An image feature processing method and system based on machine vision

Abstract

The application provides a kind of image feature processing method and system based on machine vision, it is related to digital processing technical field, method includes: through image acquisition device, extraction device parameter information is collected, full angle image acquisition is carried out, image acquisition information set is obtained, feature processing is carried out, and target object structure feature is obtained;Image acquisition information set is divided into three components, three-component segmented image set is obtained, is handled in partition, obtains multiple image acquisition information subset, in combination with acquisition device parameter information, image feature processing result is obtained, and display output is carried out.The technical problem that it is solved that image feature processing accuracy is low, resulting in the adaptability of image feature processing result and the feature function of image feature processing system is low, reaches to carry out grey matching, carries out partition processing, improves image feature processing accuracy, for feature function, matching determines image feature processing result, improves the adaptability of feature function image feature processing result.

Description

TECHNICAL FIELD

[0001] The application relates to the technical field of digital processing, in particular to an image feature processing method and system based on machine vision. BACKGROUND

[0002] With the continuous development of computer science, multimedia and communication technology, the development of image feature processing is greatly promoted. Image feature processing is applied to automatic driving, machine navigation, target tracking and the like. At present, in order to ensure the accuracy of image feature processing, a large amount of operation needs to be performed. In the case that the computing power of an image feature processing system is limited, the accuracy of image feature processing cannot be guaranteed.

[0003] In the prior art, the accuracy of image feature processing is low, which leads to low adaptability of the image feature processing result to the feature function of the image feature processing system. SUMMARY

[0004] The application provides an image feature processing method and system based on machine vision, which solves the technical problem of low accuracy of image feature processing, which leads to low adaptability of the image feature processing result to the feature function of the image feature processing system. The technical effects of performing gray matching, performing partition processing, improving the accuracy of image feature processing, matching and determining the image feature processing result according to the feature function, and improving the adaptability of the feature function image feature processing result are achieved.

[0005] In view of the above problems, the application provides an image feature processing method and system based on machine vision.

[0006] In a first aspect, the application provides an image feature processing method based on machine vision. The method is applied to an image feature processing system, the system is in communication connection with an image acquisition device, and the method comprises the following steps: extracting acquisition device parameter information through the image acquisition device; acquiring image acquisition information set by performing full-angle image acquisition on a target object through the image acquisition device and the acquisition device parameter information; acquiring target object structure features by performing feature processing on the image acquisition information set; acquiring three-component segmented image set by performing three-component segmentation on the image acquisition information set; acquiring a plurality of image acquisition information subsets by performing partition processing on the image acquisition information set through the target object structure features and the three-component segmented image set; acquiring image feature processing results by combining the acquisition device parameter information through the feature function of the image feature processing system and the plurality of image acquisition information subsets; and displaying and outputting the image feature processing results based on the display module of the image feature processing system.

[0007] In a second aspect of the present application, a machine vision-based image feature processing system is provided, which comprises: a parameter extraction unit configured to extract device parameter information of an image acquisition device; a full-angle acquisition unit configured to acquire full-angle image acquisition information of a target object by the image acquisition device and the device parameter information, and obtain an image acquisition information set; a structure feature acquisition unit configured to process the image acquisition information set to obtain structure features of the target object; a three-component segmentation unit configured to perform three-component segmentation on the image acquisition information set to obtain a three-component segmented image set; a partition processing unit configured to process the image acquisition information set by the structure features of the target object and the three-component segmented image set to obtain a plurality of image acquisition information subsets; an image feature acquisition unit configured to obtain an image feature processing result by combining the device parameter information with the plurality of image acquisition information subsets and the feature function of the image feature processing system; and a display output unit configured to display and output the image feature processing result based on a display module of the image feature processing system.

[0008] The one or more technical solutions provided in the present application have at least the following technical effects or advantages:

[0009] By extracting device parameter information of an image acquisition device, acquiring full-angle image acquisition information of a target object, processing the image acquisition information set to obtain structure features of the target object, performing three-component segmentation on the image acquisition information set to obtain a three-component segmented image set, processing the image acquisition information set to obtain a plurality of image acquisition information subsets, combining the device parameter information with the plurality of image acquisition information subsets and the feature function of the image feature processing system to obtain an image feature processing result, and displaying and outputting the image feature processing result based on a display module of the image feature processing system, the present application achieves the technical effects of performing gray scale matching, performing partition processing, improving image feature processing accuracy, matching and determining an image feature processing result according to a feature function, and improving the adaptability of a feature function image feature processing result. BRIEF DESCRIPTION OF DRAWINGS

[0010] Figure 1 FIG. 1 is a flowchart of a machine vision-based image feature processing method according to an embodiment of the present application;

[0011] Figure 2 FIG. 2 is a flowchart of a synthetic image acquisition information set of a machine vision-based image feature processing method according to an embodiment of the present application;

[0012] Figure 3 This is a schematic diagram illustrating the process of obtaining image feature processing results using a machine vision-based image feature processing method according to this application.

[0013] Figure 4 This is a schematic diagram of the structure of an image feature processing system based on machine vision according to this application.

[0014] Figure labeling: Parameter extraction unit 11, full-angle acquisition unit 12, structural feature acquisition unit 13, three-component segmentation unit 14, partition processing unit 15, image feature acquisition unit 16, display output unit 17. Detailed Implementation

[0015] This application provides a machine vision-based image feature processing method and system, which solves the technical problem of low image feature processing accuracy, resulting in low adaptability between image feature processing results and the feature functions of the image feature processing system. It achieves the technical effect of improving image feature processing accuracy by performing grayscale matching and partitioning, and improving the adaptability of image feature processing results to feature functions by matching and determining the image feature processing results.

[0016] Example 1

[0017] like Figure 1 As shown, this application provides an image feature processing method based on machine vision, wherein the method is applied to an image feature processing system, the system being communicatively connected to an image acquisition device, and the method includes:

[0018] Step S100: Extract the acquisition device parameter information using the image acquisition device;

[0019] Step S200: Using the image acquisition device and its parameter information, perform full-angle image acquisition on the target object to obtain an image acquisition information set;

[0020] Specifically, the image feature processing system, in conjunction with the image acquisition device, performs feature processing on the image information acquired by the image acquisition device. The image acquisition device is a real-time image acquisition device such as a camera. The parameter information of the acquisition device includes, but is not limited to, exposure compensation, ISO, white balance, aperture, and shutter speed, which are specifically determined based on the settings of the image acquisition device.

[0021] Specifically, by the image acquisition device and the acquisition device parameter information, the target object is collected full angle, the target object is the target of image feature processing, and the target object is not limited. Commonly, the target object is set to a fan, a water cup or any other structure. After completing the full angle image acquisition of the target object, the image acquisition device outputs an image acquisition information set, which includes image information of different angles of the target object. The elements of the image acquisition information set are image information of different angles and corresponding angle information. Data acquisition is performed to provide a data basis for subsequent feature analysis and processing.

[0022] Further, by the image acquisition device and the acquisition device parameter information, the target object is collected full angle, the target object is collected full angle, and the image acquisition information set is obtained. Step S200 includes:

[0023] Step S210: Obtain preset acquisition constraint information;

[0024] Step S220: By the preset acquisition constraint information, the target object and the image acquisition device are collected and limited, and the image acquisition instruction is obtained.

[0025] Step S230: Based on the image acquisition instruction and the acquisition device parameter information, the image acquisition device is used to collect the target object full angle, and the image acquisition information set is obtained.

[0026] Specifically, the preset acquisition constraint information includes distance constraint information of the image acquisition device and the target object and speed constraint information of the angle conversion movement process. The preset acquisition constraint information is a preset index parameter of the image feature processing system, and the preset acquisition constraint information is matched with the index limitation of the lateral process of the three-dimensional scanner. The actual parameter index is combined to determine. If the target object and the image acquisition device do not meet the preset acquisition constraint information, adjustment is needed. After the target object and the image acquisition device meet the preset acquisition constraint information, the image acquisition instruction is obtained, which is used to control the image acquisition device to start image acquisition. In the full angle image acquisition process, the preset acquisition constraint information is used for acquisition limitation. Based on the acquisition device parameter information, the full angle image acquisition is performed in the preset acquisition constraint information limitation, the image acquisition information set is obtained, and technical support is provided to ensure the accuracy of the image acquisition information set.

[0027] Further, as shown in Figure 2 Based on the image acquisition instruction and the acquisition device parameter information, the image acquisition device is used to collect the target object full angle, and the image acquisition information set is obtained. Step S230 includes:

[0028] Step S231: Using the image acquisition command and the acquisition device parameter information, perform full-angle image acquisition on the target object to obtain a full-angle blurred image information set;

[0029] Step S232: By setting a preset recognition accuracy, the full-angle blurred image information set is identified to obtain a detail focusing command;

[0030] Step S233: Using the detailed focus command, perform local image acquisition on the target object to obtain a set of local refined image information;

[0031] Step S234: Combine the locally refined image information set with the full-angle blurred image information set to synthesize the image acquisition information set.

[0032] Specifically, during image acquisition, it is inevitable that there will be some locations where detailed features are difficult to identify. The image acquisition device needs to be focused to improve the acquisition accuracy of these detailed feature locations. By preset the recognition accuracy, the image acquisition accuracy of the image acquisition device can be adaptively adjusted. While ensuring the accuracy of the image acquisition information set, the effectiveness of the image acquisition information set can be improved, avoiding the problem of data redundancy due to the increase in accuracy, which would lead to a decrease in the efficiency of the image feature processing system.

[0033] To further explain, the full-angle blurred image information set refers to the full-angle image information of the target object. Since the full-angle blurred image information set cannot guarantee the accuracy of the detailed features of the target object, a preset recognition precision is used. This preset recognition precision includes, but is not limited to, contour recognition precision and corner recognition precision (surface corner: curvature of the corner position, tangential corner: angle of the corner position). The preset recognition precision is a preset index parameter of the image feature processing system. Based on this preset recognition precision, recognition is performed on the full-angle blurred image information set. After determining that the position information of the detailed features of the target object cannot be guaranteed, a detail focusing command is obtained, and local image acquisition is performed on the target object to obtain a local refined image information set. Combining the position information of the detailed features, the local refined image information set and the full-angle blurred image information set are synthesized to obtain the image acquisition information set.

[0034] Step S300: Perform feature processing on the image acquisition information set to obtain the structural features of the target object;

[0035] Furthermore, feature processing is performed on the image acquisition information set to obtain the structural features of the target object. Step S300 includes:

[0036] Step S310: Import the image acquisition information set into the 3D scanner to obtain the geometric features of the target object;

[0037] Step S320: extracting a geometric contour of the image acquisition information set based on the geometric feature of the target object to obtain a contour feature of the target object;

[0038] Step S330: obtaining a structure feature of the target object based on the contour feature of the object and the geometric feature of the target object.

[0039] Specifically, to ensure the stability of the structure feature of the target object, the structure feature of the target object is determined by merging the contour feature of the object and the geometric feature of the target object, so as to reduce the waste of time resources caused by multiple calibrations of the three-dimensional scanner.

[0040] Further specifically, the three-dimensional scanner is used to detect and analyze the shape and appearance of the target object or the environment. The image acquisition information set is taken as input data and introduced into the input port of the three-dimensional scanner. The three-dimensional scanner outputs the geometric feature of the target object, i.e., a plurality of geometric point information of the target object. Based on the geometric feature of the target object, the contour of the target object is extracted. The contour of the target object is adjusted in precision based on the image acquisition information set to obtain the contour feature of the target object, i.e., a plurality of geometric point information of the contour position of the target object. The information entropy of the contour feature of the target object is greater than that of the contour of the target object. The structure feature of the target object is obtained to ensure the credibility of the contour feature of the target object in the structure feature of the target object.

[0041] Step S400: performing three-component segmentation on the image acquisition information set to obtain a three-component segmented image set;

[0042] Step S500: performing partition processing on the image acquisition information set based on the structure feature of the target object and the three-component segmented image set to obtain a plurality of image acquisition information subsets;

[0043] Step S600: obtaining an image feature processing result based on the feature function of the image feature processing system, the plurality of image acquisition information subsets, and the parameter information of the acquisition device;

[0044] Step S700: displaying and outputting the image feature processing result based on the display module of the image feature processing system.

[0045] Specifically, to ensure the accuracy of the image partition process, grayscale matching is required. The grayscale matching and other related image grayscale processing operations are preprocessing steps of image processing, which support subsequent image segmentation, image recognition, image analysis, and other related operations.

[0046] Further specifically, the image acquisition information set is subjected to three-component segmentation, the three components of the image acquisition information set are three primary color components, the three primary color components include single-color images of three primary colors of blue (B component), green (G component), and red (R component), a three-component segmented image set is obtained, the three-component segmented image set includes a blue primary color image information set, a green primary color image information set, and a red primary color image information set; the three-component segmented image set is subjected to gray scale matching, that is, an average value of brightness of the three components is calculated and obtained, the average value of brightness of the three components is set as a gray scale value, target object gray scale data set is obtained through the three-component segmented image set, the image acquisition information set is subjected to partition processing in combination with the structure feature of the target object, a plurality of image acquisition information subsets are obtained, and the segmentation accuracy of the image is improved in the partition processing process.

[0047] Further specifically, in order to improve the adaptability of the feature function of the image feature processing system and the image feature processing result, the feature function of the image feature processing system includes but is not limited to positioning function, measurement function, identification function, code reading function, defect marking function, 3D modeling function, and logic operation function, the feature function and the feature result are matched based on the acquisition device parameter information, the plurality of image acquisition information subsets, and the feature function of the image feature processing system, corresponding feedback output is obtained after the matching is completed, the image feature processing result is obtained, the image feature processing result is the output of the image feature processing system, and the image feature processing output is customized by using the display module of the image feature processing system to improve the flexibility of image feature processing.

[0048] Further, as shown in Figure 3 the image feature processing result is obtained through the feature function of the image feature processing system, the plurality of image acquisition information subsets, and the acquisition device parameter information, step S600 includes:

[0049] Step S610: Obtain the feature function of the image feature processing system, the feature function includes positioning and measurement function, code reading and identification and logic operation function, 3D modeling and defect function;

[0050] Step S620: The plurality of image acquisition information subsets are subjected to positioning and measurement through the positioning and measurement function, and target object measurement features are obtained;

[0051] Step S630: Based on the 3D modeling and defect function, 3D modeling is performed through the target object measurement features and the structure feature of the target object, and a target object 3D model is obtained;

[0052] Step S640: The target object 3D model is subjected to defect identification through the code reading and identification and logic operation function, and target object defect features are obtained;

[0053] Step S650: Obtain image feature processing results by the target object measurement features and the target object defect features, combined with the acquisition device parameter information.

[0054] Specifically, based on the feature function, the multiple image acquisition information subsets are subjected to targeted feature processing, so as to ensure the adaptation degree of the image feature processing results and the feature function.

[0055] Further specifically, the feature function of the image feature processing system includes but is not limited to positioning function, measurement function, identification function, code reading function, defect marking function, 3D modeling function, and logic operation function. The feature function is associated and bound, and the positioning measurement function, the code reading and identification and logic operation function, and the 3D modeling and defect function are determined correspondingly. The display module of the image feature processing system includes a display unit of the positioning measurement function, a display unit of the code reading and identification and logic operation function, and a display unit of the 3D modeling and defect function, so as to improve the convenience of image feature index extraction.

[0056] Further specifically, the positioning measurement function is correspondingly provided with multiple node marker anchor points, and the multiple image acquisition information subsets can be subjected to positioning measurement to output target object measurement features. The 3D modeling and defect function is correspondingly embedded with a three-dimensional modeling software, and the target object measurement features and the target object structure features can be subjected to three-dimensional modeling. Through the target object measurement features and the target object structure features, the target object can be subjected to digital modeling restoration and defect positioning marking. Based on the 3D modeling and defect function, a 3D model of the target object is obtained through 3D modeling. The target object 3D model is subjected to lossless comparison based on the code reading and identification and logic operation function, and target object defect features are obtained. The target object defect features include but are not limited to defect position features and defect shape features. Based on the acquisition device parameter information, the target object measurement features and the target object defect features are fed back to the corresponding feature function to obtain image feature processing results, and the image feature processing results are customized and matched to meet the customized needs of users from machine vision.

[0057] Further, the target object 3D model is subjected to defect identification through the code reading and identification and logic operation function to obtain target object defect features. Step S640 includes:

[0058] Step S641: Extract a code label on the target object 3D model through the code reading and identification function.

[0059] Step S642: Identify the code label on the target object 3D model to obtain rectangular square record data symbol information.

[0060] Step S643: Perform logical operation on the rectangular matrix recording data symbol information to obtain target object label information;

[0061] Step S644: Based on the target object label information, perform online networking to extract a target object lossless 3D model;

[0062] Step S645: Compare the target object 3D model with the target object lossless 3D model to obtain a structure comparison result;

[0063] Step S646: Perform defect identification through the structure comparison result to obtain target object defect features.

[0064] Further, the embodiments of the present application also include:

[0065] Step S810: Based on the feature functions of the image feature processing system, mark the target object measurement features, the rectangular matrix recording data symbol information and target object label information, the target object 3D model and the target object defect features;

[0066] Step S820: Through the positioning measurement function, perform data feedback to output the target object measurement features; through the code reading and logical operation function, perform data feedback to output the rectangular matrix recording data symbol information and target object label information; through the 3D modeling and defect function, perform data feedback to output the target object 3D model and the target object defect features;

[0067] Step S830: Through the display module of the image feature processing system, display and output the feature functions of the image feature processing system and the corresponding feedback output information.

[0068] Specifically, based on the basic information marked by the target object code label, the target object lossless 3D model is determined through networking, which provides data support for lossless comparison, and the display module is matched and output through the display module of the image feature processing system, thereby optimizing the display logic of image feature processing output.

[0069] Further specifically, the code reading function is associated with a radio frequency identification scanning device, which can scan and extract the code label on the target object 3D model to obtain the rectangular square record data symbol information on the code label of the target object, the rectangular square record data symbol information including the basic information of the target object, i.e. the component information of the target object, the color information of the target object, and other related basic information, and the rectangular square record data symbol information can be a bar code, a two-dimensional code or any rectangular square code; logical operation is performed to obtain the target object label information, which includes but is not limited to the target object model information, the target object type information, i.e. related data information; the target object label information is used as a marker search key to perform online networking to extract the target object lossless 3D model; comparison is performed to obtain the structure comparison result, i.e. the structure information of the difference between the target object 3D model and the target object lossless 3D model, to obtain the target object defect feature, thereby providing technical support for ensuring the effectiveness of the target object defect feature.

[0070] Specifically, to ensure the adaptability of the image feature processing result and the feature function, the feature function is used to perform positioning marking to obtain the marking feature, the marking feature including the target object measurement feature, the rectangular square record data symbol information and the target object label information, the target object 3D model and the target object defect feature; the feature function is associated with the corresponding output, the target object measurement feature is displayed and output on the display unit of the positioning measurement function; the rectangular square record data symbol information and the target object label information are displayed and output on the display unit of the code reading and logical operation function; the target object 3D model and the target object defect feature are displayed and output on the display unit of the 3D modeling and defect function, thereby providing technical support for customized matching image feature processing output.

[0071] In summary, the image feature processing method and system based on machine vision provided by the present application have the following technical effects:

[0072] Due to the adoption of the image acquisition device, the target object is subjected to full-angle image acquisition, image acquisition information set is acquired, and structural features of the target object are acquired through feature processing. The image acquisition information set is subjected to three-component segmentation, three-component segmented image set is acquired, and the image acquisition information set is subjected to partition processing, and multiple image acquisition information subsets are acquired. The image feature processing result is acquired through the feature function of the image feature processing system and the multiple image acquisition information subsets in combination with the acquisition device parameter information. The image feature processing result is displayed and output based on the display module of the image feature processing system. The image feature processing method and system based on machine vision are provided, which achieves the technical effects of gray scale matching, partition processing, improved image feature processing accuracy, matching and determining the image feature processing result according to the feature function, and improved adaptability of the feature function image feature processing result.

[0073] Due to the adoption of the image acquisition device, the target object is subjected to full-angle image acquisition, image acquisition information set is acquired, and structural features of the target object are acquired through feature processing. The image acquisition information set is subjected to three-component segmentation, three-component segmented image set is acquired, and the image acquisition information set is subjected to partition processing, and multiple image acquisition information subsets are acquired. The image feature processing result is acquired through the feature function of the image feature processing system and the multiple image acquisition information subsets in combination with the acquisition device parameter information. The image feature processing result is displayed and output based on the display module of the image feature processing system. The image feature processing method and system based on machine vision are provided, which achieves the technical effects of gray scale matching, partition processing, improved image feature processing accuracy, matching and determining the image feature processing result according to the feature function, and improved adaptability of the feature function image feature processing result.

[0074] Due to the adoption of the image acquisition device, the target object is subjected to full-angle image acquisition, image acquisition information set is acquired, and structural features of the target object are acquired through feature processing. The image acquisition information set is subjected to three-component segmentation, three-component segmented image set is acquired, and the image acquisition information set is subjected to partition processing, and multiple image acquisition information subsets are acquired. The image feature processing result is acquired through the feature function of the image feature processing system and the multiple image acquisition information subsets in combination with the acquisition device parameter information. The image feature processing result is displayed and output based on the display module of the image feature processing system. The image feature processing method and system based on machine vision are provided, which achieves the technical effects of gray scale matching, partition processing, improved image feature processing accuracy, matching and determining the image feature processing result according to the feature function, and improved adaptability of the feature function image feature processing result.

[0075] Embodiment Two

[0076] Based on the same inventive concept as the image feature processing method based on machine vision in the foregoing embodiments, as shown in Figure 4 The application provides an image feature processing system based on machine vision, wherein the system comprises:

[0077] The parameter extraction unit 11 is configured to extract acquisition device parameter information through an image acquisition device.

[0078] The full-angle acquisition unit 12 is configured to perform full-angle image acquisition on the target object by the image acquisition device and the acquisition device parameter information, and obtain an image acquisition information set;

[0079] The structural feature acquisition unit 13 is configured to perform feature processing on the image acquisition information set, and obtain a target object structural feature;

[0080] The three-component segmentation unit 14 is configured to perform three-component segmentation on the image acquisition information set, and obtain a three-component segmented image set;

[0081] The partition processing unit 15 is configured to perform partition processing on the image acquisition information set by the target object structural feature and the three-component segmented image set, and obtain a plurality of image acquisition information subsets;

[0082] The image feature acquisition unit 16 is configured to obtain an image feature processing result by the feature function of the image feature processing system, the plurality of image acquisition information subsets, and the acquisition device parameter information;

[0083] The display output unit 17 is configured to display and output the image feature processing result based on a display module of the image feature processing system.

[0084] Further, the system comprises:

[0085] The acquisition constraint information acquisition unit is configured to obtain preset acquisition constraint information;

[0086] The acquisition constraint limiting unit is configured to perform acquisition constraint limiting on the target object and the image acquisition device by the preset acquisition constraint information, and obtain an image acquisition instruction;

[0087] The full-angle image acquisition unit is configured to perform full-angle image acquisition on the target object by the image acquisition device based on the image acquisition instruction and the acquisition device parameter information, and obtain an image acquisition information set.

[0088] Further, the system comprises:

[0089] The blurred image information set acquisition unit is configured to perform full-angle image acquisition on the target object by the image acquisition instruction and the acquisition device parameter information, and obtain a full-angle blurred image information set;

[0090] An information set identifying unit is configured to identify the full-angle blurred image information set by a preset identification precision, and obtain a detailed focusing instruction;

[0091] A local image collecting unit is configured to collect a local image of the target object by the detailed focusing instruction, and obtain a local detailed image information set;

[0092] An image information set combining unit is configured to combine the local detailed image information set and the full-angle blurred image information set, and obtain the image collecting information set.

[0093] Further, the system comprises:

[0094] A geometric feature obtaining unit is configured to import the image collecting information set into a three-dimensional scanner, and obtain a geometric feature of the target object;

[0095] A geometric contour extracting unit is configured to extract a geometric contour of the image collecting information set by the geometric feature of the target object, and obtain a contour feature of the target object;

[0096] A structure feature obtaining unit is configured to obtain a structure feature of the target object by the contour feature of the object and the geometric feature of the target object.

[0097] Further, the system comprises:

[0098] A feature function obtaining unit is configured to obtain a feature function of the image feature processing system, wherein the feature function comprises a positioning measurement function, a code reading and logic operation function, and a 3D modeling and defect function;

[0099] A positioning measurement unit is configured to perform positioning measurement on the multiple image collecting information subsets by the positioning measurement function, and obtain a measurement feature of the target object;

[0100] A 3D modeling unit is configured to perform 3D modeling based on the 3D modeling and defect function by the measurement feature of the target object and the structure feature of the target object, and obtain a 3D model of the target object;

[0101] A defect identifying unit is configured to perform defect identification on the 3D model of the target object by the code reading and logic operation function, and obtain a defect feature of the target object;

[0102] The feature processing result acquisition unit is configured to acquire image feature processing results by combining the target object measurement features and the target object defect features and the parameter information of the acquisition device.

[0103] Further, the system comprises:

[0104] The code label extraction unit is configured to extract code labels on the 3D model of the target object by reading code recognition functions.

[0105] The code label identification unit is configured to identify the code labels on the 3D model of the target object and acquire rectangular square record data symbol information.

[0106] The logical operation processing unit is configured to perform logical operations on the rectangular square record data symbol information and acquire target object label information.

[0107] The online networking extraction unit is configured to perform online networking based on the target object label information and extract a lossless 3D model of the target object.

[0108] The model comparison and analysis unit is configured to compare the 3D model of the target object with the lossless 3D model of the target object and acquire a structure comparison result.

[0109] The defect identification unit is configured to perform defect identification based on the structure comparison result and acquire target object defect features.

[0110] Further, the system comprises:

[0111] The positioning marking unit is configured to position and mark the target object measurement features, the rectangular square record data symbol information and target object label information, the 3D model of the target object, and the target object defect features based on the feature functions of the image feature processing system.

[0112] The data feedback unit is configured to perform data feedback by the positioning measurement function, feed back and output the target object measurement features, perform data feedback by the code reading and logical operation functions, feed back and output the rectangular square record data symbol information and target object label information, and perform data feedback by the 3D modeling and defect functions, feed back and output the 3D model of the target object and the target object defect features.

[0113] A data display output unit is configured to display and output the feature function of the image feature processing system and the corresponding feedback output information through a display module of the image feature processing system.

[0114] The present specification and drawings are only exemplary of the present application and are not intended to limit the present application thereto. Various modifications and combinations of the present application can be made without departing from the spirit and scope of the present application. Such modifications and variations of the present application are intended to be included within the scope of the claims of the present application and their equivalents.

Claims

1. A method for processing image features based on machine vision, characterized in that, The method is applied to an image feature processing system in communication connection with an image acquisition device, and comprises the following steps: Extracting acquisition device parameter information through the image acquisition device; Collecting full-angle image information of a target object through the image acquisition device and the acquisition device parameter information; Processing the collected full-angle image information to obtain structural features of the target object; Segmenting the collected full-angle image information into three components to obtain a three-component segmented image set; Processing the collected full-angle image information through the structural features of the target object and the three-component segmented image set to obtain a plurality of image information subsets; Obtaining image feature processing results through the feature function of the image feature processing system and the plurality of image information subsets in combination with the acquisition device parameter information; Displaying and outputting the image feature processing results based on a display module of the image feature processing system; Processing the collected full-angle image information to obtain structural features of the target object, which comprises the following steps: Importing the collected full-angle image information into a three-dimensional scanner to obtain geometric features of the target object; Extracting geometric contours of the collected full-angle image information through the geometric features of the target object to obtain contour features of the target object; Obtaining the structural features of the target object through the contour features of the target object and the geometric features of the target object.

2. The method of claim 1, wherein, Collecting full-angle image information of a target object through an image acquisition device and acquisition device parameter information, which comprises the following steps: Obtaining preset acquisition constraint information; Restricting the target object and the image acquisition device through the preset acquisition constraint information to obtain image acquisition instructions; Collecting full-angle image information of a target object through an image acquisition device based on the image acquisition instructions and the acquisition device parameter information to obtain image acquisition information.

3. The method of claim 2, wherein, Collecting full-angle image information of a target object through an image acquisition device based on the image acquisition instructions and the acquisition device parameter information to obtain image acquisition information, which comprises the following steps: Collecting full-angle image information of a target object through the image acquisition instructions and the acquisition device parameter information to obtain full-angle blurred image information; Obtaining detail focusing instructions by identifying the full-angle blurred image information through a preset identification accuracy; Collecting local image information of the target object through the detail focusing instructions to obtain local refined image information; Synthesizing the image acquisition information through the local refined image information and the full-angle blurred image information.

4. The method of claim 1, wherein, Obtaining image feature processing results through the feature function of the image feature processing system and the plurality of image information subsets in combination with the acquisition device parameter information, which comprises the following steps: Obtaining the feature function of the image feature processing system, which comprises positioning measurement function, code reading and recognition and logic operation function, 3D modeling and defect function; The plurality of image acquisition information subsets are subjected to positioning measurement through the positioning measurement function, and target object measurement features are obtained; Through the target object measurement features and the target object structural features, 3D modeling is performed based on the 3D modeling and defect function, and a target object 3D model is obtained; Through the code reading and logical operation function, defect identification is performed on the target object 3D model, and target object defect features are obtained; Through the target object measurement features and the target object defect features, combined with the acquisition device parameter information, an image feature processing result is obtained.

5. The method of claim 4, wherein, Through the code reading and logical operation function, defect identification is performed on the target object 3D model, and target object defect features are obtained, and the method comprises: Through the code reading function, the code label on the target object 3D model is extracted; The code label on the target object 3D model is identified, and rectangular square record data symbol information is obtained; Logical operation is performed on the rectangular square record data symbol information, and target object label information is obtained; Based on the target object label information, online networking is performed, and a target object lossless 3D model is extracted; The target object 3D model and the target object lossless 3D model are compared, and a structure comparison result is obtained; Through the structure comparison result, defect identification is performed, and target object defect features are obtained.

6. The method of claim 5, wherein, The method further comprises: Based on the feature functions of the image feature processing system, the target object measurement features, the rectangular square record data symbol information and the target object label information, the target object 3D model and the target object defect features are marked; Through the positioning measurement function, data feedback is performed, and the target object measurement features are fed back and output; through the code reading and logical operation function, data feedback is performed, and the rectangular square record data symbol information and the target object label information are fed back and output; through the 3D modeling and defect function, data feedback is performed, and the target object 3D model and the target object defect features are fed back and output; Through the display module of the image feature processing system, the feature functions of the image feature processing system and the corresponding feedback output information are displayed and output.

7. A machine vision-based image feature processing system characterized by, The system comprises: A parameter extraction unit, the parameter extraction unit is used for extracting acquisition device parameter information through an image acquisition device; A full-angle acquisition unit, the full-angle acquisition unit is used for acquiring image acquisition information set through full-angle image acquisition of a target object by the image acquisition device and the acquisition device parameter information; A structural feature acquisition unit, the structural feature acquisition unit is used for acquiring target object structural features through feature processing of the image acquisition information set; A three-component segmentation unit, the three-component segmentation unit is used for acquiring three-component segmented image set through three-component segmentation of the image acquisition information set; A partition processing unit, the partition processing unit is used for acquiring a plurality of image acquisition information subsets through partition processing of the image acquisition information set by the target object structural features and the three-component segmented image set; An image feature acquisition unit is configured to acquire an image feature processing result by combining the plurality of image acquisition information subsets with the acquisition device parameter information through a feature function of the image feature processing system; A display output unit is configured to display and output the image feature processing result based on a display module of the image feature processing system; A geometric feature acquisition unit is configured to acquire a geometric feature of a target object by inputting the image acquisition information set into a three-dimensional scanner; A geometric contour extraction unit is configured to acquire a contour feature of the target object by performing geometric contour extraction on the image acquisition information set based on the geometric feature of the target object; A structure feature acquisition unit is configured to acquire a structure feature of the target object based on the contour feature of the object and the geometric feature of the target object.

Images