A TV cabinet injection defect detection device and its detection and control method

Abstract

The present application relates to injection molding defect detection technical field, especially a kind of TV cabinet's injection molding defect detection device and its detection and control method.The defect detection control scheme is obtained about the defect detection image information of the model area of detection loss on TV cabinet sample, pixel heterogeneity calculation of local image structure tensor is carried out to defect monitoring image information, and first defect detection result is obtained;Based on the first defect monitoring result, the pixel particle heterogeneity area of the model area of detection loss is extracted, and the source pixel saturation gamut space color of pixel particle heterogeneity area is mapped to the target pixel saturation gamut space of qualified injection molding product to analyze the pixel contrast of injection molding defect type, and second defect detection result is obtained.The present application can carry out multi-directional injection molding finished product detection to TV cabinet by injection molding defect detection device, to identify whether injection molding finished product appears injection molding defect, provide optimization basis for the injection molding production of TV cabinet, improve the injection molding detection accuracy of TV cabinet.

Description

Technical Field

[0001] This invention relates to the field of injection molding defect detection technology, and in particular to an injection molding defect detection device for TV housings and its detection and control method. Background Technology

[0002] In the current television industry, with continuous technological advancements and rising consumer demands for product quality, the injection molding quality of television casings is receiving increasing attention. Injection molding, as one of the most critical processes in television casing production, often faces various minor, localized defects, such as bubbles, cracks, shrinkage, and surface unevenness. These defects not only affect the product's appearance but can also, to some extent, impact the television's structural strength and durability, thus affecting the overall quality of the television.

[0003] Existing defect detection equipment uses vision systems with low detection accuracy, making it difficult to efficiently and accurately detect minute defects in TV casings. Furthermore, current injection molding defect detection equipment typically relies on manual intervention or simple machine vision technology to identify defects. However, manual inspection is subjective, inefficient, and prone to missing defects, leading to inaccurate results, especially in large-scale production where it's difficult to promptly identify and eliminate defective products, increasing production costs and rework rates for TV casings. While some machine vision inspection technologies have been applied to the automated identification of injection molding defects, existing systems still face problems such as low accuracy, poor adaptability, and insufficient anti-interference capabilities. Accurate judgment is often difficult, particularly regarding complex appearance features and small defects on TV casings, reducing the reliability of injection molding defect detection. Therefore, there is a need to develop more intelligent, accurate, and real-time injection molding defect detection equipment and control methods to address issues such as missed detections, unique appearance features, and minute injection molding defects in TV casings, in order to improve detection accuracy, reduce manual intervention, and optimize production efficiency. Summary of the Invention

[0004] This invention overcomes the shortcomings of the prior art and provides a device for detecting injection molding defects in TV housings and a method for detecting and controlling such defects.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] The first aspect of this invention provides a detection and control method for an injection molding defect detection device for TV housings, comprising the following steps:

[0007] S102: Construct a standard injection molding mesh model for the TV casing. Based on the standard injection molding mesh model, generate a spatial element smooth mesh model through smooth model update. Variationally fit the local process defect detection model extracted from the image data on the spatial element smooth mesh model to the spatial element smooth mesh model and perform fit analysis to obtain the detection lost model area.

[0008] S104: Construct an element scalar embedding field for the specified detection area of ​​the TV casing, determine whether the grid scalar value is greater than the isosurface threshold, segment and embed the detection loss model area into the element scalar embedding field to obtain a spatial element coordinate matrix, and control the injection molding defect detection device to perform detection based on the spatial element coordinate matrix to obtain a defect detection control scheme.

[0009] S106: Obtain defect detection image information about the missing model area on the TV casing sample through the defect detection control scheme, perform pixel variation calculation and analysis on the local image structure tensor of the defect monitoring image information, and obtain the first defect detection result;

[0010] S108: Based on the first defect monitoring results, extract the pixel particle variation region of the detected missing model area, map the source pixel saturated color gamut of the pixel particle variation region to the target pixel saturated color gamut of the qualified injection molded product, and analyze the pixel contrast of the injection molding defect type to obtain the second defect detection result.

[0011] More specifically, step S102 includes the following steps:

[0012] Obtain standardized design drawings of TV housing, construct a standard injection molding mesh model of TV housing based on the standardized design drawings, and smoothly update the interwoven mesh vertices of the standard injection molding mesh model based on injection molding rework rate analysis to obtain a smooth mesh model of spatial elements of TV housing.

[0013] The injection molding defect detection device obtains a preset detection process for standardized injection molding of TV housings. The process shooting decision points of the vision camera are extracted through the preset detection process. Based on the process shooting decision points, the spatial element smoothing model is divided into several sub-smoothing models.

[0014] The inspection log extracts multi-frame defect detection image data of the TV casing output after the visual camera of the injection molding defect detection device has fully executed the preset detection process. The Sobel operator is introduced to extract features and model each frame of defect detection image data to obtain a local process defect detection model.

[0015] Obtain the spatial element distribution on the local process defect detection model corresponding to each sub-smoothing model, and construct a variational lower bound for the model log marginal likelihood based on the spatial element distribution;

[0016] Based on the spatial element smoothing model, the baseline model variational distribution of each sub-smoothing model is preset, the variational lower bound is maximized, the posterior model variational distribution of the local defect detection model is generated, the hash misalignment function between the posterior model variational distribution and the baseline model variational distribution is calculated, and the degree of fit between each local defect detection model and each sub-smoothing model is determined based on the hash misalignment function.

[0017] If the fit is greater than the preset fit, the local defect detection model corresponding to the fit is fitted onto the current sub-smoothing model, and finally the unfitted model area in the spatial element smoothing model is peeled off and marked as the detection missing model area.

[0018] More specifically, the process of obtaining standardized design drawings of the TV casing, constructing a standard injection molding mesh model of the TV casing based on the standardized design drawings, and smoothly updating the interwoven mesh vertices of the standard injection molding mesh model based on injection molding rework rate analysis to obtain a smooth mesh model of the spatial elements of the TV casing includes the following steps:

[0019] Obtain standardized design drawings of the TV casing, and obtain multiple sets of standardized three-dimensional design parameters of the TV casing based on the standardized design drawings;

[0020] Input multiple sets of the standardized three-dimensional design parameters into SolidWorks model design software for modeling calculations to generate a standard injection-molded mesh model of the TV housing. Extract the interlaced mesh vertices and the smoothing parameters between adjacent interlaced mesh vertices of the standard injection-molded mesh model.

[0021] A Laplacian coordinate operator is introduced, and a neighborhood coordinate platform for a standard injection molding mesh model is constructed by assigning smoothing parameters between adjacent interwoven mesh vertices in the Laplacian coordinate operator. The neighborhood information of the interwoven mesh vertices is calculated through the neighborhood coordinate platform to obtain the Laplacian neighborhood coordinates of each interwoven mesh vertex.

[0022] Based on the Laplace neighborhood coordinates, a weighted average positive and negative smoothing process is performed on each interwoven mesh vertex to update and iterate the current neighborhood position of each interwoven mesh vertex in the standard injection molding mesh model, so that each interwoven mesh vertex is closer to the average position of its own neighborhood vertices, and positive and negative smoothing weight values ​​are obtained.

[0023] Obtain the detection logs of the injection molding defect detection device and the injection molding production requirements of the TV housing. Extract the historical injection molding rework rate of the TV housing in a preset time sequence from the detection logs. Preset the allowable injection molding rework rate according to the injection molding production requirements.

[0024] Calculate the difference between the historical injection molding rework rate and the allowable injection molding rework rate. Based on the difference in injection molding rework rate, preset the smoothing update iteration frequency and repeat the above-mentioned weighted average interlaced mesh vertex forward smoothing and reverse smoothing update iteration steps until the smoothing update iteration frequency is reached, and generate positive and negative smoothing weight matrices.

[0025] Based on the positive and negative smoothing weight matrix, the standard injection molding mesh model is reconstructed to obtain the spatial element smoothing mesh model of the TV casing.

[0026] More specifically, step S104 includes the following steps:

[0027] The specified inspection area of ​​the TV housing captured by the vision camera inside the injection molding defect detection device is obtained, and the spatial specification parameters of the specified inspection area are obtained.

[0028] Based on the spatial specification parameters, construct an element scalar embedding field for the specified detection area. Only plan the field area corresponding to the missing detection model area in the element scalar embedding field, mark it as a sub-detection missing scalar field, obtain the fixed scalar value of the sub-detection missing scalar field, and preset the isosurface embedding threshold according to the fixed scalar value.

[0029] The region of the missing model is divided into N subordinate model bodies. The vertex mesh scalar value of each model vertex on each subordinate model body is obtained. The vertex mesh scalar value of each model vertex is checked to see if it is greater than the isosurface threshold.

[0030] If the vertex mesh scalar value is greater than the isosurface embedding threshold, then input a binary value of 1 for the vertex of the model; if the vertex mesh scalar value is greater than the isosurface embedding threshold, then input a binary value of 0 for the vertex of the model, and generate a binary vertex status code.

[0031] Based on the programming translation of the binary vertex status code, the embedding boundary of the missing scalar field of each sub-model body is determined, resulting in multiple scalar embedding boundaries. Linear interpolation calculation is then performed on each scalar embedding boundary based on the grid scalar value to obtain the corresponding element scalar embedding isosurface for each sub-model body.

[0032] Connect the corresponding element scalars of all the auxiliary model bodies to embed isosurfaces, and finally generate a spatial element coordinate matrix of the missing model area located in the specified detection area. Control the flipping mechanism to perform fixed-point shooting defect detection on the missing model area of ​​the TV shell in the specified detection area according to the spatial element coordinate matrix, and obtain the defect detection control scheme.

[0033] More specifically, step S106 includes the following steps:

[0034] Obtain a TV housing sample that has been injection molded, and use an injection molding defect detection device to execute the defect detection control scheme to detect the TV housing sample, so as to obtain defect detection image information about the area of ​​the missing model on the TV housing sample.

[0035] The Sobel operator is used to perform pixel-level edge detection on the defect detection image information to obtain the actual horizontal vector gradient and the actual vertical vector gradient of the detected lost model region. The second moment of the defect detection image information is calculated based on the actual horizontal vector gradient and the actual vertical vector gradient to generate the local image structure tensor of the detected lost model region.

[0036] The actual image resolution of the defect detection image information is obtained. Based on the actual image resolution, the feature decomposition of the local pixel structure is performed on the local image structure tensor to obtain a series of structural feature values ​​and structural feature vectors of the local graphic structure tensor. The actual pixel structure flatness of the detection missing model region is determined according to the series of structural feature values ​​and structural feature vectors.

[0037] Based on the standard injection molding mesh model of the TV housing, the injection molding defect detection device executes the defect detection control scheme to capture standard image information of the TV housing, and calculates the pixel structure flatness of the standard graphic information, which is defined as the reference pixel structure flatness.

[0038] If the actual pixel structure flatness is higher than the reference pixel structure flatness, the TV casing in the area of ​​the missing model is calibrated as a qualified injection molded product; if the actual pixel structure flatness is lower than the reference pixel structure flatness, the TV casing in the area of ​​the missing model is calibrated as a defective injection molded product, and the first defect detection result is obtained.

[0039] More specifically, step S108 includes the following steps:

[0040] If the first defect detection result shows that the TV casing is a defective injection molded product, then the pixel particle mutation area of ​​the defective injection molded product is extracted from the defect detection image information of the defective injection molded product based on the actual pixel structure flatness.

[0041] The color saturation range of the pixel particle mutation region is obtained and defined as the first color saturation range. Based on the first color saturation range, the pixel saturation color gamut space of the pixel particle mutation region in the defective injection molded product is constructed and defined as the source pixel saturation color gamut space.

[0042] The color saturation range of the pixel particle mutation region expressed in the qualified injection molded product is obtained and defined as the second color saturation range. The pixel saturation color gamut space of the pixel particle mutation region in the defective injection molded product is constructed based on the second color saturation range and defined as the target pixel saturation color gamut space.

[0043] By acquiring color mapping function tables for color saturation expression of different injection molding defect types and predetermined Gamma correction intervals for each type of injection molding defect through big data networks, color mapping rules are established based on the color mapping function tables.

[0044] A linear transformation algorithm is introduced. Based on the color mapping rule, each saturated color in the source pixel saturated color gamut space is mapped to the target pixel saturated color gamut space in the linear transformation algorithm to generate a linear color saturation mapping equation.

[0045] Solve the linear color saturation mapping equation to obtain a series of mapped chromaticity values. Calculate and determine the restored pixel contrast of the pixel particle variation region based on the series of mapped chromaticity values, and obtain the pixel contrast of the pixel particle variation region expressed in the qualified injection molded product, which is defined as the initial pixel contrast.

[0046] The Gamma correction formula is introduced to calculate the correction between the restored pixel contrast and the initial pixel contrast, and the actual Gamma correction value is obtained. If the actual Gamma correction value is within the predetermined Gamma correction range, the injection molding defect type corresponding to the predetermined Gamma correction range is directly output to obtain the second defect detection result.

[0047] A second aspect of the present invention provides an injection molding defect detection device for TV housings, the injection molding defect detection device comprising:

[0048] Image acquisition module, which is used to capture defect image data of TV housing injection molded finished products from multiple angles;

[0049] The image processing module is responsible for introducing the Sobel operator to calculate and extract feature information from defect detection image data of different frames;

[0050] The flipping module is used to clamp the TV casing within the specified detection area and flip it in multiple directions so that the image acquisition module can fully capture the detection loss model area of ​​the TV casing.

[0051] The model building module is used to perform modeling and calculation processing on the defect detection image data extracted by the image processing module to output a local process defect detection model.

[0052] The binary encoding module, the data processing module is used to check whether the vertex mesh scalar value of each model vertex on the auxiliary model body is greater than the isosurface embedding threshold of the sub-detection lost scalar field, and thus program the model vertex of the auxiliary model body with a binary status code of 0 or 1.

[0053] This invention addresses the technical deficiencies in the prior art, and its beneficial technical effects are as follows:

[0054] A standard injection molding mesh model for a TV casing is constructed. Based on this standard mesh model, a spatial element smoothing mesh model is generated through smoothing updates. A local process defect detection model extracted from image data on the spatial element smoothing mesh model is variationally fitted to the model, and a fitting degree analysis is performed to obtain the region of the missing model. An element scalar embedding field is constructed for the specified detection region of the TV casing. The region of the missing model is segmented and embedded into the element scalar embedding field based on whether the mesh scalar value is greater than the isosurface threshold, resulting in a spatial element coordinate matrix. An injection molding defect detection device is controlled based on this spatial element coordinate matrix to obtain a defect detection control scheme. Defect detection image information about the region of the missing model on the TV casing sample is obtained through the defect detection control scheme. Pixel variation calculations and analyses are performed on the local image structure tensor of the defect monitoring image information to obtain a first defect detection result. Based on the first defect monitoring result, pixel particle variation regions of the region of the missing model are extracted. The source pixel saturated color gamut of the pixel particle variation region is mapped to the target pixel saturated color gamut of the qualified injection molded product to analyze the pixel contrast of the injection molding defect type, resulting in a second defect detection result. This invention enables multi-directional inspection of TV housing injection molded products using an injection molding defect detection device to identify whether injection molding defects exist, providing an optimization basis for the injection molding production of TV housings and effectively improving the injection molding quality and efficiency of TV housings. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other embodiments can be obtained from these drawings without creative effort.

[0056] Figure 1 A schematic diagram of the overall structure of this device is shown;

[0057] Figure 2 A schematic diagram of the internal structure of this device is shown;

[0058] Figure 3 A bottom view of the structure of this device is shown. Detailed Implementation

[0059] To better understand the above-mentioned objectives, features, and advantages of the present invention, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments. It should be noted that, unless otherwise specified, the embodiments and features described in these embodiments can be combined with each other.

[0060] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0061] The first aspect of this invention provides a detection and control method for an injection molding defect detection device for TV housings, such as... Figure 1 As shown, it includes the following steps:

[0062] S102: Construct a standard injection molding mesh model for the TV casing. Based on the standard injection molding mesh model, generate a spatial element smooth mesh model through smooth model update. Variationally fit the local process defect detection model extracted from the image data on the spatial element smooth mesh model to the spatial element smooth mesh model and perform fit analysis to obtain the detection lost model area.

[0063] S104: Construct an element scalar embedding field for the specified detection area of ​​the TV casing, determine whether the grid scalar value is greater than the isosurface threshold, segment and embed the detection loss model area into the element scalar embedding field to obtain a spatial element coordinate matrix, and control the injection molding defect detection device to perform detection based on the spatial element coordinate matrix to obtain a defect detection control scheme.

[0064] S106: Obtain defect detection image information about the missing model area on the TV casing sample through the defect detection control scheme, perform pixel variation calculation and analysis on the local image structure tensor of the defect monitoring image information, and obtain the first defect detection result;

[0065] S108: Based on the first defect monitoring results, extract the pixel particle variation region of the detected missing model area, map the source pixel saturated color gamut of the pixel particle variation region to the target pixel saturated color gamut of the qualified injection molded product, and analyze the pixel contrast of the injection molding defect type to obtain the second defect detection result.

[0066] More specifically, step S102 includes the following steps:

[0067] Obtain standardized design drawings of TV housing, construct a standard injection molding mesh model of TV housing based on the standardized design drawings, and smoothly update the interwoven mesh vertices of the standard injection molding mesh model based on injection molding rework rate analysis to obtain a smooth mesh model of spatial elements of TV housing.

[0068] The injection molding defect detection device obtains a preset detection process for standardized injection molding of TV housings. The process shooting decision points of the vision camera are extracted through the preset detection process. Based on the process shooting decision points, the spatial element smoothing model is divided into several sub-smoothing models.

[0069] The inspection log extracts multi-frame defect detection image data of the TV casing output after the visual camera of the injection molding defect detection device has fully executed the preset detection process. The Sobel operator is introduced to extract features and model each frame of defect detection image data to obtain a local process defect detection model.

[0070] Obtain the spatial element distribution on the local process defect detection model corresponding to each sub-smoothing model, and construct a variational lower bound for the model log marginal likelihood based on the spatial element distribution;

[0071] Based on the spatial element smoothing model, the baseline model variational distribution of each sub-smoothing model is preset, the variational lower bound is maximized, the posterior model variational distribution of the local defect detection model is generated, the hash misalignment function between the posterior model variational distribution and the baseline model variational distribution is calculated, and the degree of fit between each local defect detection model and each sub-smoothing model is determined based on the hash misalignment function.

[0072] If the fit is greater than the preset fit, the local defect detection model corresponding to the fit is fitted onto the current sub-smoothing model, and finally the unfitted model area in the spatial element smoothing model is peeled off and marked as the detection missing model area.

[0073] It should be noted that TV casings come in a variety of shapes and sizes. Some TV casings have multiple irregular surfaces, grooves, fine fixing holes, and textures. Because these features are uniquely positioned on the TV casing, the vision camera modules of some existing injection molding defect detection devices struggle to capture images of these defects globally according to preset control decisions. This results in these injection molding defects being missed after molding, leading to localized missed detections and significantly impacting the usability of the TV casing. Therefore, eliminating these missed detection areas is crucial. To address this, this method constructs a smoothed spatial element mesh model of the TV casing. The aim is to more smoothly represent the spatial distribution of points, lines, and surfaces of the solid after injection molding, thereby maximizing the accuracy of the injection molding defect detection device's ability to detect these unique appearance features in spatial positions. Then, the local process defect detection model in the TV casing defect image captured by the injection molding defect detection device after executing the preset detection process is variationally fitted to the spatial element smooth mesh model. The model area that is not fitted is the missed area that the injection molding defect detection device cannot detect globally for some unique appearance features on the TV casing after reasonable control of the operation. That is, the detection missing model area. This may be caused by low control accuracy or improper control of the injection molding defect detection device. Therefore, the control analysis and optimization of the subsequent injection molding defect detection device can be carried out based on the extracted detection missing model area, so as to eliminate the phenomenon of the injection molding defect detection device missing unique appearance features and improve the richness and accuracy of multi-dimensional detection of injection molding defects.

[0074] It should be noted that, regarding the regional fitting of the local process defect detection model onto the spatial element smoothed mesh model, this method utilizes the spatial element distribution on the local process defect detection model to construct a variational lower bound for the model's logarithmic marginal likelihood. This provides a matching degree inference reference for the precise alignment and fitting of the two, making the model's fitting in space more accurate. Maximizing this variational lower bound means maximizing the matching degree between the point, line, and surface element distribution on the local process defect detection model and the spatial element smoothed mesh model. The larger the variational lower bound, the closer the point, line, and surface element distribution on the local process defect detection model is to the element distribution of a certain region of the spatial element smoothed mesh model. Therefore, after maximizing the variational lower bound, calculating the hash misalignment function between the posterior model variational distribution of the local defect detection model and the baseline model variational distribution determines the fitting degree. If the fitting degree is greater than the preset fitting degree, it indicates that the local defect detection model belongs to a specified spatial region on the spatial element smoothed mesh model. Therefore, the local defect detection model should be fitted onto the current sub-smoothing model. The model region that is not fitted is the region that is missed on the TV casing. This method can trace the missed areas of TV housing based on the existing control and detection results of the injection molding defect detection device, thus providing a reliable analytical basis for the subsequent image vision detection control of TV housing by the injection molding defect detection device, avoiding blind spot detection errors, eliminating global missed detection, and improving the detection and control quality of TV housing defects by the vision camera.

[0075] More specifically, the process of obtaining standardized design drawings of the TV casing, constructing a standard injection molding mesh model of the TV casing based on the standardized design drawings, and smoothly updating the interwoven mesh vertices of the standard injection molding mesh model based on injection molding rework rate analysis to obtain a smooth mesh model of the spatial elements of the TV casing includes the following steps:

[0076] Obtain standardized design drawings of the TV casing, and obtain multiple sets of standardized three-dimensional design parameters of the TV casing based on the standardized design drawings;

[0077] Input multiple sets of the standardized three-dimensional design parameters into SolidWorks model design software for modeling calculations to generate a standard injection-molded mesh model of the TV housing. Extract the interlaced mesh vertices and the smoothing parameters between adjacent interlaced mesh vertices of the standard injection-molded mesh model.

[0078] A Laplacian coordinate operator is introduced, and a neighborhood coordinate platform for a standard injection molding mesh model is constructed by assigning smoothing parameters between adjacent interwoven mesh vertices in the Laplacian coordinate operator. The neighborhood information of the interwoven mesh vertices is calculated through the neighborhood coordinate platform to obtain the Laplacian neighborhood coordinates of each interwoven mesh vertex.

[0079] Based on the Laplace neighborhood coordinates, a weighted average positive and negative smoothing process is performed on each interwoven mesh vertex to update and iterate the current neighborhood position of each interwoven mesh vertex in the standard injection molding mesh model, so that each interwoven mesh vertex is closer to the average position of its own neighborhood vertices, and positive and negative smoothing weight values ​​are obtained.

[0080] Obtain the detection logs of the injection molding defect detection device and the injection molding production requirements of the TV housing. Extract the historical injection molding rework rate of the TV housing in a preset time sequence from the detection logs. Preset the allowable injection molding rework rate according to the injection molding production requirements.

[0081] Calculate the difference between the historical injection molding rework rate and the allowable injection molding rework rate. Based on the difference in injection molding rework rate, preset the smoothing update iteration frequency and repeat the above-mentioned weighted average interlaced mesh vertex forward smoothing and reverse smoothing update iteration steps until the smoothing update iteration frequency is reached, and generate positive and negative smoothing weight matrices.

[0082] Based on the positive and negative smoothing weight matrix, the standard injection molding mesh model is reconstructed to obtain the spatial element smoothing mesh model of the TV casing.

[0083] It should be noted that due to the diverse shapes of TV casings and the poor visual detection and control performance of some injection molding defect detection devices, smooth areas such as irregular surfaces or inner grooves are difficult for the device's visual camera to detect, resulting in blind spots. This causes distortion and other smoothness anomalies in the spatial outline texture display of the images of TV casings captured by the visual camera, leading to significant deviations in the subsequent tracing of missed areas on the TV casing. This affects the optimization of the device's control accuracy and reduces the quality of the device's injection molding defect detection and identification of TV casings. Therefore, it is necessary to provide a model that maintains and reflects the high smoothness of the standard outline texture of TV casings as much as possible to further accurately trace the missed areas when the injection molding defect detection device detects TV casings, thereby ensuring the accuracy and overall rationality of tracing the missed model areas. This method first constructs a standard injection-molded mesh model of a TV casing. Then, using the smoothing parameters between adjacent interwoven mesh vertices in this model, coordinates are assigned in the Laplacian coordinate operator. This assigns Laplacian neighborhood coordinates to each interwoven mesh vertex. The Laplacian neighborhood coordinates reflect the neighborhood relationship between the interwoven mesh vertex and its surrounding neighbors, thus determining the smoothing trend of each model detail's corresponding vertex in the standard model and revealing the general direction of smoothing optimization. Next, weighted average forward and reverse smoothing processes are performed on each interwoven mesh vertex based on the Laplacian neighborhood coordinates. This iteratively updates the current neighborhood position of each interwoven mesh vertex in the standard injection-molded mesh model. By updating the position of each vertex, it moves it towards the average position of its neighboring vertices, thereby achieving a smoothing effect. Specifically, forward smoothing adjusts each vertex with a positive weight value, causing its position to move closer to the center of its neighboring vertices. Conversely, reverse smoothing updates with a negative weight value to slightly reverse the oversmoothing effect of forward smoothing. This helps reduce the oversmoothing phenomenon that may occur in forward smoothing, preserves the details of the original shape, and thus achieves gradual smoothing optimization in the general direction of the trend towards smoothness, without destroying the overall structure of the original standard model. The final output positive and negative smoothing weight values ​​are the specific values ​​for optimizing the smoothness of the standard model, which can further make the surface of the mesh model smoother.

[0084] It should be noted that because the injection molding defect detection device uses a non-smooth TV housing model as its control basis, defect detection errors occur in the TV housing. This leads to a significant increase in the rework rate of TV housing injection molding production. Therefore, the injection molding rework rate depends on the accuracy of the smoothness of the TV housing model. Thus, this method limits the update frequency of forward and reverse smoothing based on the difference between the historical injection molding rework rate and the allowable injection molding rework rate. This allows the smoothing optimization of the model to approach the state constraint of the allowable injection molding rework rate as much as possible. This significantly improves the smoothness of the visual camera of the injection molding defect detection device in detecting points, lines, surfaces and other elements on the TV housing in space, ensuring the accuracy of tracing the source of detection blind spots such as missing model areas, and reducing the increase in rework rate caused by subsequent injection molding defect detection errors.

[0085] More specifically, step S104 includes the following steps:

[0086] The specified inspection area of ​​the TV housing captured by the vision camera inside the injection molding defect detection device is obtained, and the spatial specification parameters of the specified inspection area are obtained.

[0087] Based on the spatial specification parameters, construct an element scalar embedding field for the specified detection area. Only plan the field area corresponding to the missing detection model area in the element scalar embedding field, mark it as a sub-detection missing scalar field, obtain the fixed scalar value of the sub-detection missing scalar field, and preset the isosurface embedding threshold according to the fixed scalar value.

[0088] The region of the missing model is divided into N subordinate model bodies. The vertex mesh scalar value of each model vertex on each subordinate model body is obtained. The vertex mesh scalar value of each model vertex is checked to see if it is greater than the isosurface threshold.

[0089] If the vertex mesh scalar value is greater than the isosurface embedding threshold, then input a binary value of 1 for the vertex of the model; if the vertex mesh scalar value is greater than the isosurface embedding threshold, then input a binary value of 0 for the vertex of the model, and generate a binary vertex status code.

[0090] Based on the programming translation of the binary vertex status code, the embedding boundary of the missing scalar field of each sub-model body is determined, resulting in multiple scalar embedding boundaries. Linear interpolation calculation is then performed on each scalar embedding boundary based on the grid scalar value to obtain the corresponding element scalar embedding isosurface for each sub-model body.

[0091] Connect the corresponding element scalars of all the auxiliary model bodies to embed isosurfaces, and finally generate a spatial element coordinate matrix of the missing model area located in the specified detection area. Control the flipping mechanism to perform fixed-point shooting defect detection on the missing model area of ​​the TV shell in the specified detection area according to the spatial element coordinate matrix, and obtain the defect detection control scheme.

[0092] It should be noted that certain TV casings with irregular shapes, grooves, fine fixing holes, and textures require high-precision and high-resolution vision control to accurately detect defects in these unique features. However, some injection molding defect detection equipment still lacks the precision and rationality of vision control for TV casings. This makes it difficult for the vision camera to accurately explore and capture detailed images along the spatial elements of these features when the flipping mechanism is fully rotated. Consequently, the output images cannot display and identify more subtle hidden injection molding defects, reducing the detection accuracy of the injection molding defect detection device and causing a sharp increase in the error rate of detail detection. Therefore, controlling the fine detection of these unique features by the injection molding defect detection device is particularly important, as it directly affects the injection molding production quality of TV casings. To address this, this method constructs an element scalar embedding field for a specified detection area of ​​the TV casing. Spatial elements such as points, lines, and surfaces of the detected missing model area are mapped and embedded into this element scalar embedding field, ultimately generating a coordinate matrix of these spatial elements. This allows the vision camera to perform detailed traversal imaging of the detected missing model area of ​​the TV casing within the specified detection area based on the positional distribution described by this spatial element coordinate matrix. This enables the vision camera to acquire more detailed images of hidden defects in the detected missing model area, achieving a refined defect detection control effect. Compared to traditional techniques, this method improves the phenomenon that vision cameras are prone to global detection omissions or failure to capture unique appearance features, effectively improving the accuracy and reliability of defect detection for TV casings.

[0093] It should be noted that dividing the detected lost model region into N subordinate model bodies and mapping them using the vertex mesh scalar values ​​of each model vertex on each subordinate model body ensures that the detected lost model region can be inspected individually, improving the subdivision of the mapped lost model region. On the other hand, it reduces the error rate of the mapping embedding, improving the precision of detail defect detection. If the vertex mesh scalar value is greater than the isosurface embedding threshold, it indicates that the model vertex is located inside the surface of the sub-detected lost scalar field, so each vertex is set to a binary value of 1. If the vertex mesh scalar value is less than the isosurface embedding threshold, it indicates that the model vertex is located outside the surface of the sub-detected lost scalar field, so it is set to a binary value of 0. This describes the specific mapping point of the detected lost model region corresponding to the specified detection region. Based on the programming translation of these binary vertex status codes, the scalar embedding boundary can be quickly determined. Linear interpolation based on the vertex mesh scalar values ​​of the subordinate model bodies is then applied to each of the scalar embedding boundaries and connected to form the spatial element coordinate matrix of the detected lost model region located in the specified detection region. This method can quickly and accurately map and embed the spatial elements such as points, lines, and surfaces contained in the traced-out detection model area into the corresponding designated detection area of ​​the TV casing detected by the vision camera. This allows these spatial elements to be represented by a coordinate matrix in the designated detection area, providing a more authoritative basis for the flipping operation control of TV casing defect detection. As a result, the flipping mechanism of the injection molding defect detection device accurately flips the TV casing, enabling the vision camera to perform detailed detection of hidden defects in its unique appearance features. This improves the detail and reliability of TV casing defect detection and avoids defect detection errors and omissions.

[0094] More specifically, step S106 includes the following steps:

[0095] Obtain a TV housing sample that has been injection molded, and use an injection molding defect detection device to execute the defect detection control scheme to detect the TV housing sample, so as to obtain defect detection image information about the area of ​​the missing model on the TV housing sample.

[0096] The Sobel operator is used to perform pixel-level edge detection on the defect detection image information to obtain the actual horizontal vector gradient and the actual vertical vector gradient of the detected lost model region. The second moment of the defect detection image information is calculated based on the actual horizontal vector gradient and the actual vertical vector gradient to generate the local image structure tensor of the detected lost model region.

[0097] The actual image resolution of the defect detection image information is obtained. Based on the actual image resolution, the feature decomposition of the local pixel structure is performed on the local image structure tensor to obtain a series of structural feature values ​​and structural feature vectors of the local graphic structure tensor. The actual pixel structure flatness of the detection missing model region is determined according to the series of structural feature values ​​and structural feature vectors.

[0098] Based on the standard injection molding mesh model of the TV housing, the injection molding defect detection device executes the defect detection control scheme to capture standard image information of the TV housing, and calculates the pixel structure flatness of the standard graphic information, which is defined as the reference pixel structure flatness.

[0099] If the actual pixel structure flatness is higher than the reference pixel structure flatness, the TV casing in the area of ​​the missing model is calibrated as a qualified injection molded product; if the actual pixel structure flatness is lower than the reference pixel structure flatness, the TV casing in the area of ​​the missing model is calibrated as a defective injection molded product, and the first defect detection result is obtained.

[0100] It should be noted that during the injection molding process of TV casings, some subtle injection defects may occur that are difficult for cameras to detect. For example, a tiny crack may appear on an irregularly shaped surface. Existing injection defect detection devices may fail to capture these minute cracks due to inadequate resolution control of the visual camera, thus overlooking the defect. This could affect the TV casing's performance and quality during installation and use, increasing rework rates and costs. Therefore, precise detection of subtle injection defects on TV casings is necessary to determine the product's quality. To address this, this method first acquires defect image data of the missing model region of the TV casing by executing a defect detection control scheme through an injection molding defect detection device. Since the pixel features such as edges, textures, and contours in the image are gradient reflections of local brightness changes, the Sobel operator is used to extract the actual horizontal and vertical vector gradients of the missing model region at the pixel level from the defect detection image information. The image gradient is the basis of the structure tensor, which represents the degree of change in a local region. The gradient value is usually larger at edges and corners, while the second moment reflects the changes and interrelationships of the gradient. Constructing the second moment of the gradient can effectively capture the change patterns of local image regions. Therefore, the second moment of the defect detection image information is further calculated based on the gradient to generate the local image structure tensor of the missing model region. The structure tensor can further describe the local geometric features of the defect detection image information.

[0101] It should be noted that the capture of defect detection images is based on the resolution of the visual camera; therefore, the decomposition of local features of the image should be based on this resolution. Among these features, structural eigenvalues ​​reveal the main direction and magnitude of change in a local image region. The magnitude and differences of these eigenvalues ​​help distinguish different types of regional changes, i.e., whether a local image region is flat; structural feature vectors provide directional information about the local structure. Therefore, based on this series of structural eigenvalues ​​and structural feature vectors, the actual pixel structure flatness of the detected lost model region can be accurately determined. This actual pixel structure flatness reflects the degree of pixel anomaly in the local image of the detected lost model region. If the actual pixel structure flatness is higher than the reference pixel structure flatness, it indicates that the pixel structure is relatively stable and no anomalies have occurred, meaning that the TV casing does not have minor injection molding defects that alter the pixel structure of the image; therefore, the TV casing is a qualified injection molded product. If the actual pixel structure flatness is lower than the reference pixel structure flatness, it indicates that the defect detection image of the TV casing exhibits pixel anomalies, meaning that minor injection molding defects appear on the TV casing; therefore, the TV casing is a defective injection molded product. This method can calculate pixel variations of the structural tensor based on the captured defect detection image, thereby quickly identifying and detecting minor injection molding defects on the TV casing. This achieves accurate injection molding defect detection for TV casing products, improving the reliability and accuracy of TV casing defect detection.

[0102] More specifically, step S108 includes the following steps:

[0103] If the first defect detection result shows that the TV casing is a defective injection molded product, then the pixel particle mutation area of ​​the defective injection molded product is extracted from the defect detection image information of the defective injection molded product based on the actual pixel structure flatness.

[0104] The color saturation range of the pixel particle mutation region is obtained and defined as the first color saturation range. Based on the first color saturation range, the pixel saturation color gamut space of the pixel particle mutation region in the defective injection molded product is constructed and defined as the source pixel saturation color gamut space.

[0105] The color saturation range of the pixel particle mutation region expressed in the qualified injection molded product is obtained and defined as the second color saturation range. The pixel saturation color gamut space of the pixel particle mutation region in the defective injection molded product is constructed based on the second color saturation range and defined as the target pixel saturation color gamut space.

[0106] By acquiring color mapping function tables for color saturation expression of different injection molding defect types and predetermined Gamma correction intervals for each type of injection molding defect through big data networks, color mapping rules are established based on the color mapping function tables.

[0107] A linear transformation algorithm is introduced. Based on the color mapping rule, each saturated color in the source pixel saturated color gamut space is mapped to the target pixel saturated color gamut space in the linear transformation algorithm to generate a linear color saturation mapping equation.

[0108] Solve the linear color saturation mapping equation to obtain a series of mapped chromaticity values. Calculate and determine the restored pixel contrast of the pixel particle variation region based on the series of mapped chromaticity values, and obtain the pixel contrast of the pixel particle variation region expressed in the qualified injection molded product, which is defined as the initial pixel contrast.

[0109] The Gamma correction formula is introduced to calculate the correction between the restored pixel contrast and the initial pixel contrast, and the actual Gamma correction value is obtained. If the actual Gamma correction value is within the predetermined Gamma correction range, the injection molding defect type corresponding to the predetermined Gamma correction range is directly output to obtain the second defect detection result.

[0110] It should be noted that if the initial defect detection result indicates that the TV casing is a defective injection-molded product, it is necessary to further determine the type of injection molding defect to provide a strong basis for improvement in subsequent TV casing injection molding production optimization. Therefore, this method constructs a pixel color saturation color mapping rule using a color mapping function that expresses the color saturation of different injection molding defect types. Then, it linearly transforms each saturated color in the source pixel saturation color gamut of the pixel particle mutation region to the target pixel saturation color gamut. This achieves the effect of restoring the color saturation of the mutation pixel region on the defective injection-molded product to the corresponding color saturation that a qualified injection-molded product should possess. Furthermore, based on the mapped chromaticity manifestation of the restored color saturation, the type of injection molding defect causing the change in color saturation can be determined. The Gamma correction formula is used to adjust the image contrast, making the image display more consistent with human visual perception. Gamma correction enables the pixel chromaticity display output of the pixel particle mutation region expressed in a qualified injection-molded product to be the restored pixel display. Therefore, the Gamma correction value can serve as a key feature for tracing the type of injection molding defect. This method can trace the types of minor injection molding defects on TV housings, enabling rapid and reliable injection molding defect identification and helping to improve the optimization accuracy of TV housing injection molding processes.

[0111] A second aspect of the present invention provides a device for detecting injection molding defects in TV housings, such as... Figure 1-3 As shown, the injection molding defect detection device includes:

[0112] Image acquisition module 1011, the image acquisition module is used to capture defect image data of TV housing injection molded finished product from multiple angles;

[0113] Image processing module 1012, which is responsible for introducing the Sobel operator to calculate and extract feature information of defect detection image data of different frames;

[0114] The flipping module 1013 is used to clamp the TV casing within the specified detection area and flip it in multiple directions so that the image acquisition module can fully capture the detection loss model area of ​​the TV casing.

[0115] The model building module 1014 is used to perform modeling and calculation processing on the defect detection image data extracted by the image processing module to output a local process defect detection model.

[0116] The binary encoding module 1015, the data processing module is used to check whether the vertex mesh scalar value of each model vertex on the auxiliary model body is greater than the isosurface embedding threshold of the sub-detection lost scalar field, and thus program the model vertex of the auxiliary model body with a binary status code of 0 or 1.

[0117] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in the present invention should be included within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A detection and control method for injection molding defects in TV housings, characterized in that, Includes the following steps: S102: Construct a standard injection molding mesh model for the TV casing. Based on the standard injection molding mesh model, generate a spatial element smooth mesh model through smooth model update. Variationally fit the local process defect detection model extracted from the image data on the spatial element smooth mesh model to the spatial element smooth mesh model and perform fit analysis to obtain the detection lost model area. S104: Construct an element scalar embedding field for the specified detection area of ​​the TV casing, determine whether the grid scalar value is greater than the isosurface threshold, segment and embed the detection loss model area into the element scalar embedding field to obtain a spatial element coordinate matrix, and control the injection molding defect detection device to perform detection based on the spatial element coordinate matrix to obtain a defect detection control scheme. S106: Obtain defect detection image information about the missing model area on the TV casing sample through the defect detection control scheme, perform pixel variation calculation and analysis on the local image structure tensor of the defect monitoring image information, and obtain the first defect detection result; S108: Based on the first defect monitoring results, extract the pixel particle variation region of the detected missing model area, map the source pixel saturated color gamut of the pixel particle variation region to the target pixel saturated color gamut of the qualified injection molded product, and analyze the pixel contrast of the injection molding defect type to obtain the second defect detection result.

2. The detection and control method of the injection molding defect detection device for TV housing according to claim 1, characterized in that, Step S102 specifically includes the following steps: Obtain standardized design drawings of TV housing, construct a standard injection molding mesh model of TV housing based on the standardized design drawings, and smoothly update the interwoven mesh vertices of the standard injection molding mesh model based on injection molding rework rate analysis to obtain a smooth mesh model of spatial elements of TV housing. The injection molding defect detection device obtains a preset detection process for standardized injection molding of TV housings. The process shooting decision points of the vision camera are extracted through the preset detection process. Based on the process shooting decision points, the spatial element smoothing model is divided into several sub-smoothing models. The inspection log extracts multi-frame defect detection image data of the TV casing output after the visual camera of the injection molding defect detection device has fully executed the preset detection process. The Sobel operator is introduced to extract features and model each frame of defect detection image data to obtain a local process defect detection model. Obtain the spatial element distribution on the local process defect detection model corresponding to each sub-smoothing model, and construct a variational lower bound for the model log marginal likelihood based on the spatial element distribution; Based on the spatial element smoothing model, the baseline model variational distribution of each sub-smoothing model is preset, the variational lower bound is maximized, the posterior model variational distribution of the local defect detection model is generated, the hash misalignment function between the posterior model variational distribution and the baseline model variational distribution is calculated, and the degree of fit between each local defect detection model and each sub-smoothing model is determined based on the hash misalignment function. If the fit is greater than the preset fit, the local defect detection model corresponding to the fit is fitted onto the current sub-smoothing model, and finally the unfitted model area in the spatial element smoothing model is peeled off and marked as the detection missing model area.

3. The detection and control method of the injection molding defect detection device for TV housing according to claim 2, characterized in that, The process of obtaining standardized design drawings for the TV casing, constructing a standard injection molding mesh model for the TV casing based on the standardized design drawings, and smoothly updating the interwoven mesh vertices of the standard injection molding mesh model based on injection molding rework rate analysis to obtain a smooth mesh model of the spatial elements of the TV casing includes the following steps: Obtain standardized design drawings of the TV casing, and obtain multiple sets of standardized three-dimensional design parameters of the TV casing based on the standardized design drawings; Input multiple sets of the standardized three-dimensional design parameters into SolidWorks model design software for modeling calculations to generate a standard injection-molded mesh model of the TV housing. Extract the interlaced mesh vertices and the smoothing parameters between adjacent interlaced mesh vertices of the standard injection-molded mesh model. A Laplacian coordinate operator is introduced, and a neighborhood coordinate platform for a standard injection molding mesh model is constructed by assigning smoothing parameters between adjacent interwoven mesh vertices in the Laplacian coordinate operator. The neighborhood information of the interwoven mesh vertices is calculated through the neighborhood coordinate platform to obtain the Laplacian neighborhood coordinates of each interwoven mesh vertex. Based on the Laplace neighborhood coordinates, a weighted average positive and negative smoothing process is performed on each interwoven mesh vertex to update and iterate the current neighborhood position of each interwoven mesh vertex in the standard injection molding mesh model, so that each interwoven mesh vertex is closer to the average position of its own neighborhood vertices, and positive and negative smoothing weight values ​​are obtained. Obtain the detection logs of the injection molding defect detection device and the injection molding production requirements of the TV housing. Extract the historical injection molding rework rate of the TV housing in a preset time sequence from the detection logs. Preset the allowable injection molding rework rate according to the injection molding production requirements. Calculate the difference between the historical injection molding rework rate and the allowable injection molding rework rate. Based on the difference in injection molding rework rate, preset the smoothing update iteration frequency and repeat the above-mentioned weighted average interlaced mesh vertex forward smoothing and reverse smoothing update iteration steps until the smoothing update iteration frequency is reached, and generate positive and negative smoothing weight matrices. Based on the positive and negative smoothing weight matrix, the standard injection molding mesh model is reconstructed to obtain the spatial element smoothing mesh model of the TV casing.

4. The detection and control method of the injection molding defect detection device for TV housing according to claim 1, characterized in that, Step S104 specifically includes the following steps: The specified inspection area of ​​the TV housing captured by the vision camera inside the injection molding defect detection device is obtained, and the spatial specification parameters of the specified inspection area are obtained. Based on the spatial specification parameters, construct an element scalar embedding field for the specified detection area. Only plan the field area corresponding to the missing detection model area in the element scalar embedding field, mark it as a sub-detection missing scalar field, obtain the fixed scalar value of the sub-detection missing scalar field, and preset the isosurface embedding threshold according to the fixed scalar value. The region of the missing model is divided into N subordinate model bodies. The vertex mesh scalar value of each model vertex on each subordinate model body is obtained. The vertex mesh scalar value of each model vertex is checked to see if it is greater than the isosurface threshold. If the vertex mesh scalar value is greater than the isosurface embedding threshold, then input a binary value of 1 for the vertex of the model; if the vertex mesh scalar value is greater than the isosurface embedding threshold, then input a binary value of 0 for the vertex of the model, and generate a binary vertex status code. Based on the programming translation of the binary vertex status code, the embedding boundary of the missing scalar field of each sub-model body is determined, resulting in multiple scalar embedding boundaries. Linear interpolation calculation is then performed on each scalar embedding boundary based on the grid scalar value to obtain the corresponding element scalar embedding isosurface for each sub-model body. Connect the corresponding element scalars of all the auxiliary model bodies to embed isosurfaces, and finally generate a spatial element coordinate matrix of the missing model area located in the specified detection area. Control the flipping mechanism to perform fixed-point shooting defect detection on the missing model area of ​​the TV shell in the specified detection area according to the spatial element coordinate matrix, and obtain the defect detection control scheme.

5. The detection and control method of the injection molding defect detection device for TV housing according to claim 1, characterized in that, Step S106 specifically includes the following steps: Obtain a TV housing sample that has been injection molded, and use an injection molding defect detection device to execute the defect detection control scheme to detect the TV housing sample, so as to obtain defect detection image information about the area of ​​the missing model on the TV housing sample. The Sobel operator is used to perform pixel-level edge detection on the defect detection image information to obtain the actual horizontal vector gradient and the actual vertical vector gradient of the detected lost model region. The second moment of the defect detection image information is calculated based on the actual horizontal vector gradient and the actual vertical vector gradient to generate the local image structure tensor of the detected lost model region. The actual image resolution of the defect detection image information is obtained. Based on the actual image resolution, the feature decomposition of the local pixel structure is performed on the local image structure tensor to obtain a series of structural feature values ​​and structural feature vectors of the local graphic structure tensor. The actual pixel structure flatness of the detection missing model region is determined according to the series of structural feature values ​​and structural feature vectors. Based on the standard injection molding mesh model of the TV housing, the injection molding defect detection device executes the defect detection control scheme to capture standard image information of the TV housing, and calculates the pixel structure flatness of the standard image information, which is defined as the reference pixel structure flatness. If the actual pixel structure flatness is higher than the reference pixel structure flatness, the TV casing in the area of ​​the missing model is calibrated as a qualified injection molded product; if the actual pixel structure flatness is lower than the reference pixel structure flatness, the TV casing in the area of ​​the missing model is calibrated as a defective injection molded product, and the first defect detection result is obtained.

6. The detection and control method of the injection molding defect detection device for TV housing according to claim 1, characterized in that, Step S108 specifically includes the following steps: If the first defect detection result shows that the TV casing is a defective injection molded product, then the pixel particle mutation area of ​​the defective injection molded product is extracted from the defect detection image information of the defective injection molded product based on the actual pixel structure flatness. The color saturation range of the pixel particle mutation region is obtained and defined as the first color saturation range. Based on the first color saturation range, the pixel saturation color gamut space of the pixel particle mutation region in the defective injection molded product is constructed and defined as the source pixel saturation color gamut space. The color saturation range of the pixel particle mutation region expressed in the qualified injection molded product is obtained and defined as the second color saturation range. The pixel saturation color gamut space of the pixel particle mutation region in the defective injection molded product is constructed based on the second color saturation range and defined as the target pixel saturation color gamut space. By acquiring color mapping function tables for color saturation expression of different injection molding defect types and predetermined Gamma correction intervals for each type of injection molding defect through big data networks, color mapping rules are established based on the color mapping function tables. A linear transformation algorithm is introduced. Based on the color mapping rule, each saturated color in the source pixel saturated color gamut space is mapped to the target pixel saturated color gamut space in the linear transformation algorithm to generate a linear color saturation mapping equation. Solve the linear color saturation mapping equation to obtain a series of mapped chromaticity values. Calculate and determine the restored pixel contrast of the pixel particle variation region based on the series of mapped chromaticity values, and obtain the pixel contrast of the pixel particle variation region expressed in the qualified injection molded product, which is defined as the initial pixel contrast. The Gamma correction formula is introduced to calculate the correction between the restored pixel contrast and the initial pixel contrast, and the actual Gamma correction value is obtained. If the actual Gamma correction value is within the predetermined Gamma correction range, the injection molding defect type corresponding to the predetermined Gamma correction range is directly output to obtain the second defect detection result.

7. A device for detecting injection molding defects in TV housings, characterized in that, The injection molding defect detection device includes: Image acquisition module, which is used to capture defect image data of TV housing injection molded finished products from multiple angles; The image processing module is responsible for introducing the Sobel operator to calculate and extract feature information from defect detection image data of different frames; The flipping module is used to clamp the TV casing within the specified detection area and flip it in multiple directions so that the image acquisition module can fully capture the detection loss model area of ​​the TV casing. The model building module is used to perform modeling and calculation processing on the defect detection image data extracted by the image processing module to output a local process defect detection model. The binary encoding module, the data processing module is used to check whether the vertex mesh scalar value of each model vertex on the auxiliary model body is greater than the isosurface embedding threshold of the sub-detection lost scalar field, and thus program the model vertex of the auxiliary model body with a binary status code of 0 or 1.

Images