CN104881856A - Method and apparatus for detecting regular shapes in images - Google Patents
Method and apparatus for detecting regular shapes in images Download PDFInfo
- Publication number
- CN104881856A CN104881856A CN201410069286.9A CN201410069286A CN104881856A CN 104881856 A CN104881856 A CN 104881856A CN 201410069286 A CN201410069286 A CN 201410069286A CN 104881856 A CN104881856 A CN 104881856A
- Authority
- CN
- China
- Prior art keywords
- shape
- regular shape
- pixel
- image
- regular
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title abstract description 55
- 238000001514 detection method Methods 0.000 claims description 75
- 238000009827 uniform distribution Methods 0.000 claims description 2
- 238000010586 diagram Methods 0.000 description 43
- 230000008569 process Effects 0.000 description 24
- 230000006870 function Effects 0.000 description 15
- 238000012360 testing method Methods 0.000 description 13
- 238000004590 computer program Methods 0.000 description 10
- 238000005516 engineering process Methods 0.000 description 8
- 238000001914 filtration Methods 0.000 description 4
- 230000002093 peripheral effect Effects 0.000 description 4
- 238000012545 processing Methods 0.000 description 4
- 230000005291 magnetic effect Effects 0.000 description 3
- 238000003860 storage Methods 0.000 description 3
- 230000000007 visual effect Effects 0.000 description 3
- PCTMTFRHKVHKIS-BMFZQQSSSA-N (1s,3r,4e,6e,8e,10e,12e,14e,16e,18s,19r,20r,21s,25r,27r,30r,31r,33s,35r,37s,38r)-3-[(2r,3s,4s,5s,6r)-4-amino-3,5-dihydroxy-6-methyloxan-2-yl]oxy-19,25,27,30,31,33,35,37-octahydroxy-18,20,21-trimethyl-23-oxo-22,39-dioxabicyclo[33.3.1]nonatriaconta-4,6,8,10 Chemical compound C1C=C2C[C@@H](OS(O)(=O)=O)CC[C@]2(C)[C@@H]2[C@@H]1[C@@H]1CC[C@H]([C@H](C)CCCC(C)C)[C@@]1(C)CC2.O[C@H]1[C@@H](N)[C@H](O)[C@@H](C)O[C@H]1O[C@H]1/C=C/C=C/C=C/C=C/C=C/C=C/C=C/[C@H](C)[C@@H](O)[C@@H](C)[C@H](C)OC(=O)C[C@H](O)C[C@H](O)CC[C@@H](O)[C@H](O)C[C@H](O)C[C@](O)(C[C@H](O)[C@H]2C(O)=O)O[C@H]2C1 PCTMTFRHKVHKIS-BMFZQQSSSA-N 0.000 description 2
- 230000003321 amplification Effects 0.000 description 2
- 238000004422 calculation algorithm Methods 0.000 description 2
- 238000004364 calculation method Methods 0.000 description 2
- 238000004519 manufacturing process Methods 0.000 description 2
- 238000003199 nucleic acid amplification method Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 238000004891 communication Methods 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000003708 edge detection Methods 0.000 description 1
- 230000000694 effects Effects 0.000 description 1
- 230000005611 electricity Effects 0.000 description 1
- 238000003384 imaging method Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000009434 installation Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 239000013307 optical fiber Substances 0.000 description 1
- 230000009467 reduction Effects 0.000 description 1
- 239000004065 semiconductor Substances 0.000 description 1
- 238000004513 sizing Methods 0.000 description 1
- 238000013316 zoning Methods 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The invention provides a regular shape detecting method for detecting regular shapes in images and an apparatus thereof. The regular shape detecting method comprises the steps of acquiring image data; detecting a candidate regular shape area in an image; generating a gradient direction histogram in line with the pixels inside the candidate regular shape area; and detecting regular shapes inside the candidate regular shape area on the basis of the generated gradient direction histogram. By means of the regular shape detecting method and the apparatus disclosed in the embodiments of the invention, any types of regular shapes inside each candidate regular shape area can be detected once on the basis of the gradient direction histograms generated in the follow-on regular shape areas in the image. Moreover, according to the regular shape detecting method in the embodiments of the invention, regular shapes can be detected in any types of images including black and white images, gray scale images, color images and parallax images.
Description
Technical field
The present invention relates generally to image procossing, particularly the method and apparatus of regular shape in detected image.
Background technology
The various object with regular shape is there is, such as, various traffic signs shown in Fig. 1 in real world.
Regular shape refers to equal and in each, angle number the is equal shape in each limit.
Propose some detected rule shapes or there is the method for certain symmetric shape.
Such as, in U.S. Patent Application Publication US2006098877A1 disclosed in 2006, introduce a kind of SHAPE DETECTION technology, this technology obtains gradient image from input image data, gradient intensity vector is used to utilize ballot method to obtain the possible center of regular shape, and the angle of gradient vector is multiplied by the limit number of regular polygon thus the gradient direction on all limits is rotated on same direction, then carry out ballot method and vector corresponding for each gradient direction is carried out adding up determine the center of corresponding regular polygon.
Again such as, in U.S. Patent Application Publication US2009110286A1 disclosed in 2009, describe a kind of SHAPE DETECTION technology, this technology detects symmetrical image-region along specific symmetrical line, in the symmetrical image-region detected, only detect the object with known form.
There are the needs to one or more and even whole technology of regular shape that can detect hope more fast.
Summary of the invention
According to an aspect of the present invention, provide a kind of regular shape detection method for regular shape in detected image, regular shape is equal and in each, angle number the is equal shape in each limit, and this regular shape detection method can comprise: obtain view data; Candidate rule shape area in detected image; For the pixel in candidate rule shape area, generate gradient orientation histogram; And based on the gradient orientation histogram generated, detect the regular shape in this candidate rule shape area.
According to a further aspect in the invention, provide a kind of regular shape pick-up unit for regular shape in detected image, regular shape is equal and in each, angle number the is equal shape in each limit, and this regular shape pick-up unit can comprise: image obtaining parts, obtains view data; Candidate rule shape area detection part, the candidate rule shape area in detected image; Gradient orientation histogram obtaining parts, for the pixel in candidate rule shape area, generates gradient orientation histogram; And regular shape detection part, based on the gradient orientation histogram generated, detect the regular shape in this candidate rule shape area.
Utilize the regular shape detection method according to the embodiment of the present invention and device, can based on the gradient orientation histogram for the follow-up regular shape Area generation in image, one-time detection goes out the regular shape of any kind in each candidate rule shape area.And, according to the regular shape detection method of the embodiment of the present invention, detected rule shape from black white image, gray level image, coloured image to the image of any kind of anaglyph can be applicable to.
According to a further aspect in the invention, provide a kind of regular shape detection method for regular shape in detected image, wherein regular shape is equal and in each, angle number the is equal shape in each limit, and this regular shape detection method can comprise: obtain the edge image obtained by the edge pixel in image; Based on edge image, calculated product component; Utilize integrogram, calculate the radial symmetry degree in the toroidal region centered by each pixel; Based on the radial symmetry degree in the toroidal region centered by each pixel, detect candidate rule shape area; And in this candidate rule shape area, detected rule shape.
Above-mentioned regular shape detection method is introduced integrogram and is carried out the radial symmetry degree of regional area in computed image based on toroidal, to obtain the regular shape region of candidate.The introducing of integrogram can improve the computing velocity of radial symmetry degree greatly, and toroidal is particularly suitable for frame selects the region that any regular shape profile may exist, regular shape can be improved thus in candidate rule shape, detect ratio, improve counting yield, save computational resource.
Accompanying drawing explanation
Below in conjunction with accompanying drawing in the detailed description of the embodiment of the present invention, these and/or other side of the present invention and advantage will become clearly and be easier to understand, wherein:
Fig. 1 shows the traffic sign example with regular shape.
Fig. 2 shows some examples of regular shape.
Fig. 3 shows according to an embodiment of the invention for the functional configuration block diagram of the regular shape pick-up unit 1000 of regular shape in detected image.
Fig. 4 shows according to an embodiment of the invention for the overview flow chart of the regular shape detection method 2000 of regular shape in detected image.
Fig. 5 shows an example of input picture.
Fig. 6 shows the process flow diagram of the illustrative methods 2200 being carried out the candidate rule shape area in detected image according to an embodiment of the invention by the image-region that symmetry radial in detected image is higher.
The example of the edge image that Fig. 7 obtains after showing and carrying out rim detection for the image shown in Fig. 5.
Fig. 8 shows in a present embodiment schematic diagram that can be applied to the toroidal window detecting candidate rule shape area.
Fig. 9 shows the schematic diagram of the integral image calculated from the edge image shown in Fig. 7.
Figure 10 shows and utilizes integral image to calculate by pixel A, B, C, D as the schematic diagram of the number method of the pixel in the square rectangular frame 2 of four corner pixel.
Radial symmetry degree in Figure 11 (a), (b), (c) show and calculate centered by each pixel based on integral image and toroidal toroidal region (as Suo Shi Figure 11 (a)), and obtain the schematic diagram of radial symmetry degree image (Figure 11 (c)), radial symmetry degree wherein in radial symmetry degree image in the toroidal region of the brightness instruction of each pixel 3 centered by this pixel, radial symmetry degree is higher, and the brightness of this pixel is higher.
Figure 12 shows the schematic diagram of radial symmetry degree figure, and it is the amplification of Figure 11 (c).
Figure 13 (a) shows the radial symmetry degree figure after threshold filtering, and Figure 13 (b) shows the gray-scale map corresponding with the radial symmetry degree figure after threshold filtering.
Figure 14 (a), (b), (c) show a candidate rule shape area in image and the schematic diagram to the gradient orientation histogram that it generates.
Figure 15 shows a kind of schematic diagram verifying the illustrative methods of the validity of the pixel in candidate rule shape area.
Figure 16 shows according to an embodiment of the invention based on the gradient orientation histogram generated, and detects the process flow diagram of the illustrative methods 2400 of the regular shape in candidate rule shape area.
Figure 17 shows the visuality display example of regular shape testing result.
Figure 18 shows according to an embodiment of the invention for the functional configuration block diagram of the regular shape pick-up unit 3000 of regular shape in detected image.
Figure 19 shows according to an embodiment of the invention for the overview flow chart of the regular shape detection method 4000 of regular shape in detected image.
Figure 20 shows the block diagram of the exemplary computer system 600 be suitable for for realizing embodiment of the present invention.
Embodiment
In order to make those skilled in the art understand the present invention better, below in conjunction with the drawings and specific embodiments, the present invention is described in further detail.
The present invention is directed to the detection of regular shape.Regular shape is equal and in each, angle number the is equal shape in each limit.Regular shape herein comprises circle.Circle can be considered as along with limit number increases, the limitation of regular polygon sequence.Fig. 2 shows some examples of regular shape, and wherein top is divided into the profile of regular shape, and bottom is divided into the entirety of regular shape.These regular shapes can come across the many occasions in real world, such as, road traffic scene shown in Fig. 1.
To be described in the following order:
1, the first embodiment of regular shape pick-up unit
2, the first embodiment of regular shape detection method
3, the second embodiment of regular shape pick-up unit
4, the second embodiment of regular shape detection method
1, the first embodiment of regular shape pick-up unit
Fig. 3 shows according to an embodiment of the invention for the functional configuration block diagram of the regular shape pick-up unit 1000 of regular shape in detected image.
As shown in Figure 3, regular shape pick-up unit 1000 can comprise: image obtaining parts 1100, candidate rule shape area detection part 1200, gradient orientation histogram obtaining parts 1300, regular shape detection part 1400.
Image obtaining parts 1100 is configured to obtain view data.
Image obtaining parts 1100 self can be equipped with image capturing component (such as camera), also image can be obtained from outside, image can comprise gray level image and/or anaglyph, here gray level image is the concept of broad sense, scope contains from black white image to coloured image, and the type of the present invention to image does not limit.Such as, image obtaining parts 1100 can with the wired or wireless connections of monocular camera, binocular camera, many orders camera etc., with receive from its transmit image.
As a kind of example of application scenarios, binocular camera can be installed on the position near such as vehicle mirrors, it can be taken simultaneously obtain left-eye image and eye image, such one of left-eye image and eye image can be used as gray level image, and can calculate anaglyph by left-eye image and eye image.Thus based on gray-scale map and/or anaglyph, the regular shape pick-up unit of vehicle interior or exterior arrangement can carry out road traffic sign detection.In addition, the messaging device of vehicle interior configuration can also configure pedestrian detection, vehicle detection etc.Vehicle control module can receive the result of road traffic sign detection and/or pedestrian detection, vehicle detection, and sends corresponding control signal to vehicle.
It should be noted that, the position that on vehicle, binocular camera or other camera are installed and number can be arranged as required.Such as, binocular camera can be equipped on vehicle front, to take vehicle front scene.But, additionally camera can also be settled, to take rear view of vehicle scene at vehicle rear.Alternatively, camera can also be settled, to take left and right vehicle wheel both sides scene in vehicle left side or right side.In another example, wide-angle and/or image mosaic function can be incorporated in the camera, to make the angle of a phase function shooting wider, even can realize 360 degree of full-shape shoot functions.
Candidate rule shape area detection part 1200 is configured to the candidate rule shape area in detected image.
In one example, the regular shape region of candidate can be detected by embodying stronger symmetric regional area in detected image.
As detecting the method example of stronger symmetric regional area, such as, can adopt the technology being detected symmetrical image-region by the symmetry detected about particular line introduced in such as US2009110286A1.
According to one embodiment of present invention, symmetrical regional area can be detected by the radial symmetry detecting regional area.The detailed introduction realizing example in this respect will be provided hereinafter.
Gradient orientation histogram obtaining parts 1300 is configured to for the pixel in candidate rule shape area, generates gradient orientation histogram.
Regular shape detection part 1400 is configured to, based on the gradient orientation histogram generated, detect the regular shape in this candidate rule shape area.
The testing result of regular shape detection part 1400 can comprise at least one in the direction of the center of regular shape, the type of regular shape, regular shape.
Such testing result can export in a variety of manners.Such as, in the example detecting traffic sign, can be presented on the display screen of the convenient viewing of driver with visual form, remind driver with speech form.Also the information of this testing result can be outputted to vehicle control module in addition, by the type of vehicle control module according to the traffic sign recognized, automatically carry out Driving control, such as, when " (stopping) stop " indicates, automatically control to slow down and even stop.In addition, such testing result can also output to follow-up processing module, comes such as color combining identifying information, Letter identification information etc. and provides such as further testing result.
It should be noted that, the parts of above-mentioned regular shape pick-up unit 1000 can realize with software program, such as, realized in conjunction with RAM and ROM etc. and the software code that wherein runs by the CPU in multi-purpose computer.Software program can be stored on the storage mediums such as such as flash memory, floppy disk, hard disk, CD, is operationally loaded into cause CPU on such as random access storage device RAM and performs.In addition, except on multi-purpose computer, can also be realized by the cooperation between special IC and software.Described integrated circuit comprises by such as MPU(microprocessing unit), DSP(digital signal processor), FPGA(field programmable gate array), ASIC(special IC) etc. at least one realize.Such multi-purpose computer or special IC etc. such as can be loaded on vehicle; and communicate with the imaging device be arranged on such as vehicle such as camera; so that the gray level image obtain camera shooting and/or stereo-picture carry out process to obtain Traffic Sign Recognition or testing result; and the driving to vehicle can also control according to Traffic Sign Recognition or testing result alternatively, such as provide warning message, self-actuating brake or start emergency protecting equipment etc.In addition, all parts of regular shape pick-up unit 1000 can realize with special hardware, such as specific field programmable gate array, special IC etc.In addition, all parts of regular shape pick-up unit 1000 also can utilize the combination of software and hardware to realize.
Structure and the quantity of the unit in above-mentioned regular shape pick-up unit 1000 are not construed as limiting scope of the present invention.According to one embodiment of present invention, image obtaining parts 1100, candidate rule shape area detection part 1200, gradient orientation histogram obtaining parts 1300, regular shape detection part 1400 can merge into one independently parts perform and realize corresponding function and operation, or image obtaining parts 1100, candidate rule shape area detection part 1200, gradient orientation histogram obtaining parts 1300, regular shape detection part 1400 can be split as further less unit to realize their respective function and operation.
In addition, it should be noted that, structure shown in Fig. 3 is not exclusive formula, contrary regular shape pick-up unit 1000 can comprise miscellaneous part, such as display unit, for showing the testing result of regular shape detection part 1400, and such as communication component, for will for information about and/or intermediate processing results be delivered to outside etc.
2, the first embodiment of regular shape detection method
Fig. 4 shows according to an embodiment of the invention for the overview flow chart of the regular shape detection method 2000 of regular shape in detected image.Each step of this regular shape detection method 2000 can the parts of such as correspondence as shown in Figure 3 perform, and can certainly be performed by universal or special calculation element.
As shown in Figure 4, in step S2100, obtain view data.
It should be noted that, the present invention for view data form or obtain channel do not limit.
Such as, this view data can be that original being taken by digital camera obtains, and also can be that the photo taken by analogue camera obtains through analog-digital commutator process.In addition, also can be obtained by image-scanning device such as scanner scanning paper etc.Again such as, also can be the image generated by word or image procossing application software on the computing device.
In addition, view data can be raw image data, also can be through pretreated view data, such as, can be through the view data after noise reduction, also the binary image data after binary conversion treatment can be through, or edge image after edge extracting etc.
Fig. 5 shows an example of input picture.
In step S2200, the candidate rule shape area in detected image.
According to one embodiment of present invention, the candidate rule shape area in detected image can be realized by the image-region that symmetry radial in detected image is higher.
Fig. 6 shows the process flow diagram of the illustrative methods 2200 being carried out the candidate rule shape area in detected image according to an embodiment of the invention by the image-region that symmetry radial in detected image is higher.This illustrative methods 2200 may be used for the step S2200 shown in Fig. 6.
In step S2210, the edge pixel in detected image, obtains edge image.
In one example, Sobel operator such as can be utilized to carry out computed image gradient, thus obtain the gradient image of input picture.Then can carry out detection of edge pixels based on gradient image, the introduction about Sobel operator can be published in the article being entitled as " Asymptotic confidence intervals for indirect effects in structural equationmodels " of Sociological methodology in nineteen eighty-two with reference to ME Sobel.
Can utilize and anyly be suitable for the method at edge in detected image and carry out detection of edge pixels.In one example, canny detecting device can also be adopted to realize the rim detection in image, obtain the image of edge pixel composition, hereinafter can be simply referred to as edge image.
The example of the edge image that Fig. 7 obtains after showing and carrying out rim detection for the image shown in Fig. 5.
Edge image shown in Fig. 7 is bianry image, and namely edge pixel value is 1, and non-edge pixels value is 0.But, this is only example, and in fact edge image also can be gray level image or coloured image or anaglyph, and wherein the pixel value of non-edge pixels is 0, and the value of edge pixel can be in certain span.
For ease of describing, unless specifically stated otherwise, otherwise will be described for binary edge map hereinafter.
Get back to Fig. 6, in step S2220, in edge image, calculate the radial symmetry degree in the toroidal region centered by each pixel.
The thought of this step is, regular shape generally has good radial symmetry degree relative to its center.Therefore candidate rule shape area can be detected by the radial symmetry degree of surveyed area.Further consider, the limit of regular shape can be placed within annulus, such as, within incircle and circumscribed circle.Utilize this toroidal to detect candidate rule shape area, the impact of noise for Edge detected shape of regular shape inside in image can be reduced.
Fig. 8 shows in a present embodiment schematic diagram that can be applied to the toroidal window detecting candidate rule shape area.
In practical application, do not need strictly to utilize the annulus be made up of incircle and the circumscribed circle of shape to be detected, as long as but this annulus has the limit of shape to be detected in geosphere residence as far as possible.
In one example, if having the priori of the size of regular shape to be detected, then this priori can be utilized to limit cylindrical and/or the inner circle radius of annulus.
Such as, if know the side size range of the triangle traffic sign in Fig. 5 in advance, then can calculate or estimate the radius R of its circumscribed circle and the radius r of incircle, thus in addition certain nargin (δ can be distinguished
outwardand δ
in) when obtain exradius (the such as R+ δ of annulus
outward) and inner circle radius (such as r-δ
in).
In one example, by the annulus of pre-sizing in the enterprising line slip of edge image, the number of the pixel in accumulative toroidal centered by each pixel (part can getting rid of toroidal falls into the situation outside image), the number of accumulative pixel is more, represents that the radial symmetry degree in this toroidal delineation region is higher.
In order to calculate the number of pixels in certain region fast, according to a preferred embodiment of the present invention, integrogram is introduced.
Integral image (integral image), also known as summation area table (summed area table, abbreviate SAT), be one fast and effectively to calculate in the sub-rectangular areas of a grid and Data Structure and Algorithm.About the detailed introduction of integral image can with reference to the following webpage zh.wikipedia.org/zh-cn/ integrogram in wikipedia.
Any point (x in integral image, y) value refers to the gray-scale value sum of points all in the rectangular area formed from the upper left corner of image to this point, when bianry image, the number of the pixel (namely there is the pixel of valid pixel value) namely in this rectangular area.
Fig. 9 shows the schematic diagram of the integral image calculated from the edge image shown in Fig. 7.
Utilize integral image, the number of the pixel in image regional area can be calculated fast.
Figure 10 shows and utilizes integral image to calculate by pixel A, B, C, D as the schematic diagram of the number method of the pixel in the square rectangular frame 2 of four corner pixel.
The integrated value V of pixel A
ait is the number of the pixel in region 1.
By pixel A, B, C, D as the number V of the pixel in the square rectangular frame 2 of four corner pixel
{ ABCD}formula (1) below can be utilized to calculate.
V
{ABCD}=V
C+V
A–V
B–V
D(1)
Wherein, V
b, V
c, V
drepresent the integrated value of pixel B, C, D respectively.
Radial symmetry degree in Figure 11 (a), (b), (c) show and calculate centered by each pixel based on integral image and toroidal toroidal region (as Suo Shi Figure 11 (a)), and obtain the schematic diagram of radial symmetry degree image (Figure 11 (c)), radial symmetry degree wherein in radial symmetry degree image in the toroidal region of the brightness instruction of each pixel 3 centered by this pixel, radial symmetry degree is higher, and the brightness of this pixel is higher.
As shown in Figure 11 (b), in one example, the difference of the number of the pixel in region 2 that the number of the pixel in region 1 that can live with outer circle of the radial symmetry degree in each toroidal region and interior circle are lived indicates.
Figure 12 shows the schematic diagram of radial symmetry degree figure, and it is the amplification of Figure 11 (c).
Corresponding integral image and corresponding radial symmetry degree figure is described above for binary edge map, but the embodiment of the present invention also can be applied to gray level image by simple change, the situation of coloured image and anaglyph, such as when gray level image, any point (x in integral image, y) value refers to the gray-scale value sum of points all in the rectangular area formed from the upper left corner of image to this point, radial symmetry degree in toroidal region centered by each pixel can utilize in toroidal region gray-scale value sum a little represent.
In addition, in another example, can such as to the scaling process that the radial symmetry degree calculated in above-mentioned example is normalized etc.
Get back to Fig. 6, in step S2230, based on the radial symmetry degree in the toroidal region centered by each pixel, detect candidate rule shape area.
In one example, the radial symmetry degree in the toroidal region centered by each pixel and symmetry threshold value can be compared, and radial symmetry degree be exceeded the alternatively regular shape region, toroidal region of symmetry threshold value.
About symmetry threshold value, rule of thumb can arrange or learn to obtain by learning algorithm.
In one example, symmetry threshold value can be set adaptively based on the radial symmetry angle value of all pixels in radial symmetry degree image.
Such as, symmetry threshold value can be set according to formula (2) below.
T
thresh=f (the radial symmetry angle value of all pixels in radial symmetry degree image)
(2)
Such as, the mean value of the radial symmetry angle value of all pixels in radial symmetry degree image can be calculated, using the ratio of this mean value (this ratio can be greater than, be equal to or less than) as symmetry threshold value.
Again such as, mean value and the mean square deviation of the radial symmetry angle value of all pixels in radial symmetry degree image can be calculated, and calculate symmetry threshold value based on this mean value and mean square deviation.
Figure 13 (a) shows the radial symmetry degree figure after threshold filtering, each pixel (having the pixel of valid pixel value, corresponding bright spot wherein) in this radial symmetry degree figure is corresponding to the center of a toroidal (namely as the follow-up regular shape region of subsequent analysis object).Figure 13 (b) shows the gray-scale map corresponding with the radial symmetry degree figure after threshold filtering.
Get back to Fig. 6, after detecting the candidate rule shape area in image, process terminates.
The illustrative methods of the candidate rule shape area in detected image is described above with reference to figure 6-Figure 13.
It should be noted that, the detection method of the candidate rule shape area in Fig. 6-Figure 13 is used as example, and should not be construed as limitation of the present invention.Other any method that can detect candidate rule shape area may be used to the present invention.
Get back to Fig. 4, in step S2200, after detecting the candidate rule shape area obtained in image, proceed to step S2300.
In step S2300, for the pixel in candidate rule shape area, generate gradient orientation histogram.
Figure 14 (a), (b), (c) show a candidate rule shape area in image and the schematic diagram to the gradient orientation histogram that it generates.
For a candidate rule shape area, such as the dotted circle area in the gray-scale map shown in Figure 14 (a) is taken out, obtain the independent candidate rule shape area shown in Figure 14 (b), for this candidate rule shape area, the gradient of each pixel in zoning, obtains the histogram (being simply referred to as gradient orientation histogram herein) in the direction of the gradient of each pixel.
As previously mentioned, the method for the gradient of relevant calculation pixel such as can adopt sobel operator.
In one example, in compute gradient direction histogram process or before, the validity of the pixel in candidate rule shape area can be verified, and remove do not meet validity require pixel.
Figure 15 shows a kind of schematic diagram verifying the illustrative methods of the validity of the pixel in candidate rule shape area.
In this illustrative methods, for each pixel such as P1 in candidate rule shape area, calculate the gradient direction (direction as shown in the arrow 1 that label 1 indicates) of this pixel.
Based on the gradient direction 1 of this pixel, determine by this pixel and with the straight line of this gradient direction tangent, as shown in the dotted line that label 2 indicates.
Calculate the distance D1 of center C to this straight line 2 of this candidate rule shape area.
Determine this distance D1 whether in preset range, if such as know that the inscribed circle radius of shape to be detected is R in advance, then can judge whether this distance D1 is less than R+ δ, wherein δ is a little positive numerical value.
If this distance is not at preset range, then do not consider this pixel when generating gradient orientation histogram.
In Figure 15 example shown, the distance D1 be associated with pixel P1, in preset range, therefore verify that pixel P1 is valid pixel, and the distance D2 be associated with pixel P2 is outside preset range, therefore being regarded as inactive pixels, will not pixel P2 being considered when producing gradient orientation histogram.
In Figure 15, reference direction is level direction to the right, by when being rotated clockwise to gradient direction place straight line from reference direction the angle of process, as the angle of the gradient direction of this pixel of instruction.The angle that the gradient direction of pixel P1 is associated is θ, and the angle that the gradient direction of pixel P2 is associated is 0.In this example, ballot direction and gradient direction conllinear, according to the relative space relation of gradient direction with centroid, from pixel along gradient direction or reverse direction, select the direction pointing to centroid to be direction of voting.
Figure 14 (c) shows for the candidate rule shape area shown in Figure 14 (b), the schematic diagram of the gradient orientation histogram calculated, wherein transverse axis represents the angle value of gradient direction, and axle of another dimension represents that value is the number of the pixel of each angle value on transverse axis.
Get back to Fig. 4, for the pixel in candidate rule shape area in step S2300, after generating gradient orientation histogram, proceed to step S2400.
In step S2400, based on the gradient orientation histogram generated, detect the regular shape in this candidate rule shape area.
It should be noted that, here " detecting the regular shape in this candidate rule shape area " should do the understanding of broad sense, can refer to detect in this candidate rule shape area whether there is certain ad hoc rules shape (as triangle) or some ad hoc rules shape (triangle as shown in Figure 1, circular, square, hexagon), also can refer to detect in this candidate rule shape area whether there is any regular shape, there is which kind of regular shape, detect the type (limit number) of the regular shape existed, center, direction (angle such as reference direction), the regular shape in all candidate rule shape area in certain part (such as right side) in detected image can be referred to, also can refer to detect the regular shape in all candidate rule region-of-interests in whole image, can also refer to detect the regular shape in some particular candidate regular shape regions.
Be described for any regular shape detected in this candidate rule shape area below.
Figure 16 shows according to an embodiment of the invention based on the gradient orientation histogram generated, and detects the process flow diagram of the illustrative methods 2400 of the regular shape in candidate rule shape area.
As shown in figure 16, in step S2410, based on gradient orientation histogram, whether there is uniform distribution, judge whether there is circle in this candidate rule shape area.Should uniform principle carry out based on the gradient direction distribution of circle in step.
In step S2420, determine the peak value in gradient orientation histogram.Such as, can based on the comparing of intended pixel quantity threshold, determine peak value.
In step S2430, calculate the minimum angles interval between peak value.
In step S2440, based on the minimum angles interval between peak value, determine the largest possible number thereof on the limit of regular shape.
For a regular shape, the peak value of its gradient direction should correspond to the angle of the gradient direction representing each bar limit, and the interval between peak value, should meet 360 degree/k, and wherein k is limit number.
Such as, for equilateral triangle, the interval between the peak value of corresponding gradient orientation histogram should be 360/3=120.For square, the interval between the peak value of corresponding gradient orientation histogram should be 360/4=90.
Thus, under the condition obtaining the minimum angles interval β between peak value, the largest possible number thereof on the limit of regular shape may be calculated the result that rounds up to 360/ β gained real-value.
In step S2450, for limit number from any one regular shape of 3 to this maximum number k, all angles position of peak value is mated with the angle of the regular shape of corresponding sides number, to determine whether there is the regular shape of this limit number.
In addition, in one example, after determining to there is ad hoc rules shape, by checking in gradient orientation histogram the minimum angles value in the angle that is associated with this given shape, the anglec of rotation of this ad hoc rules shape relative to reference direction can be determined.
Particularly, such as, be the regular shape of 3 for limit number, can for each peak t, whether the inspection position of t+120, the position of t+240 exist peak value.If all there is peak value in the position of t, t+120, t+240, then there is triangle.And can determine that this triangle has turned t degree clockwise relative to being oriented of reference direction.
Similarly, such as, be the regular shape of 4 for limit number, for each peak t, can check whether the position of t+90, t+180, t+270 exists peak value.If all there is peak value in the position of t, t+90, t+180, t+270, then there is square.And can determine that this square has turned t degree clockwise relative to being oriented of reference direction.
By that analogy, such as, be the regular shape of k for limit number, for each peak t, t+360/k, t+2*360/k can be checked ..., whether the position of t+ (k-1) * 360/k exists peak value.If at t, t+360/k, t+2*360/k ..., all there is peak value in the position of t+ (k-1) * 360/k, then there is positive k limit shape.And can determine that this positive k limit shape has turned t degree clockwise relative to being oriented of reference direction.
After regular shape detected from a candidate rule shape area, the instruction regular shape position in the picture, center of this candidate rule shape area can be utilized.
In figure 16, after step S2450, process can terminate.
It should be noted that, as required, can carry out various amendment to the example shown in Figure 16, such as, when the target clearly detected is circular, the process shown in Figure 16 directly can terminate after step S2410.
Again such as, under the target clearly detected is triangle, square and hexagonal situation, can without the need to performing step S2430, S2440, and directly for any one in the regular shape of limit number 3,4,6, the all angles position of peak value is mated with the angle of the regular shape of corresponding sides number, to determine whether there is the regular shape of this limit number.
Get back to Fig. 4, after step S2400, process can terminate.
In one example, regular shape testing result can be exported in a variety of manners.Such as, in the example of road traffic sign detection, can be presented on the display screen of the convenient viewing of driver with visual form, the image of display as shown in figure 17, wherein label 1,2,3 indicates the object with regular shape detected of highlighted display.Or, driver can be reminded with speech form.Also the information of this testing result can be outputted to vehicle control module in addition, by the type of vehicle control module according to the traffic sign recognized, automatically carry out Driving control, such as, when " stop(stopping) " indicates, automatically control to slow down and even stop.In addition, such testing result can also output to follow-up processing module, comes such as color combining identifying information, Letter identification information etc. and provides such as further testing result.
According to the regular shape detection method of the embodiment of the present invention, can based on the gradient orientation histogram for the follow-up regular shape Area generation in image, one-time detection goes out the regular shape of any kind in each candidate rule shape area.
And, according to the regular shape detection method of the embodiment of the present invention, detected rule shape from black white image, gray level image, coloured image to the image of any kind of anaglyph can be applicable to.
3, the second embodiment of regular shape pick-up unit
Figure 18 shows according to an embodiment of the invention for the functional configuration block diagram of the regular shape pick-up unit 3000 of regular shape in detected image.
As shown in figure 18, regular shape pick-up unit 3000 can comprise: edge image obtaining parts 3100, integrogram calculating unit 3200, radial symmetry degree obtaining parts 3300, candidate rule shape area detection part 3400 Sum fanction SHAPE DETECTION parts 3500.
Edge image obtaining parts 3100 is configured to obtain the edge image obtained by the edge pixel in image.Here edge image should be interpreted broadly, and is construed as the set of edge pixel, and does not need separate visual to turn to edge image.Similarly, radial symmetry degree image mentioned in this article also should do extensive understanding, is construed as the set of the radial symmetry degrees of data of pixel, does not need equally to be visualized as separately radial symmetry degree image.Similarly, integrogram also should be interpreted broadly, and is construed as the set of the integrated value data of each pixel, does not need equally to be visualized as separately radial symmetry degree image.
Integrogram calculating unit 3200 is configured to based on edge image, calculated product component.
Radial symmetry degree obtaining parts 3300 is configured to utilize integrogram, calculates the radial symmetry degree in the toroidal region centered by each pixel.
Candidate rule shape area detection part 3400 is configured to the radial symmetry degree in the toroidal region centered by based on each pixel, detects candidate rule shape area.
Regular shape detection part 3500 is configured in this candidate rule shape area, detected rule shape.
About function and the operation of all parts, the description can carried out with reference to composition graphs 4-Figure 17 above, repeats no more here.
It should be noted that, the function of detected rule shape in this candidate rule shape area of regular shape detection part 3500 both can utilize gradient orientation histogram to realize by step S2300 and step S2400 as shown in Figure 4 like that.Also the method for other any detected rule shape can be utilized to realize, such as, regular shape detection method in aforementioned US2006098877A1 and US2009110286A1.
4, the second embodiment of regular shape detection method
Figure 19 shows according to an embodiment of the invention for the overview flow chart of the regular shape detection method 4000 of regular shape in detected image.
In step S4100, obtain the edge image obtained by the edge pixel in image.
In step S4200, based on edge image, calculated product component.
In step S4300, utilize integrogram, calculate the radial symmetry degree in the toroidal region centered by each pixel.
In step S4400, based on the radial symmetry degree in the toroidal region centered by each pixel, detect candidate rule shape area.
In step S4500, in this candidate rule shape area, detected rule shape.
About the detailed realization of each step, the description can carried out with reference to composition graphs 4-Figure 17 above, repeats no more here.
It should be noted that, the detected rule shape in this candidate rule shape area in step S3500 both can utilize gradient orientation histogram to realize by step S2300 and step S2400 as shown in Figure 4 like that.Also the method for other any detected rule shape can be utilized to realize, such as, regular shape detection method in aforementioned US2006098877A1 and US2009110286A1.
The regular shape detection method of the second embodiment shown in Figure 18 and Figure 19 and the key of device are the radial symmetry degree introduced integrogram and carry out regional area in computed image based on toroidal, obtain the regular shape region of candidate, the introducing of integrogram can improve the computing velocity of radial symmetry degree greatly, toroidal is particularly suitable for frame and selects the region that any regular shape profile may exist, regular shape can be improved thus in candidate rule shape, detect ratio, improve counting yield, save computational resource.
5, for carrying out the computing system of regular shape detection
The present invention can also be implemented by a kind of computing system for carrying out regular shape detection.Figure 20 shows the block diagram of the exemplary computer system 600 be suitable for for realizing embodiment of the present invention.As shown in figure 20, computing system 600 can comprise: CPU(CPU (central processing unit)) 601, RAM(random access memory) 602, ROM(ROM (read-only memory)) 603, system bus 604, hard disk controller 605, keyboard controller 606, serial interface controller 607, parallel interface controller 608, display controller 69, hard disk 610, keyboard 611, serial peripheral equipment 612, concurrent peripheral equipment 613 and display 614.In such devices, what be coupled with system bus 604 has CPU601, RAM602, ROM603, hard disk controller 605, keyboard controller 606, serialization controller 607, parallel controller 608 and display controller 609.Hard disk 610 is coupled with hard disk controller 605, keyboard 611 is coupled with keyboard controller 606, serial peripheral equipment 612 is coupled with serial interface controller 607, and concurrent peripheral equipment 613 is coupled with parallel interface controller 608, and display 614 is coupled with display controller 609.Should be appreciated that the structured flowchart described in Figure 20 is only used to the object of example, instead of limitation of the scope of the invention.In some cases, can increase or reduce some equipment as the case may be.
Person of ordinary skill in the field knows, the present invention can be implemented as system, device, method or computer program.Therefore, the present invention can be implemented as following form, that is: can be completely hardware, also can be software (comprising firmware, resident software, microcode etc.) completely, can also be the form that hardware and software combines, be commonly referred to as " circuit ", " module ", " device " or " system " herein.In addition, in certain embodiments, the present invention can also be embodied as the form of the computer program in one or more computer-readable medium, comprises computer-readable program code in this computer-readable medium.
The combination in any of one or more computer-readable medium can be adopted.Computer-readable medium can be computer-readable signal media or computer-readable recording medium.Computer-readable recording medium can be such as but be not limited to the system of electricity, magnetic, optical, electrical magnetic, infrared ray or semiconductor, device or device, or combination above arbitrarily.The example more specifically (non exhaustive list) of computer-readable recording medium comprises: the combination with the electrical connection of one or more wire, portable computer diskette, hard disk, random-access memory (ram), ROM (read-only memory) (ROM), erasable type programmable read only memory (EPROM or flash memory), optical fiber, Portable, compact disk ROM (read-only memory) (CD-ROM), light storage device, magnetic memory device or above-mentioned any appropriate.In this document, computer-readable recording medium can be any comprising or stored program tangible medium, and this program can be used by instruction execution system, device or device or be combined with it.
The data-signal that computer-readable signal media can comprise in a base band or propagate as a carrier wave part, wherein carries computer-readable program code.The data-signal of this propagation can adopt various ways, includes but not limited to the combination of electromagnetic signal, light signal or above-mentioned any appropriate.Computer-readable signal media can also be any computer-readable medium beyond computer-readable recording medium, and this computer-readable medium can send, propagates or transmit the program for being used by instruction execution system, device or device or be combined with it.
The program code that computer-readable medium comprises with any suitable medium transmission, can include but not limited to wireless, electric wire, optical cable, RF etc., or the combination of above-mentioned any appropriate.
The computer program code operated for performing the present invention can be write with one or more programming languages or its combination, described programming language comprises object oriented program language-such as Java, Smalltalk, C++, also comprises conventional process type programming language-such as " C " language or similar programming language.Program code can fully perform on the user computer, partly perform on the user computer, as one, independently software package performs, partly part performs on the remote computer or performs on remote computer or server completely on the user computer.In the situation relating to remote computer, remote computer can by the network of any kind-comprise LAN (Local Area Network) (LAN) or wide area network (WAN)-be connected to subscriber computer, or, outer computer (such as utilizing ISP to pass through Internet connection) can be connected to.
The present invention is described above with reference to the process flow diagram of the method for the embodiment of the present invention, device (system) and computer program and/or block diagram.Should be appreciated that the combination of each square frame in each square frame of process flow diagram and/or block diagram and process flow diagram and/or block diagram, can be realized by computer program instructions.These computer program instructions can be supplied to the processor of multi-purpose computer, special purpose computer or other programmable data treating apparatus, thus produce a kind of machine, these computer program instructions are performed by computing machine or other programmable data treating apparatus, create the device of the function/operation specified in the square frame in realization flow figure and/or block diagram.
Also can these computer program instructions be stored in the computer-readable medium that computing machine or other programmable data treating apparatus can be made to work in a specific way, like this, the instruction be stored in computer-readable medium just produces the manufacture (manufacture) of the command device (instruction means) of the function/operation specified in a square frame comprising in realization flow figure and/or block diagram.
Also can computer program instructions be loaded on computing machine, other programmable data treating apparatus or miscellaneous equipment, make to perform sequence of operations step on computing machine, other programmable data treating apparatus or miscellaneous equipment, to produce computer implemented process, thus make the instruction performed on computing machine or other programmable device can provide the process of the function/operation specified in the square frame in realization flow figure and/or block diagram.
Process flow diagram in accompanying drawing and block diagram show system according to multiple embodiment of the present invention, the architectural framework in the cards of method and computer program product, function and operation.In this, each square frame in process flow diagram or block diagram can represent a part for module, program segment or a code, and a part for described module, program segment or code comprises one or more executable instruction for realizing the logic function specified.Also it should be noted that at some as in the realization of replacing, the function marked in square frame also can be different from occurring in sequence of marking in accompanying drawing.Such as, in fact two continuous print square frames can perform substantially concurrently, and they also can perform by contrary order sometimes, and this determines according to involved function.Also it should be noted that, the combination of the square frame in each square frame in block diagram and/or process flow diagram and block diagram and/or process flow diagram, can realize by the special hardware based system of the function put rules into practice or operation, or can realize with the combination of specialized hardware and computer instruction.
Aforementioned description is only illustrative, can much revise and/or replace.
Such as, in description above, regular shape detection method and application of installation control the Traffic Sign Recognition in sight automatically in vehicle assistant drive or vehicle.But this is only example, and the present invention is not limited thereto, but the object that can be applied to any identification has the situation of regular shape, such as, for identifying the object with regular shape in the situations such as airport, railway station, market.
In addition, in description above, for the symmetric calculating of image local area, have employed toroidal window.This is because toroidal is suitable for detecting of any candidate contours towards, any regular shape, and be particularly suitable for the situation of the difform multiple object of one-time detection.But, in some cases, such as in object shapes to be detected with towards when all determining, also the ring-type window of this shape can directly be adopted, the ring-type window that two squares varied in size that such as employing has convenience center surround is to detect the symmetry of square candidate region, similarly, be the situation of positive k limit shape for object to be detected, the ring-type window of two positive k limit shapes varied in size with convenience center can be adopted to detect the symmetry that positive k is out of shape candidate region.
In addition, in description above, in the process of the candidate rule shape area in detected image, first carry out Image Edge-Detection.But in some cases, may not need to carry out this step, such as when view data inherently edge data, or when view data is sparse disparities diagram data, pixel major part now in view data has been edge pixel, and therefore rim detection may be no longer necessary.
Be described above various embodiments of the present invention, above-mentioned explanation is exemplary, and non-exclusive, and be also not limited to disclosed each embodiment.When not departing from the scope and spirit of illustrated each embodiment, many modifications and changes are all apparent for those skilled in the art.The selection of term used herein, is intended to explain best the principle of each embodiment, practical application or the improvement to the technology in market, or makes other those of ordinary skill of the art can understand each embodiment disclosed herein.
Claims (10)
1., for a regular shape detection method for regular shape in detected image, regular shape is equal and in each, angle number the is equal shape in each limit, and this regular shape detection method comprises:
Obtain view data;
Candidate rule shape area in detected image;
For the pixel in candidate rule shape area, generate gradient orientation histogram; And
Based on the gradient orientation histogram generated, detect the regular shape in this candidate rule shape area.
2. regular shape detection method according to claim 1, the described gradient orientation histogram based on generating, the regular shape detected in this candidate rule shape area comprises:
Based on the gradient orientation histogram of each candidate rule shape area generated, one-time detection goes out the regular shape of any kind in each candidate rule shape area.
3. regular shape detection method according to claim 1, the candidate rule shape area in described detected image comprises:
Edge pixel in detected image, obtains edge image;
In edge image, calculate the radial symmetry degree in the toroidal region centered by each pixel; And
Based on the radial symmetry degree in the toroidal region centered by each pixel, detect candidate rule shape area.
4. regular shape detection method according to claim 3, described in edge image, and the radial symmetry degree calculating the toroidal region centered by each pixel comprises:
Based on edge image, calculated product component; And
Utilize integrogram, calculate the radial symmetry degree in the toroidal region centered by each pixel.
5. regular shape detection method according to claim 1, described for the pixel in candidate rule shape area, generate gradient orientation histogram and comprise:
For each pixel in candidate rule shape area, calculate the gradient direction of this pixel;
Based on the gradient direction of this pixel, determine by this pixel and with the straight line of this gradient direction tangent;
Calculate the distance of center to this straight line of this candidate rule shape area;
Determine this distance whether in preset range; And
If this distance is not at preset range, then do not consider this pixel when generating gradient orientation histogram.
6. regular shape detection method according to claim 1, the described gradient orientation histogram based on generating, the regular shape detected in this candidate rule shape area comprises:
Based on gradient orientation histogram, whether there is uniform distribution, judge whether there is circle in this candidate rule shape area.
7. regular shape detection method according to claim 1, the described gradient orientation histogram based on generating, the regular shape detected in this candidate rule shape area comprises:
Determine the peak value in gradient orientation histogram;
Calculate the minimum angles interval between peak value;
Based on the minimum angles interval between peak value, determine the largest possible number thereof on the limit of regular shape;
For limit number from 3 to any one regular shape of this maximum number, all angles position of peak value is mated with the angle of the regular shape of corresponding sides number, to determine whether there is the regular shape of this limit number.
8. regular shape detection method according to claim 1, the described gradient orientation histogram based on generating, the regular shape detected in this candidate rule shape area comprises:
After determining to there is ad hoc rules shape, by checking in gradient orientation histogram the minimum angles value in the angle that is associated with this given shape, determine the anglec of rotation of this ad hoc rules shape relative to reference direction.
9., for a regular shape pick-up unit for regular shape in detected image, regular shape is equal and in each, angle number the is equal shape in each limit, and this regular shape pick-up unit comprises:
Image obtaining parts, obtains view data;
Candidate rule shape area detection part, the candidate rule shape area in detected image;
Gradient orientation histogram obtaining parts, for the pixel in candidate rule shape area, generates gradient orientation histogram; And
Regular shape detection part, based on the gradient orientation histogram generated, detects the regular shape in this candidate rule shape area.
10., for a regular shape detection method for regular shape in detected image, regular shape is equal and in each, angle number the is equal shape in each limit, and this regular shape detection method comprises:
Obtain the edge image obtained by the edge pixel in image;
Based on edge image, calculated product component;
Utilize integrogram, calculate the radial symmetry degree in the toroidal region centered by each pixel;
Based on the radial symmetry degree in the toroidal region centered by each pixel, detect candidate rule shape area; And
In this candidate rule shape area, detected rule shape.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410069286.9A CN104881856B (en) | 2014-02-27 | 2014-02-27 | The method and its device of regular shape in detection image |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201410069286.9A CN104881856B (en) | 2014-02-27 | 2014-02-27 | The method and its device of regular shape in detection image |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN104881856A true CN104881856A (en) | 2015-09-02 |
| CN104881856B CN104881856B (en) | 2018-01-16 |
Family
ID=53949341
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201410069286.9A Expired - Fee Related CN104881856B (en) | 2014-02-27 | 2014-02-27 | The method and its device of regular shape in detection image |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN104881856B (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105260713A (en) * | 2015-10-09 | 2016-01-20 | 东方网力科技股份有限公司 | Method and device for detecting lane line |
| CN105631881A (en) * | 2015-12-30 | 2016-06-01 | 四川华雁信息产业股份有限公司 | Target detection method and apparatus |
| CN107113421A (en) * | 2016-08-09 | 2017-08-29 | 深圳市瑞立视多媒体科技有限公司 | The detection method and device of a kind of optical system imaging quality |
| CN108734131A (en) * | 2018-05-22 | 2018-11-02 | 杭州电子科技大学 | A kind of traffic sign symmetry detection methods in image |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110249867A1 (en) * | 2010-04-13 | 2011-10-13 | International Business Machines Corporation | Detection of objects in digital images |
| CN102799859A (en) * | 2012-06-20 | 2012-11-28 | 北京交通大学 | Method for identifying traffic sign |
| CN102870138A (en) * | 2010-03-15 | 2013-01-09 | 美国亚德诺半导体公司 | Edge orientation for second derivative edge detection methods |
| CN103390167A (en) * | 2013-07-18 | 2013-11-13 | 奇瑞汽车股份有限公司 | Multi-characteristic layered traffic sign identification method |
-
2014
- 2014-02-27 CN CN201410069286.9A patent/CN104881856B/en not_active Expired - Fee Related
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102870138A (en) * | 2010-03-15 | 2013-01-09 | 美国亚德诺半导体公司 | Edge orientation for second derivative edge detection methods |
| US20110249867A1 (en) * | 2010-04-13 | 2011-10-13 | International Business Machines Corporation | Detection of objects in digital images |
| CN102799859A (en) * | 2012-06-20 | 2012-11-28 | 北京交通大学 | Method for identifying traffic sign |
| CN103390167A (en) * | 2013-07-18 | 2013-11-13 | 奇瑞汽车股份有限公司 | Multi-characteristic layered traffic sign identification method |
Non-Patent Citations (1)
| Title |
|---|
| 胡锦城: "基于稳定特征的交通标志检测与识别技术研究", 《中国优秀硕士学位论文全文数据库信息科技辑》 * |
Cited By (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105260713A (en) * | 2015-10-09 | 2016-01-20 | 东方网力科技股份有限公司 | Method and device for detecting lane line |
| CN105260713B (en) * | 2015-10-09 | 2019-06-28 | 东方网力科技股份有限公司 | A kind of method for detecting lane lines and device |
| CN105631881A (en) * | 2015-12-30 | 2016-06-01 | 四川华雁信息产业股份有限公司 | Target detection method and apparatus |
| CN105631881B (en) * | 2015-12-30 | 2019-02-12 | 四川华雁信息产业股份有限公司 | Object detection method and device |
| CN107113421A (en) * | 2016-08-09 | 2017-08-29 | 深圳市瑞立视多媒体科技有限公司 | The detection method and device of a kind of optical system imaging quality |
| CN108734131A (en) * | 2018-05-22 | 2018-11-02 | 杭州电子科技大学 | A kind of traffic sign symmetry detection methods in image |
| CN108734131B (en) * | 2018-05-22 | 2021-08-17 | 杭州电子科技大学 | Method for detecting symmetry of traffic sign in image |
Also Published As
| Publication number | Publication date |
|---|---|
| CN104881856B (en) | 2018-01-16 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10551198B2 (en) | Apparatus and method for providing guidance information using crosswalk recognition result | |
| US8611598B2 (en) | Vehicle obstacle detection system | |
| US9665782B2 (en) | Obstacle detecting apparatus and obstacle detecting method | |
| JP2017041228A (en) | Lane detection | |
| KR101279712B1 (en) | Apparatus and method for providing real-time lane detection, recording medium thereof | |
| CN108154149B (en) | License plate recognition method based on deep learning network sharing | |
| US11417080B2 (en) | Object detection apparatus, object detection method, and computer-readable recording medium | |
| KR20140030636A (en) | Traffic lane correction system, traffic lane correction apparatus and correction method | |
| JP6782433B2 (en) | Image recognition device | |
| CN104881856A (en) | Method and apparatus for detecting regular shapes in images | |
| CN105303157A (en) | Algorithm to extend detecting range for AVM stop line detection | |
| JPWO2013047088A1 (en) | Biological recognition device | |
| JP2016207191A (en) | Method and system for ground truth determination in lane departure warning | |
| KR101178508B1 (en) | Vehicle Collision Alarm System and Method | |
| WO2011039977A1 (en) | Pedestrian-crossing marking detecting method and pedestrian-crossing marking detecting device | |
| JP2010286995A (en) | Image processing system for vehicle | |
| TWI588444B (en) | Pavement detecting method, pavement detecting device and pavement detecting system | |
| KR101371875B1 (en) | A vehicle detection and distance measurement system and method between the vehicles using stereo vision | |
| US9030560B2 (en) | Apparatus for monitoring surroundings of a vehicle | |
| JP2009193130A (en) | Vehicle surrounding monitoring device, vehicle, program for vehicle surrounding monitoring and vehicle surrounding monitoring method | |
| KR100846167B1 (en) | Apparatus and method for displaying output information of variable message sign board | |
| CN112400094B (en) | Object detecting device | |
| CN105291982A (en) | Fast and robust stop line detector | |
| KR101440293B1 (en) | Apparatus and method for detecting crosswalk | |
| US9852352B2 (en) | System and method for determining colors of foreground, and computer readable recording medium therefor |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| C06 | Publication | ||
| PB01 | Publication | ||
| EXSB | Decision made by sipo to initiate substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant | ||
| CF01 | Termination of patent right due to non-payment of annual fee | ||
| CF01 | Termination of patent right due to non-payment of annual fee |
Granted publication date: 20180116 |