CN111862097A - Data enhancement method and device for micro defect detection rate - Google Patents
Data enhancement method and device for micro defect detection rate Download PDFInfo
- Publication number
- CN111862097A CN111862097A CN202011013634.2A CN202011013634A CN111862097A CN 111862097 A CN111862097 A CN 111862097A CN 202011013634 A CN202011013634 A CN 202011013634A CN 111862097 A CN111862097 A CN 111862097A
- Authority
- CN
- China
- Prior art keywords
- product image
- defect
- target
- type information
- image
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T7/00—Image analysis
- G06T7/0002—Inspection of images, e.g. flaw detection
- G06T7/0004—Industrial image inspection
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- General Physics & Mathematics (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Multimedia (AREA)
- Quality & Reliability (AREA)
- Image Processing (AREA)
Abstract
The invention provides a data enhancement method and a data enhancement device for a tiny defect detection rate, wherein the method comprises the following steps of: acquiring a product image; preprocessing the product image to obtain a product image meeting preset conditions; acquiring defect type information of the processed product image; analyzing the defect type information of the product image to obtain the type information of the target defect; and copying the target defect, and pasting the target defect at any position in the product image to finish the data enhancement of the product image. According to the data enhancement method, the number of the samples is increased by copying and pasting the defect target, on the premise of ensuring the authenticity of the increased samples, the diversity of the samples randomly appearing in the images is increased, the consistency of the sizes and the pixel densities of the training images and the test images is kept, and the micro defect detection rate of the back panel of the mobile phone is effectively improved.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to a data enhancement method for a tiny defect detection rate, a data enhancement device for the tiny defect detection rate, computer equipment and a non-transitory computer readable storage medium.
Background
At present, the industrial defect detection generally has the problems of few defect samples and small defective pixel proportion, and in order to solve the long-standing dilemma in the industrial field, technical schemes of increasing the resolution of an input image or fusing high-resolution features and high-dimensional features in a low-resolution image are proposed in the related art. However, this method using higher resolution increases computational overhead and does not address the imbalance between large and small objects.
Disclosure of Invention
The invention aims to solve the technical problems and provides a data enhancement method for the tiny defect detection rate, the number of samples is increased by copying and pasting defect targets, on the premise of ensuring the authenticity of the samples, the diversity of the samples randomly appearing in the images is increased, the consistency of the sizes and the pixel densities of the training images and the test images is kept, and the tiny defect detection rate of the mobile phone back panel is effectively improved.
The technical scheme adopted by the invention is as follows:
a data enhancement method for micro defect detection rate comprises the following steps: acquiring a product image; preprocessing the product image to obtain a product image meeting preset conditions; acquiring defect type information of the processed product image; analyzing the defect type information of the product image to obtain the type information of the target defect; and copying the target defects according to the type information of the target defects, and pasting the target defects at any position in the product image to finish the data enhancement of the product image.
According to an embodiment of the present invention, copying the target defect according to the type information of the target defect, and pasting the target defect at any position in the product image, to complete data enhancement of the product image, includes: replicating all of the target defects; pasting all the copied target defects at any position in the product image for a first preset time.
According to another embodiment of the present invention, copying the target defect according to the type information of the target defect, and pasting the target defect at any position in the product image, to complete data enhancement of the product image, includes: replicating a plurality of the target defects; pasting the copied target defects at any position in the product image for a second preset time.
According to another embodiment of the present invention, copying the target defect according to the type information of the target defect, and pasting the target defect at any position in the product image, to complete data enhancement of the product image, includes: replicating any of the target defects; pasting a third preset number of times of the copied target defect at any position in the product image.
According to another embodiment of the present invention, acquiring defect type information of the processed product image includes: describing the shape and the outline of the defect, and recording a point set and a wire frame which pass through when the shape and the outline of the defect are described; and determining the defect type of the product image according to the shape and the outline of the defect.
According to another embodiment of the present invention, a product image is acquired, comprising: moving the camera and the workpiece to a designated optical point location; setting parameters of a light source and a camera according to the optical surface information, and then shooting an original image of the workpiece; and adding the workpiece information to the original image of the workpiece to obtain the product image.
According to another embodiment of the present invention, the preprocessing the product image to obtain a product image satisfying a preset condition includes: when the product image comprises a plurality of workpieces, cutting the product image to respectively cut the workpieces; when the size of the product image is larger than a preset size, compressing the product image; when one workpiece in the product image is formed by combining a plurality of images, the images before combination are subjected to rotation and horizontal mirror image processing, and then the processed images are combined.
Corresponding to the method, the invention also provides a data enhancement device for the micro defect detection rate, which comprises the following steps: the first acquisition module is used for acquiring a product image; the processing module is used for preprocessing the product image to obtain a product image meeting preset conditions; the second acquisition module is used for acquiring the defect type information of the processed product image; the analysis module is used for analyzing the defect type information of the product image to obtain the type information of the target defect; and the enhancement module is used for copying the target defects according to the type information of the target defects and pasting the target defects at any position in the product image to finish the enhancement of the image data.
The invention also provides a computer device, which comprises a memory, a processor and a computer program stored on the memory and capable of running on the processor, wherein when the processor executes the program, the data enhancement method for the tiny defect detection rate is realized.
In response to the above method, the present invention further provides a non-transitory computer-readable storage medium having a computer program stored thereon, which when executed by a processor, implements the above data enhancement method for the small defect detection rate.
The invention has the beneficial effects that:
the invention increases the number of samples by copying and pasting the defect target, increases the diversity of the samples randomly appearing in the image on the premise of ensuring the authenticity of the increased samples, keeps the consistency of the sizes and the pixel densities of the training image and the test image, and effectively improves the detection rate of the tiny defects of the back plate of the mobile phone.
Drawings
FIG. 1 is a flow chart of a method for enhancing data of a small defect detection rate according to an embodiment of the present invention;
FIG. 2 is a block diagram illustrating an apparatus for enhancing a micro defect detection rate according to an embodiment of the present invention.
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.
Fig. 1 is a flowchart illustrating a method for enhancing data of a small defect detection rate according to an embodiment of the present invention.
As shown in fig. 1, the method for enhancing the detection rate of the micro defect according to the embodiment of the present invention may include the following steps:
and S1, acquiring a product image.
In one embodiment of the invention, acquiring a product image comprises: moving the camera and the workpiece to a designated optical point location; setting parameters of a light source and a camera according to the optical surface information, and then shooting an original image of the workpiece; and adding the workpiece information to an original image of the workpiece to obtain a product image.
That is, parameters such as exposure, a region-of-interest position, a line frequency, and a photographing delay are set for different products and different optical surfaces of the same product, and an image consistent with a reference optical surface (a reference image photographed by an optical engineer for an optimal optical scheme for each optical surface of the product) is taken.
Specifically, the steps and algorithms for obtaining the product image are as follows: firstly, moving a camera and a workpiece to a position of a specified optical surface point; then, setting light source and camera parameters (wherein the camera parameters comprise an exposure value, a gamma value, a line frequency, an area of interest and the like) according to optical surface information (related information such as an optimal focal length, an optimal angle, an optimal exposure value and the like obtained by comparing the brightness, the optimal angle, the optimal area and the like of an optical reference image and a currently shot image by an optical guidance algorithm), and triggering the camera to take a picture; then, receiving an original image of the workpiece sent back by the camera, and adding workpiece information (wherein the workpiece information comprises a workpiece number and a channel number) corresponding to the original image into the head of the original image (wherein the head information comprises the channel number, the workpiece number, a source image number and the like; and the image information comprises data information, image width and the like); and finally, storing the original image of the workpiece with the head information to obtain a product image.
And S2, preprocessing the product image to obtain the product image meeting the preset condition.
According to one embodiment of the present invention, preprocessing a product image to obtain a product image satisfying a preset condition includes: when the product image comprises a plurality of workpieces, cutting the product image to respectively cut the workpieces; when the size of the product image is larger than the preset size, compressing the product image; when one workpiece in the product image is formed by combining a plurality of images, the images before combination are rotated and horizontally mirrored, and then the processed images are combined.
Specifically, according to information such as the number of workpiece stages and the number of machine channels, algorithm operations such as cutting, compression, rotation, horizontal mirroring and vertical mirroring are carried out on the original camera image, and an image which meets the detection requirements of the depth learning model is output (wherein the image which meets the requirements is clear and visible, a fuzzy area does not exist, the image size is consistent, the workpiece exists in an area in the middle of the image, and a large number of areas with black backgrounds in the image are cut out). The number of the workpiece stages and the number of the machine station channels influence the preprocessing strategy of the acquired images, and the preprocessed final output images can be the splicing result of the images acquired on different stages or different machine stations. For example, when a plurality of workpieces are included in one product image, the workpieces in the image need to be cut out respectively; for another example, when the size of the product image is larger than the preset size, the product image needs to be compressed to be smaller (wherein, the compressed image is better when the image compression is smaller on the premise that the defect features in the image are obviously visible by naked eyes and the defect features are obviously distinguishable from the background), so as to improve the model operation speed; for another example, when a workpiece needs to be formed by combining a plurality of product images, the product images need to be combined after being rotated and mirrored.
And S3, acquiring the defect type information of the processed product image.
According to one embodiment of the invention, acquiring defect type information of a processed product image comprises the following steps: describing the shape and the outline of the defect, and recording a point set and a wire frame which pass through when the shape and the outline of the defect are described; and determining the defect type of the product image according to the shape and the outline of the defect.
Specifically, taking a product image as a mobile phone backboard as an example, a defect of the mobile phone backboard is described along a defect shape and an outline, and a defect type field is used to distinguish a marked defect type, for example, a name of the defect can be used for marking, where the defect type includes: black points, bubbles and the like, wherein different defect types are distinguished by adopting different fields, and meanwhile, a point set and a wire frame which pass through when the shape and the outline of the defect are drawn are recorded and stored in a JSON format description file for marking the defect so as to be called and analyzed subsequently.
And S4, analyzing the defect type information of the product image to obtain the type information of the target defect.
Specifically, reading and analyzing a description file for marking the defect, and extracting data information containing a target defect type field, wherein the data information comprises: and the point set and the wire frame are used for representing the outer contour information of the target defect in the image. That is, if the number of black dots in the product image is large and the number of bubbles is small, the bubbles are target defects, and the outline information of all the bubbles in the description file is acquired.
And S5, copying the target defect according to the type information of the target defect, and pasting the target defect at any position in the product image to complete the enhancement of the image data.
In the embodiment of the present invention, there are various ways to copy and paste the target defect, and three different strategies are described as examples below.
According to one embodiment of the invention, the data enhancement of the product image is completed by copying the target defect according to the type information of the target defect and pasting the target defect at any position in the product image, and the data enhancement comprises the following steps: copying any one of the target defects; pasting a copied target defect at any position in the product image for a third preset time. The third preset number may be calibrated according to an actual situation, for example, the third preset number may be calibrated according to the number of target defects, where the smaller the number of target defects is, the larger the third preset number is, and vice versa.
That is to say, the type information of the defect with a small number is obtained, the outer contour information of all bubbles is obtained by taking the bubble number as an example, the outer contour information of one bubble is selected as a copy object, and the copy object is pasted on the random position of the product image for multiple times, so that the purpose of increasing the number of defect samples can be achieved, and the data enhancement is realized.
According to another embodiment of the present invention, the data enhancement of the product image is completed by copying the target defect according to the type information of the target defect and pasting the target defect at an arbitrary position in the product image, including: replicating a plurality of the target defects; pasting the plurality of copied target defects at any position in the product image for a second preset number of times. And the second preset times can be calibrated according to the actual condition.
That is, taking the example that the number of bubbles is small, the outer contour information of all bubbles is obtained, the outer contour information of a plurality of bubbles is selected as a copy object, and the copy object is pasted on the random position of the product image for a plurality of times, so that the purpose of increasing the number of defect samples can be achieved, and data enhancement is realized.
According to another embodiment of the present invention, the data enhancement of the product image is completed by copying the target defect according to the type information of the target defect and pasting the target defect at an arbitrary position in the product image, including: copying all of the target defects; pasting all the copied target defects at any position in the product image for a first preset time. The first preset times can be calibrated according to actual conditions.
That is, taking the example that the number of bubbles is small, the outer contour information of all bubbles is obtained, the outer contour information of all bubbles is taken as a copy object, and the copy object is pasted on the random position of the product image for many times, so that the purpose of increasing the number of defect samples can be achieved, data enhancement is realized, and the original product image and the enhanced product image are retained.
By analyzing the three different copy and paste strategies, it is found that when a target defect is pasted, no defect overlap is introduced, and when the pasted target defect is not smoothed, the data enhancement effect is optimal, so that defect overlap needs to be avoided.
In one embodiment of the present invention, the position information of the defective object existing in the image is known, and when pasting, whether the position to be pasted and the position of the defective object have an intersection is calculated, if there is an intersection, the pasting is cancelled, and when there is no intersection, the pasting is performed.
In one embodiment of the present invention, the location information of the pasted target defect is also written into a new description file, and finally the enhancement of the image data is completed. The position information comprises corner point coordinates and width and height representation of the target defect, the position information of the target defect object is read from the label file during copying, and the current position point and the read width and height value are written into the label file as the position information of the pasting object during pasting.
In conclusion, the number of samples is increased by copying and pasting the defect target, the diversity of the samples randomly appearing in the image is increased on the premise of ensuring the authenticity of the increased samples, the consistency of the sizes and the pixel densities of the training image and the test image is kept, and the micro defect detection rate of the back panel of the mobile phone is effectively improved.
The invention further provides a data enhancing device for the tiny defect detection rate, which corresponds to the data enhancing method for the tiny defect detection rate of the embodiment.
FIG. 2 is a block diagram illustrating an apparatus for enhancing a micro defect detection rate according to an embodiment of the present invention.
As shown in fig. 2, the data enhancement apparatus for the micro defect detection rate according to the embodiment of the present invention may include: a first obtaining module 10, a processing module 20, a second obtaining module 30, a parsing module 40 and an enhancing module 50.
The first acquiring module 10 is used for acquiring a product image. The processing module 20 is configured to pre-process the product image to obtain a product image satisfying a preset condition. The second obtaining module 30 is used for obtaining defect type information of the processed product image. The analyzing module 40 is configured to analyze the defect type information of the product image to obtain the type information of the target defect. The enhancement module 50 is configured to copy the target defect according to the type information of the target defect, and paste the target defect at any position in the product image, thereby completing the enhancement of the image data.
According to an embodiment of the present invention, the enhancement module 50 is specifically configured to copy all of the target defects when the data enhancement of the product image is completed by copying the target defects according to the type information of the target defects and pasting the target defects at any position in the product image; pasting all the copied target defects at any position in the product image for a first preset time.
According to another embodiment of the present invention, the enhancement module 50 is specifically configured to copy a plurality of target defects when the data enhancement of the product image is completed by copying the target defects according to the type information of the target defects and pasting the target defects at any position in the product image; pasting the plurality of copied target defects at any position in the product image for a second preset number of times.
According to another embodiment of the present invention, the enhancement module 50 is specifically configured to copy any one of the target defects when the data enhancement of the product image is completed by copying the target defects according to the type information of the target defects and pasting the target defects at any position in the product image; pasting a copied target defect at any position in the product image for a third preset time.
According to an embodiment of the present invention, the second obtaining module 30, when obtaining the defect type information of the processed product image, is specifically configured to describe the shape and contour of the defect, and record a point set and a wire frame that pass when describing the shape and contour of the defect; and determining the defect type of the product image according to the shape and the outline of the defect.
According to one embodiment of the present invention, the first acquiring module 10 is specifically configured to move the camera and the workpiece to a specified optical point location when acquiring the product image; setting parameters of a light source and a camera according to the optical surface information, and then shooting an original image of the workpiece; and adding the workpiece information to an original image of the workpiece to obtain a product image.
According to an embodiment of the present invention, the processing module 20, when preprocessing the product image to obtain a product image satisfying a preset condition, is specifically configured to, when the product image includes a plurality of workpieces, perform a cutting process on the product image to respectively cut the plurality of workpieces; when the size of the product image is larger than the preset size, compressing the product image; when one workpiece in the product image is formed by combining a plurality of images, the images before combination are rotated and horizontally mirrored, and then the processed images are combined.
It should be noted that, for details not disclosed in the data enhancement apparatus for the detection rate of small defects of the present invention, please refer to details disclosed in the data enhancement method for the detection rate of small defects of the embodiment of the present invention, and detailed descriptions thereof are omitted here.
The invention increases the number of samples by copying and pasting the defect target, increases the diversity of the samples randomly appearing in the image on the premise of ensuring the authenticity of the increased samples, keeps the consistency of the sizes and the pixel densities of the training image and the test image, and effectively improves the detection rate of the tiny defects of the back plate of the mobile phone.
The invention further provides a computer device corresponding to the embodiment.
The computer device according to the embodiment of the present invention includes a memory, a processor, and a computer program stored in the memory and executable on the processor, and when the processor executes the computer program, the method for enhancing data of the small defect detection rate according to the above embodiment of the present invention may be implemented.
According to the computer equipment provided by the embodiment of the invention, when the processor executes the computer program stored on the memory, the product image is firstly acquired and preprocessed to obtain the product image meeting the preset conditions, then the defect type information of the processed product image is acquired and analyzed to obtain the type information of the target defect, and finally the target defect is copied according to the type information of the target defect and pasted at any position in the product image to complete the data enhancement of the product image. .
The invention also provides a non-transitory computer readable storage medium corresponding to the above embodiment.
A non-transitory computer-readable storage medium of an embodiment of the present invention has stored thereon a computer program that, when executed by a processor, can implement the data enhancement method of the minute defect detection rate according to the above-described embodiment of the present invention.
According to a non-transitory computer-readable storage medium of an embodiment of the present invention, when a processor executes a computer program stored thereon, firstly, obtaining a product image, preprocessing the product image to obtain a product image meeting preset conditions, then acquiring the defect type information of the processed product image, analyzing the defect type information of the product image, to obtain the type information of the target defect, and finally copying the target defect according to the type information of the target defect, and the target defect is pasted at any position in the product image to complete the data enhancement of the product image, therefore, on the premise of ensuring the authenticity of the sample to be increased, the diversity of the random appearance of the sample in the image is increased, and the consistency of the sizes and the pixel densities of the training image and the test image is kept, and the detection rate of the tiny defects of the back plate of the mobile phone is effectively improved. .
In the description of the present invention, the terms "first" and "second" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance or implying any number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. The meaning of "plurality" is two or more unless specifically limited otherwise.
In the present invention, unless otherwise expressly stated or limited, the terms "mounted," "connected," "secured," and the like are to be construed broadly and can, for example, be fixedly connected, detachably connected, or integrally formed; can be mechanically or electrically connected; either directly or indirectly through intervening media, either internally or in any other relationship. The specific meanings of the above terms in the present invention can be understood by those skilled in the art according to specific situations.
In the present invention, unless otherwise expressly stated or limited, the first feature "on" or "under" the second feature may be directly contacting the first and second features or indirectly contacting the first and second features through an intermediate. Also, a first feature "on," "over," and "above" a second feature may be directly or diagonally above the second feature, or may simply indicate that the first feature is at a higher level than the second feature. A first feature being "under," "below," and "beneath" a second feature may be directly under or obliquely under the first feature, or may simply mean that the first feature is at a lesser elevation than the second feature.
In the description herein, references to the description of the term "one embodiment," "some embodiments," "an example," "a specific example," or "some examples," etc., mean that a particular feature, structure, material, or characteristic described in connection with the embodiment or example is included in at least one embodiment or example of the invention. In this specification, the schematic representations of the terms used above are not necessarily intended to refer to the same embodiment or example. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples. Furthermore, various embodiments or examples and features of different embodiments or examples described in this specification can be combined and combined by one skilled in the art without contradiction.
Any process or method descriptions in flow charts or otherwise described herein may be understood as representing modules, segments, or portions of code which include one or more executable instructions for implementing specific logical functions or steps of the process, and alternate implementations are included within the scope of the preferred embodiment of the present invention in which functions may be executed out of order from that shown or discussed, including substantially concurrently or in reverse order, depending on the functionality involved, as would be understood by those reasonably skilled in the art of the present invention.
The logic and/or steps represented in the flowcharts or otherwise described herein, e.g., an ordered listing of executable instructions that can be considered to implement logical functions, can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, processor-containing system, or other system that can fetch the instructions from the instruction execution system, apparatus, or device and execute the instructions. For the purposes of this description, a "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device. More specific examples (a non-exhaustive list) of the computer-readable medium would include the following: an electrical connection (electronic device) having one or more wires, a portable computer diskette (magnetic device), a Random Access Memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium could even be paper or another suitable medium upon which the program is printed, as the program can be electronically captured, via for instance optical scanning of the paper or other medium, then compiled, interpreted or otherwise processed in a suitable manner if necessary, and then stored in a computer memory.
It should be understood that portions of the present invention may be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, the various steps or methods may be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one or combination of the following techniques, which are known in the art, may be used: a discrete logic circuit having a logic gate circuit for implementing a logic function on a data signal, an application specific integrated circuit having an appropriate combinational logic gate circuit, a Programmable Gate Array (PGA), a Field Programmable Gate Array (FPGA), or the like.
It will be understood by those skilled in the art that all or part of the steps carried by the method for implementing the above embodiments may be implemented by hardware related to instructions of a program, which may be stored in a computer readable storage medium, and when the program is executed, the program includes one or a combination of the steps of the method embodiments.
In addition, functional units in the embodiments of the present invention may be integrated into one processing module, or each unit may exist alone physically, or two or more units are integrated into one module. The integrated module can be realized in a hardware mode, and can also be realized in a software functional module mode. The integrated module, if implemented in the form of a software functional module and sold or used as a stand-alone product, may also be stored in a computer readable storage medium.
The storage medium mentioned above may be a read-only memory, a magnetic or optical disk, etc. Although embodiments of the present invention have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting the present invention, and that variations, modifications, substitutions and alterations can be made to the above embodiments by those of ordinary skill in the art within the scope of the present invention.
Claims (10)
1. A data enhancement method for micro defect detection rate is characterized by comprising the following steps:
acquiring a product image;
preprocessing the product image to obtain a product image meeting preset conditions;
acquiring defect type information of the processed product image;
analyzing the defect type information of the product image to obtain the type information of the target defect;
and copying the target defects according to the type information of the target defects, and pasting the target defects at any position in the product image to finish the data enhancement of the product image.
2. The method for enhancing the detection rate of the micro-defects according to claim 1, wherein the data enhancement of the product image is completed by copying the target defect according to the type information of the target defect and pasting the target defect at any position in the product image, and the method comprises the following steps:
replicating all of the target defects;
pasting all the copied target defects at any position in the product image for a first preset time.
3. The method for enhancing the detection rate of the micro-defects according to claim 1, wherein the data enhancement of the product image is completed by copying the target defect according to the type information of the target defect and pasting the target defect at any position in the product image, and the method comprises the following steps:
replicating a plurality of the target defects;
pasting the copied target defects at any position in the product image for a second preset time.
4. The method for enhancing the detection rate of the micro-defects according to claim 1, wherein the data enhancement of the product image is completed by copying the target defect according to the type information of the target defect and pasting the target defect at any position in the product image, and the method comprises the following steps:
replicating any of the target defects;
pasting a third preset number of times of the copied target defect at any position in the product image.
5. The method for enhancing the detection rate of the micro-defects according to claim 1, wherein obtaining the defect type information of the processed product image comprises:
describing the shape and the outline of the defect, and recording a point set and a wire frame which pass through when the shape and the outline of the defect are described;
and determining the defect type of the product image according to the shape and the outline of the defect.
6. The method for enhancing the detection rate of the micro-defects according to claim 1, wherein the obtaining of the product image comprises:
moving the camera and the workpiece to a designated optical point location;
setting parameters of a light source and a camera according to the optical surface information, and then shooting an original image of the workpiece;
and adding the workpiece information to the original image of the workpiece to obtain the product image.
7. The method for enhancing the micro-defect detection rate according to claim 6, wherein the preprocessing the product image to obtain a product image satisfying a preset condition comprises:
when the product image comprises a plurality of workpieces, cutting the product image to respectively cut the workpieces;
when the size of the product image is larger than a preset size, compressing the product image;
when one workpiece in the product image is formed by combining a plurality of images, the images before combination are subjected to rotation and horizontal mirror image processing, and then the processed images are combined.
8. A data enhancement apparatus for a minute defect detection rate, comprising:
the first acquisition module is used for acquiring a product image;
the processing module is used for preprocessing the product image to obtain a product image meeting preset conditions;
the second acquisition module is used for acquiring the defect type information of the processed product image;
the analysis module is used for analyzing the defect type information of the product image to obtain the type information of the target defect;
and the enhancement module is used for copying the target defects according to the type information of the target defects, pasting the target defects at any position in the product image and finishing the data enhancement of the product image.
9. A computer device comprising a memory, a processor and a computer program stored on the memory and executable on the processor, wherein the processor implements the data enhancement method for the fine defect detection rate according to any one of claims 1 to 7 when executing the program.
10. A non-transitory computer-readable storage medium having stored thereon a computer program, wherein the program, when executed by a processor, implements the data enhancement method for the minute defect detection rate according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011013634.2A CN111862097A (en) | 2020-09-24 | 2020-09-24 | Data enhancement method and device for micro defect detection rate |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011013634.2A CN111862097A (en) | 2020-09-24 | 2020-09-24 | Data enhancement method and device for micro defect detection rate |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111862097A true CN111862097A (en) | 2020-10-30 |
Family
ID=72968512
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011013634.2A Pending CN111862097A (en) | 2020-09-24 | 2020-09-24 | Data enhancement method and device for micro defect detection rate |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111862097A (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112733886A (en) * | 2020-12-24 | 2021-04-30 | 西人马帝言(北京)科技有限公司 | Sample image processing method, device, equipment and storage medium |
| CN113570565A (en) * | 2021-07-21 | 2021-10-29 | 清华大学 | Vehicle appearance defect detection method and device and storage medium |
| CN113947771A (en) * | 2021-10-15 | 2022-01-18 | 北京百度网讯科技有限公司 | Image recognition method, apparatus, device, storage medium, and program product |
| CN114419029A (en) * | 2022-03-11 | 2022-04-29 | 深圳艾灵网络有限公司 | Training method of surface defect detection model, surface defect detection method and device |
| CN117392678A (en) * | 2023-12-06 | 2024-01-12 | 常州微亿智造科技有限公司 | Defect detection method and system |
| CN117474904A (en) * | 2023-12-26 | 2024-01-30 | 北京阿丘机器人科技有限公司 | Industrial defect detection method, device, equipment and storage medium |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN110992318A (en) * | 2019-11-19 | 2020-04-10 | 上海交通大学 | Special metal flaw detection system based on deep learning |
| CN111582294A (en) * | 2019-03-05 | 2020-08-25 | 慧泉智能科技（苏州）有限公司 | Method for constructing convolutional neural network model for surface defect detection and application thereof |
-
2020
- 2020-09-24 CN CN202011013634.2A patent/CN111862097A/en active Pending
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111582294A (en) * | 2019-03-05 | 2020-08-25 | 慧泉智能科技（苏州）有限公司 | Method for constructing convolutional neural network model for surface defect detection and application thereof |
| CN110992318A (en) * | 2019-11-19 | 2020-04-10 | 上海交通大学 | Special metal flaw detection system based on deep learning |
Non-Patent Citations (1)
| Title |
|---|
| 高如新 等: "基于图像处理的手机屏幕缺陷检测系统研究", 《测控技术》 * |
Cited By (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112733886A (en) * | 2020-12-24 | 2021-04-30 | 西人马帝言(北京)科技有限公司 | Sample image processing method, device, equipment and storage medium |
| CN113570565A (en) * | 2021-07-21 | 2021-10-29 | 清华大学 | Vehicle appearance defect detection method and device and storage medium |
| CN113947771A (en) * | 2021-10-15 | 2022-01-18 | 北京百度网讯科技有限公司 | Image recognition method, apparatus, device, storage medium, and program product |
| CN114419029A (en) * | 2022-03-11 | 2022-04-29 | 深圳艾灵网络有限公司 | Training method of surface defect detection model, surface defect detection method and device |
| CN114419029B (en) * | 2022-03-11 | 2022-06-24 | 深圳艾灵网络有限公司 | Training method of surface defect detection model, surface defect detection method and device |
| CN117392678A (en) * | 2023-12-06 | 2024-01-12 | 常州微亿智造科技有限公司 | Defect detection method and system |
| CN117474904A (en) * | 2023-12-26 | 2024-01-30 | 北京阿丘机器人科技有限公司 | Industrial defect detection method, device, equipment and storage medium |
| CN117474904B (en) * | 2023-12-26 | 2024-03-12 | 北京阿丘机器人科技有限公司 | Industrial defect detection method, device, equipment and storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111862097A (en) | Data enhancement method and device for micro defect detection rate | |
| JP6871314B2 (en) | Object detection method, device and storage medium | |
| US9773302B2 (en) | Three-dimensional object model tagging | |
| US7460732B2 (en) | Method and system for object recognition using fractal maps | |
| US7257267B2 (en) | Method and system for image segmentation | |
| CN112505056A (en) | Defect detection method and device | |
| US9046496B2 (en) | Capturing method for images with different view-angles and capturing system using the same | |
| CN113176270B (en) | Dimming method, device and equipment | |
| US20030067462A1 (en) | Evaluating method, generating method and apparatus for three-dimensional shape model | |
| US11928805B2 (en) | Information processing apparatus, information processing method, and storage medium for defect inspection and detection | |
| CN116503414B (en) | Screen defect detection method, device, computer equipment and storage medium | |
| CN115937147B (en) | Defect detection parameter determining method, device, equipment and storage medium | |
| CN114359161A (en) | Defect detection method, device, equipment and storage medium | |
| CN116342589B (en) | Cross-field scratch defect continuity detection method and system | |
| CN115205288B (en) | Industrial defect detection method and device | |
| CN116681677A (en) | Lithium battery defect detection method, device and system | |
| CN113192070B (en) | Weld image processing method and system based on computer vision | |
| CN113256607B (en) | Defect detection method and device | |
| CN116703803A (en) | Method, device and system for detecting burrs at edges of battery pole pieces | |
| CN111935480B (en) | Detection method for image acquisition device and related device | |
| CN114463405A (en) | Method, device and system for accelerating surface scanning line laser 3D camera and FPGA | |
| CN115810014B (en) | Electrode cap end face defect detection method based on sub-image blocks in image | |
| CN116046791B (en) | Method and device for detecting defect of adhesive dispensing | |
| JPH06281421A (en) | Image processing method | |
| JPH0319990B2 (en) |
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: 20201030 |