Extrusion blow molding processing quality detection method and system

Abstract

The present invention relates to the field of image data processing technology, and specifically to an extrusion blow molding quality detection method and system. The present invention irradiates a bottle body with light to collect a bottle body image; performs edge detection on the bottle body image to obtain an edge line; analyzes the degree of tortuosity of the edge line and the brightness value characteristics of the pixel points on the edge line to determine the possibility that the edge line belongs to a light pattern area to obtain a light pattern area; analyzes the length characteristics of the light pattern area and the brightness change law of the pixel points in the light pattern area to determine the possibility that the light pattern area belongs to an uneven thickness area to obtain an uneven thickness area; analyzes the area of ​​the uneven thickness area to perform quality detection on the uneven thickness area. The present invention can eliminate the influence of the light pattern area generated by the bottle's own texture on the extrusion blow molding quality detection result, and improve the accuracy of the quality detection of the bottle processed by the extrusion blow molding.

Description

Technical Field

[0001] The invention relates to the technical field of image data processing, and in particular to an extrusion blow molding processing quality detection method and system. Background Art

[0002] Extrusion blow molding is a manufacturing technology widely used in plastic processing to produce hollow plastic parts such as bottles, containers and other complex-shaped products. This process combines extrusion molding and blow molding. The plastic raw material is heated and melted, extruded into a tubular preform, and then blown into the final shape with air in the mold. This process has high production efficiency, good material utilization and high repeatability. It is widely used in packaging, automotive, medical and consumer goods industries.

[0003] Bottles processed by extrusion blow molding may have uneven thickness due to changes in temperature, pressure, cooling rate and other factors during the extrusion blow molding process. When using light to illuminate the bottle body to detect thickness unevenness, scattering will occur in the uneven thickness area, resulting in light streaks on the bottle body; and scattering will also occur in the texture area of ​​the bottle body, resulting in light streaks on the bottle body; therefore, when testing the light streak area caused by uneven thickness, the light streak generated in the texture area will affect the test results, thereby reducing the accuracy of quality testing of bottles processed by extrusion blow molding. Summary of the invention

[0004] In order to solve the above technical problems, the object of the present invention is to provide an extrusion blow molding process quality detection method and system.

[0005] According to a first aspect of an embodiment of the present invention, a method for detecting the quality of extrusion blow molding is provided, and the technical solution adopted is as follows:

[0006] Expose light to the bottle body and collect the image of the bottle body;

[0007] Performing edge detection on the bottle body image to obtain edge lines;

[0008] Analyze the degree of twists and turns of the edge line and the brightness value characteristics of the pixels on the edge line to determine the possibility that the edge line belongs to the light streak area, and obtain the light streak area according to the possibility that the edge line belongs to the light streak area;

[0009] Analyze the length characteristics of the light pattern area and the brightness variation law of the pixel points in the light pattern area to determine the possibility that the light pattern area belongs to the uneven thickness area, and obtain the uneven thickness area according to the possibility that the light pattern area belongs to the uneven thickness area;

[0010] The area of ​​the uneven thickness region is analyzed, and the quality of the uneven thickness region is inspected.

[0011] In some embodiments of the present invention, the method further comprises:

[0012] Irradiate the bottle with light from multiple angles and collect multiple images of the bottle;

[0013] Analyze each of the bottle body images to obtain an area of ​​uneven thickness in each of the bottle body images;

[0014] All the uneven thickness regions are spliced ​​together to obtain a complete uneven thickness region.

[0015] In some embodiments of the present invention, all the uneven thickness regions are spliced ​​together to obtain a complete uneven thickness region, including:

[0016] Aligning all the bottle images using an ORB feature extraction algorithm to obtain an aligned image;

[0017] For the aligned images, using a connected domain analysis algorithm to identify and stitch the uneven thickness regions to obtain a stitching region;

[0018] The dilation operation in the morphological operation is used to perform hole filling processing on the spliced ​​area to obtain a complete uneven thickness area.

[0019] In some embodiments of the present invention, analyzing the degree of undulation of the edge line includes:

[0020] The number of occurrences of the maximum pixel point curvature value on the edge line and the ratio of the number of all pixel point curvature values ​​on the edge line are analyzed to obtain the degree of twists and turns of the edge line.

[0021] In some embodiments of the present invention, analyzing the brightness value characteristics of the pixel points on the edge line includes:

[0022] The average brightness value of all pixels on the edge line is analyzed to obtain the brightness value characteristics of the pixels on the edge line.

[0023] In some embodiments of the present invention, analyzing the length characteristics of the light pattern region includes:

[0024] Determine the central pixel point of the light pattern area;

[0025] Draw a straight line through the central pixel point along the extension direction of the light pattern area, and record the straight line as the length line of the light pattern area;

[0026] The number of pixel points on the length line is counted to characterize the length characteristics of the light pattern area.

[0027] In some embodiments of the present invention, analyzing the brightness variation pattern of pixels in the light pattern area includes:

[0028] In the light pattern area, a pixel with the largest brightness value is selected as a highlight pixel;

[0029] Taking the highlighted pixel point as the starting pixel point, selecting the direction closest to the edge of the light pattern area, and drawing a horizontal line perpendicular to the length line along the closest direction;

[0030] The brightness variation of the pixels on the horizontal line is analyzed to obtain the brightness variation rule of the pixels in the light pattern area.

[0031] In some embodiments of the present invention, analyzing the brightness change of the pixels on the horizontal line includes:

[0032] Analyze the brightness differences between all pixels and adjacent pixels on the horizontal line to obtain a maximum brightness difference and a minimum brightness difference;

[0033] The average of all the brightness difference values ​​is calculated, and the brightness change of the pixel points on the horizontal line is obtained by combining the maximum brightness difference value and the minimum brightness difference value.

[0034] According to a second aspect of an embodiment of the present invention, a system for detecting quality of extrusion blow molding is provided, comprising: a memory and a processor, wherein:

[0035] The memory is used to store program code;

[0036] The processor is used to read the program code stored in the memory and execute the method described in the first aspect of the embodiment of the present invention.

[0037] In some embodiments of the present invention, the processor includes: a bottle body image acquisition module, used to irradiate the bottle body with light and acquire the bottle body image;

[0038] An edge detection module, used to perform edge detection on the bottle body image to obtain an edge line;

[0039] A light streak region acquisition module is used to analyze the degree of twists and turns of the edge line and the brightness value characteristics of the pixel points on the edge line to determine the possibility that the edge line belongs to the light streak region, and obtain the light streak region according to the possibility that the edge line belongs to the light streak region;

[0040] The module for acquiring the uneven thickness region is used to analyze the length characteristics of the light pattern region and the brightness variation law of the pixels in the light pattern region, determine the possibility that the light pattern region belongs to the uneven thickness region, and obtain the uneven thickness region according to the possibility that the light pattern region belongs to the uneven thickness region;

[0041] The quality detection module is used to analyze the area of ​​the uneven thickness region and perform quality detection on the uneven thickness region.

[0042] Compared with the prior art, the extrusion blow molding quality detection method and system provided by the present invention have the following beneficial effects:

[0043] The present invention determines the possibility that the edge line belongs to the light pattern area by analyzing the degree of tortuosity of the edge line and the brightness value characteristics of the pixel points on the edge line, and obtains the light pattern area. By analyzing the edge line, the noise edge line generated by the edge of the convex area of ​​the bottle body in the edge line can be eliminated; then, the length characteristics of the light pattern area and the brightness change law of the pixel points in the light pattern area are analyzed to determine the possibility that the light pattern area belongs to the uneven thickness area, and obtain the uneven thickness area. By analyzing the characteristics of the light pattern area, the uneven thickness area and the light pattern area generated by the bottle's own texture area are accurately distinguished; the area of ​​the uneven thickness area is analyzed, and the quality of the uneven thickness area is tested, thereby eliminating the influence of the light pattern area generated by the bottle's own texture area on the quality test result of the extrusion blow molding process, and improving the accuracy of the quality test of the bottle processed by the extrusion blow molding process. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 A basic flow chart of an extrusion blow molding quality inspection method provided by an embodiment of the present invention;

[0046] Figure 2 A bottle body image provided by an embodiment of the present invention;

[0047] Figure 3 An edge line image of a bottle body image provided by an embodiment of the present invention;

[0048] Figure 4 A schematic diagram of a splicing area of ​​an area with uneven thickness provided by an embodiment of the present invention;

[0049] Figure 5 A schematic diagram of a complete thickness uneven area provided by an embodiment of the present invention;

[0050] Figure 6 A schematic diagram of the basic components of an extrusion blow molding process quality inspection system provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0051] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of the extrusion blow molding quality detection method and system proposed by the present invention, its specific implementation, structure, characteristics and effects, in conjunction with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0052] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as those commonly understood by those skilled in the art to which the present invention belongs. Terms such as "comprises", "comprising" or any other variants thereof are intended to cover non-exclusive inclusion, so that a circuit structure, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such article or device. In the absence of further restrictions, an element defined by the phrase "comprising a ..." does not exclude the presence of other identical elements in the article or device including the element.

[0053] The specific scheme of the extrusion blow molding processing quality detection method provided by the present invention is described in detail below with reference to the accompanying drawings.

[0054] See also Figure 1 , which shows the basic process of an extrusion blow molding processing quality detection method provided by an embodiment of the present invention.

[0055] like Figure 1 As shown, an embodiment of the present invention provides an extrusion blow molding processing quality detection method, which specifically includes:

[0056] S100: Expose light to the bottle body and collect an image of the bottle body.

[0057] The bottle processed by extrusion blow molding is fixed in a position, the bottle body is irradiated with light, and the bottle body is photographed by a camera to collect the image of the bottle body.

[0058] In some embodiments of the present invention, multiple images of the bottle are collected by irradiating the bottle with light from multiple angles. The specific operation method is as follows: the light moves with the camera, and an image of the bottle is captured every second. The bottle is captured at different angles until a complete cycle around the bottle is completed, and multiple images of the bottle are acquired. Figure 2 shown.

[0059] S200: Perform edge detection on the bottle body image to obtain edge lines.

[0060] When light is irradiated on the uneven thickness area and the self-textured area of ​​the bottle body, due to the refraction and scattering of light, the uneven thickness area and the self-textured area of ​​the bottle body will generate light streaks. In order to accurately detect the quality of the bottle body, it is necessary to distinguish the light streaks corresponding to the uneven thickness area and the self-textured area. Therefore, the embodiment of the present invention first performs edge detection on the bottle body image to obtain edge lines, such as Figure 3 Then the edge line can be analyzed to obtain the light pattern area.

[0061] In some embodiments of the present invention, multiple bottle body images are obtained, and this step uses the same method to obtain edge lines in all bottle body images.

[0062] S300: Analyze the degree of undulation of the edge line and the brightness value characteristics of the pixels on the edge line to determine the possibility that the edge line belongs to the light streak area, and obtain the light streak area according to the possibility that the edge line belongs to the light streak area.

[0063] Bottles manufactured by extrusion blow molding are often affected by factors such as temperature, pressure and cooling rate during the extrusion blow molding process. The thickness of the bottle body often shows obvious unevenness in certain areas. This thickness unevenness will produce different visual effects under light exposure. When light shines on thicker areas, due to the large number of plastic layers in this area, the light will propagate farther in the thick layer and experience more scattering and refraction, resulting in a blurred or uneven effect on the surface. The light scattering effect is stronger in thicker areas; on the contrary, the plastic layer in thinner areas is thinner, and the light will be clearer and less scattered when passing through these areas. In addition to thickness unevenness, the texture of the bottle body also has a significant impact on the propagation of light. Concave areas will cause light to diffuse in these areas. When light shines on concave textures, the light will be dispersed in these areas, forming halos or spots; on the contrary, convex texture areas will concentrate light in these areas. When light shines on convex areas, the light will be focused on these high points, forming a more concentrated light effect in these areas. Therefore, both uneven thickness areas and texture areas will produce light patterns, such as Figure 2 Shown in the dotted area.

[0064] The brightness of the light pattern area is relatively high, forming a clear contrast with the darker surrounding areas. At the edge line, the brightness of the light pattern area suddenly transitions from a high value to a low value. This transition produces a clear brightness difference, which makes the edge line more prominent. At the edge of the light pattern area, the brightness of the pixel point will change rapidly from the high brightness of the light pattern area to the low brightness of the surrounding area. This change is usually manifested as a brightness gradient, which can be used to identify the position and direction of the edge. In addition, for the noise edge line generated by the edge of the raised area of ​​the bottle body, such as Figure 3 The edge line in the middle dotted frame, the brightness of the pixels on the edge of the raised area of ​​the bottle body is smaller than that of the pixels on the light pattern area, and the shape is more tortuous than that of the edge of the light pattern area.

[0065] Based on the above analysis, the embodiment of the present invention determines the possibility that the edge line belongs to the light streak area by analyzing the degree of tortuosity of the edge line and analyzing the brightness value characteristics of the pixels on the edge line, and obtains the light streak area according to the possibility that the edge line belongs to the light streak area. Among them, analyzing the degree of tortuosity of the edge line further includes: analyzing the number of occurrences of the maximum pixel curvature value on the edge line and the ratio of the number of curvature values ​​of all pixels on the edge line to obtain the degree of tortuosity of the edge line. Analyzing the brightness value characteristics of the pixels on the edge line further includes: analyzing the mean brightness value of all pixels on the edge line to obtain the brightness value characteristics of the pixels on the edge line. Then, combining the degree of tortuosity of the edge line and the brightness value characteristics of the pixels on the edge line, the possibility that the edge line belongs to the light streak area is obtained. The calculation formula for constructing the possibility that the edge line belongs to the light streak area is:

[0066]

[0067]

[0068] in, Indicates The possibility that the edge line belongs to the light pattern area; Indicates The degree of undulation of the edge line; Indicates On the edge line The brightness value of each pixel; Indicates The number of pixels on the edge line; Indicates The number of times the maximum pixel curvature value appears on the edge line, for example, When the curvature value sequence of the pixel points on the edge line is (1,1,5,2,2,1,1,3,5,2,3), the number of times the maximum pixel curvature value appears is 2; Indicates The total number of curvature values ​​of all pixels on the edge line, for example, When the curvature value sequence of the pixel points on the edge line is (1,1,5,2,2,1,1,3,5,2,3), the total number of pixel curvature values ​​is 11; Represents a linear normalization function that normalizes the data value to The acquisition of the pixel point curvature value adopts the existing technology, which will not be described here.

[0069] Indicates The ratio of the number of maximum pixel curvature values ​​on the edge line to the total number of pixel curvature values. The larger the value, the higher the curvature value. The edge line has more bending points, which means that the The first edge line has a high degree of twists and turns, and may have sharp turns or sharp corners. The edge line is less likely to belong to the light pattern area; therefore, The smaller the The more regular the edge line is, the more likely it is to belong to the light pattern area. Indicates The average brightness value of all pixels on the edge line. The larger the value, the The greater the brightness of all pixels on the edge line, the The larger the edge line, the more likely it is to belong to the light pattern area.

[0070] According to the possibility that the edge line belongs to the light pattern area, the specific method of obtaining the light pattern area is as follows: The possibility that the edge line belongs to the light pattern area is judged by the threshold. When the probability that the edge line belongs to the light pattern area is greater than or equal to 0.5, the The edge line is recorded as the edge line of the light pattern area; When the probability that the edge line belongs to the light pattern area is less than 0.5, The edge lines are recorded as other area edge lines. Then, the light pattern area is obtained from the light pattern area edge lines obtained by threshold judgment according to the contour extraction algorithm.

[0071] In some embodiments of the present invention, multiple bottle body images are obtained, and this step uses the same method to obtain the light pattern areas in all the bottle body images.

[0072] S400: Analyze the length characteristics of the light pattern area and the brightness variation law of the pixels in the light pattern area to determine the possibility that the light pattern area belongs to the uneven thickness area, and obtain the uneven thickness area according to the possibility that the light pattern area belongs to the uneven thickness area.

[0073] The light streak region includes not only the light streaks generated by the thickness non-uniform region, but also the light streaks generated by the texture region. Therefore, it is necessary to further determine the region with non-uniform thickness in the light streak region based on the characteristics of the two regions.

[0074] For the light pattern area generated by the uneven thickness area, when the light irradiates the uneven thickness area of ​​the bottle body, the light will experience more scattering and refraction when passing through the uneven thickness plastic layer, and the scattering effect accumulates more when the light passes through the thicker material layer, so the length of the light pattern in the uneven thickness area is longer; while for the light pattern area generated by the texture area, since the concave-convex texture changes the distribution of light, the concave area will diffuse the light to form a larger light spot or halo, while the convex area will focus the light on the high point to form a concentrated and obvious light pattern, so the length of the light pattern in the texture area is shorter, because the light is mainly concentrated or diffused in the local area and does not involve a long path. In addition, for the light pattern area generated by the uneven thickness area, the change in thickness causes the light to be continuously refracted and scattered during the propagation process, and the thicker area causes the light to experience more scattering and absorption during the propagation process, causing the light to become more blurred, so the edge brightness of the uneven thickness area gradually decays; while for the light pattern area generated by the texture area, since the light propagates more concentratedly or diffusely in the texture area, rather than being excessively absorbed or scattered in the thickness change, the edge brightness of the texture area is relatively concentrated.

[0075] Based on the above analysis, the embodiment of the present invention determines the possibility that the light streak area belongs to the uneven thickness area by analyzing the length characteristics of the light streak area and analyzing the brightness change law of the pixels in the light streak area, and obtains the uneven thickness area according to the possibility that the light streak area belongs to the uneven thickness area. Among them, analyzing the length characteristics of the light streak area further includes: determining the central pixel point of the light streak area; drawing a straight line through the central pixel point along the length extension direction of the light streak area, and recording the line segment part of the straight line inside the light streak area as the length line of the light streak area; counting the number of pixels on the length line to characterize the length characteristics of the light streak area. Among them, as Figure 2 The length extension direction of the light pattern area is the vertical direction.

[0076] Analyzing the brightness variation of pixels in the light streak area further includes: in the light streak area, selecting a pixel with the largest brightness value as a highlighted pixel; taking the highlighted pixel as the starting pixel, selecting the direction closest to the edge of the light streak area, and drawing a horizontal line perpendicular to the length line along the closest direction; analyzing the brightness variation of pixels on the horizontal line to obtain the brightness variation of pixels in the light streak area. Furthermore, analyzing the brightness variation of pixels on the horizontal line includes: analyzing the brightness difference between all pixels on the horizontal line and adjacent pixels to obtain the maximum brightness difference and the minimum brightness difference, wherein the adjacent pixels of a pixel are pixels extending toward the edge on the horizontal line; calculating the mean of all brightness differences, combining the maximum brightness difference and the minimum brightness difference, to obtain the brightness variation of pixels on the horizontal line. Then, based on the length characteristics of the light streak area and the brightness variation of pixels in the light streak area, an area with uneven thickness is obtained. The calculation formula for constructing an area with uneven thickness is:

[0077]

[0078] in, Indicates The possibility that the light pattern area belongs to the area with uneven thickness; Indicates The number of pixels on the length line of the light pattern area; Indicates In the light pattern area, the horizontal line The brightness difference between a pixel and its adjacent pixels; Indicates the number of pixels on the horizontal line; Indicates The maximum brightness difference of all pixels on the horizontal line in the light pattern area; Indicates The minimum brightness difference of all pixels on the horizontal line in a light pattern area.

[0079] Indicates The average value of the brightness difference between each pixel point and the adjacent pixel points on the horizontal line in the light pattern area. The larger the value, the more it means that the brightness of the pixels along the horizontal line is not concentrated, but changes gradually. Indicates The maximum value of the brightness difference between each pixel and the adjacent pixel on the horizontal line in the light pattern area is Indicates The minimum value of the brightness difference between each pixel and the adjacent pixel on the horizontal line in the light pattern area; Indicates The attenuation degree of the brightness of the pixel point in the light pattern area extending toward the edge. The larger the value, the greater the attenuation degree. The greater the possibility that the thickness of the light pattern area is uneven. The larger the The longer the light pattern length of each light pattern area, the more consistent it is with the light pattern characteristics produced in the area with uneven thickness.

[0080] According to the possibility that the light pattern area belongs to the uneven thickness area, the uneven thickness area is obtained. The specific method is: The probability that the light pattern area belongs to the area with uneven thickness is judged by threshold. When the probability that the first light pattern area belongs to the thickness uneven area is greater than or equal to 0.5, The light pattern area is recorded as the uneven thickness area; When the probability that the first light pattern area belongs to the thickness uneven area is less than 0.5, The light pattern area is recorded as the texture area of ​​the bottle.

[0081] In some embodiments of the present invention, a plurality of bottle body images are obtained, and in this step, each bottle body image is analyzed by the same method to obtain the uneven thickness region of each bottle body image.

[0082] After obtaining the uneven thickness area of ​​each bottle body image, all the uneven thickness areas are spliced ​​to obtain a complete uneven thickness area. Further, all bottle body images are aligned using the ORB feature extraction algorithm to obtain an aligned image, ensuring that the light pattern areas caused by uneven thickness in different bottle body images can be accurately spliced; for the aligned image, the connected domain analysis algorithm is used to identify and splice the uneven thickness areas to obtain the spliced ​​area. The spliced ​​area is as follows: Figure 4 As shown; the expansion operation in the morphological operation is used to fill the holes in the spliced ​​area, expand the boundary of the spliced ​​area, and obtain a complete uneven thickness area. The complete uneven thickness area is shown in Figure 5 Shown in the dotted area.

[0083] S500: Analyze the area of ​​the region with uneven thickness and perform quality inspection on the region with uneven thickness.

[0084] Through the above steps, the complete uneven thickness area of ​​the bottle body is obtained. Based on the complete uneven thickness area, the uneven thickness area is evaluated. When the area of ​​the uneven thickness area is greater than or equal to 1, it means that the uneven thickness area affects the detection of the extrusion blow molding quality. The uneven thickness area is marked as an abnormal thickness area; when the area of ​​the uneven thickness area is less than 1, it means that the uneven thickness area does not affect the detection of the extrusion blow molding quality. The uneven thickness area is marked as a normal area. The area acquisition method adopts the existing technology, such as obtaining the area size within the contour of the uneven thickness area by curve fitting method, which will not be repeated here.

[0085] At this point, an accurate extrusion blow molding process quality test result is obtained, and the extrusion blow molding process quality test is completed.

[0086] Based on the same inventive concept as the above method, this embodiment also provides an extrusion blow molding process quality detection system.

[0087] See also Figure 6 , which shows the basic composition of an extrusion blow molding processing quality detection system provided by an embodiment of the present invention.

[0088] like Figure 6 As shown, an extrusion blow molding process quality detection system includes: a memory 10 and a processor 20, wherein:

[0089] A memory 10, used for storing program codes;

[0090] The processor 20 is used to read the program code stored in the memory 10, and execute the steps of irradiating the bottle body with light to collect the bottle body image; performing edge detection on the bottle body image to obtain the edge line; analyzing the degree of undulation of the edge line and the characteristics of the brightness value of the pixels on the edge line to determine the possibility that the edge line belongs to the light pattern area, and obtaining the light pattern area according to the possibility that the edge line belongs to the light pattern area; analyzing the length characteristics of the light pattern area and the brightness variation law of the pixels in the light pattern area to determine the possibility that the light pattern area belongs to the uneven thickness area, and obtaining the uneven thickness area according to the possibility that the light pattern area belongs to the uneven thickness area; analyzing the area of ​​the uneven thickness area to perform quality detection on the uneven thickness area.

[0091] Furthermore, the processor 20 includes a bottle body image acquisition module 21, an edge detection module 22, a light pattern area acquisition module 23, a thickness uneven area acquisition module 24 and a quality detection module 25. Among them:

[0092] The bottle body image acquisition module 21 is used to irradiate the bottle body with light and acquire the bottle body image;

[0093] The edge detection module 22 is used to perform edge detection on the bottle body image to obtain edge lines;

[0094] The light streak region acquisition module 23 is used to analyze the degree of undulation of the edge line and the brightness value characteristics of the pixels on the edge line to determine the possibility that the edge line belongs to the light streak region, and obtain the light streak region according to the possibility that the edge line belongs to the light streak region;

[0095] The thickness non-uniform region acquisition module 24 is used to analyze the length characteristics of the light pattern region and the brightness variation law of the pixels in the light pattern region to determine the possibility that the light pattern region belongs to the thickness non-uniform region, and obtain the thickness non-uniform region according to the possibility that the light pattern region belongs to the thickness non-uniform region;

[0096] The quality detection module 25 is used to analyze the area of ​​the region with uneven thickness and perform quality detection on the region with uneven thickness.

[0097] It should be noted that the sequence of the above embodiments of the present invention is for description only and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0098] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. A method for detecting the quality of extrusion blow molding, characterized in that: The method comprises: Expose light to the bottle body and collect the image of the bottle body; Performing edge detection on the bottle body image to obtain edge lines; Analyze the degree of twists and turns of the edge line and the brightness value characteristics of the pixels on the edge line to determine the possibility that the edge line belongs to the light streak area, and obtain the light streak area according to the possibility that the edge line belongs to the light streak area; Analyze the length characteristics of the light pattern area and the brightness variation law of the pixels in the light pattern area to determine the possibility that the light pattern area belongs to the uneven thickness area, and obtain the uneven thickness area according to the possibility that the light pattern area belongs to the uneven thickness area; Analyzing the area of ​​the region with uneven thickness and performing quality inspection on the region with uneven thickness; Analyze the degree of twists and turns of the edge line, including: The number of occurrences of the maximum pixel point curvature value on the edge line and the ratio of the number of all pixel point curvature values ​​on the edge line are analyzed to obtain the degree of twists and turns of the edge line.

2. The extrusion blow molding processing quality detection method according to claim 1, characterized in that: The method further comprises: Irradiate the bottle with light from multiple angles and collect multiple images of the bottle; Analyze each of the bottle body images to obtain an area of ​​uneven thickness in each of the bottle body images; All the uneven thickness regions are spliced ​​together to obtain a complete uneven thickness region.

3. The extrusion blow molding processing quality detection method according to claim 2, characterized in that: All the uneven thickness areas are spliced ​​together to obtain a complete uneven thickness area, including: Aligning all the bottle images using an ORB feature extraction algorithm to obtain an aligned image; For the aligned images, using a connected domain analysis algorithm to identify and stitch the uneven thickness regions to obtain a stitching region; The dilation operation in the morphological operation is used to perform hole filling processing on the spliced ​​area to obtain a complete uneven thickness area.

4. The method for detecting the quality of extrusion blow molding according to claim 1, characterized in that: Analyzing the brightness value characteristics of the pixel points on the edge line includes: The average brightness value of all pixels on the edge line is analyzed to obtain the brightness value characteristics of the pixels on the edge line.

5. The extrusion blow molding processing quality detection method according to claim 1, characterized in that: Analyze the length characteristics of the light pattern area, including: Determine the central pixel point of the light pattern area; Draw a straight line through the central pixel point along the extension direction of the light pattern area, and record the straight line as the length line of the light pattern area; The number of pixel points on the length line is counted to characterize the length characteristics of the light pattern area.

6. The method for detecting the quality of extrusion blow molding according to claim 5, characterized in that: Analyzing the brightness variation law of the pixels in the light pattern area includes: In the light pattern area, a pixel with the largest brightness value is selected as a highlight pixel; Taking the highlighted pixel point as the starting pixel point, selecting the direction closest to the edge of the light pattern area, and drawing a horizontal line perpendicular to the length line along the closest direction; The brightness variation of the pixels on the horizontal line is analyzed to obtain the brightness variation rule of the pixels in the light pattern area.

7. The method for detecting the quality of extrusion blow molding according to claim 6, characterized in that: Analyzing the brightness changes of the pixels on the horizontal line includes: Analyze the brightness differences between all pixels and adjacent pixels on the horizontal line to obtain a maximum brightness difference and a minimum brightness difference; The average of all the brightness difference values ​​is calculated, and the brightness change of the pixel points on the horizontal line is obtained by combining the maximum brightness difference value and the minimum brightness difference value.

8. An extrusion blow molding quality inspection system, characterized in that: The system comprises: a memory and a processor, wherein: The memory is used to store program code; The processor is configured to read the program code stored in the memory and execute the method according to any one of claims 1 to 7.

9. The extrusion blow molding processing quality detection system according to claim 8, characterized in that: The processor comprises: A bottle body image acquisition module is used to irradiate the bottle body with light and acquire the bottle body image; An edge detection module, used to perform edge detection on the bottle body image to obtain an edge line; The light streak region acquisition module is used to analyze the degree of twists and turns of the edge line and the brightness value characteristics of the pixel points on the edge line to determine the possibility that the edge line belongs to the light streak region, and obtain the light streak region according to the possibility that the edge line belongs to the light streak region; analyzing the degree of twists and turns of the edge line includes: Analyze the number of occurrences of the maximum pixel point curvature value on the edge line and the ratio of the number of all pixel point curvature values ​​on the edge line to obtain the degree of twists and turns of the edge line; The module for acquiring the uneven thickness region is used to analyze the length characteristics of the light pattern region and the brightness variation law of the pixels in the light pattern region, determine the possibility that the light pattern region belongs to the uneven thickness region, and obtain the uneven thickness region according to the possibility that the light pattern region belongs to the uneven thickness region; The quality detection module is used to analyze the area of ​​the uneven thickness region and perform quality detection on the uneven thickness region.

Images