A detection method

By adding image data collection of the inner wall of the bottle mouth and the inner edge of the bottle mouth end face to the image acquisition component, the problem that these areas are ignored in the existing technology is solved, a more comprehensive and accurate bottle defect detection is achieved, and the efficiency is improved. Sealing performance and strength.

CN115159093BActive Publication Date: 2025-01-21CHONGQING HUANSHI HI TECH CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202210852569.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-07-19
Publication Date
2025-01-21
Estimated Expiration
2042-07-19

AI Technical Summary

Technical Problem

In the prior art, the image acquisition component cannot fully detect the edges and corners of the inner wall of the bottle mouth and the inner edge of the bottle mouth end face when detecting the bottle body defects, resulting in some minor defects being ignored, affecting the sealing performance and strength of the bottle body.

Method used

The image acquisition component with multiple image acquisition modules is used to collect image data of the mouth, neck, body and bottom of the bottle, especially the image data collection of the inner wall of the bottle mouth and the inner edge of the end face of the bottle. , using image processing systems for defect detection.

Benefits of technology

By increasing the detection of the inner wall of the bottle mouth and the inner edge of the bottle mouth end surface, defects in the bottle body can be more comprehensively detected, the accuracy and comprehensiveness of detection can be improved, and sealing performance and strength problems caused by subtle defects can be reduced.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN115159093B_ABST
    Figure CN115159093B_ABST
Patent Text Reader

Abstract

The present invention relates to the technical field of bottle inspection, and specifically discloses a detection method, which uses an image acquisition component with multiple image acquisition modules to collect image data of the bottle mouth, bottleneck, bottle body and bottle bottom of the bottle body, and uses an image processing system to analyze and calculate the collected image data to determine the defects of the bottle body, and the image data collection of the bottle mouth includes the image data collection of the bottle mouth end face, the bottle mouth inner wall, the outer edge corners of the bottle mouth end face and the inner edge corners of the bottle mouth end face. This solution is used to solve the problem that the image acquisition component in the prior art cannot detect bottle defects more comprehensively.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of bottle detection, and in particular to a detection method. Background Art

[0002] After the bottle is produced, it is necessary to test the quality of the bottle before packaging. There are different testing methods for the bottle according to different needs. For example, if the air tightness of the bottle is high but the appearance and size requirements are not high, it is only necessary to focus on whether the bottle has cracks, breaks, and bursts that affect the air tightness. For bottles with high requirements for both air tightness and appearance, it is necessary to detect whether there are defects that affect the air tightness of the bottle, and it is also necessary to detect the shape and size specifications of the bottle, impurities, and defects such as air lines caused by molding. Regardless of the requirements, the existing technology mostly uses visual inspection combined with an image processing system with artificial intelligence deep learning to realize bottle defect detection. This detection method is based on the image data of the bottle collected by the image acquisition component to perform defect detection, and the image processing system judges the defects of the bottle based on the received image data.

[0003] In order to ensure the comprehensiveness of image data acquisition, the bottle body needs to be rotated 360° during the image data acquisition process to ensure that multiple parts of the bottle body can be photographed. The specific bottle body detection process is as follows: the two guide wheels on the lifting seat lift the bottle body upward from both sides of the bottle body (the two guide wheels are parallel to the bottle body, and there is an air avoidance space between the two guide wheels to facilitate the subsequent rotation of the bottle body while supporting the bottle body. The guide wheels can use bearings or rollers, as long as the guide wheels can rotate by themselves), until the bottle body is lifted up to the friction wheel directly above the two guide wheels. The rotation of the friction wheel generates friction between the friction wheel and the bottle body, and this friction force causes the bottle body to rotate with the cooperation and support of the two self-rotating guide wheels. During the rotation of the bottle body, multiple image acquisition modules of the image acquisition component respectively acquire image data of the bottle bottom, bottle body, bottleneck and bottle mouth of the bottle body.

[0004] At present, for the inspection of bottles, when the requirements for the bottles are not high, the data collection of the bottle bottom, bottle body, bottleneck and bottle mouth mentioned above can basically meet the inspection requirements. However, for some bottles with high requirements, the amount of image collection data of the bottle body, bottle bottom, bottleneck or bottle mouth is only increased during the inspection to improve the accuracy of the inspection. However, there are still cases where bottle users believe that the bottles have cracks, breaks, bursts and other defects, which cause problems in use. For example, some pharmaceutical bottles (referred to as pharmaceutical bottles) need to contain solvents, powders or granular drugs during use. The storage of these drugs requires extremely high sealing. Once cracks or difficult-to-find breaks or cracks appear in the bottle, they will affect the sealing performance and strength of the bottle during use. For example, in the existing market, there have been cases where syringe bottles used to contain vaccines were recalled in batches because cracks were found in the bottles after the vaccine was produced. In addition, some drug bottles need to be freeze-dried for storage, which makes the environment in which the bottles are located very harsh. Although the probability of bottle rupture during the freeze-drying process is extremely low, it will be a very serious problem once it occurs, because it will cause the product in the bottle to be lost, and even cause overflow of product and broken glass slag to contaminate the internal part of the equipment. Summary of the invention

[0005] The present invention is intended to provide a detection method to solve the problem that the image acquisition component in the prior art cannot detect bottle defects more comprehensively.

[0006] In order to achieve the above object, the present invention adopts the following technical scheme:

[0007] A detection method uses an image acquisition component with multiple image acquisition modules to acquire image data of the bottle mouth, bottleneck, bottle body and bottle bottom of a bottle body, and uses an image processing system to analyze and calculate the acquired image data to determine the defects of the bottle body. The image data acquisition of the bottle mouth includes the image data acquisition of the bottle mouth end face, the bottle mouth inner wall, the outer edge corners of the bottle mouth end face and the inner edge corners of the bottle mouth end face.

[0008] The principle and advantage of this solution are: in actual application, by increasing the image data collection of the inner wall of the bottle mouth and the inner edge corners of the bottle mouth end face, the position of the inner wall of the bottle mouth and the position of the inner edge corners of the bottle mouth end face can be detected; in the prior art, for the detection of the bottle body, because the position of the bottle mouth is used in conjunction with the bottle cap when the bottle body is used, and the bottle cap has a reinforcing effect on the sealing and strength of the bottle mouth, it is often easily overlooked, so that when the bottle mouth of the bottle body is detected, the defects of the outer surface of the bottle mouth are basically detected, and the defects of the inner wall of the bottle mouth are not detected, and the position of the inner edge corner of the bottle mouth end face is more easily ignored, but because the position of the inner edge corner of the bottle mouth end face is located between the bottle mouth end face and the bottle mouth end face, the inner edge corner of the bottle mouth end face is more likely to be ignored. The corner of the inner wall of the bottle mouth is very small compared to the size of the entire bottle mouth and is easily overlooked. In addition, the wall thickness of the bottle mouth of a pharmaceutical bottle is generally thicker, which increases the chances that the inner edge corner of the bottle mouth end face is overlooked. According to the inventor's research, although the inner edge corner of the bottle mouth end face is small, once cracks appear at the inner edge corner of the bottle mouth end face, the sealing performance of the bottle mouth will be greatly reduced in harsh environments. If there is dirt on the bottle mouth, the dirt will easily enter the bottle body, thereby affecting the quality of the contents of the bottle body. This solution only increases the image data collection of the inner wall of the bottle mouth and the inner edge corner of the bottle mouth end face, so that the originally ignored parts are detected, thereby improving the comprehensiveness of the detection.

[0009] Preferably, as an improvement, when collecting image data of the bottle mouth, side lighting is used for illumination, and the bottle mouth end face and the outer edge corners of the bottle mouth end face are simultaneously collected by the same image acquisition module. The image data collected from the bottle mouth end face and the outer edge corners of the bottle mouth end face show bright areas and dark areas. The image processing system takes the center of the bottle mouth as the origin, and selects the detection area of ​​the bright area and the detection area of ​​the dark area through the radius setting, the wall thickness setting of the mouth edge and the arc length setting of the area. Then, the image processing system eliminates the interference in the detection area before performing defect detection in the detection area.

[0010] Beneficial effects: The side-lighting method of this solution makes the collected image data of the bottle mouth end face and the outer edge corners of the bottle mouth end face show obvious dark areas and bright areas. There are two dark areas, and the line connecting the two dark areas is perpendicular to the line connecting the bright area and the light source, and the bright area is close to the light source; this solution can more clearly see the reflection of cracks on the bottle mouth surface and the edges of the bottle mouth in the dark area, and can also accurately detect surface dirt and cracks in the bright area, thereby improving the accuracy of detection. Compared with the prior art method of illuminating the bottle mouth camera with both a light source placed at the bottom of the bottle body and a light source on the side of the bottle body, this solution makes the dark area darker and the cracks in the dark area more obvious after the bottom light source is removed, thereby improving the accuracy of detection. At the same time, this solution saves the bottom light source and reduces the cost.

[0011] In addition, the method of first removing interference from the detection area and then performing defect detection on the detection area is conducive to eliminating interference caused by reflections and further improving the accuracy of detection.

[0012] Preferably, as an improvement, the image processing system eliminates interference with the detection area in the following manner:

[0013] First, the detection area is expanded to obtain the expanded area, and then the expanded area is segmented according to the threshold processing to obtain the suspected interference area. The suspected interference area is compared in width. When the width of the suspected interference area exceeds the width setting value of the image processing system, it is determined to be an interference area; if the width of the suspected interference area meets the width setting value of the image processing system, the suspected interference area is further compared in area, aspect ratio and angle to determine the interference; then defect detection is performed on the remaining area after subtracting the interference area from the detection area.

[0014] Beneficial effects: Since interference is mostly caused by reflections, such as the reflection of the friction wheel on the bottle body, such as the reflection of the guide wheel on the bottle body, and these reflections are often continuous areas with a larger area than the defects, and the reflective area is significantly different from defects such as dirt or cracks in aspect ratio and distribution angle relative to the center of the bottle mouth. This solution first compares the width of the suspected interference area, and then compares the area, aspect ratio and angle of the suspected interference area to maximize the elimination of interference and avoid mistaking defects for interference, further improving the accuracy of detection.

[0015] Preferably, as an improvement, the image acquisition module for acquiring the edge corners of the bottle mouth end face and the outer edge corners of the bottle mouth end face is offset by 5-10 mm from the central axis of the bottle body and its shooting direction is 3-5° with the central axis of the bottle body.

[0016] Beneficial effects: This solution can make the reflective effect better when collecting image data, and the defects in the dark areas of the collected image data, such as cracks, breaks, scratches, etc., are clearer and more obvious, further improving the accuracy of detection.

[0017] Preferably, as an improvement, before the image data of the bottle body is collected, the lifting seat drives the bottle body placed on the two guide wheels to rise. When the bottle body rises, the bottle body pushes the friction wheel to rise, and the friction wheel rotates only after it rises to the designed position; after the image data of the bottle body is collected, the friction wheel presses the bottle body to fall.

[0018] Beneficial effect: In the prior art, the bottle body to be inspected is continuously conveyed forward by the rotary conveyor. During the conveying process, the bottle body is quickly lifted up by the lifting seat to the position of the friction wheel for rotation inspection. If the speed of bottle body inspection is too slow, the bottle body inspection will become a bottleneck of production, affecting the production efficiency; and if the inspection speed is increased, for example, the bottle body inspection only takes a few seconds or even about 1 second, but when the inspection speed is increased, it is found that the scrap rate of the bottle body increases suddenly. According to statistics, the increase in the scrap rate mainly comes from the situation that the bottle body falls and is scrapped after being lifted by the lifting seat, and the situation that the bottle body falls and is scrapped after detaching from the lifting seat during the falling process after the inspection.

[0019] The inventor has studied and found that the reason why the bottle body is easy to fall whether in the rising stage or the falling stage is closely related to the rapid lifting of the lifting seat. The rapid lifting of the bottle body by the guide wheel on the lifting seat will cause the smooth bottle body to have a tendency to be thrown upward due to the sudden upward supporting force (overweight), and the contact between the guide wheel itself and the bottle body is line contact, and the length of the contact line is much smaller than the height of the bottle, resulting in the bottle body being unable to be stably placed on the guide wheel and falling; and when the bottle body is detected and falls, the falling speed of the lifting seat is higher than the falling speed of the bottle body, and the bottle body is easily in a weightless state, resulting in the problem that the bottle body still cannot be stably placed on the guide wheel, and finally it is easy to fall.

[0020] In this solution, before the bottle body is inspected, the friction wheel is in the lowest position under its own gravity and the gravity of the vertical sliding frame. When the bottle body needs to be inspected, the bottle body is driven to rise by the lifting seat. After the lifting seat rises slightly, the bottle body on the lifting seat is against the bottom of the friction wheel until the bottle body pushes the friction wheel to rise to the designed position. At this time, the driver drives the friction wheel to rotate, and the friction wheel drives the bottle body to rotate. During the rotation, multiple image data of the bottle body are collected by the image acquisition component. Compared with the prior art, the friction wheel of this solution can be pressed up by the friction wheel just after the bottle body is lifted a short distance, ensuring that the inspected bottle body is confined in the triangular space enclosed by the friction wheel and the two guide wheels during most of the lifting stroke, which greatly reduces the probability of the bottle body falling during the lifting process, that is, greatly reduces the scrap rate of the bottle body after the lifting seat is accelerated.

[0021] In addition, after the bottle image data is collected, the lifting seat descends, and the friction wheel will also descend. During the descent, the friction wheel falls together with the bottle, which is equivalent to pressing down on the bottle from the top of the bottle, alleviating the weightlessness of the bottle, reducing the probability of the bottle falling during the descent, and further reducing the scrap rate of the bottle after the lifting seat is accelerated.

[0022] Preferably, as an improvement, the friction wheel rotates on a vertical sliding frame, the vertical sliding frame can slide vertically along the frame, and a push rod is fixedly connected to the lifting seat, and the push rod can resist the vertical sliding frame.

[0023] Beneficial effect: When this solution is adopted, when the lifting seat drives the bottle body to rise, the push rod also rises synchronously with the lifting seat. When the bottle body and the friction wheel are against each other, the push rod also supports the vertical sliding frame, thereby ensuring that the rising force of the vertical sliding frame and the friction wheel mainly comes from the push rod, avoiding the problem of the bottle body being broken due to excessive pressure from the friction wheel and the vertical sliding frame.

[0024] Preferably, as an improvement, a height adjustment rod is provided between the vertical sliding frame and the top rod, the height adjustment rod is mounted on the vertical sliding frame or on the top rod, and the height adjustment rod can adjust the minimum distance between the top rod and the vertical sliding frame.

[0025] Beneficial effect: When the present solution is adopted, since the contour surface where the friction wheel contacts the bottle body is elastic (for example, the friction wheel itself is a rubber wheel, or a rubber ring is sleeved on the outer diameter of the friction wheel), the height adjustment rod can adjust the minimum spacing between the push rod and the vertical sliding frame, that is, the minimum spacing between the friction wheel (the friction wheel moves synchronously with the vertical sliding frame) and the guide wheel (the guide wheel moves synchronously with the lifting seat, and the push rod moves synchronously with the lifting seat) can be adjusted by adjusting the height adjustment rod. After the minimum distance is adjusted, the pressure of the friction wheel on the bottle body can be adjusted (if the minimum spacing is increased, the support of the bottle body on the friction wheel is weakened and the support of the push rod on the vertical sliding frame is strengthened, thereby reducing the friction between the bottle body and the friction wheel. On the contrary, the force of the bottle body supporting the friction wheel is increased, thereby increasing the friction between the bottle body and the friction wheel), so as to ensure that the friction between the bottle body and the friction wheel is guaranteed without breaking the bottle body.

[0026] In addition, when this solution is adopted, after the height adjustment rod is adjusted, the distance between the guide wheel and the friction wheel on the lifting seat is changed by the adjustment of the height adjustment rod, thereby adapting to the space placement requirements of new specifications of bottles during the inspection process.

[0027] In addition, due to the setting of the height adjustment rod, the height adjustment rod can be adjusted even if the specifications of the bottle body become larger (that is, the bottle diameter becomes larger), so that the rising force of the lifting friction wheel still mainly comes from the push rod, ensuring that after the specifications of the bottle body change, the bottle body will not be broken due to excessive pressure.

[0028] Preferably, as an improvement, a first elastic member is provided between the vertical sliding frame and the frame, and the first elastic member applies a downward pulling force to the vertical sliding frame.

[0029] Beneficial effect: When this solution is adopted, due to the setting of the first elastic member, the vertical sliding frame can have an active downward force to ensure that the friction wheel can be pressed against the bottle body, and to ensure that the friction force between the friction wheel and the bottle body after rotation is sufficient to make the bottle body rotate.

[0030] In addition, even if the bottle body is detected and driven by the lifting seat to quickly descend (descending too quickly, the bottle body loses weight), due to the downward pulling tendency of the first elastic member on the vertical sliding frame, the friction wheel will still press on the bottle body to quickly descend, further alleviating the weightlessness of the bottle body during the falling process, ensuring that the bottle body is also confined in the enclosed space of the friction wheel and the two guide wheels during the descending process, ensuring that the bottle body will not fall from the guide wheels of the lifting seat when the lifting seat is rapidly descending.

[0031] Preferably, as an improvement, the bottle body to be inspected is automatically placed on the rotary conveyor by the loading device for transportation, and when the bottle body is transported to the bottom of the lifting seat, it is lifted by the lifting seat for bottle body inspection.

[0032] Beneficial effects: This solution improves the degree of automation of detection by setting up a rotary conveyor and a feeding device.

[0033] Preferably, as an improvement, after the bottle body inspection is completed, it is divided into qualified products, unqualified products in appearance and unqualified products in size.

[0034] Beneficial effect: This solution divides the inspected bottles into qualified products, unqualified products in appearance and unqualified products in size, which facilitates the subsequent re-inspection and confirmation of unqualified products in appearance through manual inspection or separate equipment inspection, thereby reducing the scrap rate of bottles. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a layout diagram of the image acquisition components of Embodiment 1 of the present invention.

[0036] Figure 2 Schematic diagram of the positions of the various surfaces of the bottle body in the first embodiment of the present invention.

[0037] Figure 3 It is the image data of the bottle mouth end face and the outer edge corners of the bottle mouth end face.

[0038] Figure 4 This is a logic diagram of the image processing system determining the suspected interference area in the first embodiment of the present invention.

[0039] Figure 5 This is an axonometric diagram of the second embodiment of the present invention.

[0040] Figure 6 It is a schematic diagram of the three-dimensional structure of the second embodiment of the present invention from another viewing angle.

[0041] Figure 7for Figure 5 Axonometric view of a rotating component in .

[0042] Figure 8 for Figure 6 Axonometric view of a rotating component in .

[0043] Fig. 9 for Figure 8 Schematic diagram of the structure of the vertical sliding frame in.

[0044] Fig.10 for Figure 7 Schematic diagram of the structure of the mobile frame and driving wheel.

[0045] Fig.11 for Figure 5 The partial structural diagram of the feeding device is shown in FIG. DETAILED DESCRIPTION

[0046] The following is further described in detail through specific implementation methods:

[0047] The figure marks in the drawings of the specification include: frame 1, rotary conveyor 2, unloading robot 61, re-inspection robot 62, packaging box 63, rotating component 3, lifting seat 31, guide wheel 32, friction wheel 33, vertical sliding frame 34, driving wheel 35, push rod 36, height adjustment rod 37, first elastic member 38, buffer 39, mobile frame 40, second elastic member 41, spacing adjustment rod 42, bottle bottom image acquisition module 51, bottle body image acquisition module 52, bottle inner wall image acquisition module 53, bottle mouth inner wall image acquisition module 54, bottle mouth image acquisition module 55, light source 56.

[0048] Embodiment 1

[0049] Combination Figures 1 to 4 A detection method uses an image acquisition component with multiple image acquisition modules to collect image data of the bottle mouth, bottleneck, bottle body and bottle bottom of the bottle body, and uses an image processing system to analyze and calculate the collected image data to determine the defects of the bottle body.

[0050] Specifically, the image acquisition component includes a bottle bottom image acquisition module 51, a bottle body image acquisition module 52, a bottle inner wall image acquisition module 53, a bottle mouth inner wall image acquisition module 54 and a bottle mouth image acquisition module 55. All image acquisition modules share a light source 56, which is fixed on the frame 1. The light source 56 is parallel to the bottle body. The positional relationship between all image acquisition modules and the bottle body and the light source 56 is as follows: Figure 1 and Figure 2 As shown, the bottle bottom image acquisition module 51 and the bottle inner wall image acquisition module 53 are symmetrically arranged with respect to the bottle body.

[0051] Bottle bottom image acquisition module 51: side-light, side-view photography of the bottle body, acquisition of image data of the bottle bottom and the bottle body, facilitating the image processing system to detect defects at the bottle bottom and the left half of the bottle body.

[0052] Bottle body image acquisition module 52: Backlit, facing the bottle body to capture the overall image data of the bottle body, so as to facilitate the image processing system to detect the dimensions of various parts of the medicine bottle.

[0053] Bottle inner wall image acquisition module 53: side-light, side-view photography of the bottle body, acquisition of image data of the bottle mouth, bottleneck and bottle body, to facilitate the image processing system to detect defects in the bottle mouth, bottleneck and right half of the bottle body.

[0054] Bottle mouth inner wall image acquisition module 54: side light, side view of the bottle body, mainly collects image data of the bottle mouth inner wall and the inner edge corners of the bottle mouth end face, which facilitates the image processing system to detect defects of the bottle mouth inner wall and the inner edge corners of the bottle mouth end face. The shooting direction of the bottle mouth inner wall image acquisition module 54 is 20°-40° with the central axis of the bottle body.

[0055] Bottle mouth image acquisition module 55: Installed in front of the bottle body, the installation position is offset by 5-10mm from the center axis of the bottle body, the shooting direction of the bottle mouth image acquisition module 55 is 3°-5° with the center axis of the bottle body, and the bottle body is photographed from the side. The bottle mouth image acquisition module 55 collects image data of the bottle mouth end face and the outer edge corners of the bottle mouth end face to facilitate the image processing system to detect defects in the bottle mouth end face and the outer edge corners of the bottle mouth end face. The image data of the bottle mouth end face and the outer edge corners of the bottle mouth end face show bright areas and dark areas, combined with Figure 3 , the number of bright areas is 1, which is facing the light source 56, and the number of dark areas is two. Figure 3 For example, on the upper and lower sides of the bottle.

[0056] The image processing system takes the center of the bottle mouth as the origin, and selects the detection area of ​​the bright area and the detection area of ​​the dark area by setting the radius, the wall thickness of the mouth edge and the arc length of the area ( Figure 3 The inspection area is the area circled by the fan-shaped ring), and then the image processing system eliminates the interference in the inspection area before performing defect detection in the inspection area.

[0057] When analyzing and calculating the received ground image data, the image processing system determines the specific type of defects based on threshold processing, contrast analysis, area comparison, length analysis, aspect ratio comparison and angle comparison. The defects include: small black spots, stains, small solids, scratches, air lines, cracks, breaks, cracks, flatness, crooked necks, slanted bottoms, short bottles, concave bottoms, high edges, etc.

[0058] When the image processing system detects the image data collected by the bottle mouth image acquisition module 55, since the image data collected by the bottle mouth is easily disturbed by reflections, in order to ensure the accuracy of the detection, the interference in the detection area is first eliminated before the defect detection of the detection area is performed.

[0059] Combination Figure 3 , the image processing system eliminates interference in the detection area in the following way: first, the detection area is expanded to obtain the expanded area (such as Figure 3 The area containing the sector ring and with white bright lines is the expanded area), and then the expanded area is segmented according to the threshold processing to obtain the suspected interference area, and the suspected interference area is compared in width. When the width of the suspected interference area exceeds the width setting value of the image processing system, it is determined to be an interference area; if the width of the suspected interference area meets the width setting value of the image processing system, the suspected interference area is further compared in area, aspect ratio and angle to determine the interference; in this embodiment, the suspected interference area is determined to be an interference area by the image processing system when the line width exceeds the setting value, the aspect ratio exceeds the setting value and the angle range exceeds the setting value. After the interference area is determined, defect detection is performed on the remaining area minus the interference area from the detection area.

[0060] This detection method realizes all-round detection of the bottle body through the design of five groups of image acquisition modules, solves the problem of neglecting the detection of the inner edge corners of the bottle mouth end face in the prior art, fills the gap in the detection of the inner edge corners of the bottle mouth end face and the inner wall of the bottle mouth, and eliminates the interference area of ​​reflection during the detection process, further improves the detection accuracy, and makes the defect detection of the bottle body more comprehensive and accurate; in addition, through the setting of side lighting and the bottle mouth image acquisition module 55 at an inclined angle, defects are more easily reflected, thereby further improving the accuracy of detection.

[0061] Embodiment 2

[0062] Embodiment 2 introduces a detection machine for detecting the bottle body on the basis of Embodiment 1 to realize fully automated rapid detection of the bottle body. The specific detection machine used is as follows:

[0063] Combination Figures 5 to 11The inspection machine includes a frame 1 and a rotary conveyor 2 mounted on the frame 1, a rotating component 3 and an image acquisition component. A loading device (not shown in the figure) is installed at the input end of the rotary conveyor 2. The bottle to be inspected can be transferred to the rotary conveyor 2 by using the loading device. The loading mechanical device can be a loading manipulator, a three-dimensional robot, a four-dimensional robot, a three-dimensional space mobile module, etc. with a suction cup loading structure. The rotating component 3 is located in the middle of the rotary conveyor line, and the rotary conveyor 2 includes two synchronously arranged chain plate rotary conveyor lines. A unloading device is provided at the output end of the rotary conveyor 2, and the unloading device includes a robot with multiple suction cups. For example, the unloading device includes a unloading robot 61 and a re-inspection robot 62 installed on the frame 1. Suction cups are fixed at the output ends of the unloading robot 61 and the re-inspection robot 62, and the bottle body is transferred by applying negative pressure to the suction cups. The unloading robot 61 and the re-inspection robot 62 both have a vertical linear module and a horizontal linear module. The moving direction of the output end of the horizontal linear module is perpendicular to the conveying direction of the rotary conveyor 2.

[0064] Specifically: the rotating component 3 includes a lifting seat 31, a friction wheel 33, a vertical sliding frame 34 and a driver. The lifting seat 31 is located in the gap between the two rotary conveyor lines. The active lifting of the lifting seat 31 can be driven by a driving mechanism. The driving mechanism can be a vertical cylinder, a vertically arranged linear module or a vertical conveyor, etc. The driving mechanism is fixedly installed on the frame 1. In this embodiment, the driving mechanism adopts a vertical linear module, and the lifting seat 31 is fixedly connected to the slider of the vertical linear module.

[0065] Two self-rotating guide wheels 32 are connected to the top of the lifting seat 31 , and there is an air space between the two guide wheels 32 . In this embodiment, the guide wheels 32 use bearings.

[0066] The vertical sliding frame 34, the friction wheel 33 and the driver are all located above the lifting seat 31 and above the rotary conveyor line. The friction wheel 33 is rotatably connected to the vertical sliding frame 34, and the vertical sliding frame 34 is vertically slidably connected to the frame 1. After the friction wheel 33 rises to the designed position, the driver drives the friction wheel 33 to rotate. The driver includes a driving motor and a driving wheel 35. The output end of the driving motor is fixedly connected to the driving wheel 35. The driving wheel 35 and the friction wheel 33 are connected by direct friction transmission. When the friction wheel 33 rises to the designed position, the friction wheel 33 and the driving wheel 35 are against each other, and the driving wheel 35 drives the friction wheel 33 to rotate by friction.

[0067] A top rod 36 is also fixedly connected to the lifting seat 31. In this embodiment, a bottom cross bar is fixed to the bottom of the vertical sliding frame 34, and a top cross bar is fixed to the top of the vertical sliding frame 34. A height adjustment rod 37 is threadedly connected to the bottom cross bar. In this embodiment, the height adjustment rod 37 adopts a screw rod. The bottom of the height adjustment rod 37 can be against the top of the top rod 36. The height adjustment rod 37 can adjust the height difference between the bottom of the height adjustment rod 37 and the friction wheel 33. A first elastic member 38 is provided between the top cross bar of the vertical sliding frame 34 and the frame 1. The first elastic member 38 gives the vertical sliding frame 34 a downward pulling force. In this embodiment, the first elastic member 38 adopts a vertically arranged spring.

[0068] A vertical buffer 39 is also installed on the frame 1. The buffer 39 is used to limit the lowest position of the vertical sliding frame 34. The buffer 39 is located directly below the top crossbar.

[0069] The driving wheel 35 is connected to a moving frame 40, a driving motor is fixed to the moving frame 40, the driving wheel 35 is rotatably connected to the moving frame 40, and the driving motor drives the driving wheel 35 to rotate. The moving frame 40 is slidably connected to the frame 1 in a transverse direction, and a second elastic member 41 is provided between the moving frame 40 and the frame 1, and the second elastic member 41 gives the moving frame 40 a pulling force close to the friction wheel 33. In this embodiment, the driving wheel 35 is located on the right side of the friction wheel 33, and the second elastic member 41 makes the driving wheel 35 have a tendency to move to the left, and the second elastic member 41 also adopts a spring. The setting of the second elastic member 41 enables the driving wheel 35 to press against the friction wheel 33 to provide the friction wheel 33 with sufficient tangential force, so as to ensure that the driving wheel 35 can drive the friction wheel 33 to rotate in a friction manner. A spacing adjustment rod 42 is provided between the mobile frame 40 and the frame 1, and the spacing adjustment rod 42 can adjust the minimum spacing between the driving wheel 35 and the friction wheel 33. In this embodiment, the spacing adjustment rod 42 is threadedly connected to the mobile frame 40 and the free left end of the spacing adjustment rod 42 can be against the frame 1. The spacing adjustment rod 42 can adjust the maximum contact force between the driving wheel 35 and the friction wheel 33 to ensure that the driving wheel 35 can drive the friction wheel 33 to rotate in a friction manner, thereby reducing the contact force between the driving wheel 35 and the friction wheel 33, and avoiding the problem that the bottle body of the lifting friction wheel 33 is subjected to excessive pressure and broken due to excessive contact force between the two.

[0070] All image acquisition modules of the image acquisition component are installed on the rack 1.

[0071] The specific process of using this inspection machine to inspect bottles is as follows:

[0072] In the first step, the loading device automatically loads the bottles to be tested.

[0073] In the second step, the bottle body is sent to the top of the lifting seat 31 by the rotary conveyor line, and the lifting seat 31 lifts the bottle body moved to the top of the lifting seat 31. When the bottle body rises, the bottle body pushes the friction wheel 33 to rise, and the push rod 36 pushes the vertical sliding frame 34 to rise. At this time, the push rod 36 and the bottle body overcome the elastic force of the first elastic member 38 and the gravity of the friction wheel 33 and the vertical sliding frame 34. Only after the friction wheel 33 rises to the designed position, the friction wheel 33 presses against the driving wheel 35, and the driving motor drives the driving wheel 35 to rotate, so that the friction wheel 33 rotates under the drive of the driving wheel 35, and the bottle body also rotates with the rotation of the friction wheel 33; during the rotation of the bottle body, the image acquisition component completes the image data acquisition. After the image data acquisition is completed, the lifting seat 31 descends, and the friction wheel 33 presses the bottle body to fall.

[0074] In the third step, the bottles that have completed the inspection fall onto the rotary conveyor line for further transportation. The bottles that have completed the inspection are determined by the image processing system as qualified products, unqualified products in appearance, and unqualified products in size. The unloading robot 61 transfers the qualified products that have been inspected to the packaging table (not shown in the figure) set next to the inspection machine. The re-inspection robot 62 transfers the unqualified products in appearance that have been detected to the packaging box 63 fixed on the inspection machine frame 1. The unqualified products in size automatically fall from the output end of the rotary conveyor line as they are transported by the rotary conveyor line and are collected by the recycling box.

[0075] Compared with the first embodiment, this embodiment not only realizes full automation of bottle detection, but also has the following effects:

[0076] First: During the inspection process, this solution enables the bottle body to be continuously sent to the rotating component 3 under the transmission of the rotary conveyor 2, which facilitates the rotary component 3 to realize assembly-line inspection of the bottle body. After the inspection, the qualified bottles are sucked by the unloading robot 61 and sent to the packaging table for subsequent packaging; the unqualified products in appearance are sucked by the re-inspection robot 62 and sent to the packaging box 63; and the unqualified products in size automatically fall into the recycling box placed under the rotary conveyor 2 as the rotary conveyor 2 rotates. This implementation realizes the inspection of the bottle body in an assembly-line manner, and after the inspection is completed, the bottles with different inspection structures are automatically processed separately, which improves the inspection speed of the bottle body, and at the same time, the unqualified products are classified, which can help reduce the final product scrap rate.

[0077] Second: before the bottle body is inspected, when the bottle body is transferred from other places to the rotary conveyor line, the position of the bottle body on the rotary conveyor line may be inconsistent, and the bottle body with a small height size is easy to fall from the rotary conveyor line. The present embodiment greatly reduces the placement requirements of the bottle body through the setting of the support rod, and also avoids the problem of scrapping of the bottle body due to falling from the rotary conveyor 2; in addition, the two correcting plates arranged at the end of the support rod enable the bottle body to achieve position correction under the guide surface of the correcting plate before entering the top of the lifting seat 31, ensuring that all bottles are lifted by the lifting seat 31 at the same position for inspection, which is conducive to ensuring the accuracy of the detection structure and reducing the difficulty of bottle body detection.

[0078] Third: When the lifting seat 31 drives the bottle body to rise, the push rod 36 also rises synchronously with the lifting seat 31. When the bottle body contacts the friction wheel 33, the push rod 36 also supports the height adjustment rod 37 on the vertical sliding frame 34, thereby ensuring that the rising force of the vertical sliding frame 34 and the friction wheel 33 mainly comes from the push rod 36, avoiding the problem of deformation or breakage of the bottle body due to excessive pressure from the friction wheel 33 and the vertical sliding frame 34 (for soft and easily deformable bottles, deformation can be greatly reduced or even avoided, and for very fragile bottles, such as glass bottles, the problem of bottle breakage can be avoided).

[0079] Fourthly, due to the setting of the height adjustment rod 37 on the vertical sliding frame 34, the minimum distance between the guide wheel 32 and the friction wheel 33 on the lifting seat 31 is changed by the adjustment of the height adjustment rod 37, and the space generated by the distance between the friction wheel 33 and the guide wheel 32 is used to accommodate the bottle body, that is, the existence of the height adjustment rod 37 enables the detection machine to meet the detection requirements of bottles with different outer diameter specifications, thereby improving the practicality of the detection machine; at the same time, the pressure of the friction wheel 33 on the bottle body can be adjusted by adjusting the height adjustment rod 37, thereby ensuring that the bottle body has sufficient friction with the friction wheel 33, thereby reducing the probability of deformation or breakage of the bottle body. When the outer diameter of the bottle to be inspected does not change much, even if the maximum ascending stroke of the lifting seat 31 is not adjusted, the inspection machine can ensure that there is enough friction between the friction wheel 33 and the bottle body under the action of the first elastic member 38 so that the bottle body rotates with the friction wheel 33 (the first elastic member 38 always has a tendency to pull the friction wheel 33 downward, and the downward pulling trend gives the friction wheel 33 a force to press the bottle body, thereby ensuring the friction between the two). It can also ensure that there is enough friction between the friction wheel 33 and the driving wheel 35 under the action of the second elastic member 41 so that the friction wheel 33 rotates with the driving wheel 35 (the second elastic member 41 always has a tendency to tighten the driving wheel 35 toward the friction wheel 33, and this tendency gives the driving wheel 35 a force to press the friction wheel 33, thereby ensuring the friction between the two). After the bottle specification changes slightly, it is only necessary to adjust the height of the height adjustment rod 37 accordingly to carry out the inspection of the new specification of the bottle. The whole method is simple and convenient to operate.

[0080] Fifth: When the outer diameter of the bottle being tested changes greatly, the maximum lifting stroke of the lifting seat 31 can be adjusted to meet the testing requirements of the new specification of the bottle; or when the maximum lifting stroke of the lifting seat 31 remains unchanged (then the highest position of the friction wheel 33 is maintained), the minimum distance between the driving wheel 35 and the friction wheel 33 can be adjusted by screwing the spacing adjustment rod 42, or the height adjustment rod 37 can be adjusted to adjust the distance between the guide wheel 32 and the friction wheel 33, thereby ensuring that the friction wheel 33 can still be pressed against the driving wheel 35 and driven by the driving wheel 35 to rotate, which can also meet the testing requirements of the new specification of the bottle, and this adjustment method is also very simple and convenient.

[0081] The above is only an embodiment of the present invention, and the common knowledge such as the known specific technical solutions and / or characteristics in the solution is not described in detail here. It should be pointed out that for those skilled in the art, without departing from the technical solution of the present invention, several modifications and improvements can be made, which should also be regarded as the protection scope of the present invention, and these will not affect the effect of the implementation of the present invention and the practicality of the patent. The scope of protection required by this application shall be based on the content of its claims, and the specific implementation methods and other records in the specification can be used to interpret the content of the claims.

Claims

1. A detection method, using an image acquisition component with multiple image acquisition modules to collect image data of the bottle mouth, bottle neck, bottle body and bottle bottom of the bottle body, and using an image processing system to analyze and calculate the collected image data to determine the defects of the bottle body, characterized in that: The image data collection of the bottle mouth includes the image data collection of the bottle mouth end face, the bottle mouth inner wall, the outer edge corners of the bottle mouth end face and the inner edge corners of the bottle mouth end face; Before the image data of the bottle is collected, the lifting seat drives the bottle placed on the two guide wheels to rise. When the bottle rises, the bottle pushes the friction wheel to rise. The friction wheel rotates only after it rises to the designed position. After the image data of the bottle is collected, the friction wheel presses the bottle to fall. The friction wheel rotates on a vertical sliding frame, which can slide vertically along the frame. A top rod is fixedly connected to the lifting seat, and the top rod can resist the vertical sliding frame. A height adjustment rod is provided between the vertical sliding frame and the top rod, the height adjustment rod is installed on the vertical sliding frame or on the top rod, the height adjustment rod can adjust the minimum distance between the top rod and the vertical sliding frame, and a first elastic member is provided between the vertical sliding frame and the frame, the first elastic member gives the vertical sliding frame a downward pulling force; A vertical buffer is also installed on the frame, and the buffer is used to limit the lowest position of the vertical sliding frame.

2. The detection method according to claim 1, characterized in that: When collecting image data of the bottle mouth, side lighting is used for illumination. The end face of the bottle mouth and the edges and corners of the outer edge of the bottle mouth are collected simultaneously using the same image acquisition module. The image data collected from the end face of the bottle mouth and the edges and corners of the outer edge of the bottle mouth show bright areas and dark areas. The image processing system takes the center of the bottle mouth as the origin, and selects the detection area of ​​the bright area and the detection area of ​​the dark area through the radius setting, the wall thickness setting of the mouth edge and the arc length setting of the area. Then the image processing system eliminates the interference in the detection area before performing defect detection in the detection area.

3. The detection method according to claim 2, characterized in that: The image processing system eliminates interference with the detection area as follows: First, the detection area is expanded to obtain the expanded area, and then the expanded area is segmented according to the threshold processing to obtain the suspected interference area. The suspected interference area is compared in width. When the width of the suspected interference area exceeds the width setting value of the image processing system, it is determined to be an interference area; if the width of the suspected interference area meets the width setting value of the image processing system, the suspected interference area is further compared in area, aspect ratio and angle to determine the interference; then defect detection is performed on the remaining area after subtracting the interference area from the detection area.

4. The detection method according to claim 2, characterized in that: The image acquisition module for collecting the edge angles of the bottle mouth end face and the outer edge angle of the bottle mouth end face is offset from the central axis of the bottle body by 5-10 mm, and its shooting direction is 3-5° with the central axis of the bottle body.

5. The detection method according to claim 1, characterized in that: The bottle to be inspected is automatically placed on the rotary conveyor by the loading device for transportation. When the bottle is transported to the bottom of the lifting seat, it is lifted by the lifting seat for bottle inspection.

6. The detection method according to claim 1, characterized in that: After the bottle body inspection is completed, it is divided into qualified products, unqualified products in appearance and unqualified products in size.

Citation Information

Patent Citations

  • Bottleneck defect detection method adopting residual analysis and dynamic threshold segmentation

    CN105334219A

  • Empty bottle online detection method and device

    CN111141750A

  • Online visual inspection device for medicinal glass bottles

    CN111282837A

  • Glass oral liquid bottle defect detection system

    CN113189120A