CN109859261A - A kind of quality determining method of fish-eye optical center localization method and device and camera module - Google Patents
A kind of quality determining method of fish-eye optical center localization method and device and camera module Download PDFInfo
- Publication number
- CN109859261A CN109859261A CN201910012636.0A CN201910012636A CN109859261A CN 109859261 A CN109859261 A CN 109859261A CN 201910012636 A CN201910012636 A CN 201910012636A CN 109859261 A CN109859261 A CN 109859261A
- Authority
- CN
- China
- Prior art keywords
- image
- fish
- optical center
- binary
- image block
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000003287 optical effect Effects 0.000 title claims abstract description 135
- 238000000034 method Methods 0.000 title claims abstract description 80
- 230000004807 localization Effects 0.000 title claims abstract description 25
- 238000012360 testing method Methods 0.000 claims abstract description 58
- 238000011282 treatment Methods 0.000 claims abstract description 29
- 241000251468 Actinopterygii Species 0.000 claims abstract description 22
- 238000006243 chemical reaction Methods 0.000 claims abstract description 22
- 230000000877 morphologic effect Effects 0.000 claims abstract description 14
- 230000007797 corrosion Effects 0.000 claims description 17
- 238000005260 corrosion Methods 0.000 claims description 17
- 238000003708 edge detection Methods 0.000 claims description 13
- 238000004422 calculation algorithm Methods 0.000 claims description 11
- 238000004590 computer program Methods 0.000 claims description 7
- 230000008569 process Effects 0.000 claims description 7
- 230000010339 dilation Effects 0.000 claims description 6
- 230000003628 erosive effect Effects 0.000 claims description 6
- 238000012545 processing Methods 0.000 claims description 6
- 238000007781 pre-processing Methods 0.000 claims description 4
- 238000001514 detection method Methods 0.000 description 7
- 230000004888 barrier function Effects 0.000 description 4
- 230000001788 irregular Effects 0.000 description 4
- 238000004364 calculation method Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 238000000265 homogenisation Methods 0.000 description 2
- 238000003384 imaging method Methods 0.000 description 2
- 238000000465 moulding Methods 0.000 description 2
- 230000035772 mutation Effects 0.000 description 2
- 230000000007 visual effect Effects 0.000 description 2
- 230000009471 action Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000903 blocking effect Effects 0.000 description 1
- 238000012937 correction Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000005516 engineering process Methods 0.000 description 1
- 230000003203 everyday effect Effects 0.000 description 1
- 239000004744 fabric Substances 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012544 monitoring process Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 238000012372 quality testing Methods 0.000 description 1
- 238000013519 translation Methods 0.000 description 1
Landscapes
- Image Analysis (AREA)
Abstract
The invention discloses a kind of fish-eye optical center localization methods, comprising the following steps: receives the test image of fish eye lens acquisition to be detected；The test image is n × m to divide, after the image block for obtaining each image block, and shielding patterns will be present removes, remaining image block is done into binary conversion treatment, the image block after binary conversion treatment is spliced into binary image；The binary image calculate using morphological method and obtains fitting circle data；The fitting circle data are fitted using least square circle approximating method, the center of circle of determining fitting circle is the fish-eye optical center to be detected.Also disclose a kind of fish-eye optical center positioning device.The optical center localization method and optical center positioning device can be accurately determined fish-eye optical center position in the case where fish eye lens blocks.A kind of quality determining method of camera module is also disclosed, this method can be accurately determined whether camera module is qualified camera module.
Description
Technical field
The present invention relates to the quality testing fields of camera module, and in particular to a kind of fish-eye optical center localization method,
A kind of fish-eye optical center positioning device and a kind of quality determining method of camera module.
Background technique
Fish eye lens is a kind of bugeye lens, fish-eye because its front lens is protruded in parabolical to camera lens front
Visual angle is up to even more than 180 °, with the proposition of the applications such as panorama automobile assistant driving system, real time monitoring and three-dimensional reconstruction,
Fish eye lens has been widely used.
It before using fish eye images, needs first accurately to correct fish-eye relevant parameter, this correction
Process is known as camera calibration in computer vision.The resulting parameter of camera calibration will directly affect later image understanding process,
And then influence the subjective feeling of practical application and the authenticity of visual experience.
The content of camera calibration includes the outwardly and inwardly parameter of determining camera, and external parameter refers to position and the appearance of camera
State, the rotation and translation etc. including camera；Inner parameter is the optical distortion parameter of the element of orientation and camera lens in camera, including
Focal length and optical center coordinate etc..Wherein, optical center is most important parameter, it directly affects the precision of calibration.
For the optical lens in conventional camera module, the pixel that optical lens acquires image is relatively uniform, generally understands root
The optical center position of optical lens is directly determined according to the brightness value of acquisition image.If application publication number is disclosed in CN108632604A
The detection method and device of camera lens optical center specifically determine area-of-interest according to the monochromatic homogeneous image of acquisition, emerging according to feeling
The boundary in interesting region and the position of area-of-interest determine the optical center position of camera lens.
Furthermore it is also possible to determine the optical center position of camera lens according to the geometry that optical lens acquires image.Such as apply for public affairs
The disclosed method for finding camera lens optical center of cloth CN106683135A specifically determines in imaging region according to acquisition image
Boundary and outer boundary, and inside and outside circle is determined according to the inner and outer boundary of imaging region, the light of camera lens is determined according to the center of circle of inside and outside circle
Heart position.
Both the above scheme can accurately determine the optical center position of camera lens, but for as shown in Figure 1 molding
Vehicle-mounted fish eye lens, this fish eye lens, since vehicle-mounted fish eye lens can be blocked by bracket, cause when rotating to a direction
The brightness value for acquiring image is uneven, and boundary is also unobvious, and therefore, above two scheme is not applicable.
In addition, the quality of camera module is detected also according to the optical center position of optical lens, when the optical center position of optical lens
It sets and is overlapped with the center of acquisition image or when within a certain error range, then it is assumed that camera module is qualified to image mould
Group, therefore, the optical center position for being accurately determined optical lens are very important.
Summary of the invention
The object of the present invention is to provide a kind of fish-eye optical center localization methods, which can be accurately
Determine fish-eye optical center position.
It is a further object of the present invention to provide a kind of fish-eye optical center positioning device, which being capable of essence
Really determine fish-eye optical center position.
For another object of the present invention there is provided a kind of quality determining method of camera module, this method can accurately really
Determine whether camera module is qualified camera module.
For achieving the above object, the present invention the following technical schemes are provided:
In a first aspect, a kind of fish-eye optical center localization method, comprising the following steps:
Receive the test image of fish eye lens acquisition to be detected；
The test image is n × m to divide, obtains each image block, and the image block that shielding patterns will be present removes
Afterwards, remaining image block is done into binary conversion treatment, the image block after binary conversion treatment is spliced into binary image；
The binary image calculate using morphological method and obtains fitting circle data；
The fitting circle data are fitted using least square circle approximating method, the center of circle of determining fitting circle is
The fish-eye optical center to be detected.
Second aspect, a kind of fish-eye optical center positioning device, comprising:
Preprocessing module is received, for receiving the test image of fish eye lens acquisition to be detected；
Occlusion culling module, for removing the characteristic area in the test image；
Fitting circle data acquisition module is fitted for calculate to the binary image using morphological method
Circle data；
Optical center determining module is fitted the fitting circle data using least square circle approximating method, and what is determined is quasi-
The center of circle for closing circle is the fish-eye optical center to be detected.
The third aspect, a kind of fish-eye optical center positioning device, including computer storage, computer processor and
The computer program that can be executed in the computer storage and on the computer processor is stored in, at the computer
Reason device performs the steps of when executing the computer program
Receive the test image of fish eye lens acquisition to be detected；
The test image is n × m to divide, obtains each image block, and the image block that shielding patterns will be present removes
Afterwards, remaining image block is done into binary conversion treatment, the image block after binary conversion treatment is spliced into binary image；
The binary image calculate using morphological method and obtains fitting circle data；
The fitting circle data are fitted using least square circle approximating method, the center of circle of determining fitting circle is
The fish-eye optical center to be detected.
Fourth aspect, a kind of quality determining method of camera module, comprising the following steps:
Fish-eye optical center in camera module to be detected is determined using above-mentioned fish-eye optical center localization method；
Or,
Fish-eye optical center in camera module to be detected is determined using above-mentioned fish-eye optical center positioning device；
More fish-eye optical center and camera module to be detected acquisition image center, when fish-eye optical center and to
The center for detecting camera module acquisition image is overlapped, or within a certain error range, then camera module to be detected is qualified camera shooting
Mould group.
Compared with prior art, the device have the advantages that are as follows:
Above-mentioned optical center localization method and optical center positioning device will have and block by carrying out block division to test image
The image block of pattern removes, and to remove interference of the shielding patterns to fitting circle data are obtained, while being obtained using morphological method
Pixel break edge data have established accurate data basis for fitting circle, in addition, using least square circle approximating method pair
The fitting circle data are fitted, and provide the robustness of capability of fitting and optical center calculating, and then simply accurately define
Fish-eye optical center position.
The quality determining method of above-mentioned camera module, accurately true using above-mentioned optical center localization method and optical center positioning device
It, can be rapidly and accurately to the matter of camera module according to the center of acquisition image on the basis of fixed fish-eye optical center position
Amount judges, and can rapidly and accurately obtain qualified camera module.
Detailed description of the invention
Fig. 1 is the photo of the vehicle-mounted pick-up mould group provided in background technique；
Fig. 2 is the flow chart for the fish-eye optical center localization method that embodiment provides；
Fig. 3 is the test image for the fish eye lens to be detected acquisition that embodiment provides, wherein Fig. 3 (a) is without containing blocking
The test image of pattern, Fig. 3 (b) are the test image containing shielding patterns；
Fig. 4 is the result figure that the image block that embodiment provides divides, wherein Fig. 4 (a) is to Fig. 3 (a) region division
Result figure, Fig. 4 (b) are the result figure to Fig. 3 (b) region division；
Fig. 5 is the binarization result figure that embodiment provides, wherein Fig. 5 (a) is after carrying out shielding patterns rejecting to Fig. 4 (a)
The result figure of image progress binary conversion treatment；Fig. 5 (b) is that image carries out at binaryzation after carrying out shielding patterns rejecting to Fig. 4 (b)
The result figure of reason；
Fig. 6 is the fringe region testing result figure that embodiment provides, wherein Fig. 6 (a) is to carry out fringe region to Fig. 5 (a)
The result figure of detection；Fig. 5 (b) is the result figure that fringe region detection is carried out to Fig. 4 (b)；
Fig. 7 is the optical center definitive result figure that embodiment provides, and Fig. 7 (a) is to determine to test image shown in Fig. 3 (a)
Optical center result figure, Fig. 7 (b) be to Fig. 3 (b) be shown in test image determine optical center result images；
Fig. 8 is the structural schematic diagram for the fish-eye optical center positioning device that embodiment provides.
Specific embodiment
To make the objectives, technical solutions, and advantages of the present invention more comprehensible, with reference to the accompanying drawings and embodiments to this
Invention is described in further detail.It should be appreciated that the specific embodiments described herein are merely illustrative of the present invention, and
The non-protection scope limiting the invention in any way.In the full text of specification, the identical member of identical reference numbers
Part.Statement "and/or" includes any of one or more of associated all column phase mesh and all combinations.
In the accompanying drawings, for ease of description, thickness, the size and shape of object are slightly exaggerated.Attached drawing is merely illustrative
And it is non-critical drawn to scale.
It will also be appreciated that term " comprising ", " including ", " having ", "comprising" and/or " including ", when in this theory
In bright book use when indicate exist stated feature, step, entirety, operations, elements, and/or components, but do not exclude the presence of or
It is attached with one or more of the other feature, step, entirety, operation, component, assembly unit and/or their combination.
It as used in this article term " substantially ", " about " and is similarly used for being used as and indicates approximate term,
And be not used as the term of expression degree, and be intended to illustrate by by those skilled in the art will appreciate that, measured value or meter
Inherent variability in calculation value.
Unless otherwise defined, otherwise all terms (including technical terms and scientific words) used herein all have with
The application one skilled in the art's is generally understood identical meaning.It will also be appreciated that term (such as in everyday words
Term defined in allusion quotation) it should be interpreted as having and their consistent meanings of meaning in the context of the relevant technologies, and
It will not be explained with idealization or excessively formal sense, unless clear herein so limit.
It should be noted that in the absence of conflict, the feature in embodiments herein and embodiment can be mutual
Combination.The application is described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.
Embodiment 1
For the vehicle-mounted camera module of molding as shown in Figure 1, in order to accurately measure flake mirror in the vehicle-mounted pick-up mould group
The optical center position of head, present embodiments provides a kind of fish-eye optical center localization method.As shown in Fig. 2, this is fish-eye
Optical center localization method the following steps are included:
S201 receives the test image of fish eye lens acquisition to be detected.
Test image is the image that fish eye lens to be detected acquires scene, which can be the colour to acquisition
Image does the image of pixel Homogenization Treatments, i.e., obtain BMP figure when, tri- channel brightness values of RGB are tuned into it is close, thus
Test image as shown in Figure 3 can be obtained.Shown in the test image of normal acquisition such as Fig. 3 (a), do not hidden in the test image
Pattern is kept off, then the edge contour of test image is rendered as regular circle shapes shape.
When fish eye lens goes to some angle, when collecting test image, when fish eye lens is blocked by barriers such as brackets, then
Shown in test image such as Fig. 3 (b), the darker shade of meeting brightness value in image, which is shielding patterns, such as A in Fig. 3 (b)
It is shown.There are when shielding patterns in test image, the edge contour of test image is irregular, this irregular edge wheel
Exterior feature directly affects the determination of optical center position.
The test image is n × m and divided, obtains each image block, and the image block of shielding patterns will be present by S202
After removing, remaining image block is done into binary conversion treatment, the image block after binary conversion treatment is spliced into binary image.
Test image is done into figure in order to exclude interference of the shielding patterns to optical center position is determined, the present embodiment in test image
As block division, the image block with shielding patterns is removed.Specifically, test image can be done to n × m division, n and m are
Natural number, and n and m can be equal, can also be unequal, citing n and m all can be 4, i.e., test image is done into 4 × 4 divisions, such as
Shown in Fig. 4 (a) and 4 (b), it can obtain 16 image blocks.The size that image block divides can determine according to the actual situation, if hiding
Gear pattern is smaller, the thinner of test image division can be obtained more image block, but should not too carefully, too many figure
The computational efficiency with shielding patterns image block is excluded as block will increase, therefore, as long as the rule that image block divides is to divide
Image block can include some shielding patterns.
In the present embodiment, according to area shared by dash area in image block come to there are the progress of the image block of shielding patterns
It rejects, specifically, the image block that shielding patterns will be present, which removes, includes:
For each image block, ratio shared by shaded area in each image block is calculated, and by the shaded area institute
The image block that the ratio accounted for is greater than shaded area proportion threshold value removes.
It is understood that ratio shared by shaded area refers in one piece of image block, pixel value is less than a certain pixel
The number of the pixel of threshold value accounts for the total number of pixel in whole image block.Wherein, it is to block that pixel threshold, which is to discriminate between pixel,
The line of demarcation of pattern or normal picture, this pixel threshold generally according to the actual situation depending on, for example, the pixel threshold
Value can be 50~70.
Shaded area proportion threshold value be judge in image block whether the line of demarcation containing shielding patterns, the shaded area ratio
Threshold value according to the actual situation depending on, when the ratio shared by the shaded area is greater than set shaded area proportion threshold value, then it is assumed that
There are shielding patterns in image block, i.e., remove the image block.
Specifically, the image block with shielding patterns can be removed by the way of modifying pixel value, for example, can incite somebody to action
Pixel value is changed to 255, realizes and removes the image block with shielding patterns.
After removing shielding patterns, that is, the distracter of determining optical center position is eliminated, that is, obtaining can be quasi-
Determine the image of optical center.For more acurrate determining optical center position, remaining image block is done into binary conversion treatment, with more clearly
The contour edge of resolution chart picture, the image block after binary conversion treatment are spliced into the binary picture as shown in Fig. 5 (a) and 5 (b)
Picture.
S203 carries out the binary image using morphological method to calculate acquisition fitting circle data.
After obtaining binary image, using binary image as process object, using morphological method to binary picture
As carrying out edge detection, it can determine the pixel of pixel value mutation, these pixels are fitting circle data.
Specifically, described to include: according to binary image acquisition fitting circle data
The binary image is detected using edge detection algorithm, obtains pixel break edge in binary image, it is described
Pixel break edge is fitting circle data.
Edge detection algorithm is mainly the edge that pixel mutation is determined using the means of filtering, may include Sobel operator
Detection method, Canny operator detection method and Laplacian operator detection method.In the present embodiment, preferably select
Laplacian operator detection method detects binary image, to determine pixel break edge.
Specifically, described to include: according to binary image acquisition fitting circle data
Expansion process is carried out to the binary image, obtains expanding image；
Corrosion treatment is carried out to the binary image, obtains corrosion image；
The expanding image and the corrosion image are done into difference operation, the data of acquisition are fitting circle data.
Dilation erosion can be partitioned into independent pixel, these pixels are fitting circle data, and the present embodiment obtains
Fitting circle data show as shown in Fig. 6 (a) and Fig. 6 (b).The fitting circle data of acquisition can guarantee the accurate of data
Property, improve the capability of fitting of optical center circle and the robustness of optical center calculating.
S204 is fitted the fitting circle data using least square circle approximating method, the circle of determining fitting circle
The heart is the fish-eye optical center to be detected.
Least square method is a kind of mathematical optimization techniques, and the quadratic sum by minimizing error finds the best of one group of data
Function matching.Some absolutely not known true value are acquired using the simplest method, and enabling the sum of square-error is a koji for minimum
The algorithms most in use of line fitting.
The method of Least Square Circle fitting is as follows:
The equation of known circle: (x-A)2+(y-B)2=R2
Fitting circle data (Xi,Yi), the distance of point in i=1....N to the center of circle are as follows:
Enable σiIndicates coordinate point to the square distance of the edge of the circle and the difference of radius squared,
σ is indicated using Q (a, b, c)iQuadratic sum,
Parameter a, b, c is asked to make Q (a, b, c) minimum according to the principle of least square, wherein A=-a/2, B=-b/2,
In the present embodiment, fitting circle data are fitted using least square circle approximating method, determining fitting circle
Shown in the center of circle such as Fig. 7 (a) and 7 (b).
Above-mentioned optical center localization method can accurately calculate optical center position by individual test image, test through application, needle
To the BMP figure of 1280*800 pixel size, the algorithm time-consuming 32ms, memory consumption 9M meet to the higher field of requirement of real-time
Scape has good practicability.
Above-mentioned optical center localization method is removed the image block with shielding patterns by carrying out block division to test image
Fall, to remove interference of the shielding patterns to fitting circle data are obtained, while pixel break edge number is obtained using morphological method
According to accurate data basis having been established for fitting circle, in addition, using least square circle approximating method to the fitting circle data
It is fitted, provides the robustness of capability of fitting and optical center calculating, and then simply accurately define fish-eye optical center
Position.Importantly, no matter whether fish eye lens is blocked by barriers such as such as brackets, above-mentioned optical center localization method can be used
Accurately determine optical center position.
Embodiment 2
As shown in figure 8, present embodiments providing a kind of fish-eye optical center positioning device 800, comprising:
Preprocessing module 801 is received, is mainly used for receiving the test image of fish eye lens acquisition to be detected；
Occlusion culling module 802 is mainly used for removing the occlusion area in the test image；
Fitting circle data acquisition module 803 is mainly used for calculating the binary image using morphological method
Obtain fitting circle data；
Optical center determining module 804 is mainly fitted the fitting circle data using least square circle approximating method, really
The center of circle of fixed fitting circle is the fish-eye optical center to be detected.
Receiving the received test image of preprocessing module 801 is to obtain BMP by the image of pixel Homogenization Treatments
When figure, tri- channel brightness values of RGB are tuned into close, this makes it possible to obtain test image as shown in Figure 3.Normal acquisition
Shown in test image such as Fig. 3 (a), there is no shielding patterns in the test image, then the edge contour of test image is rendered as rule
Round.
When the fish eye lens of vehicle-mounted pick-up mould group is gone to shown in image such as Fig. 3 (b) of some angle acquisition, due to flake mirror
Head is blocked by barriers such as brackets, and the darker shade of meeting brightness value in image, which is shielding patterns, such as A in Fig. 3 (b)
It is shown.There are when shielding patterns in test image, the edge contour of test image is irregular, this irregular edge wheel
Exterior feature directly affects the determination of optical center position.
Wherein, the occlusion culling module 802 includes:
Region division module 8021 divides for being n × m to the test image, obtains each image block；
Shielded image block rejects module 8022, for calculating ratio shared by shaded area in each image block, and by institute
The image block that ratio shared by shaded area is stated greater than shaded area proportion threshold value removes；
Binary processing module 8023, for remaining image block to be done binary conversion treatment, the image after binary conversion treatment
Block is spliced into binary image.
Test image is n × m and divided by region division module 8021, and the size that image divides can be according to the actual situation
It is fixed, if shielding patterns are smaller, the thinner of test image division can be obtained into more image block, but should not be too thin, too
More image blocks will increase the computational efficiency for excluding to have shielding patterns image block, and image block can not be divided too big, be divided
It is too big, then it is too big that the image block removed accounts for area, while removing shielding patterns, also by the class round shape of more rule
Edge removes, so that the fitting circle data obtained below are less, is unfavorable for being fitted the determination of circle drawing and optical center.Therefore,
As long as the rule that image block divides is that the image block divided can include some shielding patterns.
It is mainly to delete the image block containing shielding patterns that shielded image block, which rejects module 8022,.Specifically, according to every
The area of dash area in a image block determines whether containing shielding patterns.The ratio shared by the shaded area is greater than set
When determining shaded area proportion threshold value, then it is assumed that there are shielding patterns in image block, i.e., remove the image block.Wherein, shadow surface
Depending on accumulating proportion threshold value according to the actual situation.Ratio shared by hatched area refers to that in one piece of image block, pixel value is small
The total number of pixel in whole image block is accounted in the number of the pixel of a certain pixel threshold.Wherein, pixel threshold is to discriminate between
Pixel is the line of demarcation of shielding patterns or normal picture, this pixel threshold generally according to the actual situation depending on.
After removing shielding patterns, that is, the distracter of determining optical center position is eliminated, that is, obtaining can be quasi-
Determine the image of optical center.For more acurrate determining optical center position, remaining image block is done two by binary processing module 8023
Value processing, with the contour edge of more clearly resolution chart picture, the image block after binary conversion treatment is spliced into binary picture
Picture.
Wherein, the fitting circle data acquisition module 803 includes:
Edge detection module 8031 obtains binary picture for detecting the binary image using edge detection algorithm
The pixel break edge as in, the pixel break edge is fitting circle data；And/or
Dilation erosion module 8032 obtains expanding image for carrying out expansion and corrosion treatment to the binary image
And corrosion image, and the expanding image and the corrosion image are done into difference operation, the data of acquisition are fitting circle data.
Optical center positioning device shown in Fig. 8 had both included edge detection module 8031, also included dilation erosion module 8032.When
So, optical center positioning device can also only include edge detection module, or only include dilation erosion module.
The fitting circle data that edge detection algorithm and dilation erosion obtain can guarantee the accuracy of data, improve optical center
The robustness that round capability of fitting and optical center calculate.
In optical center determining module 804, fitting circle data are fitted using least square circle approximating method, it can fast
Speed determines the optical center circle of fitting, and the center of circle of optical center circle is optical center position.
As long as the fish-eye optical center positioning device, which handles individual test image, can accurately calculate optical center position, warp
Using test, scheme for the BMP of 1280*800 pixel size, the algorithm time-consuming 32ms, memory consumption 9M, satisfaction want real-time
Higher scene is sought, there is good practicability.
Embodiment 3
Present embodiments provide a kind of fish-eye optical center positioning device, including computer storage, computer disposal
Device and it is stored in the computer program that can be executed in the computer storage and on the computer processor, the meter
Calculation machine processor performs the steps of when executing the computer program
Receive the test image of fish eye lens acquisition to be detected；
The test image is n × m to divide, obtains each image block, and the image block that shielding patterns will be present removes
Afterwards, remaining image block is done into binary conversion treatment, the image block after binary conversion treatment is spliced into binary image；
The binary image calculate using morphological method and obtains fitting circle data；
The fitting circle data are fitted using least square circle approximating method, the center of circle of determining fitting circle is
The fish-eye optical center to be detected.
Wherein, the image block that shielding patterns will be present, which removes, includes:
For each image block, ratio shared by shaded area in each image block is calculated, and by the shaded area institute
The image block that the ratio accounted for is greater than shaded area proportion threshold value removes.
It is described to include: according to binary image acquisition fitting circle data
The binary image is detected using edge detection algorithm, obtains pixel break edge in binary image, it is described
Pixel break edge is fitting circle data；Or,
Expansion process is carried out to the binary image, obtains expanding image；The binary image is carried out at corrosion
Reason obtains corrosion image；The expanding image and the corrosion image are done into difference operation, the data of acquisition are fitting circle number
According to.
All steps and details that embodiment 3 is related to are identical as the optical center localization method that embodiment 1 provides, herein no longer
It repeats.
In the present embodiment, computer processor, which refers to, is loaded with processing routine, is able to carry out the chip of processing function.Computer
Program can be stored as series of instructions or order, computer storage on the computer storage for storing and/or transmitting
Including random access memory (RAM), read-only memory (ROM), such as hard disk or floppy disk magnetic medium or such as CD
(CD) or optical medium, the flash memory etc. of DVD (multifunctional digital code CD), it can also be nonvolatile memory.
The optical center positioning device is removed the image block with shielding patterns by carrying out block division to test image
Fall, to remove interference of the shielding patterns to fitting circle data are obtained, while pixel break edge number is obtained using morphological method
According to accurate data basis having been established for fitting circle, in addition, using least square circle approximating method to the fitting circle data
It is fitted, provides the robustness of capability of fitting and optical center calculating, and then simply accurately define fish-eye optical center
Position.Importantly, no matter whether fish eye lens is blocked by barriers such as such as brackets, above-mentioned optical center localization method can be used
Accurately determine optical center position.
Embodiment 4
Present embodiments provide a kind of quality determining method of camera module, comprising the following steps:
It is determined using the fish-eye optical center localization method that embodiment 1 provides fish-eye in camera module to be detected
Optical center；
More fish-eye optical center and camera module to be detected acquisition image center, when fish-eye optical center and to
The center for detecting camera module acquisition image is overlapped, or within a certain error range, then camera module to be detected is qualified camera shooting
Mould group.
Present embodiments provide the quality determining method of another camera module, comprising the following steps:
Flake mirror in camera module to be detected is determined using the fish-eye optical center positioning device that embodiment 2 and 3 provides
The optical center of head；
More fish-eye optical center and camera module to be detected acquisition image center, when fish-eye optical center and to
The center for detecting camera module acquisition image is overlapped, or within a certain error range, then camera module to be detected is qualified camera shooting
Mould group.
In the present embodiment, certain error range refers to optical center and picture centre at a distance of the number of pixel, this is according to reality
It is unrestricted herein depending on the situation of border.
The quality determining method accurately determined using above-mentioned optical center localization method and optical center positioning device it is fish-eye
On the basis of optical center position, according to the center of acquisition image, rapidly and accurately the quality of camera module can be judged, i.e.,
Qualified camera module can rapidly and accurately be obtained.
Technical solution of the present invention and beneficial effect is described in detail in above-described specific embodiment, Ying Li
Solution is not intended to restrict the invention the foregoing is merely presently most preferred embodiment of the invention, all in principle model of the invention
Interior done any modification, supplementary, and equivalent replacement etc. are enclosed, should all be included in the protection scope of the present invention.
Claims (11)
1. a kind of fish-eye optical center localization method, comprising the following steps:
Receive the test image of fish eye lens acquisition to be detected；
The test image is n × m to divide, it, will after the image block for obtaining each image block, and shielding patterns will be present removes
Remaining image block does binary conversion treatment, and the image block after binary conversion treatment is spliced into binary image；
The binary image calculate using morphological method and obtains fitting circle data；
The fitting circle data are fitted using least square circle approximating method, the center of circle of determining fitting circle is described
Fish-eye optical center to be detected.
2. fish-eye optical center localization method as described in claim 1, which is characterized in that the shielding patterns that will be present
Image block, which removes, includes:
For each image block, ratio shared by shaded area in each image block is calculated, and will be shared by the shaded area
The image block that ratio is greater than shaded area proportion threshold value removes.
3. fish-eye optical center localization method as described in claim 1, which is characterized in that described according to the binary picture
Include: as obtaining fitting circle data
The binary image is detected using edge detection algorithm, obtains pixel break edge in binary image, the pixel
Break edge is fitting circle data.
4. fish-eye optical center localization method as described in claim 1, which is characterized in that described according to the binary picture
Include: as obtaining fitting circle data
Expansion process is carried out to the binary image, obtains expanding image；
Corrosion treatment is carried out to the binary image, obtains corrosion image；
The expanding image and the corrosion image are done into difference operation, the data of acquisition are fitting circle data.
5. a kind of fish-eye optical center positioning device characterized by comprising
Preprocessing module is received, for receiving the test image of fish eye lens acquisition to be detected；
Occlusion culling module, for removing the occlusion area in the test image；
Fitting circle data acquisition module obtains fitting circle number for calculate to the binary image using morphological method
According to；
Optical center determining module is fitted the fitting circle data using least square circle approximating method, determining fitting circle
The center of circle be the fish-eye optical center to be detected.
6. fish-eye optical center positioning device as claimed in claim 5, which is characterized in that the occlusion culling module packet
It includes:
Region division module divides for being n × m to the test image, obtains each image block；
Shielded image block rejects module, for calculating ratio shared by shaded area in each image block, and by the shadow surface
The image block that the shared ratio of product is greater than shaded area proportion threshold value removes；
Binary processing module, for remaining image block to be done binary conversion treatment, the image block after binary conversion treatment is spliced into
Binary image.
7. fish-eye optical center positioning device as claimed in claim 5, which is characterized in that the fitting circle data acquisition mould
Block includes:
Edge detection module obtains pixel in binary image for detecting the binary image using edge detection algorithm
Break edge, the pixel break edge are fitting circle data；And/or
Dilation erosion module obtains expanding image and etch figures for carrying out expansion and corrosion treatment to the binary image
Picture, and the expanding image and the corrosion image are done into difference operation, the data of acquisition are fitting circle data.
8. a kind of fish-eye optical center positioning device, including computer storage, computer processor and it is stored in described
In computer storage and the computer program that can be executed on the computer processor, which is characterized in that the computer
Processor performs the steps of when executing the computer program
Receive the test image of fish eye lens acquisition to be detected；
The test image is n × m to divide, it, will after the image block for obtaining each image block, and shielding patterns will be present removes
Remaining image block does binary conversion treatment, and the image block after binary conversion treatment is spliced into binary image；
The binary image calculate using morphological method and obtains fitting circle data；
The fitting circle data are fitted using least square circle approximating method, the center of circle of determining fitting circle is described
Fish-eye optical center to be detected.
9. fish-eye optical center positioning device as claimed in claim 8, which is characterized in that the shielding patterns that will be present
Image block, which removes, includes:
For each image block, ratio shared by shaded area in each image block is calculated, and will be shared by the shaded area
The image block that ratio is greater than shaded area proportion threshold value removes.
10. fish-eye optical center positioning device as claimed in claim 8, which is characterized in that described according to the binaryzation
Image obtains fitting circle data
The binary image is detected using edge detection algorithm, obtains pixel break edge in binary image, the pixel
Break edge is fitting circle data；Or,
Expansion process is carried out to the binary image, obtains expanding image；Corrosion treatment is carried out to the binary image, is obtained
Obtain corrosion image；The expanding image and the corrosion image are done into difference operation, the data of acquisition are fitting circle data.
11. a kind of quality determining method of camera module, comprising the following steps:
It is determined in camera module to be detected using the described in any item fish-eye optical center localization methods of such as Claims 1 to 4
Fish-eye optical center；Or,
It is determined in camera module to be detected using the described in any item fish-eye optical center positioning devices of such as claim 5~10
Fish-eye optical center；
The center of more fish-eye optical center and camera module to be detected acquisition image, when fish-eye optical center and to be detected
The center that camera module acquires image is overlapped, or within a certain error range, then camera module to be detected is qualified camera module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910012636.0A CN109859261A (en) | 2019-01-07 | 2019-01-07 | A kind of quality determining method of fish-eye optical center localization method and device and camera module |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910012636.0A CN109859261A (en) | 2019-01-07 | 2019-01-07 | A kind of quality determining method of fish-eye optical center localization method and device and camera module |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN109859261A true CN109859261A (en) | 2019-06-07 |
Family
ID=66894090
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910012636.0A Pending CN109859261A (en) | 2019-01-07 | 2019-01-07 | A kind of quality determining method of fish-eye optical center localization method and device and camera module |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109859261A (en) |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110099220A (en) * | 2019-06-17 | 2019-08-06 | 广东中星微电子有限公司 | A kind of panorama mosaic method and device |
| CN111638043A (en) * | 2020-06-10 | 2020-09-08 | 浙江大华技术股份有限公司 | Optical center determining method and device |
| CN119211735A (en) * | 2024-10-23 | 2024-12-27 | 深圳森云智能科技有限公司 | Fisheye lens correction method, system and storage medium |
Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103413318A (en) * | 2013-08-27 | 2013-11-27 | 北京航空航天大学 | Target ball mirror surface center locating method |
| CN104182757A (en) * | 2013-05-22 | 2014-12-03 | 内蒙古大学 | Method of acquiring actual coverage area of measured target and device |
| CN104408730A (en) * | 2014-12-08 | 2015-03-11 | 北京市新技术应用研究所 | Calibration device for fish eye lens |
| CN104980730A (en) * | 2014-04-01 | 2015-10-14 | 宁波舜宇光电信息有限公司 | Method for positioning optical center on the basis of concentric circle theory |
| CN106651959A (en) * | 2016-11-15 | 2017-05-10 | 东南大学 | Optical field camera micro-lens array geometric parameter calibration method |
| US20180300957A1 (en) * | 2017-04-13 | 2018-10-18 | Facebook, Inc. | Panoramic camera systems |
-
2019
- 2019-01-07 CN CN201910012636.0A patent/CN109859261A/en active Pending
Patent Citations (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104182757A (en) * | 2013-05-22 | 2014-12-03 | 内蒙古大学 | Method of acquiring actual coverage area of measured target and device |
| CN103413318A (en) * | 2013-08-27 | 2013-11-27 | 北京航空航天大学 | Target ball mirror surface center locating method |
| CN104980730A (en) * | 2014-04-01 | 2015-10-14 | 宁波舜宇光电信息有限公司 | Method for positioning optical center on the basis of concentric circle theory |
| CN104408730A (en) * | 2014-12-08 | 2015-03-11 | 北京市新技术应用研究所 | Calibration device for fish eye lens |
| CN106651959A (en) * | 2016-11-15 | 2017-05-10 | 东南大学 | Optical field camera micro-lens array geometric parameter calibration method |
| US20180300957A1 (en) * | 2017-04-13 | 2018-10-18 | Facebook, Inc. | Panoramic camera systems |
Non-Patent Citations (5)
| Title |
|---|
| BAOSU GUO ET AL: "A novel algorithm for the extraction of machining areas based on morphological image processing", 《INTERNATIONAL JOURNAL OF COMPUTER INTEGRATED MANUFACTURING》 * |
| 冯为嘉: "基于鱼眼镜头的全方位视觉及全景立体球视觉研究", 《中国博士学位论文全文数据库 信息科技辑》 * |
| 吕楠楠: "基于图像分块的遮挡人脸识别研究", 《中国优秀硕士学位论文全文数据库 信息科技辑》 * |
| 张宝峰等: "全景立体球视觉下的鱼眼镜头参数标定方法", 《天津理工大学学报》 * |
| 王波涛等: "基于最小二乘法的鱼眼镜头轮廓提取算法", 《国外电子测量技术》 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110099220A (en) * | 2019-06-17 | 2019-08-06 | 广东中星微电子有限公司 | A kind of panorama mosaic method and device |
| CN110099220B (en) * | 2019-06-17 | 2021-04-13 | 广东中星微电子有限公司 | Panoramic stitching method and device |
| CN111638043A (en) * | 2020-06-10 | 2020-09-08 | 浙江大华技术股份有限公司 | Optical center determining method and device |
| CN119211735A (en) * | 2024-10-23 | 2024-12-27 | 深圳森云智能科技有限公司 | Fisheye lens correction method, system and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110717942B (en) | Image processing method and device, electronic equipment and computer readable storage medium | |
| CN109461181B (en) | Depth image acquisition method and system based on speckle structured light | |
| CN106780392B (en) | Image fusion method and device | |
| CN108876836B (en) | Depth estimation method, device and system and computer readable storage medium | |
| Delaunoy et al. | Photometric bundle adjustment for dense multi-view 3d modeling | |
| KR101643607B1 (en) | Method and apparatus for generating of image data | |
| EP1958158B1 (en) | Method for detecting streaks in digital images | |
| CN111160136B (en) | Standardized 3D information acquisition and measurement method and system | |
| CN109859261A (en) | A kind of quality determining method of fish-eye optical center localization method and device and camera module | |
| WO2011137140A1 (en) | Range measurement using a coded aperture | |
| CN116542982B (en) | Departure judgment device defect detection method and device based on machine vision | |
| Banerjee et al. | In-camera automation of photographic composition rules | |
| US20140146999A1 (en) | Device, method and non-transitory computer readable storage medium for detecting object | |
| CN107113421B (en) | The detection method and device of a kind of optical system imaging quality | |
| Khan et al. | High-density single shot 3D sensing using adaptable speckle projection system with varying preprocessing | |
| CN109614858B (en) | Pupil center detection method and device | |
| EP1190208B1 (en) | Method for measuring a three-dimensional object, or assembly of objects | |
| Reich et al. | A Real-Time Edge-Preserving Denoising Filter. | |
| CN114972084A (en) | Image focusing accuracy evaluation method and system | |
| CN115035281B (en) | Rapid infrared panoramic image stitching method | |
| CN115797372B (en) | Flame target extraction method based on combined image segmentation | |
| CN117058183A (en) | Image processing method and device based on double cameras, electronic equipment and storage medium | |
| Pashchenko et al. | An algorithm for the visualization of stereo images simultaneously captured with different exposures | |
| Verhoeven | Focusing on out-of-focus: assessing defocus estimation algorithms for the benefit of automated image masking | |
| Cao et al. | Depth image vibration filtering and shadow detection based on fusion and fractional differential |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| RJ01 | Rejection of invention patent application after publication | ||
| RJ01 | Rejection of invention patent application after publication |
Application publication date: 20190607 |