CN118096692A - Embedded image processing method and system based on machine vision - Google Patents
Embedded image processing method and system based on machine vision Download PDFInfo
- Publication number
- CN118096692A CN118096692A CN202410225623.2A CN202410225623A CN118096692A CN 118096692 A CN118096692 A CN 118096692A CN 202410225623 A CN202410225623 A CN 202410225623A CN 118096692 A CN118096692 A CN 118096692A
- Authority
- CN
- China
- Prior art keywords
- image
- abnormal
- data
- target
- information
- 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
- 238000003672 processing method Methods 0.000 title claims abstract description 21
- 230000002159 abnormal effect Effects 0.000 claims abstract description 173
- 230000000877 morphologic effect Effects 0.000 claims abstract description 113
- 238000012545 processing Methods 0.000 claims abstract description 60
- 238000004364 calculation method Methods 0.000 claims description 18
- 230000005856 abnormality Effects 0.000 claims description 16
- 238000002372 labelling Methods 0.000 claims description 12
- 238000007781 pre-processing Methods 0.000 claims description 10
- 238000001914 filtration Methods 0.000 claims description 4
- 238000009499 grossing Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 238000000034 method Methods 0.000 abstract description 17
- 230000000694 effects Effects 0.000 abstract description 8
- 238000005516 engineering process Methods 0.000 description 5
- 238000000605 extraction Methods 0.000 description 3
- 238000005192 partition Methods 0.000 description 3
- 238000001514 detection method Methods 0.000 description 2
- 230000006870 function Effects 0.000 description 2
- 239000000463 material Substances 0.000 description 2
- 230000004075 alteration Effects 0.000 description 1
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000002902 bimodal effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 239000003086 colorant Substances 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000006835 compression Effects 0.000 description 1
- 238000007906 compression Methods 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 230000018109 developmental process Effects 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 230000010354 integration Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000010801 machine learning Methods 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000008569 process Effects 0.000 description 1
- 238000012797 qualification Methods 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 238000010561 standard procedure Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Landscapes
- Image Processing (AREA)
Abstract
The invention discloses an embedded image processing method and system based on machine vision in the technical field of image processing, wherein the method comprises the following steps: setting image acquisition frequency based on related information of a shot target, and acquiring an image to be processed based on the image acquisition frequency to obtain target image data; extracting morphological feature data of a shot target based on the target image data to obtain morphological information of the shot target; judging whether the morphological information of the shot object is abnormal or not based on the standard image to obtain abnormal morphological data; acquiring abnormal information of morphological information of a shot object, numbering the abnormal information to form a numbering set; and determining the number of the abnormal form data based on the number set, and determining the abnormal information of the abnormal form data based on the number of the abnormal form data. The workload of image processing can be effectively reduced, the memory of the embedded system occupied during output is reduced, and the image processing effect is improved.
Description
Technical Field
The invention relates to the technical field of image processing, in particular to an embedded image processing method and system based on machine vision.
Background
In the field of machine vision, there is an image processing method that is based on a processing technology of an embedded system. The image calculation is performed by utilizing the processing capacity of an embedded processor or an image processing application specific integrated circuit ASIC, and the method is characterized by strong pertinence, low power consumption and good stability, and the processing capacity can be covered from a low end to a high end. In the field of machine vision, due to the development of image processing algorithms, typical machine vision image schemes have developed standard procedures such as image enhancement, template matching, measurement, flaw detection, and the like. Each function has a plurality of specific implementation algorithms, each step is only interacted by image data in theory, the same image is processed by the previous algorithm step and then is sent to the next processing step, and little involvement exists between the algorithm steps and the algorithm steps can be executed in parallel.
When the existing image processing technology is used, the storage space of the embedded system is usually limited, and the memory of image data is limited, so that the resolution of the image needs to be reduced to adapt to the embedded system, and the effect on image processing is affected.
For this reason, we propose an embedded image processing method and system based on machine vision to solve the above problems.
Disclosure of Invention
The invention aims to provide an embedded image processing method and system based on machine vision, so as to solve the problems in the background technology.
In order to achieve the above purpose, the present invention provides the following technical solutions: an embedded image processing method and system based on machine vision, the method comprises the following steps:
setting image acquisition frequency based on related information of a shot target, and acquiring an image to be processed based on the image acquisition frequency to obtain target image data;
Extracting morphological feature data of a shot target based on the target image data to obtain morphological information of the shot target, wherein the morphological information comprises the morphological feature data and shape data of a target object;
judging whether the morphological information of the shot object is abnormal or not based on the standard image to obtain abnormal morphological data;
Acquiring abnormal information of morphological information of a shot object, numbering the abnormal information to form a numbering set;
Determining the number of the abnormal form data based on the number set, and determining the abnormal information of the abnormal form data based on the number of the abnormal form data;
And acquiring the processed target image data based on the abnormal form data, and transmitting the target image data to a display module.
By adopting the technical scheme, the processing accuracy is improved by carrying out partition numbering processing on the image data, the acquired target image is compared with the standard image, the abnormal form data of the target object in the target image can be judged, and the number corresponding to the abnormal form data is directly output when the image is output, so that the workload of image processing can be effectively reduced, the embedded system memory occupied during output is reduced, and the processing effect on the image is improved.
Preferably, the step of setting the image acquisition frequency based on the related information of the subject includes:
Acquiring related information of a shot target, wherein the related information comprises a moving speed parameter of the shot target, an acquisition surface parameter of the shot target and a volume parameter of the shot target;
And calculating the corresponding acquisition frequency based on the movement speed parameter of the shot object, the acquisition surface parameter of the shot object and the volume parameter of the shot object.
Preferably, the step of obtaining the target image based on the image acquisition frequency to obtain the image to be processed includes: paving a machine vision acquisition point, regularly shooting a target object through the machine vision acquisition point based on the acquisition frequency, and converting a shot target into an image signal to obtain target image data.
By adopting the technical scheme, the shooting frequency is determined through the moving speed of the shot target, the number of the acquisition surfaces and the volume of the shot target, so that the completeness of the object characteristics of the image acquisition target is improved.
Preferably, the step of extracting morphological feature data of the photographed object based on the object image data to obtain morphological information of the photographed object includes:
dividing the target image data into a plurality of areas to obtain target image subareas;
Adjusting the contrast of each target image subarea, and extracting the outline features of the shot target based on the adjustment result to obtain the shape data of the shot target;
performing gray scale processing on the target image data based on the shape data, and performing binarization processing meeting preset conditions to obtain a binarized image;
And extracting feature data of the shot object based on the binarized image, and determining morphological information of the shot object according to the feature data, wherein the feature data comprises surface smoothness and surface flatness.
By adopting the technical scheme, the feature data of the shot target is extracted based on the binarization result by dividing the target image into areas and performing binarization processing, so that the accuracy and the extraction efficiency of the extracted feature data are improved.
Preferably, the step of determining whether the morphological information of the photographed object is abnormal based on the standard image, and obtaining abnormal morphological data includes:
establishing a morphological information threshold of a standard image meeting preset conditions, wherein the preset conditions refer to image data of a standard shot object acquired in advance;
Acquiring morphological information of a shot target, and comparing the morphological information of a target image with a morphological information threshold of a standard image;
Judging whether the morphological information of the target image exceeds a morphological information threshold of the standard image, if so, indicating that the morphological information of the target image is abnormal, and obtaining abnormal morphological data, wherein the morphological data comprises the shape and the characteristic data of the target image.
By adopting the technical scheme, the abnormal morphological data of the target information can be judged by comparing the morphological information threshold of the preset standard image with the morphological information of the target image, and the accuracy of the judging result is improved.
Preferably, the step of establishing the morphological information threshold of the standard image satisfying the preset condition includes:
acquiring standard data of a shot target object, and training an image model based on the standard data; identifying standard form information of a target object based on the image model to obtain a standard image;
carrying out region division on the standard image, and carrying out preprocessing on the standard image based on a region division result, wherein the preprocessing comprises image enhancement, denoising, smoothing and filtering;
Obtaining morphological information of the standard image based on the preprocessing result; a morphological information threshold of the standard image is determined based on the standard data of the target object.
By adopting the technical scheme, the corresponding standard image is automatically established through the standard data of the shot target object, so that the adaptability of the processing technology is improved.
Preferably, the step of obtaining the abnormal information of the morphological information of the shot object and numbering the abnormal information includes:
Numbering and labeling each target image subarea;
acquiring all abnormal categories of morphological information of a shot object, and uniquely numbering each abnormal category;
Classifying the degree of abnormality in each abnormality category, setting corresponding abnormality thresholds based on the classification, and uniquely numbering the thresholds;
And determining the number of the abnormal form data based on the target image subarea, the abnormal category and the abnormal degree.
Preferably, the step of determining the number of the abnormal shape data based on the target image subregion, the abnormal category, and the degree of abnormality includes:
Acquiring abnormal form data of a shot object, and determining related information of the abnormal form data, wherein the related information comprises position information, abnormal category information and abnormal degree information;
and matching corresponding numbers based on the related information of the abnormal form data to obtain the numbers of the abnormal form data.
By adopting the technical scheme, the system workload can be simplified and the processing effect can be improved by numbering and marking each subarea, each abnormal category and each abnormal degree grade and replacing abnormal form data with the numbers.
Preferably, the step of determining the degree of abnormality of the abnormal morphology data includes:
Acquiring all abnormal categories of the form data, and setting corresponding calculation modes based on the abnormal categories to obtain a calculation mode set;
acquiring the category of the abnormal form data of the shot object, and selecting a calculation mode corresponding to the category of the abnormal form data of the shot object to obtain a calculation result;
Different abnormal degree thresholds are preset, the abnormal degree threshold corresponding to the calculation result is judged, and the abnormal degree of the abnormal form data is determined.
By adopting the technical scheme, the abnormal degree of the morphological data can be effectively judged through the threshold value to which the abnormal degree belongs, and the judging result is improved.
An embedded image processing method and system based on machine vision, comprising:
the image acquisition module is used for acquiring images to be processed based on the image acquisition frequency, obtaining target image data, acquiring standard data of a target object and establishing a morphological information threshold of a standard image meeting preset conditions;
The image processing module is configured to be connected with the image acquisition module and used for processing and analyzing the target image data transmitted by the image acquisition module, extracting morphological feature data of the shot target based on the target image data and obtaining morphological information of the shot target;
The abnormal data judging module is configured to be connected with the image processing module and used for judging whether the morphological information of the shot object is abnormal or not based on the standard image to obtain abnormal morphological data;
the image labeling module is configured to be connected with the image processing module and used for numbering and labeling the abnormal information of the form information of the shot object to form a numbering set, determining the number of the abnormal form data based on the numbering set and determining the abnormal information of the abnormal form data based on the number of the abnormal form data;
and the image output module is configured to be connected with the image annotation module and used for transmitting the target image data to the display module.
By adopting the technical scheme, the acquisition frequency is set according to the related information of the target object, so that the acquisition accuracy can be increased, the abnormal form data of the shot target is determined by the abnormal data judging module, the number of the abnormal form data is transmitted, the workload can be effectively reduced, and the processing efficiency is improved.
Compared with the prior art, the invention has the beneficial effects that:
1. the partition numbering processing is carried out on the image data, so that the processing accuracy is improved, the acquired target image is compared with the standard image, the abnormal form data of the target object in the target image can be judged, and the number corresponding to the abnormal form data is directly output when the image is output, so that the workload of image processing can be effectively reduced, the embedded system memory occupied during the output is reduced, and the processing effect on the image is improved;
2. The shooting frequency is determined through the moving speed, the acquisition surface number and the volume of the shot target, so that the completeness of the image acquisition target object characteristic is improved, the target image is subjected to region division and binarization processing, the characteristic data of the shot target is extracted based on the binarization result, and the accuracy and the extraction efficiency of the extracted characteristic data are improved;
3. The acquisition frequency is set according to the related information of the target object, so that the acquisition accuracy can be improved, the abnormal form data of the shot target is determined through the abnormal data judging module, the number of the abnormal form data is transmitted, the workload can be effectively reduced, and the processing efficiency is improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed for the description of the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and that other drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a schematic flow chart of the method of the present invention;
Fig. 2 is a block diagram of the system architecture of the present invention.
In the drawings, the list of components represented by the various numbers is as follows:
1. An image acquisition module; 2. an image processing module; 3. an abnormal data judging module; 4. an image labeling module; 5. and an image output module.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Example 1
Referring to fig. 1 to 2, the present invention provides an embedded image processing method and system based on machine vision, and the technical scheme thereof: an embedded image processing method and system based on machine vision, the method comprises the following steps:
S1: setting image acquisition frequency based on related information of a photographed object, acquiring an image to be processed based on the image acquisition frequency to obtain object image data,
The step of setting the image acquisition frequency based on the related information of the subject includes: acquiring related information of a shot target, wherein the related information comprises a moving speed parameter of the shot target, an acquisition surface parameter of the shot target and a volume parameter of the shot target; calculating corresponding acquisition frequency based on the moving speed parameter of the shot target, the acquisition surface parameter of the shot target and the volume parameter of the shot target;
The formula corresponding to the corresponding acquisition frequency P obtained by calculation based on the moving speed parameter of the shot object, the acquisition surface parameter of the shot object and the volume parameter of the shot object is as follows: wherein P represents the frequency of collecting target images within a unit time delta T, V represents the moving speed parameter of the shot target, N represents the collecting surface parameter of the shot target, and T represents the volume parameter of the shot target;
The step of obtaining the target image based on the image acquisition frequency to obtain the image to be processed comprises the following steps: paving a machine vision acquisition point, regularly shooting a target object through the machine vision acquisition point based on the acquisition frequency, and converting a shot target into an image signal to obtain target image data;
The shooting frequency is determined through the moving speed of the shot target, the number of acquisition surfaces and the volume of the shot target, so that the integrity of the object characteristics of the image acquisition target is improved;
S2: extracting morphological feature data of the photographed object based on the object image data to obtain morphological information of the photographed object, wherein the morphological information includes the morphological feature data and shape data of the object,
The step of extracting morphological feature data of the photographed object based on the object image data to obtain morphological information of the photographed object includes: dividing the target image data into a plurality of areas to obtain target image subareas; adjusting the contrast of each target image subarea, and extracting the outline features of the shot target based on the adjustment result to obtain the shape data of the shot target; performing gray scale processing on the target image data based on the shape data, and performing binarization processing meeting preset conditions to obtain a binarized image; extracting feature data of the shot object based on the binarized image, and determining form information of the shot object according to the feature data, wherein the feature data comprises surface smoothness and surface flatness;
Image enhancement is used to adjust the contrast of the image, highlight important details in the image, and improve the image quality. The method is characterized in that the integral or local characteristics of the image are purposefully emphasized, the original unclear image is changed into clear or some interesting characteristics are emphasized, the difference between different object characteristics in the image is enlarged, the uninteresting characteristics are restrained, the image quality is improved, the information quantity is enriched, the image interpretation and recognition effect is enhanced, the needs of certain special analysis are met, the object image is divided into areas, binarization processing is carried out, the characteristic data of a shot object is extracted based on the binarization result, and the accuracy and the extraction efficiency of the extracted characteristic data are improved;
The process of converting a color image into a grayscale image is referred to as image graying processing. The color of each pixel in the color image has R, GB component decisions, while each component has a 255 median value, such a pixel may have 1600 tens of thousands of pixels (255 x 255) a range of color variations. The gray image is a special color image with the same components of RG and B, and the variation range of one pixel point is 255, so that in the digital image processing, various formats of images are generally converted into gray images first so that the calculation amount of the subsequent images is reduced. The description of the gray scale image, like the color image, still reflects the distribution and characteristics of the chromaticity and luminance levels throughout and locally of the image.
Graying of an image can be achieved in two ways: the first method averages R, G, B three components per pixel and then assigns this average to the three components for this pixel. The second method is that according to the color space of YUV, the physical meaning of the component of Y is the brightness of the point, the brightness level is reflected by the value, and according to the change relation of RGB and YUV color space, the correspondence of brightness Y and three color components of RGB can be established, Y=0.3 R+0.59G+0.11B, and the gray value of the image is expressed by the brightness value.
When the binary image is processed and analyzed, firstly, the gray level image is binarized to obtain a binarized image, and then the image is further processed, the aggregate property of the image is only related to the position of the point with the pixel value of 0 or 255, and the multi-level value of the pixel is not related, so that the processing is simple, and the processing and compression amount of data are small. In order to obtain an ideal binary image, a closed, connected boundary is generally used to define a non-overlapping region. All pixels with gray levels greater than or equal to the threshold are determined to belong to a particular object, with gray levels of 255 indicating that the pixel points are otherwise excluded from the object region, and gray levels of 0 indicating the background or exceptional object region. If a particular object has uniform gray values within it and it is in a uniform background with other levels of gray values, a thresholding method can be used to obtain a comparative segmentation effect. If the difference between the object and the background is not represented by gray values (e.g., different textures), the difference feature may be converted to a gray difference, and then the image may be segmented using a thresholding technique. The binarization of the dynamic adjustment threshold realization image can dynamically observe the specific result of the segmented image.
When binarizing a gray-scale image or a color image displayed in a gray-scale mode, a threshold may be manually set, or the threshold may be automatically obtained by a system to binarize the image. The more commonly used methods for calculating the threshold value include a bimodal method, a P parameter method, an iterative method, an OTSU method and the like.
S3: judging whether the morphological information of the shot object is abnormal based on the standard image to obtain abnormal morphological data,
Judging whether the morphological information of the shot object is abnormal based on the standard image, and obtaining abnormal morphological data comprises the following steps: establishing a morphological information threshold of a standard image meeting preset conditions, wherein the preset conditions refer to image data of a standard shot object acquired in advance; acquiring morphological information of a shot target, and comparing the morphological information of a target image with a morphological information threshold of a standard image; judging whether the morphological information of the target image exceeds a morphological information threshold of the standard image, if so, indicating that the morphological information of the target image is abnormal, and obtaining abnormal morphological data, wherein the morphological data comprises the shape and the characteristic data of the target image; the abnormal morphological data of the target information can be judged by comparing the morphological information threshold of the preset standard image with the morphological information of the target image, so that the accuracy of a judging result is improved;
The step of establishing a morphological information threshold of the standard image satisfying the preset condition includes: acquiring standard data of a shot target object, and training an image model based on the standard data; identifying standard form information of a target object based on the image model to obtain a standard image; carrying out region division on the standard image, preprocessing the standard image based on a region division result, wherein the preprocessing comprises image enhancement, denoising, smoothing, filtering and the like, and obtaining form information of the standard image based on the preprocessing result; determining a morphological information threshold of the standard image based on the standard data of the target object; the corresponding standard image is automatically established through the standard data of the shot target object, so that the adaptability of the processing technology is improved;
s4: obtaining abnormal information of morphological information of a shot object, numbering the abnormal information to form a numbering set, determining the number of the abnormal morphological data based on the numbering set, and determining the abnormal information of the abnormal morphological data based on the number of the abnormal morphological data;
numbering and labeling each target image subarea; acquiring all abnormal categories of morphological information of a shot object, and uniquely numbering each abnormal category; classifying the degree of abnormality in each abnormality category, setting corresponding abnormality thresholds based on the classification, and uniquely numbering the thresholds; determining the number of the abnormal form data based on the target image subregion, the abnormal category and the abnormal degree;
The step of determining the number of the abnormal shape data based on the target image subregion, the abnormal category and the degree of abnormality comprises the following steps: acquiring abnormal form data of a shot object, and determining related information of the abnormal form data, wherein the related information comprises position information, abnormal category information and abnormal degree information; matching corresponding numbers based on the related information of the abnormal form data to obtain the numbers of the abnormal form data;
The number marking is carried out on each subarea, each abnormal category and each abnormal degree grade, and the number is used for replacing abnormal form data, so that the workload of the system can be simplified, and the processing effect can be improved;
The step of determining the degree of abnormality of the abnormal morphology data includes: acquiring all abnormal categories of the form data, and setting corresponding calculation modes based on the abnormal categories to obtain a calculation mode set; acquiring the category of the abnormal form data of the shot object, and selecting a calculation mode corresponding to the category of the abnormal form data of the shot object to obtain a calculation result; different abnormal degree thresholds are preset, the abnormal degree threshold corresponding to the calculation result is judged, the abnormal degree of the abnormal form data is determined, the abnormal degree of the form data can be effectively judged through the threshold to which the abnormal degree belongs, and the judgment result is improved;
Assuming that the object is a product in terms of product defect detection, the object is to detect the qualification of the product, when scratches appear, the quality of the product is affected, that is, the abnormal shape of the product is shown, if the abnormal shape data is the surface of the product, the degree of the abnormal shape data can be determined by calculating the ratio Z of the depth Sy of the scratches and the length Cy of the scratches, and the corresponding formula is that Wherein Sz represents the total depth of the surface where the scratch is located, cz represents the length value of the surface where the scratch is located, delta > 0, eta > 0; assuming that the abnormal shape data represents that burrs exist on the product, the degree of the abnormal shape data can be determined by calculating the proportion M of the area occupied by the burrs, and the corresponding formula is as follows: Wherein My represents the occupied area of the burr, mz represents the total area of the surface where the burr is located, and beta is more than 0; if the abnormal shape data is used for representing whether the product is deformed, the degree of the abnormal shape data can be obtained by comparing the image data of the shot object with the standard image and calculating the degree of difference of deformation, and the degree of difference of deformation is related to the contour coincidence degree F of the image, and the corresponding formula is as follows: /(I) Wherein Ly represents contour data of a target object in a target image, lz represents standard contour data of the target object, and alpha is more than 0;
s5: and acquiring the processed target image data based on the abnormal form data, and transmitting the target image data to a display module.
The image data is subjected to partition numbering processing, so that the processing accuracy is improved, the acquired target image is compared with the standard image, abnormal form data of a target object in the target image can be judged, and then the number corresponding to the abnormal form data is directly output when the image is output, so that the size of the image is effectively reduced, the workload of image processing can be effectively reduced, the embedded system memory occupied during output is reduced, and the processing efficiency of the image is improved.
An embedded machine vision based image processing system comprising:
the image acquisition module 1 is used for acquiring images to be processed based on image acquisition frequency, obtaining target image data, acquiring standard data of a target object and establishing a morphological information threshold of a standard image meeting preset conditions;
The image processing module 2 is configured to be connected with the image acquisition module 1 and is used for processing and analyzing the target image data transmitted by the image acquisition module 1, extracting morphological feature data of a shot target based on the target image data and obtaining morphological information of the shot target;
An abnormal data judging module 3 configured to be connected to the image processing module 2, for judging whether the morphological information of the photographed object is abnormal based on the standard image, to obtain abnormal morphological data;
the image labeling module 4 is configured to be connected with the image processing module 2 and is used for numbering and labeling the abnormal information of the morphological information of the shot object to form a numbering set, determining the number of the abnormal morphological data based on the numbering set and determining the abnormal information of the abnormal morphological data based on the number of the abnormal morphological data;
and the image output module 5 is configured to be connected with the image labeling module 4 and used for transmitting the target image data to the display module.
It should be noted that: firstly, an image to be processed needs to be acquired, a camera or other image acquisition equipment can be used for acquiring the image, preprocessing operations such as image enhancement, denoising, smoothing, filtering and the like are performed on the acquired image so as to improve the quality and contrast of the image, the image is divided according to different areas or features, the image can be divided into a plurality of areas, each area is independently processed, interesting features such as colors, textures, shapes and the like are extracted from the image, the features can be used for subsequent object identification and classification, the machine learning, deep learning or other algorithms are used for identifying and classifying objects in the image, the existing models or training new models can be used for outputting the processed image to a display or other equipment so as to perform visualization or further processing, the acquisition frequency is set through the related information of the object, the acquisition accuracy can be increased, the abnormal form data of the shot object is determined through an abnormal data judging module 3, the number of the abnormal form data is transmitted, the work load can be effectively reduced, and the processing efficiency is improved.
The embedded system is used for image processing, for example, the ARM microprocessor or ARM+DSP is used for constructing a machine vision system, so that the human-computer interaction function is strong, the integration level is high, the instantaneity is good, the multitasking is supported, and the specific implementation method can be selected and adjusted according to specific application scenes and requirements.
In the description of the present specification, the descriptions of the terms "one embodiment," "example," "specific example," and the like, 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 present invention. In this specification, schematic representations of the above terms do not necessarily refer to the same embodiments or examples. Furthermore, the particular features, structures, materials, or characteristics described may be combined in any suitable manner in any one or more embodiments or examples.
Although embodiments of the present invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made therein without departing from the principles and spirit of the invention, the scope of which is defined in the appended claims and their equivalents.
Claims (10)
1. The embedded image processing method based on machine vision is characterized by comprising the following steps of:
setting image acquisition frequency based on related information of a shot target, and acquiring an image to be processed based on the image acquisition frequency to obtain target image data;
Extracting morphological feature data of a shot target based on the target image data to obtain morphological information of the shot target, wherein the morphological information comprises the morphological feature data and shape data of a target object;
judging whether the morphological information of the shot object is abnormal or not based on the standard image to obtain abnormal morphological data;
Acquiring abnormal information of morphological information of a shot object, numbering the abnormal information to form a numbering set;
Determining the number of the abnormal form data based on the number set, and determining the abnormal information of the abnormal form data based on the number of the abnormal form data;
And acquiring the processed target image data based on the abnormal form data, and transmitting the target image data to a display module.
2. The machine vision-based embedded image processing method according to claim 1, wherein the step of setting the image acquisition frequency based on the related information of the subject includes:
Acquiring related information of a shot target, wherein the related information comprises a moving speed parameter of the shot target, an acquisition surface parameter of the shot target and a volume parameter of the shot target;
And calculating the corresponding acquisition frequency based on the movement speed parameter of the shot object, the acquisition surface parameter of the shot object and the volume parameter of the shot object.
3. The machine vision-based embedded image processing method and system according to claim 1, wherein the step of obtaining the target image by acquiring the image to be processed based on the image acquisition frequency comprises: paving a machine vision acquisition point, regularly shooting a target object through the machine vision acquisition point based on the acquisition frequency, and converting a shot target into an image signal to obtain target image data.
4. The machine vision-based embedded image processing method according to claim 1, wherein the step of extracting morphological feature data of a subject based on the target image data to obtain morphological information of the subject comprises:
dividing the target image data into a plurality of areas to obtain target image subareas;
Adjusting the contrast of each target image subarea, and extracting the outline features of the shot target based on the adjustment result to obtain the shape data of the shot target;
performing gray scale processing on the target image data based on the shape data, and performing binarization processing meeting preset conditions to obtain a binarized image;
And extracting feature data of the shot object based on the binarized image, and determining morphological information of the shot object according to the feature data, wherein the feature data comprises surface smoothness and surface flatness.
5. The machine vision-based embedded image processing method according to claim 1, wherein the step of determining whether the morphological information of the subject is abnormal based on the standard image, and obtaining abnormal morphological data includes:
establishing a morphological information threshold of a standard image meeting preset conditions, wherein the preset conditions refer to image data of a standard shot object acquired in advance;
Acquiring morphological information of a shot target, and comparing the morphological information of a target image with a morphological information threshold of a standard image;
Judging whether the morphological information of the target image exceeds a morphological information threshold of the standard image, if so, indicating that the morphological information of the target image is abnormal, and obtaining abnormal morphological data, wherein the morphological data comprises the shape and the characteristic data of the target image.
6. The machine vision based embedded image processing method of claim 5, wherein: the step of establishing the morphological information threshold of the standard image meeting the preset condition comprises the following steps:
acquiring standard data of a shot target object, and training an image model based on the standard data; identifying standard form information of a target object based on the image model to obtain a standard image;
carrying out region division on the standard image, and carrying out preprocessing on the standard image based on a region division result, wherein the preprocessing comprises image enhancement, denoising, smoothing and filtering;
Obtaining morphological information of the standard image based on the preprocessing result; a morphological information threshold of the standard image is determined based on the standard data of the target object.
7. The machine vision-based embedded image processing method according to claim 1, wherein the step of obtaining the anomaly information of the morphological information of the object, and numbering the anomaly information comprises:
Numbering and labeling each target image subarea;
acquiring all abnormal categories of morphological information of a shot object, and uniquely numbering each abnormal category;
Classifying the degree of abnormality in each abnormality category, setting corresponding abnormality thresholds based on the classification, and uniquely numbering the thresholds;
And determining the number of the abnormal form data based on the target image subarea, the abnormal category and the abnormal degree.
8. The machine vision-based embedded image processing method according to claim 7, wherein the step of determining the number of the abnormal shape data based on the target image subregion, the abnormality category, the abnormality degree includes:
Acquiring abnormal form data of a shot object, and determining related information of the abnormal form data, wherein the related information comprises position information, abnormal category information and abnormal degree information;
and matching corresponding numbers based on the related information of the abnormal form data to obtain the numbers of the abnormal form data.
9. The machine vision-based embedded image processing method according to claim 8, wherein the step of determining the degree of abnormality of the abnormal morphology data includes:
Acquiring all abnormal categories of the form data, and setting corresponding calculation modes based on the abnormal categories to obtain a calculation mode set;
acquiring the category of the abnormal form data of the shot object, and selecting a calculation mode corresponding to the category of the abnormal form data of the shot object to obtain a calculation result;
Different abnormal degree thresholds are preset, the abnormal degree threshold corresponding to the calculation result is judged, and the abnormal degree of the abnormal form data is determined.
10. The machine vision-based embedded image processing method and system are characterized by comprising the following steps:
The image acquisition module (1) is used for acquiring images to be processed based on the image acquisition frequency, obtaining target image data, acquiring standard data of a target object and establishing a morphological information threshold of the standard image meeting preset conditions;
The image processing module (2) is configured to be connected with the image acquisition module (1) and is used for processing and analyzing the target image data transmitted by the image acquisition module (1), extracting morphological characteristic data of a shot target based on the target image data and obtaining morphological information of the shot target;
an abnormal data judging module (3) configured to be connected with the image processing module (2) and used for judging whether the morphological information of the shot object is abnormal or not based on a standard image to obtain abnormal morphological data;
The image labeling module (4) is connected with the image processing module (2) and is used for numbering and labeling the abnormal information of the morphological information of the shot object to form a numbering set, determining the number of the abnormal morphological data based on the numbering set and determining the abnormal information of the abnormal morphological data based on the number of the abnormal morphological data;
And the image output module (5) is configured to be connected with the image labeling module (4) and used for transmitting the target image data to the display module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410225623.2A CN118096692A (en) | 2024-02-29 | 2024-02-29 | Embedded image processing method and system based on machine vision |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410225623.2A CN118096692A (en) | 2024-02-29 | 2024-02-29 | Embedded image processing method and system based on machine vision |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN118096692A true CN118096692A (en) | 2024-05-28 |
Family
ID=91145186
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410225623.2A Pending CN118096692A (en) | 2024-02-29 | 2024-02-29 | Embedded image processing method and system based on machine vision |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN118096692A (en) |
-
2024
- 2024-02-29 CN CN202410225623.2A patent/CN118096692A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114937055B (en) | Image self-adaptive segmentation method and system based on artificial intelligence | |
| CN113781402B (en) | Method and device for detecting scratch defects on chip surface and computer equipment | |
| CN115082419B (en) | Blow-molded luggage production defect detection method | |
| CN110286124B (en) | Machine vision-based refractory brick measuring system | |
| CN111179243A (en) | Small-size chip crack detection method and system based on computer vision | |
| CN114419025A (en) | Fiberboard quality evaluation method based on image processing | |
| CN115345885A (en) | Method for detecting appearance quality of metal fitness equipment | |
| CN108711158B (en) | Pointer instrument image identification method based on contour fitting and radial segmentation | |
| CN108181316B (en) | Bamboo strip defect detection method based on machine vision | |
| CN115082451B (en) | Stainless steel soup ladle defect detection method based on image processing | |
| CN114820625B (en) | Automobile top block defect detection method | |
| CN110648330B (en) | Defect detection method for camera glass | |
| CN111667475B (en) | Machine vision-based Chinese date grading detection method | |
| CN116883408B (en) | Integrating instrument shell defect detection method based on artificial intelligence | |
| CN113177924A (en) | Industrial production line product flaw detection method | |
| CN114972356A (en) | Plastic product surface defect detection and identification method and system | |
| CN112561875A (en) | Photovoltaic cell panel coarse grid detection method based on artificial intelligence | |
| CN116152242B (en) | Visual detection system of natural leather defect for basketball | |
| CN111968082A (en) | Product packaging defect detection and identification method based on machine vision | |
| CN116109577A (en) | Printing label defect detection system and method | |
| CN116402822B (en) | Concrete structure image detection method and device, electronic equipment and storage medium | |
| CN114820597B (en) | Smelting product defect detection method, device and system based on artificial intelligence | |
| CN112801963B (en) | Video image occlusion detection method and system | |
| CN118096692A (en) | Embedded image processing method and system based on machine vision | |
| CN114266748A (en) | Method and device for judging integrity of surface of process plate in rail transit maintenance field |
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 |