | |
CN103886589B (en) | Object-oriented automated high-precision edge extracting method | |
CN104299008B (en) | Vehicle type classification method based on multi-feature fusion | |
CN105354866A (en) | Polygon contour similarity detection method | |
CN108090492B (en) | Contour detection method based on scale clue suppression | |
CN107886539B (en) | High-precision gear visual detection method in industrial scene | |
CN105447512A (en) | Coarse-fine optical surface defect detection method and coarse-fine optical surface defect detection device | |
CN103020582A (en) | Method for computer to identify vehicle type by video image | |
CN103413119A (en) | Single sample face recognition method based on face sparse descriptors | |
CN106033608B (en) | The objective contour detection method of bionical object smooth pursuit eye movement information processing mechanism | |
CN108710909B (en) | Counting method for deformable, rotary and invariant boxed objects | |
CN103425998A (en) | Method for identifying SAR target under shielding conditions | |
CN110222661B (en) | Feature extraction method for moving target identification and tracking | |
CN112488211A (en) | Fabric image flaw classification method | |
CN104298995A (en) | Three-dimensional face identification device and method based on three-dimensional point cloud | |
CN110232390B (en) | Method for extracting image features under changed illumination | |
CN107742302B (en) | Contour detection method based on primary visual cortex multi-scale contour fusion | |
CN110991547A (en) | Image significance detection method based on multi-feature optimal fusion | |
CN107180422A (en) | A kind of labeling damage testing method based on bag of words feature | |
CN102999909A (en) | Synthetic aperture radar (SAR) target detection method based on improved visual attention model |
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 | ||
EE01 | Entry into force of recordation of patent licensing contract | ||
EE01 | Entry into force of recordation of patent licensing contract |
Application publication date: 20180306 Assignee: GUANGXI YINGTENG EDUCATION TECHNOLOGY Co.,Ltd. Assignor: GUANGXI University OF SCIENCE AND TECHNOLOGY Contract record no.: X2023980053979 Denomination of invention: A contour detection method based on variable receptive field scale global modulation Granted publication date: 20200505 License type: Common License Record date: 20231226 |