CN115471458A

CN115471458A - Image processing method and system suitable for electronic paper surface defects

Info

Publication number: CN115471458A
Application number: CN202211019798.5A
Authority: CN
Inventors: 张金越; 王昕彤; 黄祖成; 杨根; 江旭耀; 钟名锋; 王卫军
Original assignee: Guangzhou Institute Of Advanced Technology
Current assignee: Guangzhou Institute Of Advanced Technology
Priority date: 2022-08-24
Filing date: 2022-08-24
Publication date: 2022-12-13

Abstract

The invention belongs to the technical field of electronic paper surface defect detection, and particularly relates to an image processing method, system, platform and storage medium suitable for electronic paper surface defects. The method comprises the steps of acquiring an electronic paper image with a significant defect in real time, and preprocessing the electronic paper image in real time; according to the real-time preprocessing state of the electronic paper image, combined with the constructed detection algorithm, the electronic paper surface defect detection data are generated in real time, after fusion processing, the pictures to be detected are reduced from 2n to 2, the detection time is greatly reduced, and due to the fusion effect, the gray value difference of the suspected defect position and the gray value of other normal positions is increased, so that the surface defect of the product is highlighted to the greatest extent.

Description

Image processing method and system suitable for electronic paper surface defects

Technical Field

The invention belongs to the technical field of electronic paper surface defect detection, and particularly relates to an image processing method, system, platform and storage medium suitable for electronic paper surface defects.

Background

At the present stage, the surface defect detection system for electronic paper is not put into mass production, and most of the surface defect detection system is in a research and development state. Moreover, because the number of the collected images is large, the problems of large workload, long time consumption and the like of a computer can occur when each image is detected.

Therefore, in the above current stage, the surface defect detection system for electronic paper has not been put into mass production, and most of them are in research and development. Moreover, because the number of the acquired images is large, the technical problems of large workload and long time consumption of a computer can occur when each image is detected, and a method, a system, a platform and a storage medium suitable for processing the surface defect image of the electronic paper are urgently needed to be designed and developed.

Disclosure of Invention

In order to overcome the defects and difficulties of the prior art, the present invention provides a method, a system, a platform and a storage medium for processing an image of a surface defect of electronic paper, wherein after fusion processing, the image to be detected is reduced from 2n to 2, the detection time is greatly reduced, and due to the fusion effect, the gray value difference between the position of a suspected defect and the gray values of other normal positions is increased, so as to maximally highlight the surface defect of the product.

The first purpose of the invention is to provide an image processing method suitable for the surface defect of the electronic paper;

the second purpose of the invention is to provide an image processing system suitable for the surface defect of the electronic paper;

the third purpose of the invention is to provide an image processing platform suitable for the electronic paper surface defect;

a fourth object of the present invention is to provide a computer-readable storage medium;

the first object of the present invention is achieved by: the method specifically comprises the following steps:

acquiring an electronic paper image with a significant defect in real time, and preprocessing the electronic paper image in real time;

and generating the detection data of the surface defects of the electronic paper in real time according to the real-time preprocessing state of the electronic paper image and by combining the constructed detection algorithm.

Further, the method for generating the electronic paper surface defect detection data in real time according to the real-time preprocessing state of the electronic paper image and by combining the constructed detection algorithm further comprises the following steps:

acquiring a plurality of images in the process of changing from white gray scale to black gray scale by a feature identification method;

and carrying out fusion processing on the acquired multiple images in real time.

Further, the process of fusing the acquired multiple images in real time further includes the following steps:

fusing the images of the light gray scales according to the gray value of the full white;

and fusing the dark gray level pictures according to the full black gray level value.

Further, in the process of fusing the image with the light color gray scale according to the full white gray scale value, a formula when the light color gray scale is fused is specifically as follows:

P′(n)＝255-P(n) (I)

wherein P (n) is the nth picture, and P' (n) isAs a result of the inversion of the nth picture,

the average gray value of the P' (n) picture is obtained, and P is the picture obtained after fusion.

Further, in the process of fusing the picture with the dark gray scale according to the all-black gray scale value, the formula for fusing the dark gray scale is specifically as follows:

wherein P (n) is the nth picture,

the average gray value of the P (n) pictures is shown, and P is the picture obtained after fusion.

Further, the real-time acquisition of the electronic paper image with the significant defect and the real-time preprocessing of the electronic paper image further include the following steps:

performing real-time segmentation processing on the acquired electronic paper image; and carrying out pose correction processing on the electronic paper.

The second object of the present invention is achieved by: the system specifically comprises:

the image acquisition and preprocessing unit is used for acquiring an electronic paper image with a significant defect in real time and preprocessing the electronic paper image in real time; and the significant defect detection data generation unit is used for generating the surface defect detection data of the electronic paper in real time according to the real-time preprocessing state of the electronic paper image and by combining the constructed detection algorithm.

Further, in the image acquisition preprocessing unit, there are also provided:

the system is used for segmenting and processing the acquired electronic paper image in real time; the segmentation correction module is used for carrying out pose correction processing on the electronic paper;

the significant defect detection data generation unit is further provided with:

the first acquisition module is used for acquiring a plurality of images in the process from white gray scale to black gray scale through a feature identification method; the first fusion processing module is used for performing fusion processing on the acquired multiple images in real time;

in the first fusion processing module, still be provided with:

the second fusion processing module is used for fusing the images with the light gray scales according to the full white gray value;

and the third fusion processing module is used for fusing the dark gray level pictures according to the full black gray level value.

The third object of the present invention is achieved by: the method comprises the following steps: the system comprises a processor, a memory and a control program of an image processing platform suitable for the surface defects of the electronic paper;

the control program of the image processing platform suitable for the surface defect of the electronic paper is executed in the processor, the control program of the image processing platform suitable for the surface defect of the electronic paper is stored in the memory, and the control program of the image processing platform suitable for the surface defect of the electronic paper realizes the image processing method suitable for the surface defect of the electronic paper.

The fourth object of the present invention is achieved by: the computer readable storage medium stores a control program of an image processing platform suitable for the surface defects of the electronic paper, and the control program of the image processing platform suitable for the surface defects of the electronic paper realizes the image processing method suitable for the surface defects of the electronic paper.

The method comprises the steps of acquiring an electronic paper image with a significant defect in real time, and preprocessing the electronic paper image in real time; according to the real-time preprocessing state of the electronic paper image, combined with the constructed detection algorithm, the electronic paper surface defect detection data are generated in real time, after fusion processing, the pictures to be detected are reduced from 2n to 2, the detection time is greatly reduced, and due to the fusion effect, the gray value difference of the suspected defect position and the gray value of other normal positions is increased, so that the surface defect of the product is highlighted to the greatest extent.

Drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.

FIG. 1 is a schematic view of an overall structure of an apparatus for detecting surface defects of electronic paper according to the present invention;

FIG. 2 is a schematic diagram of a feeding mechanism of an apparatus for detecting surface defects of electronic paper according to the present invention;

FIG. 3 is a schematic diagram of a positioning module of a positioning mechanism of an electronic paper surface defect detecting device according to the present invention;

FIG. 4 is a schematic diagram of a positioning mechanism for an electronic paper surface defect detecting apparatus according to the present invention;

FIG. 5 is a schematic diagram of a detecting mechanism for an electronic paper surface defect detecting apparatus according to the present invention;

FIG. 6 is a schematic view of an electronic paper lighting structure of a detecting mechanism of the electronic paper surface defect detecting device according to the present invention;

FIG. 7 is a schematic view of a feeding mechanism for an electronic paper surface defect detecting apparatus according to the present invention;

FIG. 8 is a schematic diagram of a single camera multi-electronic paper region segmentation method for processing an image of a surface defect of electronic paper according to the present invention;

FIG. 8a is a schematic diagram of a single camera with multiple electronic paper area segmentation for capturing pictures according to the present invention;

FIG. 8b is a diagram illustrating binary results of multi-electron paper region segmentation of a single camera according to the present invention;

FIG. 8c is a schematic diagram of a single camera with multiple electronic paper regions dividing the target region;

FIG. 8d is the schematic view of the electronic paper with multiple electronic paper areas divided by the single camera according to the present invention (left);

FIG. 8e is the schematic diagram of the electronic paper (right) divided by multiple electronic paper areas of the single camera according to the present invention;

FIG. 9 is a schematic diagram of a multi-gray-scale image region clipping suitable for an image processing method of electronic paper surface defects according to the present invention;

FIG. 9a is a schematic diagram of an original image captured from a multi-gray-scale picture region;

FIG. 9b is a schematic view of the electronic paper (left) cut from the multi-gray-scale image region;

FIG. 9c is a schematic view of the electronic paper (right) cut from the multi-gray level image region;

FIG. 10 is a schematic diagram of edge extraction and correction of an electronic paper suitable for an image processing method of surface defects of an electronic paper according to the present invention;

FIG. 10a is a schematic diagram of an electronic paper separated by edge extraction and correction;

FIG. 10b is a diagram illustrating the binary image of edge extraction and correction of electronic paper;

FIG. 10c is a schematic diagram of edge extraction and correction for electronic paper;

FIG. 10d is a schematic diagram showing the edge capturing line for edge extraction and correction of the electronic paper;

FIG. 10e is a schematic diagram of the correction of the edge extraction and correction of the electronic paper to obtain a transformation matrix;

FIG. 10f is a schematic diagram of the correction of an original image by a transformation matrix for edge extraction and correction of electronic paper;

FIG. 11 is a schematic diagram of an image effect of an image acquisition method for processing a defect image on an electronic paper according to the present invention;

FIG. 12 is a schematic diagram illustrating an image enhancement result of multi-gray scale fusion applied to an image processing method for surface defects of electronic paper according to the present invention;

FIG. 13 is a schematic diagram of an insignificant defect image processing method for electronic paper surface defects according to the present invention;

FIG. 14 is a schematic diagram of a defect detection process based on deep learning for an electronic paper surface defect image processing method according to the present invention;

FIG. 15 is a schematic diagram of a deep learning defect detection process based on extended samples for an electronic paper surface defect image processing method according to the present invention;

FIG. 16 is a flowchart illustrating a method for processing a defect image on an electronic paper according to the present invention;

FIG. 17 is a schematic diagram of an image processing system for electronic paper surface defects according to the present invention;

FIG. 18 is a schematic diagram of an image processing platform for electronic paper surface defects according to the present invention;

FIG. 19 is a block diagram of a computer readable storage medium according to an embodiment of the present invention;

in the figure:

1-a feeding mechanism; 2-a positioning mechanism; 201-a positioning module; 3-a detection mechanism; 4-a blanking mechanism; 5, a mechanical arm; 6-pushing the module; 601-a first pushing module; 602-a second pushing module; 7-a limiting block; 701-a first limiting block; 702-a second stopper; 8-an adsorption plate; 9-a transport mechanism; 10-camera/light source; 11-electronic paper lighting device; 1101-electronic paper drive contacts;

the objects, features and advantages of the present invention will be further explained with reference to the accompanying drawings.

Detailed Description

For better understanding of the objects, aspects and advantages of the present invention, reference will now be made to the following detailed description taken in conjunction with the accompanying drawings, and other advantages and capabilities of the present invention will become apparent to those skilled in the art from the description.

The invention is capable of other and different embodiments and its several details are capable of modification in various other respects, all without departing from the spirit and scope of the present invention.

It should be noted that, if directional indications (such as up, down, left, right, front, and back … …) are involved in the embodiment of the present invention, the directional indications are only used to explain the relative position relationship between the components, the motion situation, and the like in a specific posture (as shown in the drawing), and if the specific posture is changed, the directional indication is changed accordingly.

In addition, if there is a description of "first", "second", etc. in an embodiment of the present invention, the description of "first", "second", etc. is for descriptive purposes only and is not to be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one such feature. Secondly, technical solutions between the embodiments may be combined with each other, but it must be based on the realization of those skilled in the art, and when the technical solutions are contradictory or cannot be realized, such a combination of technical solutions should be considered to be absent and not be within the protection scope of the present invention.

Preferably, the image processing method for the electronic paper surface defect is applied to one or more terminals or servers. The terminal is a device capable of automatically performing numerical calculation and/or information processing according to a preset or stored instruction, and the hardware includes, but is not limited to, a microprocessor, an Application Specific Integrated Circuit (ASIC), a Programmable Gate Array (FPGA), a Digital Signal Processor (DSP), an embedded device, and the like.

The terminal can be a desktop computer, a notebook, a palm computer, a cloud server and other computing equipment. The terminal can be in man-machine interaction with a client in a keyboard mode, a mouse mode, a remote controller mode, a touch panel mode or a voice control device mode.

The invention provides a method, a system, a platform and a storage medium for processing an image suitable for electronic paper surface defects.

Fig. 16 is a flowchart of an image processing method suitable for electronic paper surface defects according to an embodiment of the present invention.

In this embodiment, the method for processing the surface defect image of the electronic paper may be applied to a terminal or a fixed terminal having a display function, where the terminal is not limited to a personal computer, a smart phone, a tablet computer, a desktop or all-in-one machine with a camera, and the like.

The method for processing the surface defect image of the electronic paper can also be applied to a hardware environment consisting of a terminal and a server connected with the terminal through a network. Networks include, but are not limited to: a wide area network, a metropolitan area network, or a local area network. The method for processing the surface defect image of the electronic paper, which is suitable for the embodiment of the invention, can be executed by a server, a terminal or both the server and the terminal.

For example, for a terminal needing to process an image suitable for the surface defect of the electronic paper, the function of processing the image suitable for the surface defect of the electronic paper provided by the method of the invention can be directly integrated on the terminal, or a client used for realizing the method of the invention is installed. For another example, the method provided by the present invention may also be run on a device such as a server in the form of a Software Development Kit (SDK), and an interface suitable for the electronic paper surface defect image processing function is provided in the form of an SDK, and a terminal or other devices may implement the function suitable for the electronic paper surface defect image processing function through the provided interface.

The invention is further elucidated with reference to the drawing.

As shown in fig. 1 to 19, the present invention provides an image processing method suitable for electronic paper surface defects, which specifically includes the following steps:

s1, acquiring an electronic paper image with a significant defect in real time, and preprocessing the electronic paper image in real time;

and S2, generating the detection data of the surface defects of the electronic paper in real time according to the real-time preprocessing state of the electronic paper image and by combining the constructed detection algorithm.

The method comprises the following steps of generating electronic paper surface defect detection data in real time according to the real-time preprocessing state of an electronic paper image and by combining a constructed detection algorithm, and further comprising the following steps:

s21, acquiring a plurality of images in the process from the white gray scale to the black gray scale through a feature identification method;

and S22, carrying out fusion processing on the acquired multiple images in real time.

The method for fusing the acquired images in real time further comprises the following steps:

s221, fusing the light-color gray-scale pictures according to the full-white gray-scale value;

s222, fusing the dark gray level pictures according to the full black gray level value.

In the process of fusing the images with the light gray scales according to the full white gray scale value, the formula when the light gray scales are fused is as follows:

P′(n)＝255-P(n) (I)

wherein P (n) is the nth picture, P' (n) is the result of negating the nth picture,

In the process of fusing the picture with the dark gray scale according to the all-black gray scale value, the formula when the dark gray scale is fused is specifically as follows:

wherein P (n) is the nth picture,

the average gray value of the P (n) picture is obtained, and P is the picture obtained after fusion.

The method comprises the following steps of obtaining an electronic paper image with a significant defect in real time, and preprocessing the electronic paper image in real time:

s11, segmenting and processing the acquired electronic paper image in real time; and carrying out pose correction processing on the electronic paper.

Specifically, in the embodiment of the invention, the device for detecting the surface defects of the electronic paper is provided; comprises a feeding unit (feeding mechanism), a positioning unit/waiting area (positioning mechanism), a detection unit (detection mechanism) and a discharging unit (discharging mechanism);

a schematic diagram of the feeding unit is shown in fig. 2. After the tray filled with the electronic paper is placed at the corresponding position of the feeding unit manually or by a conveying belt, the mechanical arm (or an xyz displacement platform) sequentially picks the electronic paper and moves the electronic paper to the positioning unit. When the material loading tray has no material, the system can prompt and inform staff of supplementing the material in time.

The positioning unit (as shown in the blue frame at the lower part of fig. 4) mainly has two modules, including a positioning module (as shown in fig. 3) and a transmission module, for adjusting and transmitting the pose of the electronic paper.

In the positioning module, the electronic paper obtained from the feeding unit is placed on the adsorption plate, the pushing module starts to work to push the electronic paper, when one end of the electronic paper contacts the limiting devices, the pushing module on the opposite side stops working and retracts, and when the electronic paper contacts the two limiting devices, the positioning work is completed. (electronic paper positioning work can also be performed using a UVW stage). The trachea is connected to adsorption plate below, has the adsorption efficiency, prevents that the edge warping or the removal phenomenon from appearing in the electron paper quality is too light.

After the positioning is finished, the linear module in the transmission module transmits the whole positioning module to the corresponding position of the detection unit.

The detection unit (as shown in fig. 5) is composed of a camera, a light source, an industrial personal computer and an electronic paper lighting device. The camera is used for collecting images, and the position of the camera is right above the electronic paper to be detected; because the electronic paper does not emit light, a light source is needed to illuminate the shooting environment of the electronic paper, and light support is provided for a camera to shoot images smoothly; a gray scale picture program which needs to be flashed is arranged in the electronic paper lighting equipment and is used for controlling the gray scale change of the electronic paper; the industrial personal computer is used for detecting operation of an algorithm and controlling a camera and electronic paper lighting equipment.

Under the operation of the positioning unit, the electronic paper finishes posture adjustment, the electronic paper and the positioning module are moved to a detection station together by the transmission module, electronic paper lightening equipment (the specific structure is shown in figure 6) controls an electronic paper driving contact to be pressed downwards, after the electronic paper is electrified and conducted, the electronic paper is driven to flicker in a plurality of gray scales, and when the electronic paper is flicked by one gray scale, the electronic paper lightening equipment sends a signal to the industrial personal computer, and the industrial personal computer sends an instruction according to the signal to enable the camera to acquire pictures; after the camera finishes shooting, a signal is fed back to the industrial personal computer, the industrial personal computer controls the electronic paper lighting equipment according to the signal, the electronic paper is driven to flicker to the next gray scale, the reciprocating operation of gray scale change → shooting → gray scale change is carried out, and finally the set image acquisition of the needed gray scale is finished. After image acquisition is completed, the electronic paper and the positioning module return to a waiting area, and the unit to be detected executes a blanking step after detection is completed.

The blanking unit is structured as shown in fig. 7. When the detection unit finishes detection, the detection result (namely OK or NG) of the electronic paper is fed back to the mechanical arm, and the mechanical arm grabs the electronic paper in the waiting area according to the detection result and transfers the electronic paper to the corresponding position of the blanking tray of the blanking unit. When the OK area or the NG area of the discharging tray is full of electronic paper which is detected, the system can give a prompt to inform staff of taking the materials away in time.

The defect detection of the electronic paper requires observation from display results at a plurality of gray levels.

1) Image preprocessing: including multiple e-paper segmentation and e-paper pose correction

(1) Multi-electronic paper segmentation: in the detection unit, since there are multiple cameras for image acquisition, and multiple electronic papers may be photographed in the field of view of a single camera, each electronic paper needs to be separated from the photographed image for detection of each electronic paper. The division of a plurality of pieces of electronic paper from a single picture only needs to depend on a white 1 grayscale picture, and the flow of region extraction is shown in fig. 8. Firstly, the original image (a) is subjected to binarization processing to obtain a binary image (b) of the whole image. And then calculating the circumscribed rectangle (c) of each region, and copying the region of which the size of the circumscribed rectangle exceeds a given value to obtain each electronic paper region in the same picture of the electronic paper.

For the pictures with other gray scales, since the electronic paper only has gray scale change and the position is fixed in the driving process, after the target area with the white 1 gray scale is obtained by the algorithm, the pictures are cut according to the same area on the pictures with the other gray scales, and then the pictures (as shown in fig. 9) of each piece of electronic paper under each gray scale in the visual field can be obtained.

(2) Electronic paper pose correction

When detecting the surface defect of the electronic paper, the edge width needs to be controlled to detect the edge defect. Therefore, it is necessary to develop an automatic edge extraction and correction algorithm for electronic paper. The automatic edge extraction and correction algorithm for the electronic paper to be used in the project is shown in fig. 10. Firstly, binarizing a segmented electronic paper image (b), then obtaining an image edge (c) by adopting an edge extraction algorithm, obtaining straight line boundaries (d) of four sides of the electronic paper through Hough line transformation, obtaining a perspective transformation matrix for correcting the 4 sides through the coordinate transformation relation between the intersection point of the 4 sides and the coordinates after correcting the 4 sides, and applying the perspective transformation matrix to an original image (a) to correct the position and orientation of the original image.

2) Image detection: electronic paper surface defect detection algorithm

The electronic paper surface defects include significant defects and non-significant defects.

For the significant defects, a feature identification method is adopted:

as 10 images are acquired in the process of changing from white gray scale to black gray scale during image acquisition (the 10 images are not particularly required, and the number is determined according to actual experimental effect) as shown in fig. 11.

Because the number of the collected images is large, the problems of large workload, long time consumption and the like of a computer can occur when each image is detected, and therefore, the image fusion method is provided, namely 5 images with light gray scales are fused, and 5 images with deep gray scales are fused, and the specific method is as follows:

light gray scale: since the gray value of full white is 255, the formula when the gray levels of light colors are fused is as follows:

P′(n)＝255-P(n) (I)

is the average gray value of the P' (n) picture,

and P is a picture obtained after fusion.

Dark gray scale: since the gray value of the total black is 0, the formula when the dark gray levels are fused is as follows:

wherein P (n) is the nth picture,

After the fusion, the pictures to be detected are reduced from 2n to 2, the detection time is greatly reduced, and due to the fusion effect, the gray value difference of the position of the suspected defect and the gray values of other normal positions is increased, so that the surface defect of the product is highlighted to the maximum extent.

The results after the algorithm are shown in fig. 12. The white point is more pronounced as can be seen in the figure, facilitating more accurate measurement of the size of the point.

For non-significant defects, a deep learning method is adopted:

the difference between the non-significant defect (e.g. black dot in a cluster shape, dot connecting line) and the background is small, there is no clear size definition, and it is mainly based on whether the human eye can observe obviously, as shown in fig. 13. For non-significant defects, the defect detection method cannot be used for processing, and in consideration of the obvious difference between the defect regions and other background regions, a deep learning method is often adopted for defect detection.

The defect detection process using the deep learning method is shown in fig. 14, and the specific process is as follows:

1. model training:

i. collecting a large number of data sets;

constructing a network structure, selecting a proper loss function, setting proper parameters and realizing model design;

optimizing the interior of the model through the change of the evaluation index to realize the training of the model;

saving the trained models

2. And loading the model, importing the picture to be detected into the model, judging through the model, and outputting a final detection result.

However, since it is difficult to obtain a large amount of defective image data and heavy labeling work is also required on the data, which results in high cost of the deep learning detection method, a brand-new training method for a few-sample supervised target detection model is proposed herein, and the detection accuracy of the model is improved as much as possible under the condition of limited data by the method.

For object detection tasks, it is possible for the detected object to only count in image pixels

However, during detection, the model only needs to know the characteristics of the detection target, and does not need to pay attention to the background, which indicates that a large amount of redundant information exists in the image data. Therefore, the feature information of the target object is excavated in a more targeted manner, and the detection target needs to be extracted. On the other hand, the detection target and the background are separated, and then the diversity of data can be effectively improved through the fusion of the simulated detection target and different backgrounds. And through the combination mode of the target and the background, the generated data can actively generate the label without carrying out complicated label making work. The defect detection flow using the new method is shown in fig. 15.

Namely, a mechanical arm is adopted for feeding and classified blanking; the pose of the electronic paper is fixed by the aid of the pushing module, the electronic paper is conveniently electrified during detection, high-resolution camera image acquisition is adopted, and the traditional machine vision technology is combined with an AI algorithm to detect the surface defects of the electronic paper.

A method for fusing images includes processing multiple images by means of gray level and the like, and superposing the processed images to highlight features of the images. Meanwhile, due to the reduction of the number of pictures, the detection time is also shortened.

A method for increasing deep learning training samples, the method separates the detection target and the background of the existing sample, then carries on the random combination of the detection target and the background, the random combination includes the random of the detection target and the background, the position of the detection target fused to the background is random; meanwhile, a plurality of detection targets can be randomly fused to combine a new detection target. The method effectively solves the problem that the number of training samples is insufficient when a deep learning method is used. The traditional machine vision technology is combined with an AI algorithm to detect the surface defects.

Because the electronic paper diaphragms and modules have the characteristics of multiple size ranges, multiple defect types, multiple defect standards, the need of detection under multiple pictures and the like, the electronic paper detection in China mostly adopts manual detection at present. The traditional manual detection mode has the problems that detection personnel are long in training period, large in personnel mobility, low in detection speed, poor in product consistency, large in damage to human eyes in the detection process and the like, and at the moment, a set of system needs to be developed to replace the traditional work of production line personnel. At the present stage, the invention mainly relates to a full-automatic electronic paper surface defect detection system, which transmits electronic paper to a detection unit through a feeding unit, detects defects of the electronic paper by the detection system based on AI and the traditional vision technology, and feeds back the detection result to a blanking unit for classifying the electronic paper.

1. The machine detection reduces the eye damage caused by long-term work of workers under strong light, and the consistency of detection results is high;

2. full automatic feeding, automatic classification yields and defect article have effectively saved the human cost.

3. The machine vision detection technology is combined with the AI algorithm, so that the accuracy of the detection result is improved.

The time for single detection of the new algorithm combining the image fusion and sample expansion technologies is not more than 15 seconds; without the algorithm of the technology, the time for single detection is about 36 seconds; it can be seen that the detection time is significantly shortened.

In other words, mechanical design aspects: the mechanical arm can be replaced by equipment with a grabbing function, such as an xyz displacement platform; the positioning module in the positioning device can be replaced by equipment with angle adjustment and positioning, such as a UVW platform and the like;

in the aspect of algorithm: the algorithm mentioned herein is not limited to the surface defect detection applied to electronic paper, but can also be used for surface defect detection of other objects.

Other descriptions: the mechanical structure of the invention adopts a detection station to carry out drawing, and for multi-station detection, the basic design principle is not changed, and only the positioning unit/waiting area and the detection unit are required to be adaptively modified.

In order to achieve the above object, the present invention further provides an image processing system suitable for electronic paper surface defects, as shown in fig. 17, the system specifically includes:

the image acquisition and preprocessing unit is used for acquiring an electronic paper image with a significant defect in real time and preprocessing the electronic paper image in real time; and the significant defect detection data generation unit is used for generating the electronic paper surface defect detection data in real time according to the real-time preprocessing state of the electronic paper image and by combining the constructed detection algorithm.

In the image acquisition preprocessing unit, still be provided with:

the significant defect detection data generation unit is further provided with:

in the first fusion processing module, still be provided with:

In the embodiment of the system scheme of the present invention, the specific details of the method steps involved in the processing of the surface defect image of the electronic paper are described above, and are not described herein again.

In order to achieve the above object, the present invention further provides an image processing platform suitable for electronic paper surface defects, as shown in fig. 18, including: the system comprises a processor, a memory and a control program of an image processing platform suitable for the surface defects of the electronic paper;

wherein the processor executes the e-paper-compatible surface defect image processing platform control program, the e-paper-compatible surface defect image processing platform control program is stored in the memory, and the e-paper-compatible surface defect image processing platform control program implements the e-paper-compatible surface defect image processing method steps, such as:

The details of the steps have been set forth above and will not be described herein.

In an embodiment of the present invention, the built-in processor suitable for the electronic paper surface defect image Processing platform may be composed of an integrated circuit, for example, a single packaged integrated circuit, or may be composed of a plurality of integrated circuits packaged with the same function or different functions, and includes one or more Central Processing Units (CPUs), a microprocessor, a digital Processing chip, a graphics processor, and a combination of various control chips. The processor accesses each component by using various interfaces and line connections, executes various functions and processes data suitable for the electronic paper surface defect image processing by running or executing a program or unit stored in the memory and calling data stored in the memory;

the memory is used for storing program codes and various data, is installed in an image processing platform suitable for the surface defects of the electronic paper, and realizes high-speed and automatic access to the program or the data in the running process.

The Memory includes Read-Only Memory (ROM), random Access Memory (RAM), programmable Read-Only Memory (PROM), erasable Programmable Read-Only Memory (EPROM), one-time Programmable Read-Only Memory (OTPROM), electrically Erasable rewritable Read-Only Memory (EEPROM), compact Disc Read-Only Memory (CD-ROM) or other optical Disc Memory, magnetic disk Memory, tape Memory, or any other medium readable by a computer that can be used to carry or store data.

In order to achieve the above object, the present invention further provides a computer readable storage medium, as shown in fig. 19, which stores a control program of a platform for processing surface defects of electronic paper, and the control program of the platform for processing surface defects of electronic paper implements the steps of the method for processing surface defects of electronic paper, for example:

In describing embodiments of the present invention, it should be noted that 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 that the scope of the preferred embodiments of the present invention includes additional implementations 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, such as 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, processing module-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.

In an embodiment of the present invention, to achieve the above object, the present invention further provides a chip system, where the chip system includes at least one processor, and when program instructions are executed in the at least one processor, the chip system is caused to execute the steps of the image processing method applicable to the surface defect of electronic paper, for example:

Those of ordinary skill in the art will appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, or combinations of computer software and electronic hardware. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application. It is clear to those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, apparatuses and units may refer to the corresponding processes in the foregoing method embodiments, and are not described herein again.

The above-mentioned embodiments only express several embodiments of the present invention, and the description thereof is more specific and detailed, but not construed as limiting the scope of the present invention. It should be noted that various changes and modifications can be made by those skilled in the art without departing from the spirit of the invention, and these changes and modifications are all within the scope of the invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.

Claims

1. The image processing method suitable for the electronic paper surface defect is characterized by comprising the following steps:

2. The method for processing the image of the surface defect of the electronic paper as claimed in claim 1, wherein the detection data of the surface defect of the electronic paper is generated in real time according to the real-time preprocessing state of the image of the electronic paper and the constructed detection algorithm, and further comprising the steps of:

3. The method for processing the image of the surface defect of the electronic paper according to claim 2, wherein the acquired images are fused in real time, and further comprising the following steps:

4. The method for processing the image of the surface defect of the electronic paper as claimed in claim 3, wherein in the fusing of the images with the light gray scales according to the gray scale value of the full white, a formula when the light gray scales are fused is specifically as follows:

P′(n)＝255-P(n) (1)

5. The method for processing the image of the surface defect of the electronic paper as claimed in claim 3, wherein the formula for fusing the dark gray scales is as follows:

wherein P (n) is the nth picture,

6. The method for processing the image of the surface defect of the electronic paper as claimed in claim 1, wherein the step of acquiring the image of the electronic paper with the significant defect in real time and preprocessing the image of the electronic paper in real time further comprises the steps of:

segmenting and processing the acquired electronic paper image in real time; and carrying out pose correction processing on the electronic paper.

7. An image processing system suitable for electronic paper surface defect is characterized by specifically comprising:

the image acquisition preprocessing unit is used for acquiring the electronic paper image with the significant defect in real time and preprocessing the electronic paper image in real time; and the significant defect detection data generation unit is used for generating the electronic paper surface defect detection data in real time according to the real-time preprocessing state of the electronic paper image and by combining the constructed detection algorithm.

8. The image processing system for surface defects of electronic paper as set forth in claim 7, wherein the image acquisition preprocessing unit further comprises:

the significant defect detection data generation unit is further provided with:

in the first fusion processing module, still be provided with:

9. An image processing platform suitable for electronic paper surface defects, comprising: the system comprises a processor, a memory and a control program of an image processing platform suitable for the surface defects of the electronic paper;

wherein the control program for the electronic paper surface defect image processing platform is executed on the processor, the control program for the electronic paper surface defect image processing platform is stored in the memory, and the control program for the electronic paper surface defect image processing platform realizes the method for processing the electronic paper surface defect image according to any one of claims 1 to 6.

10. A computer-readable storage medium, wherein the computer-readable storage medium stores a control program for an image processing platform for surface defects of electronic paper, and the control program for the image processing platform for surface defects of electronic paper implements the method for processing the image of surface defects of electronic paper according to any one of claims 1 to 6.