CN111145117B - Spot detection method and system based on soft mathematical morphology operator - Google Patents
Spot detection method and system based on soft mathematical morphology operator Download PDFInfo
- Publication number
- CN111145117B CN111145117B CN201911347702.6A CN201911347702A CN111145117B CN 111145117 B CN111145117 B CN 111145117B CN 201911347702 A CN201911347702 A CN 201911347702A CN 111145117 B CN111145117 B CN 111145117B
- Authority
- CN
- China
- Prior art keywords
- smq
- image
- filter
- speckle
- spot
- 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 42
- 238000001914 filtration Methods 0.000 claims abstract description 32
- 238000012545 processing Methods 0.000 claims abstract description 27
- 238000012216 screening Methods 0.000 claims abstract description 10
- 238000000034 method Methods 0.000 claims description 23
- 230000008569 process Effects 0.000 claims description 4
- 238000002474 experimental method Methods 0.000 abstract description 5
- 230000000877 morphologic effect Effects 0.000 description 11
- 230000004044 response Effects 0.000 description 9
- 238000013461 design Methods 0.000 description 6
- 230000004660 morphological change Effects 0.000 description 4
- 238000002059 diagnostic imaging Methods 0.000 description 3
- 240000009088 Fragaria x ananassa Species 0.000 description 2
- 230000008859 change Effects 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 208000020816 lung neoplasm Diseases 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000003909 pattern recognition Methods 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- PXFBZOLANLWPMH-UHFFFAOYSA-N 16-Epiaffinine Natural products C1C(C2=CC=CC=C2N2)=C2C(=O)CC2C(=CC)CN(C)C1C2CO PXFBZOLANLWPMH-UHFFFAOYSA-N 0.000 description 1
- 241001164374 Calyx Species 0.000 description 1
- 235000002566 Capsicum Nutrition 0.000 description 1
- 235000016623 Fragaria vesca Nutrition 0.000 description 1
- 235000011363 Fragaria x ananassa Nutrition 0.000 description 1
- 235000019687 Lamb Nutrition 0.000 description 1
- 206010058467 Lung neoplasm malignant Diseases 0.000 description 1
- 239000006002 Pepper Substances 0.000 description 1
- 235000016761 Piper aduncum Nutrition 0.000 description 1
- 235000017804 Piper guineense Nutrition 0.000 description 1
- 244000203593 Piper nigrum Species 0.000 description 1
- 235000008184 Piper nigrum Nutrition 0.000 description 1
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 238000013459 approach Methods 0.000 description 1
- 238000000339 bright-field microscopy Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 230000000593 degrading effect Effects 0.000 description 1
- 230000010339 dilation Effects 0.000 description 1
- 239000012634 fragment Substances 0.000 description 1
- 230000004807 localization Effects 0.000 description 1
- 201000005202 lung cancer Diseases 0.000 description 1
- 239000011159 matrix material Substances 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008447 perception Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 150000003839 salts Chemical class 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 235000013570 smoothie Nutrition 0.000 description 1
- 235000021012 strawberries Nutrition 0.000 description 1
- 238000012706 support-vector machine Methods 0.000 description 1
- 230000002123 temporal effect Effects 0.000 description 1
- 230000009466 transformation Effects 0.000 description 1
- 230000001131 transforming effect Effects 0.000 description 1
- 239000009131 zhiling Substances 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
- G06T7/155—Segmentation; Edge detection involving morphological operators
-
- 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
- G06T7/00—Image analysis
- G06T7/60—Analysis of geometric attributes
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Geometry (AREA)
- Image Processing (AREA)
- Image Analysis (AREA)
Abstract
The invention provides a spot detection method based on a soft mathematical morphology filter, which comprises the following steps: step 1: inputting a gray image; step 2: setting parameters of the filter; step 3: performing SMQ_A filtering processing on an input image; step 4: obtaining candidate spot images; step 5: screening candidate spots; step 6: performing SMQ_B filtering processing on the candidate spot images according to the mask image; step 7: obtaining a filtered image of smq_b; step 8: the spot shape and location are obtained from the SMQ B filtered image. The invention verifies that the SMBD has better anti-noise and shape restoration capability than similar speckle filters Quoit and N-Quoit based on mathematical morphology from both theoretical and experimental aspects. In addition, the invention also makes a comparison experiment with the most commonly used Log and DoH spot detectors of different types, and SMBD also obtains a remarkable result. The invention provides a spot detection system based on a soft mathematical morphology filter.
Description
Technical Field
The invention belongs to the technical field of image processing and computer vision, and particularly relates to mathematical morphology filtering, in particular to a spot detection method and system based on a soft mathematical morphology operator.
Background
Speckle generally refers to a region that is colored or gray-scale differently from the surrounding area. As the spot represents a region, compared with a simple corner point, the spot has better stability and stronger noise resistance. Speckle detection is an important technique for computer vision, which finds application in many fields: medical image processing [1] [2] [13] [14], video monitoring [15] [16], robots [17] [18] [19], aviation systems [20] [21] [22] and the like.
LoG (Laplacian of Gaussian) [4] is a commonly used spot detector which is relatively efficient for the detection of radially symmetric spots, but which has a relatively high time complexity. To reduce the temporal complexity of LoG, document [5] proposes a DoG (Difference of Gaussian) spot detector that replaces the gaussian laplace operation of LoG with a gaussian differential operation approximation. In order to improve the detection performance of LoG on radially asymmetric spots, document [6] proposes a gLoG (Generalized Laplacian of Gaussian) spot detector in which the gclog extends the detection to rotationally asymmetric structures using different gaussian kernels, thereby enabling detection of spots of rotationally symmetric and asymmetric generally elliptical structures. In addition, miao [7] et al propose a ROLG (Rank Order Laplacian of Gaussian) -based speckle detection algorithm that makes the filter robust to noise by setting a rank ordering operation.
Another relatively common spot detector is DoH (Determinant of Hessian) [4], which uses a Hessian matrix whose determinant is used to detect spots on a single scale. In order to extract smaller-scale spots from images, two spot detectors, HLoG (Hessian-based Laplacian of Gaussian) [8] and HDoG (Hessian-based Difference of Gaussian) [9], were proposed on the basis of DoH, respectively.
The above-mentioned speckle detection algorithm is not affine-invariant, so that affine-co-ordinated-invariant speckle detection algorithm is studied, wherein the more influential study is that matas [10] proposes a feature detection method of the maximum stable extremum region (Maximally Stable Extremal Regions, MSERs). The method uses the thought of watershed to detect the local area with the most stable gray level in the image, then rotates and normalizes the detection area in size, and finally the obtained local feature has affine invariance in strict sense. Rather than finding blobs that meet a specified size, the MSER is to find the blob area that meets the measure of "minimum area rate of change".
In addition to the 3 classical spot detectors of the Log, doH and MSERs, and variants thereof, there are other types of spot detection algorithms [11] [12].
Document [11] proposes a blob detector SBD (with scale invariance) and a local feature descriptor SLD based on Shearlet (Shearlet is a multi-scale framework with orientation). After transforming the image to the Shearlet domain, the SBD has scale invariant properties by finding local extremal points in the 3 x 3 domain of the neighboring scale space. That is, SBD, although denominated as a spot detector, is not actually used to detect spots in the image domain that fit a specified size, but rather to find local extremal points in the Shearlet transform domain.
Reference [12] proposes a RIFF feature detection algorithm based on radial gradient transformation, which has rotational invariance. The feature point locations are obtained by DOB (Difference of Boxes), which is to find local extremal points in the 3 x 3 domain of the neighboring scale space. That is, RIFF is a feature descriptor, and feature point detection is performed by using the DOB algorithm, which is known as a spot detector, but actually detects extreme points of a local area, not for detecting spots conforming to a specified size in an image field.
In addition to the above-described spot detectors, there is a class of spot detection filters based on mathematical morphology. Document [1] proposes a choice filter that has a relatively good shape detection capability for radially symmetric spots. Since the time complexity of the choice filter is relatively high, document [2] proposes an N-choice filter, which increases the computation speed. But these two algorithms have two major weaknesses: one is weak against noise, and the other is that the shape of the spot is not well detected, although the spot can be located.
Disclosure of Invention
The invention firstly promotes a standard mathematical morphology condition operator to a soft morphology order condition operator, designs 3 soft morphology filters by using the operator, and further provides a novel spot detection method based on the operator, which comprises the following steps:
step 1: inputting a gray image;
step 2: setting parameters of the filter;
step 3: performing SMQ_A filtering processing on an input image;
step 4: obtaining candidate spot images;
step 5: screening candidate spots;
step 6: performing SMQ_B filtering processing on the candidate spot images according to the mask image;
step 7: obtaining a filtered image of smq_b;
step 8: the spot shape and location are obtained from the SMQ B filtered image.
The filter in the step 2 of the invention comprises an SMQ_A filter and an SMQ_B filter:
wherein k is a ，k b ＝1，2，...，|D|/2；k c ，k d ＝1，2，...，|R|/2；k a ≤k b ，k c ≤k d ；
Where f is the filtering object image, k m ，k n ＝1，...，|R|/2，k m ≤k n R is a circular ring structural element, D is a circular disk structural element, (u, v) epsilon D (x, y),
and D and R are concentric.
The step 2 of the invention specifically comprises the following steps:
step 2.1: setting parameters DA, R, k of SMQ_A filter a 、k b 、k c 、k d The method comprises the steps of carrying out a first treatment on the surface of the DA is a disk structural element of the SMQ_A filter;
step 2.2: setting parameters DB, R, k of SMQ_B filter m 、k n The method comprises the steps of carrying out a first treatment on the surface of the DB is a disk structure element of the SMQ_B filter;
step 2.3: setting a speckle positioning threshold Th A1 And a speckle region threshold Th A2 For controlling the number of blob candidates.
In step 3 of the present invention, the smq_a filter is used to perform a speckle filtering on the input image I (x, y) to obtain a anchor point candidate image L (x, y):
L(x，y)＝SMQ_A(D A ，R，k a ，k b ，k c ，k d ，I，x，y)。
in step 5 of the present invention, mask image M (x, y) is generated:
ifL(x，y)＞Th A1 M(x，y)＝1
else M(x，y)＝0。
in step 6 of the present invention, set I 2 (x, y) is the image filtered by smq_b; initializing I with 0 2 (x, y) and then traversing each of the images I (x, y)The effective pixel points (x, y) are subjected to the following filtering process by SMQ_B; v (u, v) is the filtered value of SMQ_B at (u, v) ε DB (x, y);
in step 8 of the present invention, the spot shape is extracted:
if I 2 (u，v)≥Th A2 ，I 3 (u，v)＝I(u，v)
else I 3 (u，v)＝0
wherein I is 3 (x, y) is the detected speckle image.
Based on the method, the invention also provides a spot detection system based on the soft mathematical morphology filter, which comprises the following steps:
an input module for inputting a gray image;
a setting module for setting parameters of the filter;
a first processing module for performing smq_a filtering processing on an input image;
a candidate image acquisition module for acquiring a candidate speckle image;
a screening module for screening candidate spots;
a second processing module for performing SMQ_B filtering processing on the candidate spot images according to the mask image;
a filtered image acquisition module for acquiring a filtered image of smq_b;
an output module for obtaining the spot shape and location from the smq_b filtered image.
The invention provides a soft-form spot detection method (called SMBD for short) for improving the defects of the quick-response type spot filter based on the research of the quick-response type spot filter and the N-quick-response type spot filter based on mathematical morphology. According to the invention, a basic operator of standard mathematical morphology is firstly promoted to an order condition of soft morphology, and then, two soft morphology filters SMQ_A and SMQ_B are constructed by using the order condition operator, wherein the former is used for positioning the spot, and the latter is used for detecting the shape of the spot. Finally, a complete speckle detection method SMBD is designed using smq_a and smq_b filters. In order to verify the performance of the SMBD, the invention uses the method of the invention to make a comparison experiment with the 4 methods of the Quoit, N-Quoit, log and DoH on a spot image library, and the experiment shows that the SMBD is obviously better than the other 4 methods in the aspects of noise resistance and spot shape detection.
Reference to the literature
[1].Yamamoto，Shinji，et al.″Quoit Filter-A new filter based on mathematical morphology to extract the isolated shadow.and its application to automatic detection of lung cancer in x-ray CT.″Pattern Recognition，1996.，Proceedings of the 13th International Conference on.Vol.2.IEEE，1996.
[2].Miwa，Tomoko，et al.″Automatic Detection of Lung Cancers in Chest CT Images by the Variable N-Quoit Filter.″Systems and Computers in Japan 33.1(2002)：53-63.
[3].Lasse Koskinen，Jaakko T.Astola，and Yrjo A.Neuvo.″Soft morphological filters.″Image Algebra and Morphological Image Processing ll.Vol.1568.International Society for Optics and Photonics，1991.
[4].Lindeberg，Tony.″Feature detection with automatic scale selection.″International journal of computer vision 30.2(1998)：79-116.
[5].Hay，Geoffrey J.，et al.″A scale-space primer for exploring and quantifying complex landscapes.″Ecological Modelling 153.1-2(2002)：27-49.
[6].Kong，Hui，Hatice Cinar Akakin，and Sanjay E.Sarma.″A generalized Laplacian of Gaussian filter for blob detection and its applications.″IEEE transactions on cybemetics 43.6(2013)：1719-1733.
[7].Miao，Zhenwei，and Xudong Jiang.″Interest point detection using rank order LoG filter.″Pattern Recognition 46.11(2013)：2890-2901.
[8].Zhang，Min，Teresa Wu，and Kevin M.Bennett.″Small blob identification in medical images using regioRal features from optimum scale.″IEEE transactions on biomedical engineering62.4(2015)：1051-1062.
[9].Zhang，Min，et al.″Efficient small blob detection based on local convexity，intensity and shape information.″IEEE transactions on medical imaging 35.4(2016)：1127-1137.
[10].Matas，Jiri，et al.″Robust wide-baseline stereo from maximally stable extremal regions.″Image and vision computing 22.10(2004)：761-767.
[11].Duval-Poo，Miguel A.，et al.″Scale invariant and noise robust interest points with shearlets.″IEEE Transactions on Image Processing 26.6(2017)：2853-2867.
[12].Takacs，Gabriel，et al.″Fast computation of rotation-invariant image features by an approximate radial gradient transform.″IEEE Transactions on Image Processing 22.8(2013)：2970-2982.
[13].Kurzendorfer，Tanja，et al.″Cryo-Balloon Catheter Localization Based on a Support-Vector-Machine Approach.″IEEE transactions on medical imaging 35.8(2016)：1892-1902.
[14].Zhang，Min，et al.″Efricient small blob detection based on local convexity，intensity and shape information.″IEEE transactions on medicalimaging 35.4(2016)：1127-1137.
[15].Zhang，Guyue，et al.″Physical blob detector and Multi-Channel Color Shape Descriptor for human detection.″Journal of Visual Communication and Image Representation52(2018)：13-23.
[16].Afsar，Palwasha，Paulo Cortez，and Henrique Santos.″Automatic human trajectory destination prediction from video.″Expert Systems with Applications 110(2018)：41-51.
[17].Hu，Yuting，Zhiling Long，and Ghassan AlRegib.″A High-Speed，Real-Time Vision System for Texture Tracking and Thread Counting.″IEEE Signal Processing Letters 25.6(2018)：758-762.
[18].Durand-Petiteville，A.，S.Vougioukas，and D.C.Slaughter.″Real-time segmentation of strawberry flesh and calyx from images of singulated strawberries during postharvest processing.″Computers and Electronics in Agriculfure 142(2017)：298-313.
[19].Rajasekaran，Keshav，et al.″An accurate perception method for Iow contrast bright field microscopy in heterogeneous microenvironments.″Applied Sciences 7.12(2017)：1327.
[20].Ewald，Vincentius，Roger M.Groves，and Rinze Benedictus.″Transducer Placement Option of Lamb Wave SHM System for Hotspot Damage Monitoring.″Aerospace 5.2(2018)：39.
[21].Xin，Long，Delin Luo，and Han Li.″A monocular visual measurement system for UAV probe-and-drogue autonomous aerial refueling.″International Journal of Intelligent Computing and Cybernetics just-accepted(2018)：00-00.
[22].Du，Yuhan，Yicheng Jiang，and Wei Zhou.″An Accurate Two-Step ISAR Cross-Range Scaling Method for Earth-Orbit Target.″IEEE Geoscience and Remote Sensing Letters 14.11(2017)：1893-1897.
Drawings
FIG. 1 is a flow chart of the spot detection method SMBD of the present invention.
Fig. 2 is an input image.
FIG. 3 is a candidate blob anchor point image obtained after SMQ_A filtering.
Fig. 4 is an image obtained by binarizing the image of fig. 3.
FIG. 5 is a morphological filtered image of a blob at a candidate anchor point by SMQ_B.
Fig. 6 is a final obtained spot shape image.
Detailed Description
The invention will be described in further detail with reference to the following specific examples and drawings. The procedures, conditions, experimental methods, etc. for carrying out the present invention are common knowledge and common knowledge in the art, except for the following specific references, and the present invention is not particularly limited.
According to the invention, a standard mathematical morphology condition operator is promoted to a soft morphology order condition operator, 3 soft morphology filters are designed by using the operator, and a novel spot detection method SMBD is further provided on the basis. These contents are described below, respectively.
(1) Design of order condition operator of soft morphology
Koskinen et al [3] promoted the standard mathematical morphology's position operator to soft position, which is mainly characterized by being less sensitive to additive noise and to shape changes that require filtering. The definition of soft condition is as follows:
the soft morphological operators are defined on the basis of ordering statistics and they use the structural system [ B, A, k ]]Substituting structural elements of standard mathematical morphology. The structural system contains 3 parameters: a finite set of planes a and B,
one meets the following requirements
Is the total number of elements in the set. B is called the structure set, A is the hard center of B, B-A gives its soft contour, and k is its sequence number. The symbol indicates a repetition operation, defined as
For example, {2×1,2,3×3} = {1, 2, 3). Obviously, when a=Φ and k=1, the soft condition is degraded to the standarddilation。
In order to design a soft morphology query filter, a sequential morphology condition operator is defined on the basis of a soft condition operator.
Definition of order condition:
here, B is a structural element, k is a natural number satisfying 1.ltoreq.k.ltoreq.B|/2, and the order numbers are ordered from large to small. Equation (2) is in essence to change the hard center a in the structural system [ B, a, k ] in equation (1) to the empty set Φ, thereby degrading the structural system to [ B, k ]. This means that the original morphological dilation of the image f with the soft contours (B-ase:Sub>A) and the hard center ase:Sub>A becomes shape dilated with only the soft shapes of the structural elements B.
(2) Design of 3 soft morphology Quoint filters
The following 3 soft morphology choice filters are defined first using the order condition operator.
(a) Definition of SMQ filter:
(b) Definition of smq_a filter:
here, k a ，k b ＝1，2，...，|D|/2；k c ，k d ＝1，2，...，|R|/2；k a ≤k b ，k c ≤k d 。
(c) Definition of smq_b filter:
here, f isFiltering object image, k m ，k n ＝1，...，|R|/2，k m ≤k n R is a circular ring structural element, D is a circular disk structural element, (u, v) epsilon D (x, y),
and D and R are concentric.
Next, the operation principle of the above 3 soft-form choice filters will be described.
■ Working principle of SMQ filter
Intuitively, the choice filter is to first move concentric discs and rings to the object point p, and then find the maximum V in the disc field D And maximum value V in the field of circles R Through V p ＝V D -V R To determine whether the object point P belongs to the blob area. The SMQ filter is similar to the choice filter except that it does not find the maximum but rather finds the respective assigned sequence values in the disk and ring fields, respectively, thus effectively avoiding the effects of noise and morphological changes. K of SMQ filter D And k R When set to 1, it is degraded to a choice filter, and if D is set to one element, it is degraded to an N-choice filter. By adjusting order bits km and kn, the robustness of the SMQ to noise and morphological changes can be adjusted. The SMQ filter is mainly used for spot positioning.
■ Working principle of SMQ_A filter
As can be seen from the above description of the principles of SMQ, SMQ is a morphological filter that specifies order bits, and smq_a is an extension of SMQ. When the SMQ_A is used for morphological filtering of the spots by using the disc and the ring structural elements, the SMQ_A is not based on the gray value corresponding to a certain sequence bit, but based on the average value of the gray values corresponding to a certain segment of sequence bit. Thus, in theory smq_a has better shape robustness than SMQ based on only a certain order bit. The current bit k a ＝k b And k b ＝k d When smq_a is degraded to SMQ. The smq_a filter is mainly used for spot positioning.
■ Working principle of SMQ_B filter
SMQ (smoothie-rich fragment)Smq_a is only to locate the spot and cannot detect the actual shape of the spot because only the morphological changes of the spot are considered when morphological filtering the object spot, and not the morphological changes of other spots in the disc field. tosolvethisproblem,thesmq_aismodifiedtodecomposetheoriginaldiskstructuralelementDintostructuralelementswithonlyasingleelement(thedilatonprocessingbasedonthestructuralelementsofasingleelementistheimageitself),andtheSQM-aisappliedtousesuchstructuralelementstoperformmorphologicalfilteringoneachpoint(u,v)inthediskfield(butalwayskeeprfilteringonlyontheobjectpoints(x,y)). In other words, the SQM-B filter calculates the difference between each point (u, v) and (x, y) based on the R soft morphology filter response value (right part of equation 12) in the range covered by D with the point (x, y) as the center of the circle when filtering the object point (x, y), so that the smq_b can detect the shape of the region covered by D at the object point (x, y) while performing morphology filtering on the object point (x, y). With the structural element R in the form of a ring, in the order k m ～k n The soft morphology filter process (right part of equation 12) is performed on the point (x, y) in the range, which essentially changes the filtering effect of the filter on the spot morphology at that point by adjusting the filter response value of that point. Since it is in sequence bit k m ～k n The mean value between these values is used as the morphological filter response value, so that the morphological filter based on the single sequence is more stable in theory than that based on the single sequence.
(3) Design of spot detection algorithm SMBD
SMQ is a soft form condition operator-based Quot filter designed by the invention, which comprises a Quot filter and an N-Quot filter, and SMQ_A and SMQ_B are extensions to the SMQ. Next, a bright spot detection algorithm SMBD based on smq_a and smq_b filters is given, and if dark spots are to be detected, the gray scale of the input image is simply inverted.
SMBD algorithm
Step 1: setting filter parameters according to Table 1
(1) Setting parameters DA, R, k of SMQ_A filter a 、k b 、k c 、k d Where DA is the disk junction of the SMQ_A filterConstituent elements.
(2) Setting parameters DB, R, k of SMQ_B filter m 、k n Here DB is a disk structural element of the smq_b filter.
(3) Setting a speckle positioning threshold Th A1 And a speckle region threshold Th A2 These two thresholds are mainly used to control the number of blob candidates.
Table 1: estimation of SMBD parameters
Step 2: SMQ_A filtering
The input image I (x, y) is subjected to speckle filtering with an smq_a filter to obtain a anchor point candidate image L (x, y):
L(x，y)＝SMQ_A(D A ，R，k a ，k b ，k c ，k d ，I，x，y)
step 3: generating mask image M (x, y)
if L(x，y)＞Th A1 M(x，y)＝1
else M(x，y)＝0
Step 4: SMQ_B filtering
Set I 2 (x, y) is the image filtered by SMQ_B. Initializing I with 0 2 (x, y), then each effective pixel point (x, y) in the image I (x, y) is traversed, and the following filtering process is performed with smq_b. Here, v (u, v) is the filtered value of smq_b at (u, v) ∈db (x, y).
Step 5: extracting the spot shape
if I 2 (u，v)≥Th A2 ，I 3 (u，v)＝I(u，v)
elseI 3 (u，v)＝0
Here I 3 (x, y) is the detected speckle image.
The invention also provides a spot detection system based on the soft mathematical morphology filter, which comprises the following steps:
an input module for inputting a gray image;
a setting module for setting parameters of the filter;
a first processing module for performing smq_a filtering processing on an input image;
a candidate image acquisition module for acquiring a candidate speckle image;
a screening module for screening candidate spots;
a second processing module for performing SMQ_B filtering processing on the candidate spot images according to the mask image;
a filtered image acquisition module for acquiring a filtered image of smq_b;
an output module for obtaining the spot shape and location from the smq_b filtered image.
Examples
The design principles of the present invention will be further described with reference to the drawings and specific examples below:
step 1: setting all parameters in the SMBD method
Image 2 is an image obtained by superimposing 5% salt and pepper noise on the original image. The object of the present invention is to detect spots of radius 26 pixels in rows 3 to 4 of image 2, the following parameter values being given according to the parameter estimation algorithm of table 1: r is (r) A1 ＝5，r A2 ＝28，r A3 ＝9，r A4 ＝26，
k d ＝3k c ＝27，k m ＝k c ＝9，k n ＝k d ＝27，Th A1 ＝60，Th A2 ＝10。
Step 2: the smq_a filter is used to locate spots in the input image that are smaller than the radius of the given circle, and the radius of the disc can be made smaller to increase the processing speed. Since the computation of directly performing the spot filtering with SMQ_B is very large, the method uses SMQ_A to determine the approximate location of the central region of the spot. FIG. 3 is a candidate blob anchor point obtained after SMQ_A filtering, from which it can be seen that the larger the blob area, the smaller its corresponding anchor point region.
Step 3: and screening the candidate positioning points of the spots to generate a mask image.
And 4, taking a point with the median value of 1 in the mask image as the center of the SMQ_B filter, and then performing SMQ_B filtering processing on the image to obtain a filter response value image of the SMQ_B, namely the detected spot shape response value. FIG. 4 is a graph of threshold Th A1 The binarized image of FIG. 3 is processed to reject those filtered values that are less than Th A1 Is a false candidate anchor point. FIG. 5 is a morphological filtered image of a blob at a candidate anchor point by SMQ_B.
Step 5: from the filtered image of smq_b, the shape of the blob whose filter response value is greater than the threshold ThA2 is extracted from image 2. FIG. 6 is a graph of extracting from an original image that the speckle morphology filter value is greater than the threshold Th A2 Is a true spot shape of (c).
The invention verifies that the SMBD has better anti-noise and shape restoration capability than similar speckle filters Quoit and N-Quoit based on mathematical morphology from both theoretical and experimental aspects. In addition, the invention also makes a comparison experiment with the most commonly used Log and DoH spot detectors of different types, and SMBD also obtains a remarkable result.
The embodiments described herein are presented to aid the reader in understanding the principles of the invention and should be understood to be not limited to such specific statements and embodiments. All possible equivalents or modifications from the above description are considered to be within the scope of the claims.
The protection of the present invention is not limited to the above embodiments. Variations and advantages that would occur to one skilled in the art are included in the invention without departing from the spirit and scope of the inventive concept, and the scope of the invention is defined by the appended claims.
Claims (7)
1. The spot detection method based on the soft mathematical morphology filter is characterized by comprising the following steps of:
step 1: inputting a gray image;
step 2: setting parameters of the filter; the filter in step 2 comprises an smq_a filter and an smq_b filter:
wherein k is a ,k b ＝1,2，...,|D|/2；k c ,k d ＝1,2，...,|R|/2；k a ≤k b ,k c ≤k d ；
Where f is the filtering object image, k m ,k n ＝1,...,|R|/2,；k m ≤k n R is a circular ring structural element, D is a circular disk structural element, (u, v) epsilon D (x, y),
and D and R are concentric;
step 3: performing SMQ_A filtering processing on an input image;
step 4: obtaining candidate spot images;
step 5: screening candidate spots;
step 6: performing SMQ_B filtering processing on the candidate spot images according to the mask image;
step 7: obtaining a filtered image of smq_b;
step 8: the spot shape and location are obtained from the SMQ B filtered image.
2. The method for detecting speckle based on soft mathematical morphology filter of claim 1, wherein step 2 specifically comprises:
step 2.1: setting parameters DA, R, k of SMQ_A filter a 、k b 、k c 、k d The method comprises the steps of carrying out a first treatment on the surface of the DA is a disk structural element of the SMQ_A filter;
step 2.2: setting parameters DB, R, k of SMQ_B filter m 、k n The method comprises the steps of carrying out a first treatment on the surface of the DB is a disk structure element of the SMQ_B filter;
step 2.3: setting a speckle positioning threshold Th A1 And a speckle region threshold Th A2 For controlling the number of blob candidates.
3. The soft mathematical morphology filter-based speckle detection method of claim 2, wherein in step 3, the input image I (x, y) is speckle filtered with a smq_a filter to obtain a anchor point candidate image L (x, y):
L(x,y)＝SMQ-A(D A ,R,k a ,k b ,k c ,k d ,I,x,y)。
4. the method for detecting speckle based on soft mathematical morphology filter of claim 2, wherein in step 5, mask image M (x, y) is generated:
if L(x，y)＞Th A1 M(x，y)＝1
else M(x，y)＝0。
5. the method for detecting speckle based on soft mathematical morphology filter of claim 1, wherein in step 6, I is set to 2 (x, y) is the image filtered by smq_b; initializing I with 0 2 (x, y), then traversing each effective pixel point (x, y) in the image I (x, y), and performing the following filtering process with smq_b; v (u, v) is the filtered value of SMQ_B at (u, v) ε DB (x, y);
If M(x，y)＝1，
do{v(u，v)＝SMQ_B(D B ，R，k m ，k n ，I，x，y，u，v)
if v(u，v)＞I 2 (u，v)，I 2 (u，v)＝v(u，v)
}。
6. the method for detecting a speckle based on a soft mathematical morphology filter of claim 2, wherein in step 8, the speckle shape is extracted:
if I 2 (u，v)≥Th A2 ，I 3 (u，v)＝I(u，v)
else I 3 (u，v)＝0
wherein I is 3 (x, y) is the detected speckle image.
7. A speckle detection system based on a soft mathematical morphology filter, characterized in that a detection method according to any one of claims 1-6 is used, said system comprising the steps of:
an input module for inputting a gray image;
a setting module for setting parameters of the filter;
a first processing module for performing smq_a filtering processing on an input image;
a candidate image acquisition module for acquiring a candidate speckle image;
a screening module for screening candidate spots;
a second processing module for performing SMQ_B filtering processing on the candidate spot images according to the mask image;
a filtered image acquisition module for acquiring a filtered image of smq_b;
an output module for obtaining the spot shape and location from the smq_b filtered image.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911347702.6A CN111145117B (en) | 2019-12-24 | 2019-12-24 | Spot detection method and system based on soft mathematical morphology operator |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201911347702.6A CN111145117B (en) | 2019-12-24 | 2019-12-24 | Spot detection method and system based on soft mathematical morphology operator |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN111145117A CN111145117A (en) | 2020-05-12 |
| CN111145117B true CN111145117B (en) | 2023-06-09 |
Family
ID=70519625
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201911347702.6A Active CN111145117B (en) | 2019-12-24 | 2019-12-24 | Spot detection method and system based on soft mathematical morphology operator |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111145117B (en) |
Families Citing this family (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113450331B (en) * | 2021-06-30 | 2022-08-23 | 青岛卓信通智能科技有限公司 | Special-shaped component pin detection method |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107292869A (en) * | 2017-06-07 | 2017-10-24 | 西安电子科技大学 | Image Speckle detection method based on anisotropic Gaussian core and gradient search |
| CN110135312A (en) * | 2019-05-06 | 2019-08-16 | 电子科技大学 | A kind of quick small target detecting method based on classification LCM |
Family Cites Families (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8164039B2 (en) * | 2008-04-30 | 2012-04-24 | Board Of Regents, The University Of Texas System | Method and apparatus for detecting spiculated masses in mammography |
-
2019
- 2019-12-24 CN CN201911347702.6A patent/CN111145117B/en active Active
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107292869A (en) * | 2017-06-07 | 2017-10-24 | 西安电子科技大学 | Image Speckle detection method based on anisotropic Gaussian core and gradient search |
| CN110135312A (en) * | 2019-05-06 | 2019-08-16 | 电子科技大学 | A kind of quick small target detecting method based on classification LCM |
Non-Patent Citations (1)
| Title |
|---|
| 基于序位滤波的斑点形态滤波器;张艳勤;童卫青;何荣炜;计算机系统应用(第009期);全文 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN111145117A (en) | 2020-05-12 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Arslan et al. | A color and shape based algorithm for segmentation of white blood cells in peripheral blood and bone marrow images | |
| US20170017667A1 (en) | Vehicle inspection methods and systems | |
| US9454817B2 (en) | Relating to image processing | |
| Panda et al. | New binary Hausdorff symmetry measure based seeded region growing for retinal vessel segmentation | |
| Plissiti et al. | A review of automated techniques for cervical cell image analysis and classification | |
| Zhang et al. | Robust and fast vessel segmentation via Gaussian derivatives in orientation scores | |
| Ravichandran et al. | A fast enhancement/thresholding based blood vessel segmentation for retinal image using contrast limited adaptive histogram equalization | |
| Sran et al. | Integrating saliency with fuzzy thresholding for brain tumor extraction in MR images | |
| CN111145117B (en) | Spot detection method and system based on soft mathematical morphology operator | |
| Choudhary et al. | Curvelet and fast marching method‐based technique for efficient artifact detection and removal in dermoscopic images | |
| CN114581658A (en) | Target detection method and device based on computer vision | |
| Hashim et al. | Optic disc boundary detection from digital fundus images | |
| Gwetu et al. | Segmentation of retinal blood vessels using normalized Gabor filters and automatic thresholding | |
| Khan et al. | A fast and accurate Iris segmentation method using an LoG filter and its zero-crossings | |
| Ali et al. | Segmenting retinal blood vessels with Gabor filter and automatic binarization | |
| Soares et al. | Microaneurysms detection using a novel neighborhood analysis | |
| Králik et al. | Canny edge detector algorithm optimization using 2D spatial separable convolution | |
| Tong et al. | Blob detection based on soft morphological filter | |
| Qu et al. | Contour detection based on SUSAN principle and surround suppression | |
| Wen et al. | Feature-level image fusion for SAR and optical images | |
| Guo et al. | Robust vessel detection and segmentation in ultrasound images by a data-driven approach | |
| Youmaran et al. | Improved identification of iris and eyelash features | |
| Srikrishna et al. | Realization of Human Eye Pupil Detection System using Canny Edge Detector and Circular Hough Transform Technique | |
| Orlando et al. | Reviewing Preprocessing and Feature Extraction Techniques for Retinal Blood Vessels Segmentation in Fundus Images | |
| Agrahari et al. | Automatic Polyp Detection using DSC Edge Detector and HOG Features. |
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 |