A method and device for all-angle appearance inspection of plastic products

Abstract

The present invention provides a method and device for full-angle appearance detection of a plastic product, which is applied to the field of safety observation of plastic products. The method comprises: obtaining first basic information of a first plastic product, the first basic information including a first actual demand, constructing a first standard three-dimensional model, and using a depth camera to collect an image of the first plastic product to obtain a first depth image set. The images are sequentially input into a three-dimensional reconstruction processor to obtain a first three-dimensional reconstruction result. The first three-dimensional reconstruction result is compared with the first standard three-dimensional model to obtain a plurality of distinguishing features. It is sequentially judged whether the plurality of distinguishing features meet a preset error threshold, and a plurality of judgment results are obtained respectively. According to the plurality of judgment results, a first appearance detection result of the first plastic product is generated. The method solves the technical problem that the existing technology lacks a standard defect detection method for plastic products, resulting in large detection errors, low detection efficiency, and inability to accurately evaluate defect features.

Description

Technical Field

[0001] The present invention relates to the field of appearance inspection of plastic products, and in particular to a method and device for all-angle appearance inspection of plastic products. Background Art

[0002] Plastics are widely used in daily life and are indispensable components in home appliances, automobiles, and mobile phones. Plastics are also often used as raw materials for high-end runways. However, plastic products often have defects during the manufacturing or construction process, which affects the appearance and performance of the product.

[0003] However, in the prior art, defect characteristics of plastic products are often detected by human eyes. Due to the limitations of human eyes, more obvious defects can be discovered in time, but less obvious defect characteristics are difficult to detect. In addition, when evaluating defects, the evaluation of defect characteristics is inaccurate due to human factors.

[0004] Therefore, there is a technical problem in the prior art that there is a lack of a standard defect detection method for plastic products, which leads to large detection errors, low detection efficiency and inability to accurately evaluate defect characteristics. Summary of the invention

[0005] The present application provides a method and device for all-angle appearance inspection of plastic products, which is used to solve the technical problem in the prior art that there is a lack of a standard defect detection method for plastic products, resulting in large detection errors, low detection efficiency and inability to accurately evaluate defect characteristics.

[0006] In view of the above problems, the present application provides a method for detecting the appearance of a plastic product from all angles.

[0007] In a first aspect of the present application, a method for detecting the appearance of a plastic product from all angles is provided, the method comprising: obtaining first basic information of a first plastic product, wherein the first basic information comprises a first actual demand; constructing a first standard three-dimensional model according to the first actual demand; acquiring an image of the first plastic product using a depth camera to obtain a first depth image set, wherein the first depth image set comprises a plurality of RGB-D images; using the plurality of RGB-D images as input information, and sequentially inputting the plurality of RGB-D images into a three-dimensional reconstruction processor to obtain a first three-dimensional reconstruction result; comparing the first three-dimensional reconstruction result with the first standard three-dimensional model to obtain a first comparison result, wherein the first comparison result comprises a plurality of distinguishing features; sequentially judging whether the plurality of distinguishing features meet a preset error threshold, and obtaining a plurality of judgment results respectively; generating a first appearance detection result of the first plastic product according to the plurality of judgment results.

[0008] According to a second aspect of the present application, a full-angle appearance detection device for a plastic product is provided, the device comprising: a first obtaining unit, used to obtain first basic information of a first plastic product, wherein the first basic information includes a first actual demand; a first construction unit, used to construct a first standard three-dimensional model according to the first actual demand; a second obtaining unit, used to collect an image of the first plastic product using a depth camera to obtain a first depth image set, wherein the first depth image set includes multiple RGB-D images; a third obtaining unit, used to use the multiple RGB-D images as input information, and sequentially input them into a three-dimensional reconstruction processor to obtain a first three-dimensional reconstruction result; a first comparison unit, used to compare the first three-dimensional reconstruction result with the first standard three-dimensional model to obtain a first comparison result, wherein the first comparison result includes multiple distinguishing features; a first judgment unit, used to sequentially judge whether the multiple distinguishing features meet a preset error threshold, and obtain multiple judgment results respectively; a first generation unit, used to generate a first appearance detection result of the first plastic product according to the multiple judgment results.

[0009] The third aspect of the present application provides an electronic device, comprising: a processor and a memory, wherein the processor is coupled to the memory; the memory is used to store programs, and the processor is used to implement the steps of the method described in the first aspect when calling and executing.

[0010] A fourth aspect of the present application provides a computer-readable storage medium having a computer program stored thereon, and when the computer program is executed by a processor, the steps of the method described in the first aspect are implemented.

[0011] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0012] The method provided in the embodiment of the present application obtains the first basic information of the first plastic product, the first basic information includes the first actual demand, constructs the first standard three-dimensional model, uses the depth camera to collect the image of the first plastic product, obtains the first depth image set, sequentially inputs the three-dimensional reconstruction processor, obtains the first three-dimensional reconstruction result, compares the first three-dimensional reconstruction result with the first standard three-dimensional model, obtains the first comparison result, sequentially judges whether the multiple distinguishing features meet the preset error threshold, obtains multiple judgment results respectively, and generates the first appearance detection result of the first plastic product according to the multiple judgment results. The method solves the technical problem that the existing plastic product lacks a standard defect detection method, resulting in large detection errors, low detection efficiency and inability to accurately evaluate defect features. The method realizes the technical effect of quickly outputting accurate detection results for the appearance of plastic products through comprehensive detection, and the output detection results are more intuitive, so that construction personnel can quickly and specifically deal with the defects of plastic products.

[0013] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 A schematic diagram of a flow chart of a method for all-angle appearance inspection of a plastic product provided in this application;

[0015] Figure 2 A schematic diagram of a flow chart of obtaining a three-dimensional reconstruction processor in a method for inspecting the appearance of a plastic product at all angles provided in this application;

[0016] Figure 3 A schematic diagram of a process for obtaining a first marking instruction in a method for inspecting the appearance of a plastic product at all angles provided in the present application;

[0017] Figure 4 A schematic diagram of the structure of a full-angle appearance inspection device for plastic products is provided for this application;

[0018] Figure 5 This is a schematic diagram of the structure of an exemplary electronic device of the present application.

[0019] Explanation of the accompanying drawings: first obtaining unit 11, first constructing unit 12, second obtaining unit 13, third obtaining unit 14, first comparing unit 15, first judging unit 16, first generating unit 17, bus 300, memory 301, processor 302, communication interface 303, bus architecture 304. DETAILED DESCRIPTION

[0020] The present application provides a method and device for all-angle appearance inspection of plastic products, which is used to solve the technical problem in the prior art that there is a lack of a standard defect detection method for plastic products, resulting in large detection errors, low detection efficiency and inability to accurately evaluate defect characteristics.

[0021] In response to the above technical problems, the overall idea of ​​the technical solution provided by this application is as follows:

[0022] The method provided in the embodiment of the present application acquires an actual three-dimensional image of a plastic product by collecting RGB-D images of the plastic product and performing three-dimensional modeling. A standard three-dimensional model is established according to the actual needs of the plastic product, and the actual three-dimensional image of the plastic product is compared with the standard three-dimensional model to obtain multiple distinguishing features. The appearance inspection result of the plastic product is finally output by analyzing and processing multiple distinguishing features. The technical problem that the plastic products in the prior art lack a standard defect detection method, resulting in large detection errors, low detection efficiency and inability to accurately evaluate defect features is solved. The comprehensive detection of the appearance of the plastic product is achieved to quickly output accurate detection results, and the output detection results are more intuitive, so that construction personnel can quickly and specifically deal with the defects of the plastic product.

[0023] After introducing the basic principles of the present application, the technical solutions in the present application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all of the embodiments of the present application. It should be understood that the present application is not limited to the example embodiments described here. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present application. It should also be noted that, for the convenience of description, only the parts related to the present application are shown in the accompanying drawings, rather than all of them.

[0024] Embodiment 1

[0025] like Figure 1 As shown, the present application provides a method for detecting the appearance of a plastic product at all angles, the method comprising:

[0026] S100: Obtaining first basic information of a first plastic product, wherein the first basic information includes a first actual demand;

[0027] Specifically, plastics are widely used in daily life. Since plastic products often have defects during the manufacturing or construction process, the appearance and performance of the products are affected. By obtaining the first basic information of the first plastic product, wherein the first plastic product is a plastic product after the plastic material has been processed. The first basic information is the demand information of the plastic product and the use environment information. Among them, the demand information of the plastic product includes the appearance information such as the color, size and specifications of the plastic product. Among them, the first basic information includes the first actual demand, and the first actual demand is the appearance demand such as the size and color of the plastic material after construction.

[0028] S200: constructing a first standard three-dimensional model according to the first actual demand;

[0029] Specifically, according to the first actual demand, that is, the appearance demand such as the size and color of the plastic material after construction, a standard three-dimensional model is constructed, and the standard three-dimensional model is the standard effect of the plastic product after construction. The three-dimensional model can be constructed by various types of three-dimensional modeling software. By constructing the first standard three-dimensional model as the standard for judging the plastic material after construction, the comparison between the first standard three-dimensional model and the effect after construction is more visual, and it is more convenient to compare the construction effect of the plastic product.

[0030] S300: using a depth camera to capture an image of the first plastic product to obtain a first depth image set, wherein the first depth image set includes a plurality of RGB-D images;

[0031] Specifically, the image of the first plastic product is collected by a depth camera, wherein the depth camera collects the image by tilting an ultra-depth camera 20 degrees to sample the image, and then a set of 360-degree rotating vacuum suction high-speed platforms are designed to achieve 360-degree free-angle shooting and detection to collect the image of the first plastic product. The image of the first plastic product is obtained to form a first depth image set, and the first depth image set contains multiple RGB-D images of the first plastic product. The RGB-D image is a Depth image obtained on the basis of an ordinary three-channel color image. Each pixel value of the Depth image is the actual distance between the sensor and the object, and the RGB image and the Depth image have a one-to-one correspondence. The first depth image set is obtained based on the obtained RGB-D image to provide support for the subsequent establishment of a three-dimensional model of the first plastic product.

[0032] S400: using the plurality of RGB-D images as input information and sequentially inputting them into a 3D reconstruction processor to obtain a first 3D reconstruction result;

[0033] Specifically, the image of the first plastic product captured by the depth camera, i.e., the RGB-D image, is used as input information and input into a three-dimensional reconstruction processor for processing, wherein the three-dimensional reconstruction processor includes two processing layers, and the two processing layers sequentially construct the geometric relationship of the image and assign colors to the constructed geometric relationship. One of the processing layers obtains the actual distance between the sensor and the object for each pixel value according to the Depth image in the RGB-D image to realize the construction of the geometric relationship of the first plastic product. The result processed by the processing layer can obtain the geometric relationship of the first plastic product. Then, the second processing layer combines the geometric relationship of the first plastic product output by the first processing layer, and performs color assignment processing on the geometric relationship of the first plastic product. The result after processing is the first three-dimensional reconstruction result. The first three-dimensional reconstruction result includes both the real geometric relationship of the first plastic product and the real color embodiment of the first plastic product, and can intuitively obtain the actual characteristics of the first plastic product, providing technical support for the subsequent comparison with the standard three-dimensional model.

[0034] S500: Compare the first 3D reconstruction result with the first standard 3D model to obtain a first comparison result, wherein the first comparison result includes a plurality of distinguishing features;

[0035] Specifically, the first three-dimensional reconstruction result is compared with the first standard three-dimensional model, and the model features of the first three-dimensional reconstruction result are obtained and collected to form a first actual model feature set, wherein the first actual model feature set includes multiple actual features. The model feature set of the first three-dimensional reconstruction result includes the color features of different regions of the first three-dimensional reconstruction result and the size specifications of the model of the first three-dimensional reconstruction result, corner burrs, stain areas, etc. The model features of the first standard three-dimensional model are collected to form a first standard model feature set, wherein the first standard model feature set includes multiple standard features. The first standard three-dimensional model feature set includes the color features of different regions of the first standard three-dimensional model and the model size specifications of the first standard three-dimensional model. The multiple actual features are compared with the multiple standard features to obtain the multiple distinguishing features, and the multiple distinguishing features refer to the characteristic points of the poor appearance of the first plastic product, that is, the points of poor appearance.

[0036] S600: sequentially determining whether the plurality of distinguishing features meet a preset error threshold, and obtaining a plurality of determination results respectively;

[0037] S700: Generate a first appearance inspection result of the first plastic product according to the multiple judgment results.

[0038] Specifically, it is determined in sequence whether the multiple distinguishing features meet the preset error threshold, wherein the preset error threshold is the error threshold preset during the device processing, and each feature has a preset error threshold, and the preset error threshold includes the error thresholds of multiple appearance features. When the distinguishing feature meets the preset error threshold, it means that although the distinguishing feature in the first three-dimensional reconstruction result is different from the distinguishing feature in the first standard three-dimensional model, the difference is within the allowable range. When the distinguishing feature does not meet the preset error threshold, it means that the difference between the distinguishing feature in the first three-dimensional reconstruction result and the distinguishing feature in the first standard three-dimensional model is more obvious, so the distinguishing feature is the point of unqualified actual appearance. Due to the influence of various factors in the construction process of the first plastic product, the first plastic product cannot be completely consistent with the first standard three-dimensional model. Therefore, by setting the preset error threshold, the distinguishing features within the allowable range can be directly filtered out, and the points of unqualified actual appearance can be screened out for output. After completing the judgment of all distinguishing features and outputting the judgment results of each distinguishing feature, the appearance inspection of the first plastic product is completed, and the judgment results of all distinguishing features are collectively output to obtain the final first appearance inspection result. It realizes the technical effect of comprehensively testing the appearance of plastic products and outputting corresponding test results, and the output test results are more intuitive, so that construction personnel can quickly and specifically deal with the defects of plastic products.

[0039] like Figure 2 As shown, the method provided in the embodiment of the present application includes, before completing step S400:

[0040] S410: constructing a first processing layer using a Kinect Fusion algorithm, wherein the first processing layer is used for geometric reconstruction processing;

[0041] S420: constructing a second processing layer by using a threshold-limited texture map, wherein the second processing layer is used for texture reconstruction processing;

[0042] S430: constructing the three-dimensional reconstruction processor according to the first processing layer and the second processing layer.

[0043] Specifically, the three-dimensional reconstruction processor is used to take multiple RGB-D images as input information to obtain the constructed three-dimensional reconstruction results. The three-dimensional reconstruction processor includes two processing layers, namely the first processing layer and the second processing layer. The first processing layer is used for geometric reconstruction, and the first processing layer is constructed by the Kinect Fusion algorithm. The Kinect Fusion algorithm is a real-time reconstruction tracking and reconstruction algorithm based on an RGB-D camera and suitable for any ambient lighting. The algorithm is mainly divided into four steps when working, generating a three-dimensional point cloud, point cloud preprocessing, point cloud pose estimation, and point cloud fusion. The geometric reconstruction processing of the first plastic product image can be completed through the processing of the above steps. The image obtained after processing by the first processing layer can only reflect the geometric relationship of the first plastic product image, and then the obtained geometric relationship is input into the second processing layer. The second processing layer is used for texture reconstruction processing, that is, adding texture information to the geometric relationship, wherein the second processing layer uses a threshold-limited texture map to construct. The construction of the three-dimensional reconstruction processor is completed according to the first processing layer and the second processing layer obtained above.

[0044] Step S400 in the method provided in the embodiment of the present application includes:

[0045] S440: extracting RGB images and Depth images from the multiple RGB-D images in sequence to form multiple RGB images and multiple Depth images respectively;

[0046] S450: extracting depth data of the plurality of Depth images in sequence to form a plurality of groups of depth data;

[0047] S460: Processing the multiple sets of depth data using the first processing layer to obtain a first geometric reconstruction model;

[0048] S470: extracting color data of the plurality of RGB images in sequence to form a plurality of groups of color data;

[0049] S480: Performing texture mapping on the first geometric reconstruction model using the second processing layer according to the multiple sets of color data to obtain the first three-dimensional reconstruction result.

[0050] Specifically, the RGB image and the Depth image in the multiple RGB-D images are extracted in turn. Each pixel value of the Depth image is the actual distance between the sensor and the object, that is, the depth data. The actual distance data of the sensor from the object of the multiple Depth images are extracted in turn to form multiple groups of depth data. The multiple groups of depth data are processed by the first processing layer obtained. When working, the processing layer generates a three-dimensional point cloud from the multiple groups of depth data, and pre-processes the generated point cloud. After completing the point cloud pose estimation and point cloud fusion, the construction of the geometric model can be completed. The geometric model constructed by the first processing layer is the first geometric reconstruction model. At this time, the first geometric reconstruction model can only reflect the geometric relationship of the first plastic product image. Subsequently, the color data of the extracted multiple RGB images are combined into multiple groups of color data. The RGB image is composed of three component images, in which each pixel corresponds to the three color components of red, green and blue. The color of each pixel is obtained according to the superposition of the three color components of red, green and blue. The second processing layer is used to texture map the first geometric reconstruction model with the multiple sets of color data, that is, to assign color to each pixel in the first geometric reconstruction model. After the second processing layer, each pixel in the first geometric reconstruction model has a corresponding color. The output image at this time is the first three-dimensional reconstruction result, which can fully display the color characteristics of the first plastic product and clearly display the geometric characteristics of the first plastic product. By obtaining the first three-dimensional reconstruction result, the three-dimensional construction of the first plastic product is completed, so that the characteristics of the first plastic product can be clearly displayed.

[0051] like Figure 3 As shown, step S600 in the method provided in the embodiment of the present application includes:

[0052] S610: extracting a first distinguishing feature of the plurality of distinguishing features;

[0053] S620: Matching a first preset error threshold of the first distinguishing feature according to the preset error threshold;

[0054] S630: Determine whether the first distinguishing feature meets the first preset error threshold;

[0055] S640: If the first distinguishing feature meets the first preset error threshold, obtain a first marking instruction, wherein the first marking instruction is used to mark the first distinguishing feature as qualified.

[0056] Specifically, extract the first distinguishing feature of the plurality of distinguishing features, the first distinguishing feature is a single feature among the plurality of distinguishing features, each feature has a preset error threshold, such as surface stains and defects, the error threshold specifically specifies the stain area, stain position, etc., such as corner burr defects, the error threshold specifically specifies the maximum size of the burr, etc. According to the preset error threshold, match the first preset error threshold of the first distinguishing feature, wherein the first preset error threshold is the corresponding error threshold of the first distinguishing feature. Determine whether the first distinguishing feature meets the first preset error threshold. When the first distinguishing feature meets the first preset error threshold, the distinguishing feature is within the allowable range, and obtain a first marking instruction, which is used to mark the first distinguishing feature as qualified. For example, the first distinguishing feature is a burr defect, the error threshold specifically specifies that the maximum size of the burr is 1 square meter, and the first distinguishing feature burr defect obtained is 0.5 square meters. At this time, the first distinguishing feature meets the first preset error threshold, and the first distinguishing feature can be judged as qualified, and the qualified mark of the first distinguishing feature, i.e., the first marking instruction, is output. It is realized that the appearance defects are reasonably screened according to the appearance distinguishing feature, and the waste of resources caused by excessive processing of the distinguishing features with less influence is avoided.

[0057] Step S630 in the method provided in the embodiment of the present application includes:

[0058] S631: If the first distinguishing feature does not meet the first preset error threshold, obtain a second marking instruction;

[0059] S632: performing a defect mark on the first distinguishing feature according to the second marking instruction to obtain a first defect feature;

[0060] S633: Evaluate the defect grade of the first defect feature based on a preset defect rating standard to obtain a first defect level;

[0061] S634: Matching a first customer group of the first defect level based on a preset customer rating standard.

[0062] Specifically, if the first distinguishing feature does not meet the first preset error threshold, the distinguishing feature has exceeded the preset error threshold, that is, the impact of the distinguishing feature is large and constitutes a defect, and a second marking instruction is obtained, wherein the second marking instruction is used to mark the distinguishing feature as a defect feature, and the defect feature after the mark is obtained is the first defect feature. The first defect feature is rated for defect grade based on the preset defect rating standard to obtain the first defect level of the first defect feature. The preset defect rating standard is a rating standard formed by the first defect feature exceeding the preset error threshold range. Exemplarily, the first distinguishing feature is a burr defect, and the error threshold specifically stipulates that the maximum size of the burr is 1 square meter. The first distinguishing feature burr defect obtained is 2 square meters. At this time, the burr defect exceeds the preset range of 1 square meter. If the first plastic product has a larger area, the impact of the defect of 1 square meter is small. At this time, the preset defect rating is low. The preset defect rating standard can be set according to the actual situation of the first plastic product. Based on the preset customer rating standard, the first customer group of the first defect level is matched, wherein the preset customer rating standard is the different rating standards for defects of different customers, and the first customer group is the specific requirements of customers for the first plastic product. By matching the first defect level and the requirements of customers in the first customer group, the defects are processed on demand. The technical effect of formulating a personalized defect product processing plan for customer needs in view of the appearance defects is achieved, saving resources and avoiding waste.

[0063] Step S100 in the method provided in the embodiment of the present application includes:

[0064] S110: Determine a first actual application environment according to the first basic information;

[0065] S120: Obtaining a first aging test plan according to the first actual application environment;

[0066] S130: testing the first plastic product according to the first aging test scheme to obtain a first test result;

[0067] S140: Determine a first duration of appearance failure of the first plastic product according to the first test result;

[0068] S150: Using the first time period as a replacement period for the first plastic product.

[0069] Specifically, the first actual application environment is determined through the first basic information, and the first actual application environment is the actual use environment of the plastic product. According to the actual use environment of the plastic product, the first aging test scheme is obtained, and the first aging test scheme is a process of simulating the aging of the plastic product in the actual use environment conditions and conducting corresponding conditions to strengthen the experiment. The first plastic product is tested through the first aging test scheme to obtain the final experimental results, and the first duration of the appearance failure of the first plastic product is determined according to the experimental results, that is, the duration of the appearance failure of the first plastic product in the actual use environment conditions is obtained through the aging experiment, and the first duration is used as the replacement cycle of the first plastic product, that is, the duration of the appearance failure of the first plastic product in the actual use environment conditions is obtained through the aging experiment as the replacement cycle of the first plastic product. The acquisition of the first duration provides guidance for the actual application of plastic products, regular replacement and maintenance, and ensures safe use.

[0070] Step S120 in the method provided in the embodiment of the present application includes:

[0071] S121: Obtain a first environment feature set according to the first actual application environment, wherein the first environment feature set includes a plurality of environment features;

[0072] S122: Collecting external factors that affect the appearance of plastic products based on big data to form a first external factor set, wherein the first external factor set includes multiple factors;

[0073] S123: Traversing the multiple factors according to the multiple environmental features to form a first actual influencing factor set, wherein the first actual influencing factor set includes multiple actual factors;

[0074] S124: according to the first actual application environment, sequentially obtaining the factor level of each actual factor in the multiple actual factors to form multiple actual levels;

[0075] S125: constructing an actual factor-level list according to the multiple actual factors and the multiple actual levels, and generating the first aging test plan according to the actual factor-level list.

[0076] Specifically, environmental features in the first actual application environment are obtained, and the environmental features are combined into an environmental feature set, namely, a first environmental feature set. For example, the temperature range in the actual use environment is 0-38°C, and the temperature range is used as an element in the environmental feature set to construct the first environmental feature set. External factors affecting the appearance of plastic products are collected through big data. For example, the external factors affecting the appearance of plastic products include temperature, sunlight, humidity and other external environmental factors, and all the external factors affecting the appearance of plastic products are combined into a first external factor set. According to multiple environmental features, multiple factors are traversed to obtain a first actual influencing factor set. The first actual influencing factor set is the factors in the environmental features that affect the appearance of plastic products. The first actual influencing factor contains multiple actual influencing factors. According to the first actual application environment, the factor level of each actual factor in the actual factor is obtained in turn, and the factor levels of all actual factors are obtained to form multiple actual levels. The factor level is the severity of the impact of each factor on the appearance of the plastic product. According to multiple actual factors and multiple actual levels, an actual factor-level list is constructed, that is, a list of actual influencing factors and influencing degrees. According to the above list, a first aging test scheme is obtained. The first aging test scheme is used to achieve the actual influencing factors and influencing degrees in a targeted manner. By determining the environmental factors that affect aging in actual applications of plastic products, and then determining the severity of each factor, that is, the factor level, the first aging test plan can be obtained to facilitate accurate simulation of subsequent aging tests.

[0077] In summary, the method provided in the embodiment of the present application obtains the first basic information of the first plastic product, the first basic information includes the first actual demand, and obtains the actual three-dimensional image of the plastic product. A standard three-dimensional model is established according to the actual demand of the plastic product, and the first plastic product image is collected by a depth camera to obtain a first depth image set, which is sequentially input into a three-dimensional reconstruction processor to obtain a first three-dimensional reconstruction result. The actual three-dimensional image of the plastic product is compared with the standard three-dimensional model to obtain a plurality of distinguishing features. It is sequentially judged whether the plurality of distinguishing features meet the preset error threshold, and a plurality of judgment results are obtained respectively. According to the plurality of judgment results, a first appearance detection result of the first plastic product is generated. The technical problem that the plastic product lacks a standard defect detection method in the prior art, resulting in large detection errors, low detection efficiency and inability to accurately evaluate defect features is solved. The appearance of the plastic product is comprehensively detected and accurate detection results are quickly output, and the output detection results are more intuitive, so that construction personnel can quickly and specifically handle the defects of the plastic product.

[0078] Embodiment 2

[0079] Based on the same inventive concept as the method for detecting the appearance of a plastic product at all angles in the aforementioned embodiment, Figure 4As shown, the present application provides a full-angle appearance inspection device for plastic products, the device comprising:

[0080] A first obtaining unit 11 is used to obtain first basic information of a first plastic product, wherein the first basic information includes a first actual demand;

[0081] A first construction unit 12, configured to construct a first standard three-dimensional model according to the first actual demand;

[0082] A second obtaining unit 13 is configured to collect an image of the first plastic product using a depth camera to obtain a first depth image set, wherein the first depth image set includes a plurality of RGB-D images;

[0083] A third obtaining unit 14 is used to input the plurality of RGB-D images as input information into a three-dimensional reconstruction processor in sequence to obtain a first three-dimensional reconstruction result;

[0084] A first comparison unit 15, configured to compare the first 3D reconstruction result with the first standard 3D model to obtain a first comparison result, wherein the first comparison result includes a plurality of distinguishing features;

[0085] A first judging unit 16 is used to judge in sequence whether the plurality of distinguishing features meet a preset error threshold, and obtain a plurality of judging results respectively;

[0086] The first generating unit 17 is configured to generate a first appearance detection result of the first plastic product according to the plurality of determination results.

[0087] Furthermore, the device also includes:

[0088] A second construction unit is used to construct a first processing layer using a KinectFusion algorithm, wherein the first processing layer is used for geometric reconstruction processing;

[0089] A first processing unit, configured to construct a second processing layer using a threshold-limited texture map, wherein the second processing layer is used for texture reconstruction processing;

[0090] The third construction unit is used to construct the three-dimensional reconstruction processor according to the first processing layer and the second processing layer.

[0091] Furthermore, the device also includes:

[0092] A first extraction unit is used to sequentially extract RGB images and Depth images from the multiple RGB-D images to form multiple RGB images and multiple Depth images respectively;

[0093] A second processing unit is used to extract depth data of the plurality of Depth images in sequence to form a plurality of groups of depth data;

[0094] a fourth obtaining unit, configured to process the plurality of sets of depth data using the first processing layer to obtain a first geometric reconstruction model;

[0095] A third processing unit is used to extract the color data of the multiple RGB images in sequence to form multiple groups of color data;

[0096] A fifth obtaining unit is used to perform texture mapping on the first geometric reconstruction model using the second processing layer according to the multiple groups of color data to obtain the first three-dimensional reconstruction result.

[0097] Furthermore, the device also includes:

[0098] A first extraction unit, configured to extract a first distinguishing feature of the plurality of distinguishing features;

[0099] A first matching unit, configured to match a first preset error threshold of the first distinguishing feature according to the preset error threshold;

[0100] A second judging unit, configured to judge whether the first distinguishing feature meets the first preset error threshold;

[0101] A sixth obtaining unit is configured to obtain a first marking instruction if the first distinguishing feature meets the first preset error threshold, wherein the first marking instruction is used to perform a qualified marking on the first distinguishing feature.

[0102] Furthermore, the device also includes:

[0103] A seventh obtaining unit, configured to obtain a second marking instruction if the first distinguishing feature does not meet the first preset error threshold;

[0104] an eighth obtaining unit, configured to perform a defect mark on the first distinguishing feature according to the second marking instruction to obtain a first defect feature;

[0105] a ninth obtaining unit, configured to evaluate the defect grade of the first defect feature based on a preset defect rating standard to obtain a first defect level;

[0106] The second matching unit is configured to match the first customer group of the first defect level based on a preset customer rating standard.

[0107] Furthermore, the device also includes:

[0108] a tenth obtaining unit, configured to determine a first actual application environment according to the first basic information;

[0109] an eleventh obtaining unit, configured to obtain a first aging test plan according to the first actual application environment;

[0110] a twelfth obtaining unit, configured to test the first plastic product according to the first aging test scheme to obtain a first test result;

[0111] a thirteenth obtaining unit, configured to determine a first duration of appearance failure of the first plastic product according to the first test result;

[0112] The fourth processing unit is used to use the first time period as a replacement period of the first plastic product.

[0113] Furthermore, the device also includes:

[0114] A fourteenth obtaining unit, configured to obtain a first environment feature set according to the first actual application environment, wherein the first environment feature set includes a plurality of environment features;

[0115] A fifteenth obtaining unit, configured to collect external factors affecting the appearance of plastic products based on big data to form a first external factor set, wherein the first external factor set includes a plurality of factors;

[0116] A sixteenth obtaining unit, configured to traverse the multiple factors according to the multiple environmental characteristics to form a first actual influencing factor set, wherein the first actual influencing factor set includes multiple actual factors;

[0117] a fifth processing unit, configured to sequentially obtain, according to the first actual application environment, a factor level of each actual factor in the multiple actual factors to form multiple actual levels;

[0118] The fourth construction unit is used to construct an actual factor-level list according to the multiple actual factors and the multiple actual levels, and generate the first aging test plan according to the actual factor-level list.

[0119] Embodiment 3

[0120] Based on the same inventive concept as the method for detecting the appearance of a plastic product at all angles in the aforementioned embodiment, the present application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, the method in embodiment one is implemented.

[0121] Exemplary Electronic Devices

[0122] Reference below Figure 5 To describe the electronic device of the present application.

[0123] Based on the same inventive concept as the method for detecting the appearance of a plastic product at all angles in the aforementioned embodiment, the present application also provides an electronic device, including: a processor and a memory, wherein the processor is coupled to the memory; the memory is used to store programs, and the processor is used to execute the steps of the method described in embodiment one by calling.

[0124] The electronic device 300 includes: a processor 302, a communication interface 303, and a memory 301. Optionally, the electronic device 300 may also include a bus architecture 304. The communication interface 303, the processor 302, and the memory 301 may be interconnected via the bus architecture 304; the bus architecture 304 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus architecture 304 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 5 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0125] The processor 302 may be a CPU, a microprocessor, an ASIC, or one or more integrated circuits for controlling the execution of the program of the present application.

[0126] The communication interface 303 uses any transceiver-like device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), wired access networks, etc.

[0127] The memory 301 may be a ROM or other types of static storage devices that can store static information and instructions, a RAM or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical disc, laser disc, optical disc, digital versatile disc, Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory may exist independently and be connected to the processor via the bus architecture 304. The memory may also be integrated with the processor.

[0128] The memory 301 is used to store computer-executable instructions for executing the solution of the present application, and the execution is controlled by the processor 302. The processor 302 is used to execute the computer-executable instructions stored in the memory 301, thereby realizing a method for detecting the appearance of a plastic product at all angles provided in the above embodiment of the present application.

[0129] Those skilled in the art will understand that the various digital numbers such as the first and second involved in this application are only for the convenience of description, and are not used to limit the scope of this application, nor do they indicate a sequence of precedence. "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. The character " / " generally indicates that the objects associated with each other are in an "or" relationship. "At least one" refers to one or more. At least two refers to two or more. "At least one", "any one" or similar expressions refer to any combination of these items, including any combination of single items (individuals) or plural items (individuals). For example, at least one item (individual, kind) of a, b, or c can represent: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0130] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server or data center. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more available media integrated. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive (SSD)), etc.

[0131] The various illustrative logic units and circuits described in this application can be implemented or operated by a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field programmable gate array (FPGA) or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or any combination of the above. The general-purpose processor can be a microprocessor, and optionally, the general-purpose processor can also be any conventional processor, controller, microcontroller or state machine. The processor can also be implemented by a combination of computing devices, such as a digital signal processor and a microprocessor, a plurality of microprocessors, one or more microprocessors in conjunction with a digital signal processor core, or any other similar configuration.

[0132] The steps of the method or algorithm described in this application can be directly embedded in hardware, software units executed by a processor, or a combination of the two. The software unit can be stored in RAM memory, flash memory, ROM memory, EPROM memory, EEPROM memory, register, hard disk, removable disk, CD-ROM or any other form of storage medium in the art. Exemplarily, the storage medium can be connected to the processor so that the processor can read information from the storage medium and write information to the storage medium. Optionally, the storage medium can also be integrated into the processor. The processor and the storage medium can be arranged in an ASIC, and the ASIC can be arranged in a terminal. Optionally, the processor and the storage medium can also be arranged in different components in the terminal. These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are performed on the computer or other programmable device to produce a computer-implemented process, so that the instructions executed on the computer or other programmable device provide for implementing the process in the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0133] Although the present application has been described in conjunction with specific features and embodiments thereof, it is obvious that various modifications and combinations may be made thereto without departing from the spirit and scope of the present application. Accordingly, this specification and the accompanying drawings are merely exemplary illustrations of the present application and are deemed to have covered any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application is intended to include these modifications and variations.

Claims

1. A method for detecting the appearance of a plastic product from all angles, characterized in that: The method comprises: Obtaining first basic information of a first plastic product, wherein the first basic information includes a first actual demand; According to the first actual demand, construct a first standard three-dimensional model; Using a depth camera to capture an image of the first plastic product to obtain a first depth image set, wherein the first depth image set includes a plurality of RGB-D images; The plurality of RGB-D images are sequentially input into a 3D reconstruction processor as input information to obtain a first 3D reconstruction result; Comparing the first three-dimensional reconstruction result with the first standard three-dimensional model to obtain a first comparison result, wherein the first comparison result includes a plurality of distinguishing features; sequentially determining whether the plurality of distinguishing features meet a preset error threshold, and obtaining a plurality of determination results respectively; generating a first appearance inspection result of the first plastic product according to the multiple judgment results; The multiple RGB-D images are used as input information and sequentially input into a three-dimensional reconstruction processor, which includes: Constructing a first processing layer using a Kinect Fusion algorithm, wherein the first processing layer is used for geometric reconstruction processing; constructing a second processing layer using a threshold-limited texture map, wherein the second processing layer is used for texture reconstruction processing; According to the first processing layer and the second processing layer, the three-dimensional reconstruction processor is constructed; The step of taking the plurality of RGB-D images as input information and sequentially inputting them into a three-dimensional reconstruction processor to obtain a first three-dimensional reconstruction result includes: Sequentially extracting RGB images and Depth images from the multiple RGB-D images to form multiple RGB images and multiple Depth images respectively; Extracting depth data of the plurality of Depth images in sequence to form a plurality of groups of depth data; Processing the plurality of sets of depth data using the first processing layer to obtain a first geometric reconstruction model; Extracting color data of the multiple RGB images in sequence to form multiple groups of color data; According to the multiple sets of color data, the first geometric reconstruction model is texture mapped using the second processing layer to obtain the first three-dimensional reconstruction result.

2. The method according to claim 1, characterized in that The step of sequentially determining whether the plurality of distinguishing features meet a preset error threshold comprises: extracting a first distinguishing feature of the plurality of distinguishing features; According to the preset error threshold, matching a first preset error threshold of the first distinguishing feature; Determining whether the first distinguishing feature meets the first preset error threshold; If the first distinguishing feature meets the first preset error threshold, a first marking instruction is obtained, wherein the first marking instruction is used to perform a qualified marking on the first distinguishing feature.

3. The method according to claim 2, characterized in that The determining whether the first distinguishing feature meets the first preset error threshold further includes: If the first distinguishing feature does not meet the first preset error threshold, obtaining a second marking instruction; Performing a defect mark on the first distinguishing feature according to the second marking instruction to obtain a first defect feature; Performing a defect grade assessment on the first defect feature based on a preset defect rating standard to obtain a first defect level; Based on a preset customer rating standard, a first customer group of the first defect level is matched.

4. The method according to claim 1, characterized in that The method further comprises: Determining a first actual application environment according to the first basic information; Obtaining a first aging test plan according to the first actual application environment; According to the first aging test plan, the first plastic product is tested to obtain a first test result; Determining a first duration of appearance failure of the first plastic product according to the first test result; The first duration is used as a replacement period of the first plastic product.

5. The method according to claim 4, characterized in that The step of obtaining a first aging test scheme according to the first actual application environment includes: Obtaining a first environment feature set according to the first actual application environment, wherein the first environment feature set includes a plurality of environment features; Collecting external factors that affect the appearance of plastic products based on big data to form a first external factor set, wherein the first external factor set includes multiple factors; Traversing the multiple factors according to the multiple environmental features to form a first actual influencing factor set, wherein the first actual influencing factor set includes multiple actual factors; According to the first actual application environment, sequentially obtaining the factor level of each actual factor in the multiple actual factors to form multiple actual levels; An actual factor-level list is constructed according to the multiple actual factors and the multiple actual levels, and the first aging test plan is generated according to the actual factor-level list.

6. A plastic product all-angle appearance inspection device, characterized in that: The device comprises: A first obtaining unit, configured to obtain first basic information of a first plastic product, wherein the first basic information includes a first actual demand; A first construction unit, configured to construct a first standard three-dimensional model according to the first actual demand; A second obtaining unit, configured to collect an image of the first plastic product using a depth camera to obtain a first depth image set, wherein the first depth image set includes a plurality of RGB-D images; A third obtaining unit is used to input the plurality of RGB-D images as input information into a three-dimensional reconstruction processor in sequence to obtain a first three-dimensional reconstruction result; A first comparison unit, configured to compare the first three-dimensional reconstruction result with the first standard three-dimensional model to obtain a first comparison result, wherein the first comparison result includes a plurality of distinguishing features; A first judgment unit is used to judge in sequence whether the plurality of distinguishing features meet a preset error threshold, and obtain a plurality of judgment results respectively; A first generating unit, configured to generate a first appearance detection result of the first plastic product according to the plurality of judgment results; The device also includes: A second construction unit is used to construct a first processing layer using a Kinect Fusion algorithm, wherein the first processing layer is used for geometric reconstruction processing; A first processing unit, configured to construct a second processing layer using a threshold-limited texture map, wherein the second processing layer is used for texture reconstruction processing; A third construction unit, configured to construct the three-dimensional reconstruction processor according to the first processing layer and the second processing layer; The device also includes: A first extraction unit is used to sequentially extract RGB images and Depth images from the multiple RGB-D images to form multiple RGB images and multiple Depth images respectively; A second processing unit is used to extract depth data of the plurality of Depth images in sequence to form a plurality of groups of depth data; a fourth obtaining unit, configured to process the plurality of sets of depth data using the first processing layer to obtain a first geometric reconstruction model; A third processing unit is used to extract the color data of the multiple RGB images in sequence to form multiple groups of color data; A fifth obtaining unit is used to perform texture mapping on the first geometric reconstruction model using the second processing layer according to the multiple groups of color data to obtain the first three-dimensional reconstruction result.

7. An electronic device, characterized in that: include: A processor and a memory, wherein the processor is coupled to the memory; the memory is used to store programs, and the processor is used to execute the steps of the method according to any one of claims 1 to 5 by calling.

8. A computer-readable storage medium, characterized in that: The storage medium stores a computer program, which implements the steps of the method according to any one of claims 1 to 5 when executed by a processor.

Images