CN113923358A - Online automatic focusing method and system in flying shooting mode - Google Patents
Online automatic focusing method and system in flying shooting mode Download PDFInfo
- Publication number
- CN113923358A CN113923358A CN202111174787.XA CN202111174787A CN113923358A CN 113923358 A CN113923358 A CN 113923358A CN 202111174787 A CN202111174787 A CN 202111174787A CN 113923358 A CN113923358 A CN 113923358A
- Authority
- CN
- China
- Prior art keywords
- focusing
- image
- definition
- module
- product
- 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
- 238000000034 method Methods 0.000 title claims abstract description 64
- 238000001514 detection method Methods 0.000 claims abstract description 63
- 238000004458 analytical method Methods 0.000 claims abstract description 31
- 230000000007 visual effect Effects 0.000 claims abstract description 15
- 230000008569 process Effects 0.000 claims description 25
- 230000007547 defect Effects 0.000 claims description 24
- 230000033001 locomotion Effects 0.000 claims description 24
- 238000004519 manufacturing process Methods 0.000 claims description 22
- 238000011156 evaluation Methods 0.000 claims description 18
- 239000007788 liquid Substances 0.000 claims description 14
- 230000007246 mechanism Effects 0.000 claims description 13
- 230000006870 function Effects 0.000 claims description 11
- 230000008859 change Effects 0.000 claims description 8
- 230000003993 interaction Effects 0.000 claims description 8
- 230000000694 effects Effects 0.000 abstract description 8
- 238000005516 engineering process Methods 0.000 abstract description 6
- 238000013461 design Methods 0.000 abstract description 3
- 230000009286 beneficial effect Effects 0.000 abstract 1
- 239000000047 product Substances 0.000 description 70
- 238000003384 imaging method Methods 0.000 description 13
- 238000007689 inspection Methods 0.000 description 10
- 230000007423 decrease Effects 0.000 description 5
- 238000011179 visual inspection Methods 0.000 description 4
- 238000010586 diagram Methods 0.000 description 3
- 238000009826 distribution Methods 0.000 description 3
- 230000003287 optical effect Effects 0.000 description 3
- 230000010354 integration Effects 0.000 description 2
- 238000012545 processing Methods 0.000 description 2
- 230000004044 response Effects 0.000 description 2
- 241001156002 Anthonomus pomorum Species 0.000 description 1
- 238000010521 absorption reaction Methods 0.000 description 1
- 238000010923 batch production Methods 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 239000012467 final product Substances 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 229910052751 metal Inorganic materials 0.000 description 1
- 239000002184 metal Substances 0.000 description 1
- 150000002739 metals Chemical class 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 239000004033 plastic Substances 0.000 description 1
- 229920003023 plastic Polymers 0.000 description 1
- OIGNJSKKLXVSLS-VWUMJDOOSA-N prednisolone Chemical compound O=C1C=C[C@]2(C)[C@H]3[C@@H](O)C[C@](C)([C@@](CC4)(O)C(=O)CO)[C@@H]4[C@@H]3CCC2=C1 OIGNJSKKLXVSLS-VWUMJDOOSA-N 0.000 description 1
- 230000004043 responsiveness Effects 0.000 description 1
- 230000035945 sensitivity Effects 0.000 description 1
- 239000000126 substance Substances 0.000 description 1
Images
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/60—Control of cameras or camera modules
- H04N23/67—Focus control based on electronic image sensor signals
-
- G—PHYSICS
- G02—OPTICS
- G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
- G02B7/00—Mountings, adjusting means, or light-tight connections, for optical elements
- G02B7/28—Systems for automatic generation of focusing signals
- G02B7/282—Autofocusing of zoom lenses
-
- G—PHYSICS
- G02—OPTICS
- G02B—OPTICAL ELEMENTS, SYSTEMS OR APPARATUS
- G02B7/00—Mountings, adjusting means, or light-tight connections, for optical elements
- G02B7/28—Systems for automatic generation of focusing signals
- G02B7/36—Systems for automatic generation of focusing signals using image sharpness techniques, e.g. image processing techniques for generating autofocus signals
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
- H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
- H04N23/50—Constructional details
- H04N23/55—Optical parts specially adapted for electronic image sensors; Mounting thereof
Landscapes
- Physics & Mathematics (AREA)
- Engineering & Computer Science (AREA)
- Multimedia (AREA)
- Signal Processing (AREA)
- General Physics & Mathematics (AREA)
- Optics & Photonics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Studio Devices (AREA)
- Automatic Focus Adjustment (AREA)
Abstract
The invention provides an online automatic focusing method and system in a flying shooting mode. The system comprises an image acquisition module and an online focusing module, wherein the image acquisition module acquires an image of a product by utilizing a fly-shooting technology, the online focusing module is used for performing focusing control input according to definition analysis of the acquired image, the focal length of the image acquisition module is adjusted according to the definition of the product until the corresponding definition standard is reached, the acquired image is transmitted to a visual detection module, and the visual detection module detects the image of the product and outputs a detection result. The invention has the beneficial effects that: the method has the advantages of reducing complex mechanical structure design, ensuring the stability of a shooting system, setting a focusing value through software, quickly responding within a millisecond range, and realizing the technical effect of on-line automatic focusing in a continuous operation mode.
Description
Technical Field
The invention relates to the field of product detection, in particular to an online automatic focusing method and system in a flying shooting mode.
Background
In the industrial batch production process, the product defect detection is used as a main application scene in the field of industrial quality inspection, and the quality of a final product is influenced by the defect detection result. The traditional industrial quality inspection field depends on manual visual inspection, the manual inspection consumes long time, subjective deviation is easy to appear, and secondary damage is easy to be caused to products in the quality inspection process. With the development of deep learning technology, computer vision is widely applied to the field of industrial quality inspection, and defect detection is carried out on the surface of a product through an image recognition technology. The image acquisition module is used as the input of machine vision and is a key factor in the detection system, the higher the image quality is, the lower the processing difficulty of the vision module is, the better the detection result is, and meanwhile, the stability and the reliability of the system are higher. The image definition is used as an important index for measuring the quality of an image, and the main influence factors of the image definition depend on the type selection of a lens and the focusing precision. In the industrial batch quality inspection process, the image of the product is easy to be burnt due to the fact that the product or the machine is uneven and has poor depth information. The detection image with the inaccurate focusing is often low in definition, blurred in visual effect, incapable of obtaining defect information and seriously interfered with the defect detection effect, so that the camera needs to be refocused.
The process of focusing the camera is essentially the process of evaluating the image definition. At present, focusing modes mainly comprise active focusing and passive focusing. An active focusing solution, a type of which determines the target distance according to reflection by means of external substances with strong absorption on a distance measuring source, such as infrared rays, laser and the like, then adjusts the distance, and drives a motor to control a product or a camera lens to be adjusted within the range of depth of field. The mode has strict requirements on the design and precision of a mechanical structure, easily causes product damage in the detection process, and has the advantages of high focusing difficulty, low speed, unstable image imaging quality of the product and incapability of ensuring image definition. In addition, for some products, the imaging effect is limited by the material and size of the product, the leveling precision of an optical system and a workbench, and the image acquisition system driven by the mechanical structure has poor image quality and can not meet the quality inspection requirement on the defect detection effect.
In the traditional offline focusing, a motor structure is driven according to different searching methods, target images are obtained by traversing different focusing positions, the definition of all the images is calculated, and the optimal image is selected. The whole searching process needs more than ten frames of continuous images to finish focusing, and the image with the best quality and the highest definition is selected as the final detection input, so that the focusing efficiency is low, the searching process is long, and meanwhile, the motor driving mechanism has more repeated motions, and the mode is only suitable for visual detection in a fixed shooting mode. In order to improve the operating efficiency, most of the current industrial automatic quality inspection adopts a flying-shooting detection technology to realize high-efficiency mass production, and the traditional fixed-point focusing technology cannot meet the requirement of industrial mass quality inspection.
Patent CN113296242A discloses an auto-focus imaging lens capable of realizing rapid zooming and large depth of field using requirement, which uses a liquid lens structure to rapidly realize the auto-focus process of the lens, and uses a small number of lenses, thereby reducing the volume and cost of the lens, and having high resolution. The liquid lens adjusts the surface curvature thereof by changing the voltage to realize the focusing function. Do the transport main part through the SCARA robot in patent CN113103215A, it is fast to have, characteristics such as precision height, the efficiency of equipment improves greatly, the camera that will fly in the system of clapping is placed in the robot end, the scope in operation space has been guaranteed, the realization of flying the function of clapping has been guaranteed under the condition that does not influence the space use, product simple structure, it is small and exquisite, can satisfy C electron, household electrical appliances, toys, the application demand of trades such as plastics five metals, cooperation high response camera and stroboscopic light source controller, can realize grabbing the discernment of thing and the adjustment of angle under the condition of taking a picture incessantly, can realize the dynamic snatching. In CN113252007A, an initial image is obtained by using a preliminarily set flying shooting control parameter, the similarity of the initial image to a target image is graded through a plurality of image comparison algorithms, then the flying shooting control parameter corresponding to the highest grading is predicted based on the flying shooting control parameter and the grading through a Bayesian optimizer, and finally an actual image used for detection is shot by using the flying shooting control parameter corresponding to the highest grading. But mainly adjust the relevant parameter of flying to clapping among the technical content of flying to clap at present, but need guarantee to carry out clear acquisition to the image of product in the middle of the detection of product and just can guarantee the accurate detection to the product, fly clap and traditional decide clap and have great difference, need guarantee to carry out clear acquisition to the product image in real time.
Disclosure of Invention
In order to solve the technical problems, the invention discloses an online automatic focusing method and an online automatic focusing system aiming at the defocusing caused by the depth disturbance of the unevenness of a detection system, the error of a mechanical mechanism and the like and solving the defocusing problem caused by the spatial position deviation in the batch quality inspection process in a flying shooting mode, and the technical scheme of the invention is implemented as follows:
an online automatic focusing method in a flying shooting mode specifically comprises the following steps: s1: carrying out real-time image acquisition on products on a product production line through a mobile flyswatter; s2: evaluating the definition of the acquired image; s3: carrying out on-line automatic focusing according to the definition evaluation result; s4: detecting defects of the acquired images and transmitting the defect detection results to a production line end; the online automatic focusing specifically comprises the following steps: s3-1: focusing to enable the product to be detected to be located in the field depth range and to be in an accurate focusing state; s3-2: in a flying shooting mode, acquiring an image of a next frame of product to be detected, calculating an image definition evaluation result, and storing a current focusing value; s3-3: calculating a definition difference value between adjacent image frames in the continuous detection process; s3-4: judging the tendency of the fuzzy degree according to the difference of the definition of the two frames of images in S3-3, if the current frame is clearer than the previous frame or the change degree of the fuzzy degree is in an allowable range, using the current focusing value, otherwise, calculating the focusing step value according to the difference of the definition, and controlling the lens to apply the focusing value, wherein the focusing direction is positive; s3-5: continuing to capture the next image; s3-6: calculating the definition difference after fine adjustment, if the definition analysis result is a definition tendency, indicating that the current focusing direction is correct, and otherwise, applying a step value in the opposite direction; s3-7: steps S3-2 through S3-6 are repeated.
Preferably, the algorithm of image sharpness includes a gray gradient, a statistical feature, a frequency domain analysis or an information entropy method.
Preferably, the constraint between camera and depth of field is
The utility model provides an online automatic focusing system under flying to shoot mode, according to the above-mentioned online automatic focusing method under flying to shoot mode, including image acquisition module, online focusing module and visual detection module, wherein image acquisition module includes liquid adjustable focus camera lens and high resolution camera, image acquisition module gathers the image of product through flying to shoot, online focusing module is right according to the definition evaluation the focus of liquid adjustable focus camera lens is adjusted, visual detection module is right the image of gathering in the image acquisition module detects the discernment to detect the result output.
Preferably, the flying racket further comprises a motion control module, wherein the motion control module comprises a motion control mechanism, and the motion control mechanism controls the moving flying racket.
Preferably, the online focusing system further comprises a software system, the software system is provided with the image definition algorithm, the image definition analysis function and the focusing value, and the software system takes definition evaluation as control input of the online focusing module.
Preferably, the device further comprises a hardware control module, wherein the hardware control module comprises the motion control module.
Preferably, the system further comprises a human-computer interaction interface, and the human-computer interaction interface controls the software system and the hardware control module.
The technical scheme of the invention can solve the technical problems of the prior art that the product detection system is uneven, the defocusing is caused by depth disturbance, errors of a mechanical mechanism and the like, and the defocusing is easily caused by the spatial position deviation in the batch quality detection process; according to the technical scheme, the detection platform is directly arranged on the production line, the image acquisition module consisting of the liquid-state adjustable-focus lens and the high-resolution camera is arranged on the detection platform, automatic on-line focusing is realized according to the set focusing function to ensure that the image is clearly acquired, the acquired image is subjected to defect detection through the visual defect detection module, and the generated detection result is output to the production line end, so that the complex mechanical structure design can be reduced, the stability of a shooting system is ensured, the focusing value can be set through software, quick response is realized within a millisecond range, and the technical effect of on-line automatic focusing in a continuous operation mode is realized.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly described below, it is obvious that the drawings in the following description are only one embodiment of the present invention, and for those skilled in the art, other drawings can be obtained according to the drawings without creative efforts.
In which like parts are designated by like reference numerals. It should be noted that the terms "front," "back," "left," "right," "upper" and "lower" used in the following description refer to directions in the drawings, and the terms "bottom" and "top," "inner" and "outer" refer to directions toward and away from, respectively, the geometric center of a particular component.
FIG. 1 is a schematic diagram of an on-line auto-focusing process;
FIG. 2 is a schematic diagram of a system architecture;
FIG. 3 is a schematic diagram of an on-line auto-focusing process;
fig. 4 is a schematic view of focal plane and depth of field imaging.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
Example 1
In a preferred embodiment 1, as shown in fig. 1, an online auto-focusing method in a flying shooting mode specifically includes the following steps: s1: carrying out real-time image acquisition on products on a product production line through a mobile flyswatter; s2: evaluating the definition of the acquired image; s3: carrying out on-line automatic focusing according to the definition evaluation result; s4: detecting defects of the acquired images and transmitting the defect detection results to a production line end; the online automatic focusing specifically comprises the following steps: s3-1: focusing to enable the product to be detected to be located in the field depth range and to be in an accurate focusing state; s3-2: in a flying shooting mode, acquiring an image of a next frame of product to be detected, calculating an image definition evaluation result, and storing a current focusing value; s3-3: calculating a definition difference value between adjacent image frames in the continuous detection process; s3-4: judging the tendency of the fuzzy degree according to the difference of the definition of the two frames of images in S3-3, if the current frame is clearer than the previous frame or the change degree of the fuzzy degree is in an allowable range, using the current focusing value, otherwise, calculating the focusing step value according to the difference of the definition, and controlling the lens to apply the focusing value, wherein the focusing direction is positive; s3-5: continuing to capture the next image; s3-6: calculating the definition difference after fine adjustment, if the definition analysis result is a definition tendency, indicating that the current focusing direction is correct, and otherwise, applying a step value in the opposite direction; s3-7: steps S3-2 through S3-6 are repeated.
In embodiment 1, automatic online focusing is performed in the fly-shooting mode to realize rapid and accurate detection of products on the production line, and the method can directly detect the products produced on the production line. The flying shooting technology is to acquire an image of a product in the moving process of the product and then use the acquired image for detecting the defects of the product. However, unlike the fixed shooting, the product to be shot is always in motion during the flying shooting process, but because a large number of devices are available in the production line, the devices may have a certain influence on the flying shooting effect, for example, the vibration of the machine may affect the definition of the product image. In embodiment 1, the detection of the product is performed by flying swatting, which specifically includes the following steps: s1: carrying out real-time image acquisition on products on a product production line through a mobile flyswatter; s2: evaluating the definition of the acquired image; s3: carrying out on-line automatic focusing according to the definition evaluation result until an image with satisfactory definition is obtained; s4: the defect detection is carried out on the collected image, and the defect detection result is transmitted to the production line end. In the embodiment 1, automatic focusing is realized through the definition evaluation result, and the clear image of the product is always acquired in the flying shooting process. And (4) image definition analysis, namely evaluating the information content of the image. Clear high-quality images have rich image content and sharp edge information, the boundaries of the out-of-focus images are often not clear, a large amount of detail information is lost, and the image quality is poor. And in the out-of-focus and in-focus state of the contrast image, the difference of characteristic information is compared, the contrast of the clear image is high, the edge gradient is obvious, and the contained information is rich. Image sharpness algorithms are roughly classified into four categories: grayscale gradient, statistical features, frequency domain analysis, and information entropy methods. The gray gradient method analyzes the definition of an image according to the gray value of a picture of the image. The statistical characteristic method analyzes the definition of the image by performing statistical histogram analysis on the distribution of the image gray values of focusing and defocusing of the image; the information entropy method analyzes the images of the commodities by counting the gray values of the images. The frequency domain analysis realizes the definition analysis of the image by carrying out component analysis on the definition and the fuzziness of the image. For the analysis method of the image definition, different methods can be selected according to the conditions of shape, color and the like of a product, or multiple methods are set to select the optimal analysis result according to the output optimal result and the optimal detection effect. In the process of flying shooting, on-line automatic focusing needs to be realized according to definition, and with the depth of field range as a reference, as shown in fig. 4, an image of a focusing plane is clear, and a clear image can be obtained. Outside the depth of field region, imaging blur is exacerbated. If the direction indicated by the arrow is the change direction of the focusing value, the image imaging definition gradually blurs from A to D along with the unidirectional increase or decrease of the focusing value, when the product is applied to an actual project, the product is located in the range of depth of field in an initial state, after a period of time, the image definition gradually decreases along with the error of a mechanical mechanism and the like due to the fact that a detection system is uneven, and finally the complete defocusing is caused. Based on the optical characteristics, the image sharpness evaluation result obtained by the image sharpness analysis method is used as a feedback control input of the focusing control, and the specific method can be as follows: s3-1: focusing initially to enable the product to be detected to be located in the field depth range and to be in an accurate focusing state; s3-2: in a flying shooting mode, acquiring an image of a next frame of product to be detected, calculating an image definition evaluation result, and storing a current focusing value; s3-3: calculating a definition difference value between adjacent image frames in the continuous detection process; s3-4: judging the tendency of the fuzzy degree according to the difference of the definition of the two frames of images in S3-3, if the current frame is clearer than the previous frame or the change degree of the fuzzy degree is in an allowable range, using the current focusing value, otherwise, calculating the focusing step value according to the difference of the definition, and controlling the lens to apply the focusing value, wherein the focusing direction is positive; s3-5: continuing to capture the next image; s3-6: calculating the definition difference after fine adjustment, if the definition analysis result is a definition tendency, indicating that the current focusing direction is correct, and otherwise, applying a step value in the opposite direction; s3-7: steps S3-2 through S3-6 are repeated. Through the steps, clear image acquisition of the product in the flying shooting mode is realized, and clear acquisition of the image of the product to be detected on a production line is guaranteed, so that efficient detection of the product is realized. The method can be used for detecting products with different sizes, shapes and colors, and can be realized by mainly correspondingly adjusting the specific position between the camera and the product to be detected. In embodiment 1, the minimum influence on the effect of the aerial photography in the production line is realized through the interaction of image detection and adjustment of the online focal length.
In a preferred embodiment, the image sharpness algorithm includes gray scale gradient, statistical features, frequency domain analysis, or entropy method.
In this specific embodiment, the grayscale gradient method includes a Tenengrad function, a Laplacian function, a Brenner gradient method, a variance function, and the like, and this method mostly uses the grayscale value change or edge gradient information to determine the sharpness of the image, and a larger grayscale value of the whole image indicates that the image is sharper, otherwise the image is blurred. The statistical characteristic method is used for performing statistical histogram analysis according to different distribution of gray values of the out-of-focus focusing image. Similarly, the information entropy method counts the distribution of gray values as an image sharpness index. The frequency domain analysis method comprises a full-frequency-band integration method and a threshold integration method, and more high-frequency components of the image which is clearly focused exist, and more low-frequency components of the image which is blurred are focused. In practical application, a proper definition evaluation method is selected according to sensitivity, responsiveness and consistency.
In a preferred embodiment, the constraint between the camera and the depth of field is
In the specific implementation mode, the flatness of the product and the workbench is considered, and even if the flatness precision and accuracy of the machine reach the precision processing degree, all unit products cannot be guaranteed to be on the same plane, the imaging of the area beyond the depth of field is fuzzy, the characteristic information is not clear, and the product is located in the depth of field, namely the focusing process. Depth of field (Depth of field) and camera lens have the following constraints: δ is the allowable circle diameter of confusion, F is the lens aperture value, F is the lens focal length, M is the magnification, and the depth of field decreases rapidly as the magnification increases. In the traditional scheme, the mechanism has large interference on the environment in the motor motion process, images are often out of focus, in fuzzy imaging, the imaging is slightly blurred, so that the characteristic information is incomplete, and in a serious way, the image characteristic information is completely lost.
Example 2
In a preferred embodiment 2, as shown in fig. 2, an online auto-focusing system in a flying photography mode, according to the online auto-focusing method in the flying photography mode in embodiment 1, includes an image acquisition module, an online focusing module, and a visual detection module, where the image acquisition module includes a liquid focus-adjustable lens and a high-resolution camera, the image acquisition module acquires an image of a product through flying photography, the online focusing module adjusts a focal length of the liquid focus-adjustable lens according to sharpness evaluation, and the visual detection module detects and identifies the image acquired in the image acquisition module and outputs a detection result.
In embodiment 2, as shown in fig. 2, a system is formed according to the online focusing method in the fly-shooting mode in embodiment 1, the system is directly disposed on a production line of a product and can directly detect the product produced on the production line, the system includes an image acquisition module, an online focusing module, and a city department detection module, the image acquisition module includes a liquid adjustable focusing lens and a high resolution camera, wherein the liquid adjustable focusing lens has the advantages of low power consumption, small size, and electronically controlled focusing as a novel imaging device, and is often used for acquiring depth information of a scene through a defocusing and focusing mode. Therefore, in embodiment 2, an image acquisition system composed of a liquid adjustable focus lens optical system and a high-resolution camera is used to perform fly-shooting imaging on a target product, and image sharpness analysis and online automatic focusing are performed. The online focusing module is used for realizing online adjustment of a focusing distance according to the definition analysis result of an image obtained by the image acquisition module as input, one or more image definition analysis methods such as a liquid adjustable focusing lens and a high-resolution camera can be selected from the online focusing module, the acquired image is transmitted to the online focusing module, the online focusing module is used for analyzing the definition of the image according to the gray gradient analysis method, and as shown in fig. 3, in batch detection, the deviation generated by mechanical motion of a platform and the error of a product per se do not exist between the product and a mechanism, so that the height h of the product detected at different moments t from the camera has a certain deviation. In the actual detection process, a proper image definition analysis function is selected as feedback input of automatic focusing according to product characteristics, definition is obtained according to edge definition or histogram analysis, and focusing of the liquid-state adjustable focusing lens is controlled in a feedback mode. With the depth of field range as a reference, as shown in fig. 4, the image of the focal plane is clear, and a clear image can be obtained. Outside the depth of field region, imaging blur is exacerbated. The direction indicated by a blue arrow is a focusing value change direction, the image imaging definition gradually blurs from A to D from the blur level along with the unidirectional increase or decrease of the focusing value, when the product is applied to an actual project, the product is located in the field depth range in an initial state, after a period of time, the image definition gradually decreases along with the error of a mechanical mechanism and the like due to the fact that a detection system is uneven, and finally the complete defocusing is caused. The online focusing of the system comprises the following steps:
the method comprises the following steps: initializing a working platform, focusing to ensure that the target is positioned in the range of depth of field and is in an accurate focusing state, and recording the definition result of the current frame image as F0；
Acquiring a next frame of target image in a flying shooting mode, calculating an image definition evaluation result, and storing a current focusing value;
step three: calculating definition difference F between adjacent image frames in continuous detection processt-Ft-1；
Step four: and judging the blurring degree trend according to the difference delta F of the image sharpness of the two frames in the third step, and using the current focusing value if the current frame is clearer than the previous frame or the blurring degree change degree is in an allowable range. Otherwise, the focusing step value epsilon is calculated according to the definition difference value1The software controls the liquid lens to apply a focusing value, and the direction is positive;
step five: the platform continues to operate and captures the next image;
step six: calculating the definition difference after fine adjustment, if the definition analysis result is a definition tendency, indicating that the current focusing direction is correct, otherwise applying a step value epsilon in the reverse direction2；
Step seven: and repeating the steps from two to six.
Can guarantee through image acquisition module and the online module of focusing to produce the product on the line and carry out quick clear image acquisition, clear image transmission to the visual inspection module that will acquire afterwards, can set up corresponding product defect detecting system in the visual inspection module, the visual inspection module can carry out the defect detection and have the ability of self-study and upgrading to the product image, the renewal to the defect database can be realized according to the continuous update of detecting the sample, the realization can detect and filter the new defect of product, guarantee the intellectuality of system. The system in embodiment 2 may be disposed on a detection platform, and the detection platform may be directly connected to a production line end. For example, the system in embodiment 2 is disposed on a lip beautifying instrument production line, and the image of the lip beautifying instrument on the production line is acquired by flying photography, and the image pixels of the lip beautifying instrument at 800 × 1200 can realize clear visual defect detection on the lip beautifying instrument, so that when the image pixels of the lip beautifying instrument do not reach 800 × 1200, the on-line focusing module adjusts the focal length of the liquid focus-adjustable lens until the acquired lip beautifying instrument image reaches 800 × 1200, and transmits the relevant image to the visual detection module for product defect detection, thereby ensuring that high-precision visual defect detection can be performed on the product.
In a preferred embodiment, the flying racket further comprises a motion control module, wherein the motion control module comprises a motion control mechanism, and the motion control mechanism controls the moving flying racket.
In this concrete embodiment, realize the control to image acquisition module through motion control module, in the middle of the process to product image acquisition, the product carries out quick removal on producing the line, consequently need carry out the removal of the position that does not stop to image acquisition module, and motion control module can be fine controls image acquisition module, guarantees the smoothness nature and the mobility of image acquisition module motion.
In a preferred embodiment, the system further comprises a software system, wherein an image definition algorithm, an image definition analysis function and a focusing value are set in the software system, and the software system takes definition evaluation as control input of the online focusing module.
In this specific embodiment, a software system is further provided in the system, and the software system controls software in the whole system, such as an image sharpness analysis method and an image sharpness analysis function in an online focusing module, product detection software in a visual detection module, and the like, and the software system realizes the accuracy and efficiency of the whole system.
In a preferred embodiment, the system further comprises a hardware control module, and the hardware control module comprises a motion control module.
In this specific embodiment, the hardware control module controls the motion control module, and the hardware control module can precisely control the motion control module. The hardware control module has good maneuvering performance, avoids interference on equipment on a production line, and ensures that the system has wide applicability.
In a preferred embodiment, the system further comprises a human-computer interaction interface, and the human-computer interaction interface controls the software system and the hardware control module.
In the specific implementation mode, the man-machine interaction interface can accurately control the software system and the hardware control module, perfect matching of the hardware control module and the software system is realized, meanwhile, the man-machine interaction interface can realize manual accurate operation on the system, and the accuracy of product detection and the accuracy of mechanical operation can be improved.
It should be understood that the above-described embodiments are merely exemplary of the present invention, and are not intended to limit the present invention, and that any modification, equivalent replacement, or improvement made without departing from the spirit and principle of the present invention shall fall within the protection scope of the present invention.
Claims (8)
1. An online automatic focusing method in a flying shooting mode is characterized by comprising the following steps: s1: carrying out real-time image acquisition on products on a product production line through a mobile flyswatter; s2: evaluating the definition of the acquired image; s3: carrying out on-line automatic focusing according to the definition evaluation result; s4: detecting defects of the acquired images and transmitting the defect detection results to a production line end; the online automatic focusing specifically comprises the following steps: s3-1: focusing to enable the product to be detected to be located in the field depth range and to be in an accurate focusing state; s3-2: in a flying shooting mode, acquiring an image of a next frame of product to be detected, calculating an image definition evaluation result, and storing a current focusing value; s3-3: calculating a definition difference value between adjacent image frames in the continuous detection process; s3-4: judging the tendency of the fuzzy degree according to the difference of the definition of the two frames of images in S3-3, if the current frame is clearer than the previous frame or the change degree of the fuzzy degree is in an allowable range, using the current focusing value, otherwise, calculating the focusing step value according to the difference of the definition, and controlling the lens to apply the focusing value, wherein the focusing direction is positive; s3-5: continuing to capture the next image; s3-6: calculating the definition difference after fine adjustment, if the definition analysis result is a definition tendency, indicating that the current focusing direction is correct, and otherwise, applying a step value in the opposite direction; s3-7: steps S3-2 through S3-6 are repeated.
2. The method of claim 1, wherein the method comprises: the image definition algorithm comprises a gray gradient, a statistical characteristic, a frequency domain analysis or an information entropy method.
3. The method of claim 1, wherein the method comprises: the constraint between camera and depth of field is:
4. an on-line automatic focusing system in a flying shooting mode, an on-line automatic focusing method in a flying shooting mode according to claims 1-3, characterized in that: including image acquisition module, online module and the visual detection module of focusing, wherein image acquisition module includes but liquid focusing lens and high resolution camera, image acquisition module gathers the image of product through flying to shoot, it is right according to the definition evaluation to online module of focusing the focus of but liquid focusing lens is adjusted, the visual detection module is right the image of gathering in the image acquisition module detects the discernment to with the testing result output.
5. The system of claim 4, wherein the system further comprises: still include the motion control module, the motion control module includes motion control mechanism, motion control mechanism control removes the flyswatter.
6. The system of claim 4, wherein the system further comprises: the system also comprises a software system, wherein the software system is provided with the image definition algorithm, an image definition analysis function and a focusing value, and the software system takes definition evaluation as control input of the online focusing module.
7. The system of claim 4, wherein the system further comprises: the device also comprises a hardware control module, wherein the hardware control module comprises the motion control module.
8. The system of claim 4, wherein the system further comprises: the system also comprises a human-computer interaction interface which controls the software system and the hardware control module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111174787.XA CN113923358A (en) | 2021-10-09 | 2021-10-09 | Online automatic focusing method and system in flying shooting mode |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111174787.XA CN113923358A (en) | 2021-10-09 | 2021-10-09 | Online automatic focusing method and system in flying shooting mode |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113923358A true CN113923358A (en) | 2022-01-11 |
Family
ID=79238454
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111174787.XA Pending CN113923358A (en) | 2021-10-09 | 2021-10-09 | Online automatic focusing method and system in flying shooting mode |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113923358A (en) |
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114754598A (en) * | 2022-06-14 | 2022-07-15 | 深圳中宝新材科技有限公司 | Method and device for reducing power consumption of gold bonding wire production line imaging system |
| CN114813794A (en) * | 2022-02-18 | 2022-07-29 | 成都飞机工业（集团）有限责任公司 | 3D printing nondestructive testing method based on robot |
| CN115484371A (en) * | 2022-09-13 | 2022-12-16 | 菲特（天津）检测技术有限公司 | Image acquisition method, image acquisition device and readable storage medium |
| CN115672786A (en) * | 2022-11-28 | 2023-02-03 | 广州市普理司科技有限公司 | Intelligent plane detection system and detection method for printed matter |
| CN117782998A (en) * | 2024-02-27 | 2024-03-29 | 宁德时代新能源科技股份有限公司 | Battery detection method and system |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110069190A1 (en) * | 2009-09-18 | 2011-03-24 | Altek Corporation | Fast focusing method for digital camera |
| CN102707545A (en) * | 2012-06-21 | 2012-10-03 | 武汉烽火众智数字技术有限责任公司 | DFF-based auto-focusing method |
| CN106901712A (en) * | 2017-02-28 | 2017-06-30 | 广州医软智能科技有限公司 | A kind of microcirculation imaging device |
| CN110392200A (en) * | 2018-04-20 | 2019-10-29 | 杭州海康威视数字技术股份有限公司 | Automatically the method and apparatus focused |
| CN110441312A (en) * | 2019-07-30 | 2019-11-12 | 上海深视信息科技有限公司 | A kind of surface defects of products detection system based on multispectral imaging |
| CN112903694A (en) * | 2021-01-18 | 2021-06-04 | 苏州华兴源创科技股份有限公司 | Flying shoot detection system and method |
| WO2021134179A1 (en) * | 2019-12-30 | 2021-07-08 | 深圳市大疆创新科技有限公司 | Focusing method and apparatus, photographing device, movable platform and storage medium |
-
2021
- 2021-10-09 CN CN202111174787.XA patent/CN113923358A/en active Pending
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20110069190A1 (en) * | 2009-09-18 | 2011-03-24 | Altek Corporation | Fast focusing method for digital camera |
| CN102707545A (en) * | 2012-06-21 | 2012-10-03 | 武汉烽火众智数字技术有限责任公司 | DFF-based auto-focusing method |
| CN106901712A (en) * | 2017-02-28 | 2017-06-30 | 广州医软智能科技有限公司 | A kind of microcirculation imaging device |
| CN110392200A (en) * | 2018-04-20 | 2019-10-29 | 杭州海康威视数字技术股份有限公司 | Automatically the method and apparatus focused |
| CN110441312A (en) * | 2019-07-30 | 2019-11-12 | 上海深视信息科技有限公司 | A kind of surface defects of products detection system based on multispectral imaging |
| WO2021134179A1 (en) * | 2019-12-30 | 2021-07-08 | 深圳市大疆创新科技有限公司 | Focusing method and apparatus, photographing device, movable platform and storage medium |
| CN112903694A (en) * | 2021-01-18 | 2021-06-04 | 苏州华兴源创科技股份有限公司 | Flying shoot detection system and method |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114813794A (en) * | 2022-02-18 | 2022-07-29 | 成都飞机工业（集团）有限责任公司 | 3D printing nondestructive testing method based on robot |
| CN114813794B (en) * | 2022-02-18 | 2023-10-03 | 成都飞机工业（集团）有限责任公司 | Method for acquiring scanning photo required by robot 3D printing nondestructive testing |
| CN114754598A (en) * | 2022-06-14 | 2022-07-15 | 深圳中宝新材科技有限公司 | Method and device for reducing power consumption of gold bonding wire production line imaging system |
| CN115484371A (en) * | 2022-09-13 | 2022-12-16 | 菲特（天津）检测技术有限公司 | Image acquisition method, image acquisition device and readable storage medium |
| CN115672786A (en) * | 2022-11-28 | 2023-02-03 | 广州市普理司科技有限公司 | Intelligent plane detection system and detection method for printed matter |
| CN117782998A (en) * | 2024-02-27 | 2024-03-29 | 宁德时代新能源科技股份有限公司 | Battery detection method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN113923358A (en) | Online automatic focusing method and system in flying shooting mode | |
| CN108152869B (en) | Small step focusing method suitable for bionic vision rapid focusing | |
| CN108106603B (en) | Variable focus lens system with multi-stage extended depth of field image processing | |
| CN109451244B (en) | Automatic focusing method and system based on liquid lens | |
| CN105938243B (en) | More multiplying power microscope quick focusing methods in a kind of TFT LCD detections | |
| US10462351B2 (en) | Fast auto-focus in imaging | |
| CN107529011B (en) | A kind of motorized zoom lens control method | |
| CN109521547B (en) | Variable-step-length automatic focusing method and system | |
| CN104079827B (en) | A kind of optical field imaging weighs focusing method automatically | |
| CN109085113B (en) | Automatic focusing method and device for cervical exfoliated cell detection device | |
| US10534164B2 (en) | Digital microscope and focusing method thereof | |
| CN111083365B (en) | Method and device for rapidly detecting optimal focal plane position | |
| CN106210520B (en) | A kind of automatic focusing electronic eyepiece and system | |
| CN108259753A (en) | A kind of camera auto-focusing method and device that climbing method is improved based on defocus estimation | |
| CN104181685A (en) | Automatic digital slide focusing device and method based on microscope | |
| CN103529543A (en) | Automatic microscope focusing method | |
| CN112203012A (en) | Image definition calculation method, automatic focusing method and system | |
| CN110428463A (en) | The method that image automatically extracts center during aspherical optical element defocus blur is fixed | |
| CN114785959B (en) | Automatic focusing method and device for fluorescence microscope, storage medium and electronic equipment | |
| CN106154688A (en) | A kind of method and device of auto-focusing | |
| CN113899698A (en) | Real-time focusing and centering adjustment method and device for in-situ test platform | |
| CN116540502B (en) | Automatic focusing method and system for LDI equipment | |
| US11249225B2 (en) | Tunable acoustic gradient lens system utilizing amplitude adjustments for acquiring images focused at different z-heights | |
| JP2015102694A (en) | Alignment device, microscopic system, alignment method, and alignment program | |
| CN109318235B (en) | Quick focusing method of robot vision servo system |
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 | ||
| CB02 | Change of applicant information |
Country or region after: China Address after: Room 709-3, Building 2, No. 21 Intercity Road, High tech Zone, Suzhou City, Jiangsu Province, 215000 Applicant after: Suzhou Shenshi Information Technology Co.,Ltd. Address before: 200241, room 1027, building B, 555 Dongchuan Road, Shanghai, Minhang District Applicant before: SHANGHAI DEEPSIGHT INFORMATION TECHNOLOGY CO.,LTD. Country or region before: China |
|
| CB02 | Change of applicant information |