Curvature detection device and curvature detection method

Abstract

The invention provides a curvature detection device and a curvature detection method, and relates to the technical field of bar detection. The curvature detection device comprises a conveying mechanism, a discharging mechanism, a collecting mechanism and a detection mechanism, the conveying mechanism comprises a conveying table and a conveying assembly, and the conveying table is used for bearing a to-be-detected bar; the discharging mechanism is used for containing a to-be-detected bar and can convey the to-be-detected bar to the conveying table. Wherein the conveying table is configured to be capable of receiving to-be-detected bars output from the discharging mechanism, the conveying assembly can drive the to-be-detected bars to roll along the conveying table and towards the collecting mechanism, the collecting mechanism can collect end image information of the to-be-detected bars, and the detecting mechanism detects the end image information of the to-be-detected bars according to the end image information of the to-be-detected bars. And determining a curvature detection result of the to-be-detected bar. By capturing and analyzing the end image information of the to-be-detected bar, the automatic detection of the quality of the to-be-detected bar is realized, the detection standard is unified, and the detection accuracy and efficiency are improved.

Description

Technical Field

[0001] The present invention generally relates to the technical field of bar material detection, and in particular, to a curvature detection device and a curvature detection method. Background Art

[0002] Since the curvature or straightness of the bar plays a key role in the assembly of parts, products with certain precision requirements need to be tested for the curvature of the bar before assembly. The existing bar curvature detection has the following forms:

[0003] The first method is manual visual inspection or manual measurement using tools such as a ruler or a dial indicator. This method will overly rely on the subjective experience of the operator, and it is difficult to distinguish bars with unclear curvature with the naked eye, which is easy to miss and difficult to ensure the reliability of the test results.

[0004] The second method is to transfer the bar to a special curvature measuring device for measurement. Specifically, the curvature measuring device has a contact or non-contact sensor to collect the shape data of the bar to be measured, but this method has a cumbersome and time-consuming detection process and is not suitable for occasions with a high degree of automation requirements.

[0005] In addition, the above two methods are only applicable to off-line detection, which affects production continuity and production efficiency. Moreover, off-line detection cannot reflect product quality in a timely manner, resulting in a relatively high scrap rate after assembly. Summary of the invention

[0006] The invention provides a curvature detection device and a curvature detection method, which can improve detection accuracy and detection production efficiency.

[0007] According to a first aspect of the present invention, there is provided a curvature detection device, comprising:

[0008] A conveying mechanism, comprising a conveying platform and a conveying assembly, wherein the conveying platform is used to carry the bar to be tested;

[0009] A material unloading mechanism, used for accommodating the bar to be inspected and capable of conveying the bar to be inspected to the conveying platform;

[0010] A collecting mechanism, located on a side of the unloading mechanism away from the conveying assembly;

[0011] A detection mechanism, located on at least one side of the conveying mechanism;

[0012] Wherein, the conveyor platform is configured to be able to receive the rod to be detected output from the unloading mechanism, and the conveying assembly can drive the rod to be detected to roll along the conveyor platform and toward the collection mechanism, so that the collection mechanism can collect end image information of the rod to be detected, and the detection mechanism determines the curvature detection result of the rod to be detected based on the end image information of the rod to be detected.

[0013] In some embodiments, the delivery assembly comprises:

[0014] A conveying air nozzle is capable of blowing air toward the conveying platform to make the bar to be inspected roll relative to the conveying platform.

[0015] In some embodiments, along the first direction, the height of the conveying gas nozzle is equal to the height of the conveying platform, and the conveying gas nozzle and the top surface of the conveying platform along the first direction are arranged in parallel;

[0016] Alternatively, along the first direction, the height of the conveying gas nozzle is greater than the height of the conveying platform, and the conveying gas nozzle and the top surface of the conveying platform along the first direction are arranged at an angle.

[0017] In some embodiments, two first baffles are respectively provided on both sides of the conveying platform along the third direction, the bar to be detected is located between the two first baffles, the first baffles extend along the second direction, and the first baffles are used to limit the bar to be detected;

[0018] The first direction, the second direction and the third direction are perpendicular to each other, and the second direction is the extension direction of the conveying platform.

[0019] In some embodiments, the side surface of the gas delivery nozzle along the second direction and the side surface of the first baffle along the third direction are vertically arranged;

[0020] And / or, a central axis of the gas delivery nozzle along the second direction is the same as a distance between the two first baffles along the third direction.

[0021] In some embodiments, the collection mechanism includes:

[0022] A photographing component, used for photographing the end position of the bar to be inspected;

[0023] A light source is arranged directly opposite to the shooting component;

[0024] A first sensor, located between the photographing component and the unloading mechanism, for identifying the bar to be detected;

[0025] The second sensor is located at a side of the photographing component away from the first sensor, and is used to identify the bar to be inspected after being photographed by the photographing component.

[0026] In some embodiments, a picking mechanism is further included. The picking mechanism is located on a side of the collecting mechanism away from the unloading mechanism. The picking mechanism is used to classify and collect the bars that have completed the inspection.

[0027] In some embodiments, the picking mechanism includes:

[0028] A first collection box and a first air nozzle are located on both sides of the conveying platform along the third direction and are arranged correspondingly, and the first air nozzle can blow the inspected rods into the first collection box;

[0029] A second collection box and a second air nozzle are located on both sides of the first collection box along the second direction and are arranged correspondingly, and the second air nozzle can blow the inspected rods into the second collection box;

[0030] Wherein, one of the first collection box and the second collection box is used to collect qualified bars among the bars that have been inspected, and the other is used to collect defective bars among the bars that have been inspected;

[0031] The second direction and the third direction are perpendicular to each other, and the second direction is an extension direction of the conveying platform.

[0032] In some embodiments, two second baffles are respectively provided on both sides of the conveying platform along the third direction, the second baffles are located between the first collecting box, the first air nozzle, the second collecting box and the second air nozzle, the rod to be inspected is located between the two second baffles, and the second baffles are used to limit the rod to be inspected.

[0033] In some embodiments, the detection mechanism includes:

[0034] An image processing unit, used for processing the image information of the end of the rod to be detected collected by the collection mechanism to extract the moving track of the rod to be detected;

[0035] A detection and judgment unit, used for analyzing the motion trajectory of the bar to be detected obtained by the image processing unit and obtaining a curvature detection result of the bar to be detected, wherein the curvature detection result includes qualified bars and defective bars;

[0036] A result statistics unit, used for performing result statistics according to the curvature detection result obtained by the detection and judgment unit;

[0037] The control unit is used to control the picking mechanism to classify and collect defective bars and qualified bars among the bars that have completed the inspection according to the curvature inspection result.

[0038] In some embodiments, the unloading mechanism comprises:

[0039] Mounting frame;

[0040] A lower hopper is rotatably connected to the mounting frame, the lower hopper is used to accommodate the bar to be tested, the lower hopper is at least partially a conical structure, the small end of the lower hopper is arranged toward the conveying platform, and the small end of the lower hopper is provided with a lowering port;

[0041] A first driving assembly, connected to the lower hopper, the first driving assembly being capable of driving the lower hopper to swing relative to the mounting frame;

[0042] A receiving plate, wherein a receiving groove is provided on one side of the receiving plate facing the feeding port;

[0043] The second driving component is arranged on the mounting frame and connected to the receiving plate. The second driving component can drive the receiving plate to move in the direction close to the discharge port, and is used for the accommodating groove to be arranged opposite to the discharge port, so that the rod to be inspected falls into the accommodating groove through the discharge port.

[0044] According to a second aspect of the present invention, an embodiment of the present invention further provides a curvature detection method, comprising the following steps:

[0045] Acquire multiple image information of the bar to be detected, wherein the multiple image information is image information of the end of the bar to be detected at different positions collected when the bar to be detected is driven by the conveying component to roll along the conveying platform;

[0046] Performing image processing on multiple image information to obtain multiple characteristic image points;

[0047] Connect multiple feature image points to obtain the motion trajectory of the bar to be detected;

[0048] The bending test result of the bar to be tested is determined according to the maximum and minimum values ​​of the motion trajectory.

[0049] In some embodiments, obtaining a plurality of image information of the bar to be inspected includes the following steps:

[0050] Obtain video information of the bar to be inspected within a preset time;

[0051] The video frame of the video information is extracted to obtain a plurality of image information.

[0052] In some of the embodiments, the image processing method for performing image processing on multiple image information includes at least one of a background difference method, an optical flow method, a frame difference method, and a machine learning target detection algorithm.

[0053] In some embodiments, performing image processing on the plurality of image information comprises the following steps:

[0054] Performing image processing on multiple image information and preset background image information to obtain multiple characteristic image points;

[0055] The preset background image information is an image when there is no bar to be detected on the conveying platform.

[0056] In some implementation modes, the preset background image information is obtained before obtaining the multiple image information or by calling a database.

[0057] In some implementations, obtaining a plurality of characteristic image points comprises the following steps:

[0058] When the image information is a point image, the coordinates corresponding to the point image are determined as feature image points;

[0059] When the feature image point is a regional image, the contour of the regional image is obtained, the minimum circumscribed rectangle is determined according to the contour of the regional image, and the coordinates corresponding to the center point of the minimum circumscribed rectangle are determined as the feature image point.

[0060] In some embodiments, determining the curvature test result of the bar to be tested includes the following steps:

[0061] According to the difference between the maximum value and the minimum value of the motion trajectory, the curvature of the bar to be tested is determined;

[0062] Compare the curvature of the bar to be tested with the preset curvature;

[0063] Determine whether the curvature of the bar to be tested is less than or equal to a preset curvature;

[0064] If yes, determine that the bending test result of the bar to be tested is a qualified bar;

[0065] If not, it is determined that the bending test result of the bar to be tested is a defective bar.

[0066] In some embodiments, after determining the curvature test result of the bar to be tested, the following steps are included:

[0067] According to the curvature test results of the bars to be tested, the control picking mechanism will classify and collect the defective bars and qualified bars among the bars that have completed the test.

[0068] An embodiment of the present invention has the following advantages or beneficial effects:

[0069] The curvature detection device provided in this embodiment uses a feeding mechanism to output the bar to be detected to the conveyor platform. Under the driving action of the conveyor assembly, the bar to be detected can roll along the conveyor platform. When the bar to be detected passes through the collection mechanism, the collection mechanism can collect the end image information of the bar to be detected. As the bar to be detected rolls, if the curvature of the circumferential side of the bar to be detected is inconsistent, it will be directly reflected in the end position of the bar to be detected, that is, the curvature change of the bar to be detected is converted into the up and down jump amplitude of the end side when the bar to be detected rolls. The curvature of the bar to be detected can be characterized by detecting the motion trajectory of the end side of the bar to be detected. The detection mechanism determines the curvature detection result of the bar to be detected based on the end image information of the bar to be detected.

[0070] By capturing and analyzing the motion trajectory of the end image of the bar to be inspected, the curvature of the bar to be inspected can be quickly and accurately inspected, thereby realizing automated inspection of the quality of the bar to be inspected, unifying the inspection standards, and solving the problems of low efficiency and low accuracy of manual inspection. This not only frees inspectors from heavy and repetitive work, but also improves the accuracy and efficiency of inspection and saves labor costs.

[0071] The curvature detection method provided in this embodiment collects image information of the ends of the rod to be detected at different positions when the conveying component is driven to roll along the conveying platform, thereby obtaining multiple image information of the rod to be detected, and then performs image processing on the multiple image information to obtain multiple feature image points, and then connects the multiple feature image points to obtain the motion trajectory of the rod to be detected; according to the maximum and minimum values ​​of the motion trajectory, the curvature detection result of the rod to be detected is determined, thereby realizing automated quality detection of the rod to be detected, unifying the detection standards, effectively improving the efficiency and accuracy of the detection, and saving labor costs. BRIEF DESCRIPTION OF THE DRAWINGS

[0072] In order to better understand the present invention, reference may be made to the embodiments shown in the following drawings. The components in the drawings are not necessarily to scale, and related elements may be omitted in order to emphasize and clearly illustrate the technical features of the present invention. In addition, related elements or components may have different settings as known in the art. In addition, in the drawings, the same reference numerals represent the same or similar components in each of the drawings. The above and other features and advantages of the present invention will become more apparent by describing in detail its exemplary embodiments with reference to the drawings.

[0073] in:

[0074] Figure 1 The figure shows a schematic diagram of the structure of a bar material to be detected by a bending detection device according to an embodiment of the present invention;

[0075] Figure 2The structure of the bending detection device according to one embodiment of the present invention is shown in FIG. Figure 1 ;

[0076] Figure 3 The structure of the bending detection device according to one embodiment of the present invention is shown in FIG. Figure 2 ;

[0077] Figure 4 The figure shows the structure of the feeding mechanism in the bending detection device according to one embodiment of the present invention. Figure 1 ;

[0078] Figure 5 The figure shows the structure of the feeding mechanism in the bending detection device according to one embodiment of the present invention. Figure 2 ;

[0079] Figure 6 Shown is a cross-sectional view of a material feeding mechanism in a curvature detection device according to an embodiment of the present invention;

[0080] Figure 7 It shows Figure 6 A local enlarged view at point A;

[0081] Figure 8 The figure shows the motion trajectory of the bar to be tested by the bending detection device according to one embodiment of the present invention. Figure 1 ;

[0082] Fig. 9 The figure shows the motion trajectory of the bar to be tested by the bending detection device according to one embodiment of the present invention. Figure 2 ;

[0083] Fig.10 FIG. 1 is a flow chart of a method for detecting curvature according to an embodiment of the present invention.

[0084] The reference numerals are described as follows:

[0085] 100. Rod to be tested; 101. Oil pipe; 102. Pen tip;

[0086] 1. Conveying mechanism; 2. Unloading mechanism; 3. Collecting mechanism; 4. Detecting mechanism; 5. Picking mechanism; 6. Base;

[0087] 11. conveying platform; 111. first baffle plate; 112. second baffle plate;

[0088] 12. Conveying assembly; 121. Conveying air nozzle; 122. First fixing seat;

[0089] 21. mounting frame; 22. material discharging hopper; 221. material discharging port; 23. first driving assembly; 231. first driving source; 232. first connecting rod; 233. second connecting rod; 234. third connecting rod; 24. material receiving plate; 241. containing groove; 25. second driving assembly;

[0090] 31. Shooting component; 32. Light source; 33. First sensor; 34. Second sensor; 35. Display;

[0091] 51. First collecting box; 52. First gas nozzle; 53. Second collecting box; 54. Second gas nozzle. DETAILED DESCRIPTION

[0092] The following will be combined with the accompanying drawings in the exemplary embodiments of the present invention to clearly and completely describe the technical solutions in the exemplary embodiments of the present invention. The exemplary embodiments described herein are only for illustrative purposes and are not intended to limit the scope of protection of the present invention. Therefore, it should be understood that various modifications and changes can be made to the exemplary embodiments without departing from the scope of protection of the present invention.

[0093] In the description of the present invention, unless otherwise clearly specified and limited, the terms "first" and "second" are used for descriptive purposes only and cannot be understood as indicating or implying relative importance; the term "plurality" refers to two or more; the term "and / or" includes any and all combinations of one or more associated listed items. In particular, reference to "the / the" object or "an" object is also intended to indicate one of a possible plurality of such objects.

[0094] Unless otherwise specified or explained, the terms "connection", "fixation", etc. should be understood in a broad sense. For example, "connection" can be a fixed connection, a detachable connection, an integral connection, an electrical connection, or a signal connection; "connection" can be a direct connection or an indirect connection through an intermediate medium. For those skilled in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0095] Further, in the description of the present invention, it should be understood that the directional words such as "upper", "lower", "inner", "outer" and the like described in the exemplary embodiments of the present invention are described at the angles shown in the accompanying drawings and should not be understood as limiting the exemplary embodiments of the present invention. It should also be understood that, in the context, when it is mentioned that an element or feature is connected to another element (one or more) "upper", "lower", or "inner", "outer", it can not only be directly connected to the other (one or more) elements "upper", "lower", "inner", "outer", but also can be indirectly connected to the other (one or more) elements "upper", "lower", "inner", "outer" through an intermediate element.

[0096] Example embodiments will now be described more fully with reference to the accompanying drawings. However, example embodiments can be implemented in a variety of forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that the present invention will be comprehensive and complete and fully convey the concepts of the example embodiments to those skilled in the art. The same reference numerals in the figures represent the same or similar structures, and thus their detailed description will be omitted.

[0097] This embodiment provides a curvature detection device for detecting the curvature of a rod 100 to be detected, wherein the rod 100 to be detected is at least partially cylindrical, that is, the rod 100 to be detected may be entirely cylindrical or partially cylindrical, for example. Figure 1 As shown, the rod 100 to be detected may have an optional refill, which includes an oil tube 101 and a pen tip 102. The pen tip 102 is arranged at the end of the oil tube 101 along the axial direction of the oil tube 101. The diameter of the pen tip 102 is smaller than the diameter of the pen shaft, and the position of the pen tip 102 is specifically the end of the rod 100 to be detected.

[0098] like Figure 2-Figure 3 As shown, the curvature detection device includes a base 6 and a conveying mechanism 1, a feeding mechanism 2, a collecting mechanism 3 and a detecting mechanism 4 arranged on the base 6. The conveying mechanism 1 includes a conveying platform 11 and a conveying assembly 12. The conveying platform 11 is used to carry the bar material 100 to be detected. The feeding mechanism 2 is used to accommodate the bar material 100 to be detected and can convey the bar material 100 to the conveying platform 11. The collecting mechanism 3 is located on the side of the feeding mechanism 2 away from the conveying assembly 12, and the detecting mechanism 4 is located on at least one side of the conveying mechanism 1. Among them, the conveying platform 11 is configured to be able to receive the bar material 100 to be detected output from the feeding mechanism 2, and the conveying assembly 12 can drive the bar material 100 to be detected to roll along the conveying platform 11 and toward the collecting mechanism 3, so that the collecting mechanism 3 can collect the end image information of the bar material 100 to be detected, and the detecting mechanism 4 determines the curvature detection result of the bar material 100 to be detected according to the end image information of the bar material 100 to be detected.

[0099] The curvature detection device provided in this embodiment uses the unloading mechanism 2 to output the bar material 100 to be detected to the conveying platform 11. Under the driving action of the conveying assembly 12, the bar material 100 to be detected can roll along the conveying platform 11. When the bar material 100 to be detected passes through the collecting mechanism 3, the collecting mechanism 3 can collect the end image information of the bar material 100 to be detected. As the bar material 100 to be detected rolls, if the curvature of the side surface of the bar material 100 to be detected is inconsistent, it will be directly reflected in the end position of the bar material 100 to be detected, that is, the curvature change of the oil pipe 101 in the bar material 100 to be detected is converted into the jumping amplitude of the pen tip 102. By detecting the position of the pen tip 102 of the bar material 100 to be detected, the curvature of the pen shaft in the bar material 100 to be detected can be characterized. The detecting mechanism 4 determines the curvature detection result of the bar material 100 to be detected based on the end image information of the bar material 100 to be detected.

[0100] By capturing and analyzing the image information of the end of the rod 100 to be inspected, the curvature of the rod 100 to be inspected can be quickly and accurately inspected, thereby realizing automated inspection of the quality of the rod 100 to be inspected, unifying the inspection standards, and solving the problems of low efficiency and low accuracy of manual inspection. This can not only free the inspectors from heavy and repetitive work, but also improve the accuracy and efficiency of inspection and save labor costs.

[0101] It should be noted that although Figure 2 The rod 100 to be detected is shown in the figure, but the structure of the curvature detection device provided in this embodiment does not include the rod 100 to be detected.

[0102] It should be understood that the number of rods 100 to be inspected can be one or more. When the unloading mechanism 2 accommodates multiple rods 100 to be inspected, the unloading mechanism 2 can output the multiple rods 100 to be inspected one by one to the conveying platform 11 to ensure the accuracy of the inspection result of each rod 100 to be inspected.

[0103] In one embodiment, the base 6 is similar to a rectangular frame structure, and the height direction of the base 6 is defined as the first direction, the length direction of the base 6 is defined as the second direction, the second direction is the extension direction of the conveying platform 11, and the width direction of the base 6 is the third direction, wherein the first direction, the second direction and the third direction are perpendicular to each other.

[0104] The top surface of the conveying platform 11 along the first direction is a smooth plane, which is conducive to the rolling of the bar 100 to be inspected along the conveying platform 11 .

[0105] In one embodiment, Figure 4-Figure 5As shown, the unloading mechanism 2 includes a mounting frame 21, a unloading hopper 22 and a first driving assembly 23. The mounting frame 21 is arranged on the base 6, and the mounting frame 21 is used to install the unloading hopper 22 and the first driving assembly 23. The unloading hopper 22 is used to accommodate the bar 100 to be detected. The unloading hopper 22 is at least partially a conical structure. The small end of the unloading hopper 22 is arranged toward the conveying platform 11. The small end of the unloading hopper 22 is provided with an unloading port 221. The unloading port 221 can be a through hole, such as a rectangular through hole. The size of the through hole is slightly larger than the size of one bar 100 to be detected, but smaller than the size of two bars 100 to be detected, so that the bars 100 to be detected can pass through the unloading port 221 one by one. The inclined surface of the conical structure plays a guiding role, so that the bar 100 to be detected is guided and accommodated along the inclined surface toward the unloading port 221, so that the bar 100 to be detected can be discharged from the unloading port 221.

[0106] One end of the lower hopper 22 is rotatably connected to the mounting frame 21, and the other end is connected to the first driving assembly 23. For example, the lower opening 221 and the mounting frame 21 are connected via a rotating shaft, and the first driving assembly 23 can drive the lower hopper 22 to swing relative to the mounting frame 21. If a plurality of bars 100 to be detected are closely arranged in the lower hopper 22, when the lower hopper 22 shakes, the plurality of bars 100 to be detected become loose under the guidance of the inclined surface of the conical structure, which is conducive to the discharge of the bars 100 to be detected through the lower opening 221.

[0107] Specifically, the first driving assembly 23 includes a first driving source 231, a first connecting rod 232 and a second connecting rod 233. The first driving source 231 may be a motor. One end of the first connecting rod 232 is connected to the output end of the first driving source 231, and the other end is rotatably connected to the second connecting rod 233. One end of the second connecting rod 233 away from the first connecting rod 232 is connected to the lower hopper 22. The first driving source 231 can drive the first connecting rod 232 to rotate. As the first connecting rod 232 rotates, the second connecting rod 233 is driven to move along the third direction. The second connecting rod 233 drives the lower hopper 22 to swing relative to the mounting frame 21, thereby shaking the lower hopper 22.

[0108] By controlling the rotation speed of the first driving source 231 , the time interval between two adjacent bars 100 to be inspected being discharged from the discharge port 221 is controlled, so that one bar 100 to be inspected passes through the collecting mechanism 3 and the inspecting mechanism 4 for collection and inspection.

[0109] It is understandable that the number of the second connecting rod 233 can be one or more. Specifically, when the number of the second connecting rod 233 is one, the second connecting rod 233 can be connected to the approximately middle position of the lower hopper 22, and the connection position between the second connecting rod 233 and the lower hopper 22 and the lower material port 221 are staggered to avoid blocking the lower material port 221.

[0110] Specifically, when the number of the second connecting rods 233 is two, the two second connecting rods 233 are arranged on both sides of the mounting frame 21 along the second direction, and the first driving assembly 23 further includes a third connecting rod 234, which is extended along the second direction, and the third connecting rod 234 is arranged between the two second connecting rods 233, and the two second connecting rods 233 are fixedly connected to both sides of the lower hopper 22 along the second direction to ensure the stability of the swing of the lower hopper 22. At the same time, the third connecting rod 234 plays the role of an intermediate transmission, and the synchronous movement of the two second connecting rods 233 is realized through the third connecting rod 234 by using a first driving source 231, so as to ensure the swing balance of the lower hopper 22.

[0111] In one embodiment, Figure 5-Figure 7 As shown, the material unloading mechanism 2 includes a material receiving plate 24 and a second driving assembly 25. The material receiving plate 24 is provided with a receiving groove 241 on one side facing the material unloading port 221. The second driving assembly 25 is provided on the mounting frame 21 and connected to the material receiving plate 24. The second driving assembly 25 can drive the material receiving plate 24 to move toward or away from the material unloading port 221.

[0112] When the second driving component 25 drives the receiving plate 24 to move toward the direction close to the unloading port 221, the accommodating groove 241 is arranged opposite to the unloading port 221. At this time, the unloading port 221 is in an open state, so that the rod 100 to be inspected falls into the accommodating groove 241 through the unloading port 221, and then the conveying component 12 blows the rod 100 to be inspected in the accommodating groove 241 onto the conveying platform 11.

[0113] When the second driving assembly 25 drives the receiving plate 24 to move in the direction away from the discharge port 221, the accommodating groove 241 is offset from the discharge port 221. At this time, the discharge port 221 is in a closed state, and the rod 100 to be inspected cannot fall into the accommodating groove 241 through the discharge port 221. This means that the previous rod 100 to be inspected has not completely passed through the collecting mechanism 3 and the detecting mechanism 4, and the discharge mechanism 2 is in a waiting state for unloading, so as to avoid the situation where the collecting mechanism 3 collects multiple rods 100 to be inspected, causing detection confusion and affecting the detection accuracy.

[0114] The second driving assembly 25 includes a second driving source and a connecting shaft, and the output end of the second driving source is connected to the receiving plate 24 via the connecting shaft. The second driving source can be a cylinder, a motor, or the like.

[0115] In one embodiment, Figure 2 As shown, the conveying assembly 12 includes a first fixed seat 122 and a conveying air nozzle 121. The first fixed seat 122 is located on one side of the conveying platform 11 along the second direction. The conveying air nozzle 121 is arranged at the top of the first fixed seat 122 along the first direction. The conveying air nozzle 121 can blow air toward the conveying platform 11 to make the rod 100 to be inspected roll relative to the conveying platform 11.

[0116] When the rod 100 to be tested is a pen core, the weight of the rod 100 to be tested is relatively light. By blowing air toward the conveying platform 11 through the conveying air nozzle 121, the rod 100 to be tested can be blown and made to roll along the conveying platform 11, saving driving energy and reducing detection costs.

[0117] In one embodiment, along the first direction, when the height of the first fixing seat 122 is approximately equal to the height of the conveying platform 11, the height of the conveying air nozzle 121 is equal to the height of the conveying platform 11, and the conveying air nozzle 121 and the top surface of the conveying platform 11 along the first direction are arranged in parallel.

[0118] In this configuration, the conveying air nozzle 121 is flush with the conveying platform 11 , and the airflow direction output by the conveying air nozzle 121 is along the extension direction of the conveying platform 11 , which is beneficial for the bar 100 to be tested to roll directly on the conveying platform 11 along the second direction.

[0119] In another embodiment, along the first direction, when the height of the first fixed seat 122 is greater than the height of the conveying platform 11, the height of the conveying air nozzle 121 is greater than the height of the conveying platform 11, and the conveying air nozzle 121 and the top surface of the conveying platform 11 along the first direction are set at an angle.

[0120] For example, the conveying gas nozzle 121 is tilted downward relative to the conveying platform 11, so that there is a certain angle between the direction of the airflow output by the conveying gas nozzle 121 and the extension direction of the conveying platform 11. The force of the airflow output by the conveying gas nozzle 121 on the rod 100 to be inspected can be decomposed into a pressure F1 along a first direction and a driving force F2 along a second direction. The driving force F2 enables the rod 100 to be inspected to roll on the conveying platform 11 along the second direction. The pressure F1 can press the rod 100 to be inspected against the conveying platform 11, avoiding the situation where the rod 100 to be inspected is blown away and causes position displacement, thereby ensuring the detection and conveying accuracy of the rod 100 to be inspected.

[0121] It is understandable that the conveying assembly 12 includes but is not limited to pneumatic rolling, and the conveying assembly 12 can also adopt a friction rolling method. For example, the pen head 102 in the bar material 100 to be detected is at least partially made of magnetic material, and the conveying assembly 12 includes a moving part and a magnet. The moving part drives the magnet to move along the second direction, and the magnet attracts the pen head 102 of the bar material 100 to be detected, thereby driving the rolling and movement of the bar material 100 to be detected. Of course, the conveying assembly 12 can also be other methods, as long as it can achieve the rolling and movement of the bar material 100 to be detected, it is within the protection scope of this embodiment.

[0122] In one embodiment, Figure 2As shown, two first baffles 111 are respectively provided on both sides of the conveying platform 11 along the third direction, and the bar 100 to be detected is located between the two first baffles 111. The first baffle 111 extends along the second direction, and the first baffle 111 is used to limit the bar 100 to be detected.

[0123] In this way, the extension direction of the first baffle 111 is consistent with the conveying direction of the rod 100 to be detected. The first baffle 111 not only guides the rod 100 to be detected, but also limits the position of the rod 100 to be detected, thereby avoiding position deviation of the rod 100 to be detected during the conveying process, ensuring the position stability of the rod 100 to be detected when passing through the collection mechanism 3, thereby ensuring the detection accuracy of the rod 100 to be detected.

[0124] In one embodiment, the side surface of the gas delivery nozzle 121 along the second direction and the side surface of the first baffle 111 along the third direction are vertically arranged.

[0125] In this way, the conveying air nozzle 121 is arranged parallel to the rod 100 to be detected on the side along the second direction, and the airflow direction output by the conveying air nozzle 121 is arranged perpendicular to the rod 100 to be detected, which is conducive to the rolling of the rod 100 to be detected along the second direction.

[0126] In one embodiment, the distance between the center axis of the gas delivery nozzle 121 along the second direction and the distance between the two first baffles 111 along the third direction is the same.

[0127] That is, the central axis of the conveying air nozzle 121 along the second direction is collinear with the central axis of the rod 100 to be detected along the second direction, and the conveying air nozzle 121 and the rod 100 to be detected are aligned to avoid the situation where the rod 100 to be detected becomes skewed due to the blowing deviation of the conveying air nozzle 121.

[0128] In one embodiment, Figure 2 As shown, the acquisition mechanism 3 includes a shooting component 31 and a light source 32. The shooting component 31 is used to shoot the end position of the rod 100 to be detected, so as to collect the end image information of the rod 100 to be detected. The light source 32 and the shooting component 31 are arranged on both sides of the conveying platform 11 along the third direction, or the light source 32 is arranged directly above the conveying platform 11 along the first direction. The light source 32 is arranged directly opposite to the shooting component 31, and is used to fill light for the shooting component 31 to ensure the accuracy of the image captured by the shooting component 31.

[0129] Specifically, the shooting component 31 includes a camera, a lens, a shooting light source, and a light source controller. The lens is arranged on the camera, wherein the camera can be a planar array camera, the lens can be a FA high-magnification lens, and the shooting light source can be a bar light source. The shooting light source is used to fill in the lens with light to improve the shooting quality. The light source controller is electrically connected to the shooting light source and is used to adjust the opening and closing or brightness of the shooting light source.

[0130] Specifically, the shooting component 31 also includes an adjustment platform, which is connected to the camera. The adjustment platform can drive the camera to move along the first direction, the second direction and the third direction to achieve precise adjustment of the camera position, and can better collect the end image information of the rod 100 to be inspected.

[0131] The adjustment platform includes a first adjustment driving source, a first adjustment platform, a second adjustment driving source, a second adjustment platform, a third adjustment driving source and a third adjustment platform. The output end of the first adjustment driving source is connected to the first adjustment platform, and the first adjustment driving source can drive the first adjustment platform to move in the second direction. The output end of the second adjustment driving source is connected to the second adjustment platform, and the second adjustment driving source can drive the second adjustment platform to move in the third direction. The output end of the third adjustment driving source is connected to the third adjustment platform, and the third adjustment driving source can drive the third adjustment platform to move in the first direction.

[0132] It is understandable that the first adjustment drive source, the second adjustment drive source and the third adjustment drive source can be motors or manual screws. This embodiment does not limit the automatic adjustment or manual adjustment mode of each adjustment drive source, and can be adjusted according to actual production needs. In addition, this embodiment takes the first adjustment platform, the second adjustment platform and the third adjustment platform as an example of stacking the first adjustment platform, the second adjustment platform and the third adjustment platform in sequence from bottom to top along the first direction. The setting positions of each adjustment platform can be interchanged. As long as the adjustment platform can realize the movement of the shooting component 31 along the first direction, the second direction and the third direction, it is within the protection scope of this embodiment.

[0133] In one embodiment, Figure 2 As shown, the acquisition mechanism 3 further includes a first sensor 33 and a second sensor 34, wherein the first sensor 33 may be a photoelectric sensor or an infrared sensor, and the first sensor 33 is located between the shooting component 31 and the unloading mechanism 2, and is used to identify the bar material 100 to be detected. The second sensor 34 is located on a side of the shooting component 31 away from the first sensor 33, and is used to identify the bar material 100 to be detected after the shooting component 31 has completed shooting.

[0134] When the bar material 100 to be detected rolls along the conveyor 11 and passes through the first sensor 33, the bar material 100 to be detected just enters the shooting field of view of the shooting component 31, and the first sensor 33 sends a high-level signal to trigger the shooting component 31 to start recording. When the bar material 100 to be detected continues to roll along the conveyor 11 and passes through the second sensor 34, the second sensor 34 is triggered, which means that the bar material 100 to be detected just leaves the shooting field of view of the shooting component 31, and the second sensor 34 sends a high-level signal to trigger the shooting component 31 to stop recording.

[0135] It can be understood that the time difference between the bar 100 to be detected triggering the first sensor 33 and the second sensor 34 is the duration of the video recording of the bar 100 to be detected by the shooting component 31.

[0136] It can be understood that by controlling the rotation speed of the first driving source 231 in the unloading mechanism 2 to control the unloading speed of the unloading mechanism 2, it is ensured that only one rod 100 to be inspected appears in the shooting field of view of the shooting component 31 at the same time, that is, the previous rod 100 to be inspected moves out of the shooting field of view of the shooting component 31 before the next rod 100 to be inspected can enter the shooting field of view of the shooting component 31.

[0137] The acquisition mechanism 3 further includes a display 35 , and the image information of the end of the bar material 100 to be inspected acquired by the photographing component 31 can be displayed on the display 35 .

[0138] In one embodiment, the detection mechanism 4 includes an image processing unit, and the image processing unit is used to process the end image information of the rod material 100 to be detected collected by the collection mechanism 3 to extract the movement trajectory of the rod material 100 to be detected.

[0139] Specifically, when the conveying component 12 is driven to roll along the conveying platform 11, multiple image information of the rod 100 to be detected is obtained by collecting end image information of the rod 100 to be detected at different positions, and then the multiple image information is processed to obtain multiple feature image points, and then the multiple feature image points are connected to obtain the motion trajectory of the rod 100 to be detected.

[0140] It is understandable that the video information of the rod 100 to be detected can be collected within a preset time, and the video information is the moving image of the pen core in the rolling process of the rod 100 to be detected. Each frame of video information corresponds to an image information, and by extracting the video frame of the video information, the multiple image information is made into continuous image information, which can more accurately fit the motion trajectory of the rod 100 to be detected, and further improve the detection accuracy of the rod 100 to be detected.

[0141] The preset time can be obtained by the time interval between the first sensor 33 and the second sensor 34 identifying the bar 100 to be detected. For example, the time when the bar 100 to be detected passes through the first sensor 33 and triggers the first sensor 33 is T1, and the time when the bar 100 to be detected passes through the second sensor 34 and triggers the second sensor 34 is T2, then the preset time is T2-T1.

[0142] The video information of the rod 100 to be inspected specifically refers to the image of the tip of the pen tip 102 in the rod 100 to be inspected, or the image of the area of ​​the pen tip 102 .

[0143] Since the bar material 100 to be detected is a moving object, the multiple image information is multiple video frames in the image sequence of the bar material 100 to be detected, and multiple characteristic image points are obtained by performing image processing on the multiple image information and the preset background image information; wherein the preset background image information is an image when there is no bar material 100 to be detected on the collection conveyor 11. Then, the video information of the bar material 100 to be detected collected by the collection mechanism 3 and the preset background image information are compared frame by frame, and the difference position between each frame of video information and the preset background image information is found by comparing the two and performing a differential operation, and then the differential result is thresholded to find the position of the pen head 102 of the bar material 100 to be detected in each frame of video information, thereby extracting the motion area of ​​the end of the bar material 100 to be detected, so as to obtain the end motion trajectory of the bar material 100 to be detected.

[0144] Exemplarily, if the image information is a point image, for example, the point image is the tipmost position of the pen tip 102 in the rod 100 to be detected, then the tipmost position corresponds to a coordinate point, and the coordinates corresponding to the point image are determined as feature image points to characterize the movement area of ​​the pen tip 102 of the rod 100 to be detected.

[0145] Exemplarily, if the image information is not a point image but a regional image, the regional image is an image in a region near the pen tip 102 of the rod 100 to be detected, and the regional image includes an image corresponding to at least part of the pen tip 102 of the rod to be detected. The contour of the regional image is obtained, and the minimum circumscribed rectangle is determined according to the contour of the regional image, and the coordinates corresponding to the center point of the minimum circumscribed rectangle are determined as feature image points. Then, the coordinates of the feature image points of each frame of video information are plotted on an image, and the motion trajectory of the pen tip 102 in the rod 100 to be detected can be obtained.

[0146] The detection mechanism 4 also includes a detection and judgment unit, which is used to analyze the motion trajectory of the bar 100 to be detected obtained by the image processing unit and obtain the curvature detection result of the bar 100 to be detected, and the curvature detection result includes qualified bars and defective bars.

[0147] like Figure 8-Figure 9 As shown, the bending test result of the bar material 100 to be tested is determined according to the maximum and minimum values ​​of the motion trajectory. Since the bar material 100 to be tested moves along the second direction and from the first sensor 33 to the second sensor 34 within the preset time, the motion trajectory of the bar material 100 to be tested changes periodically, and the motion trajectory is approximately a sine curve. The maximum value of the motion trajectory is the peak of the sine curve, and the minimum value of the motion trajectory is the trough of the sine curve. The difference between the maximum and minimum values ​​of the motion trajectory is the peak amplitude of the sine curve. Figure 8 As shown, the peak amplitude is about 2 mm. The larger the peak amplitude, the greater the curvature of the bar 100 to be tested. Fig. 9 As shown, the peak amplitude is about 1 mm, and the smaller the peak amplitude is, the smaller the curvature of the bar 100 to be detected is. If the bar 100 to be detected is completely straight without curvature, the motion trajectory of the bar 100 to be detected is approximately a straight line.

[0148] The curvature of the rod 100 to be detected is determined according to the difference between the maximum value and the minimum value of the motion trajectory; the curvature of the rod 100 to be detected is compared with the preset curvature; if the curvature of the rod 100 to be detected is less than or equal to the preset curvature, it means that the curvature of the rod 100 to be detected is relatively small, and the curvature test result of the rod 100 to be detected is determined to be a qualified rod; if the curvature of the rod 100 to be detected is greater than the preset curvature, it means that the curvature of the rod 100 to be detected is relatively large, and the curvature test result of the rod 100 to be detected is determined to be a defective rod.

[0149] The detection mechanism 4 also includes a result statistics unit, which is used to perform result statistics based on the curvature detection results obtained by the detection judgment unit. The curvature detection results are then stored in a database. Specifically, the database is updated in real time to record the current total production, the total number of qualified products, the total number of defective products, and the qualified product rate, and the image collected by the collection mechanism 3, the curvature detection results analyzed by the detection mechanism 4, and the statistical production data are displayed in real time through the display screen 35. Then it is stored in the database in the form of Excel, which is convenient for viewing real-time data at the production site, tracing back defects, and improving production quality. This facilitates real-time monitoring and improvement of the production process and improves product quality.

[0150] The detection mechanism 4 also includes a result transmission unit and a control unit. The result transmission unit is used to send the curvature detection result obtained by the detection and judgment unit to the control unit. Exemplarily, the result transmission unit packages the quality detection result in a binary form (for example, qualified products are 1, and defective products are 0) and sends it to the control unit. The transmission mode of the result transmission unit can be serial communication, network communication (such as Socket communication) or other standard communication protocols. The specific selection depends on factors such as application requirements, transmission distance and data rate, and this embodiment does not limit this.

[0151] In one embodiment, Figure 2 As shown, the curvature detection device also includes a picking mechanism 5, which is located on the side of the collecting mechanism 3 away from the unloading mechanism 2. The picking mechanism 5 classifies and collects defective bars and qualified bars among the bars that have completed the detection.

[0152] Among them, after the rod 100 to be tested is tested by the collection mechanism 3 and the testing mechanism 4, the curvature test result of the rod 100 to be tested can be determined, wherein the curvature test result includes qualified rods and defective rods. According to different curvature test results, the control unit is used to control the picking mechanism 5 to classify and collect qualified rods and defective rods according to the curvature test result, and remove defective rods in time, which can efficiently and automatically screen out rods with curvature exceeding the error range, thereby reducing labor costs.

[0153] Specifically, Figure 2 As shown, the picking mechanism 5 includes a first collection box 51, a first air nozzle 52, a second collection box 53, and a second air nozzle 54. The first collection box 51 and the first air nozzle 52 are located on both sides of the conveying platform 11 along the third direction and are correspondingly arranged. The first air nozzle 52 can blow the bars that have been tested into the first collection box 51. The second collection box 53 and the second air nozzle 54 are located on both sides of the first collection box 51 along the second direction and are correspondingly arranged. The second air nozzle 54 can blow the bars that have been tested into the second collection box 53. Among them, one of the first collection box 51 and the second collection box 53 is used to collect qualified bars among the bars that have been tested, and the other is used to collect defective bars among the bars that have been tested.

[0154] For example, when the rod 100 to be inspected passes through the second sensor 34, the second sensor 34 is triggered and sends a trigger signal to the control unit. If the rod 100 to be inspected is a defective rod, the control unit controls the first air nozzle 52 to blow air to blow the inspected rod into the first collection box 51, which is used to collect defective rods; if the rod 100 to be inspected is a qualified rod, the control unit controls the second air nozzle 54 to blow air to blow the inspected rod into the second collection box 53, which is used to collect qualified rods.

[0155] The first air nozzle 52 and the second air nozzle 54 may both be fan-shaped air nozzles, and the first air nozzle 52 and the second air nozzle 54 may be arranged parallel to or at an angle to the top surface of the conveyor platform 11 along the first direction. The airflow directions of the first air nozzle 52 and the second air nozzle 54 are arranged at an angle, for example, the angle between the airflow directions of the first air nozzle 52 and the second air nozzle 54 is 90°, so that the output directions of the qualified bars and the defective bars do not interfere with each other.

[0156] In one embodiment, two second baffles 112 are respectively provided on both sides of the conveying platform 11 along the third direction, the second baffles 112 are located between the first collecting box 51, the first air nozzle 52, the second collecting box 53 and the second air nozzle 54, and the rod 100 to be inspected is located between the two second baffles 112, and the second baffles 112 are used to limit the rod 100 to be inspected.

[0157] In this way, the extension direction of the second baffle 112 is consistent with the conveying direction of the qualified rods. The second baffle 112 not only guides the qualified rods, but also limits the position of the qualified rods to avoid positional deviation of the qualified rods when the second air nozzle 54 blows and outputs them.

[0158] The curvature detection device provided in this embodiment can detect the curvature of the bar material based on machine vision. Python can be used as the development language, and the program can be written in combination with the open source visual function library OpenCV and PYQT5 to realize the functions required by the standardized detection system. The entire software interface may include an image display area, a basic menu area, a detection result display area, a software button operation area, and other parts. The curvature detection device can not only realize the real-time display function of the motion trajectory of the bar material 100 to be detected, and realize the function of automatically determining whether the curvature is qualified, but also realize the function of setting basic parameters in the menu area.

[0159] This embodiment also provides a method for detecting curvature. Fig.10 As shown, the curvature detection method includes the following steps:

[0160] S1, obtaining multiple image information of the rod 100 to be detected, wherein the multiple image information is image information of the end of the rod 100 to be detected at different positions collected when the rod 100 to be detected is driven by the conveying component 12 to roll along the conveying platform 11;

[0161] S2, performing image processing on the multiple image information to obtain multiple characteristic image points;

[0162] S3, connecting multiple feature image points to obtain the motion trajectory of the bar 100 to be detected;

[0163] S4. Determine the curvature test result of the bar 100 to be tested according to the maximum value and the minimum value of the motion trajectory.

[0164] Among them, the end image information of the rod 100 to be detected specifically refers to the image information of the pen tip 102 in the rod to be detected, and the different positions of the rod 100 to be detected specifically refer to the various positions of the pen tip 102 when the rod 100 to be detected rolls at least one circle along the second direction on the conveyor 11.

[0165] The curvature detection method provided in this embodiment collects image information of the end of the rod 100 to be detected at different positions when the conveying component 12 is driven to roll along the conveying platform 11, thereby obtaining multiple image information of the rod 100 to be detected, and then performs image processing on the multiple image information to obtain multiple characteristic image points, and then connects the multiple characteristic image points to obtain the motion trajectory of the rod 100 to be detected; according to the maximum and minimum values ​​of the motion trajectory, the curvature detection result of the rod 100 to be detected is determined, thereby realizing automatic quality detection of the rod 100 to be detected, unifying the detection standards, effectively improving the efficiency and accuracy of the detection, and saving labor costs.

[0166] The curvature detection method provided in this embodiment can be applied to the curvature detection device provided in this embodiment.

[0167] In one embodiment, obtaining a plurality of image information of the bar material 100 to be inspected includes the following steps:

[0168] Acquire video information of the bar material 100 to be inspected within a preset time;

[0169] The video frame of the video information is extracted to obtain a plurality of image information.

[0170] The preset time can be obtained by the time interval between the first sensor 33 and the second sensor 34 identifying the bar 100 to be detected. For example, the time when the bar 100 to be detected passes through the first sensor 33 and triggers the first sensor 33 is T1, and the time when the bar 100 to be detected passes through the second sensor 34 and triggers the second sensor 34 is T2, then the preset time is T2-T1.

[0171] The video information of the rod 100 to be inspected specifically refers to the image of the tip of the pen tip 102 in the rod 100 to be inspected, or the image of the area of ​​the pen tip 102 .

[0172] By collecting video information of the rod 100 to be detected within a preset time, the video information is a moving image of the pen core in the rolling process of the rod 100 to be detected. Each frame of video information corresponds to an image information. By extracting the video frames of the video information, multiple image information are made into continuous image information, which can more accurately fit the motion trajectory of the rod 100 to be detected, and further improve the detection accuracy of the rod 100 to be detected.

[0173] In another embodiment, the multiple image information of the rod 100 to be inspected may be intermittent image information. For example, when the conveying assembly 12 is driven to roll along the conveying platform 11, the collecting mechanism 3 collects the end image information of the rod 100 to be inspected at different positions at an interval time, and two adjacent interval times may be the same or different.

[0174] Among them, the processing method for processing multiple image information can adopt the background difference method. The background difference method builds a background model, compares the current frame image with the background model, finds out the pixels with large differences, and these pixels are considered to be part of the moving target, thereby detecting the moving target.

[0175] Specifically, performing image processing on multiple image information includes the following steps:

[0176] Performing image processing on multiple image information and preset background image information to obtain multiple characteristic image points;

[0177] The preset background image information is an image when the conveying platform 11 is not provided with the rod 100 to be inspected.

[0178] Since the rod 100 to be detected is a moving object, the multiple image information is multiple video frames in the image sequence of the rod 100 to be detected, and the preset background image information is used as the background reference model. The video information of the rod 100 to be detected collected by the acquisition mechanism 3 and the preset background image information are compared frame by frame, and the difference position between each frame of video information and the preset background image information is found by comparing the two and performing a differential operation, and then the differential result is thresholded to find the position of the pen head 102 of the rod 100 to be detected in each frame of video information, so as to extract the motion area of ​​the end of the rod 100 to be detected, so as to obtain the end motion trajectory of the rod 100 to be detected.

[0179] Exemplarily, the preset background image information is obtained before obtaining the multiple image information or by calling a database.

[0180] For example, the preset background image may be that there is no bar material 100 to be detected rolling in the shooting field of view of the acquisition mechanism 3, and the acquisition mechanism 3 acquires video information of the conveyor platform 11 within a preset time. Then, within the same preset time, the video information of the bar material 100 to be detected and the video information of the bar material 100 not to be detected are compared frame by frame to find the difference position, and the obtained image information is used as the movement area of ​​the bar material 100 to be detected.

[0181] Specifically, the cv2.createBackgroundSubtractor() method of the Python standard library OpenCV is used for implementation. cv2.createBackgroundSubtractorMOG2 is a function in the OpenCV library that is used to create a background subtractor. This function is based on the Gaussian Mixture Model (GMM) and is particularly suitable for dealing with lighting changes and dynamic backgrounds. Its main parameters include: history, varThreshold, and detectShadows. Among them, history is the number of historical frames of the background model, which is used to determine the update speed of the background model. If it represents the number of past frames, when history is 1, it is the difference between two frames. varThreshold is the variance threshold of the pixel belonging to the background, which is used to determine whether the pixel belongs to the background, or the threshold of the square distance between the pixel and the sample, which is used to determine whether the pixel is close to the sample. detectShadows is whether to detect shadows, and the default is False.

[0182] Exemplarily, the preset background image may also be obtained by retrieving an image previously saved in a database as the preset background image.

[0183] In some other embodiments, a preset background image may be constructed as a background model by averaging the first few frames of a video sequence or using other statistical methods.

[0184] This background difference method is simple to implement, easy to program and debug. Since the background of the conveyor platform 11 provided in this embodiment is fixed or changes slowly when there is no bar 100 to be detected, it can provide more accurate moving target information for the bar 100 to be detected.

[0185] It can be understood that the image processing method for multiple image information includes at least one of a background difference method, an optical flow method, a frame difference method, and a machine learning target detection algorithm.

[0186] In one embodiment, obtaining a plurality of characteristic image points comprises the following steps:

[0187] When the image information is a point image, the coordinates corresponding to the point image are determined as feature image points;

[0188] When the feature image point is a regional image, the contour of the regional image is obtained, the minimum circumscribed rectangle is determined according to the contour of the regional image, and the coordinates corresponding to the center point of the minimum circumscribed rectangle are determined as the feature image point.

[0189] Exemplarily, if the image information is a point image, for example, the point image is the tipmost position of the pen tip 102 in the rod 100 to be detected, then the tipmost position corresponds to a coordinate point, and the coordinate point can be used as a feature image point to characterize the movement area of ​​the pen tip 102 of the rod 100 to be detected.

[0190] Exemplarily, if the image information is not a point image but an area image, the area image is an image in an area near the pen tip 102 of the rod 100 to be inspected, and the area image includes an image corresponding to at least part of the pen tip 102 of the inspected rod.

[0191] The shape of this area image can be regular or irregular. For irregular area images, the minimum bounding rectangle can be determined according to the outer contour of the area image. The minimum bounding rectangle can also be called the minimum boundary rectangle, the minimum containing rectangle, the minimum outer bounding rectangle, etc. The minimum bounding rectangle refers to the maximum range of several two-dimensional shapes (such as points, lines, polygons) represented by two-dimensional coordinates, that is, the rectangle with the maximum horizontal coordinate, the minimum horizontal coordinate, the maximum vertical coordinate, and the minimum vertical coordinate of each vertex of the given two-dimensional shape. The minimum bounding rectangle contains the two-dimensional shape image represented by the given contour, that is, the minimum bounding rectangle is the two-dimensional form of the minimum bounding box.

[0192] Specifically, the contour of the regional image is obtained using the cv2.findContours() method of the Python standard library OpenCV. The contour of the regional image can be regarded as a curve connecting continuous points (along the boundary) with the same color or intensity. This function is often used for object detection and recognition in image processing and computer vision. Its main parameters include: image, mode method. Among them, image represents the input image and must be a binary image. mode represents the contour retrieval mode. Commonly used modes include: only detecting external contours, detecting all contours but not establishing a hierarchical relationship between contours, detecting all contours and organizing them into a two-level hierarchy, etc. method represents the contour approximation method. Commonly used methods include: storing all contour points, compressing elements in the horizontal, vertical and diagonal directions, etc.

[0193] After determining the minimum circumscribed rectangle, the coordinates corresponding to the center point of the minimum circumscribed rectangle are used as feature image points, and the center point of the rectangle is specifically the midpoint of the diagonal line of the minimum circumscribed rectangle. Then, the coordinates of the feature image points of each frame of video information are plotted on an image, and the motion trajectory of the pen head 102 in the rod 100 to be detected can be obtained.

[0194] Since the bar material 100 to be detected moves in the second direction from the first sensor 33 to the second sensor 34 within the preset time, the motion trajectory of the bar material 100 to be detected changes periodically, and the motion trajectory is approximately a sine curve. The maximum value of the motion trajectory is the peak of the sine curve, and the minimum value of the motion trajectory is the trough of the sine curve. The difference between the maximum and minimum values ​​of the motion trajectory is the peak amplitude of the sine curve. The greater the peak amplitude, the greater the curvature of the bar material 100 to be detected; the smaller the peak amplitude, the smaller the curvature of the bar material 100 to be detected. If the bar material 100 to be detected is completely straight and has no curvature, the motion trajectory of the bar material 100 to be detected is approximately a straight line.

[0195] In one embodiment, determining the curvature test result of the bar 100 to be tested includes the following steps:

[0196] Determine the curvature of the bar 100 to be tested according to the difference between the maximum value and the minimum value of the motion trajectory;

[0197] Comparing the curvature of the bar 100 to be tested with a preset curvature;

[0198] Determine whether the curvature of the bar 100 to be tested is less than or equal to a preset curvature;

[0199] If yes, it is determined that the bending test result of the bar material 100 to be tested is a qualified bar material;

[0200] If not, it is determined that the bending test result of the bar material 100 to be tested is a defective bar material.

[0201] Specifically, by linearly modeling the peak amplitude and the curvature of the bar 100 to be detected, the curvature of the bar 100 to be detected can be obtained. In this embodiment, the curvature of the bar 100 to be detected can be obtained through the peak amplitude by virtue of the linear or nonlinear relationship between the peak amplitude and the curvature of the bar 100 to be detected, and the functional relationship between the two will not be described in detail.

[0202] If the curvature of the rod 100 to be tested is less than or equal to the preset curvature, it means that the curvature of the rod 100 to be tested is relatively small, and the curvature test result of the rod 100 to be tested is determined to be a qualified rod; if the curvature of the rod 100 to be tested is greater than the preset curvature, it means that the curvature of the rod 100 to be tested is relatively large, and the curvature test result of the rod 100 to be tested is determined to be a defective rod.

[0203] In some embodiments, after determining the curvature test result of the bar 100 to be tested, the result is counted, and the curvature test result is stored in a database. Specifically, the database is updated in real time to record the current total production, the total number of qualified products, the total number of defective products, and the qualified product rate, and the image collected by the collection mechanism 3, the curvature test result obtained by the analysis of the detection mechanism 4, and the statistical production data are displayed in real time on the display screen. Then, it is stored in the database in the form of Excel, which is convenient for viewing real-time data at the production site, tracing back bad problems, and improving production quality. This facilitates real-time monitoring and improvement of the production process and improves product quality.

[0204] It should be noted that since most of the testing equipment at the production site is offline, the result statistics unit in this example stores the test results locally in the form of Excel. With the automation, intelligence and progress of the workshop, the collected data can be stored in the cloud synchronously, and the unified management of data of multiple devices, multiple production lines and even multiple factories can be achieved, and an intelligent defect traceability system can be established to facilitate product problem tracing, make the production process more transparent, and facilitate production management.

[0205] In one embodiment, after determining the curvature test result of the bar 100 to be tested, the following steps are included:

[0206] According to the curvature test result of the bar 100 to be tested, the picking mechanism 5 is controlled to classify and collect the defective bars and qualified bars among the tested bars.

[0207] Exemplarily, the result transmission unit transmits the curvature detection result to the control unit, and the control unit is used to control the picking mechanism 5 to remove the defective bars from the qualified bars. Specifically, if the curvature detection result of the bar 100 to be detected is a defective bar, when the bar 100 to be detected passes through the second sensor 34, the second sensor 34 is triggered and sends a trigger signal to the control unit, and the control unit controls the first air nozzle 52 to blow air to blow the bars that have completed the detection into the first collection box 51, and the first collection box 51 is used to collect defective bars; if the curvature detection result of the bar 100 to be detected is a qualified bar, the control unit controls the second air nozzle 54 to blow air to blow the bars that have completed the detection into the second collection box 53, and the second collection box 53 is used to collect qualified bars.

[0208] The curvature detection method provided in this embodiment is based on a curvature detection device of machine vision, which tracks and images the motion trajectory of the end of the bar 100 to be detected, calculates the peak amplitude of its motion trajectory, and constructs a model between the peak amplitude and the curvature of the bar, thereby realizing automatic, efficient and real-time detection of the curvature of the bar, and can timely eliminate defective products and count production data. Compared with manual detection methods, it has the advantages of simple operation, high efficiency and high accuracy.

[0209] It should be noted that the embodiments of the present invention are shown in the drawings and described in this specification is only an example of the principle of the present invention. It should be clearly understood by those skilled in the art that the principle of the present invention is not limited to any details or any components of the device shown in the drawings or described in the specification.

[0210] It should be understood that the present invention is not limited in its application to the detailed structure and arrangement of the components proposed in this specification. The present invention can have other embodiments and can be implemented and executed in a variety of ways. The aforementioned variations and modifications fall within the scope of the present invention. It should be understood that the present invention disclosed and defined in this specification extends to all alternative combinations of two or more individual features mentioned or evident in the text and / or the drawings. All these different combinations constitute multiple alternative aspects of the present invention. The embodiments described in this specification illustrate the best mode known for implementing the present invention and will enable those skilled in the art to utilize the present invention.

[0211] Those skilled in the art will readily appreciate other embodiments of the present invention after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present invention that follow the general principles of the present invention and include common knowledge or customary techniques in the art that are not disclosed by the present invention. The specification and example embodiments are to be considered exemplary only, and the true scope and spirit of the present invention are indicated by the appended claims.

[0212] It should be understood that the present invention is not limited to the precise structure described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of protection of the present invention is limited only by the appended claims.

Claims

1. A curvature detection device, characterized in that: include: A conveying mechanism, comprising a conveying platform and a conveying assembly, wherein the conveying platform is used to carry the bar to be tested; A material unloading mechanism, used for accommodating the bar to be inspected and capable of conveying the bar to be inspected to the conveying platform; A collecting mechanism, located on a side of the unloading mechanism away from the conveying assembly; A detection mechanism, located on at least one side of the conveying mechanism; Wherein, the conveyor platform is configured to be able to receive the rod to be detected output from the unloading mechanism, and the conveying assembly can drive the rod to be detected to roll along the conveyor platform and toward the collection mechanism, so that the collection mechanism can collect end image information of the rod to be detected, and the detection mechanism determines the curvature detection result of the rod to be detected based on the end image information of the rod to be detected.

2. The curvature detection device according to claim 1, characterized in that: The conveying assembly comprises: A conveying air nozzle is capable of blowing air toward the conveying platform to make the bar to be inspected roll relative to the conveying platform.

3. The curvature detection device according to claim 2, characterized in that: Along the first direction, the height of the conveying gas nozzle is equal to the height of the conveying platform, and the conveying gas nozzle and the top surface of the conveying platform along the first direction are arranged in parallel; Alternatively, along the first direction, the height of the conveying gas nozzle is greater than the height of the conveying platform, and the conveying gas nozzle and the top surface of the conveying platform along the first direction are arranged at an angle.

4. The curvature detection device according to claim 3, characterized in that: Two first baffles are respectively arranged on both sides of the conveying platform along the third direction, the bar to be detected is located between the two first baffles, the first baffles extend along the second direction, and the first baffles are used to limit the bar to be detected; The first direction, the second direction and the third direction are perpendicular to each other, and the second direction is the extension direction of the conveying platform.

5. The curvature detection device according to claim 4, characterized in that: The side surface of the gas delivery nozzle along the second direction and the side surface of the first baffle along the third direction are vertically arranged; And / or, a central axis of the gas delivery nozzle along the second direction is the same as a distance between the two first baffles along the third direction.

6. The curvature detection device according to claim 1, characterized in that: The collection mechanism includes: A photographing component, used for photographing the end position of the bar to be inspected; A light source is arranged opposite to the shooting component; A first sensor, located between the photographing component and the unloading mechanism, for identifying the bar to be detected; The second sensor is located at a side of the photographing component away from the first sensor, and is used to identify the bar to be inspected after being photographed by the photographing component.

7. The curvature detection device according to any one of claims 1 to 6, characterized in that: It also includes a picking mechanism, which is located on a side of the collecting mechanism away from the unloading mechanism, and is used to classify and collect the bars that have completed the inspection.

8. The curvature detection device according to claim 7, characterized in that: The picking mechanism comprises: A first collection box and a first air nozzle are located on both sides of the conveying platform along the third direction and are arranged correspondingly, and the first air nozzle can blow the inspected rods into the first collection box; A second collection box and a second air nozzle are located on both sides of the first collection box along the second direction and are arranged correspondingly, and the second air nozzle can blow the inspected rods into the second collection box; Wherein, one of the first collection box and the second collection box is used to collect qualified bars among the bars that have been inspected, and the other is used to collect defective bars among the bars that have been inspected; The second direction and the third direction are perpendicular to each other, and the second direction is an extension direction of the conveying platform.

9. The curvature detection device according to claim 8, characterized in that: Two second baffles are respectively provided on both sides of the conveying platform along the third direction, the second baffles are located between the first collecting box, the first air nozzle, the second collecting box and the second air nozzle, the rod to be inspected is located between the two second baffles, and the second baffles are used to limit the rod to be inspected.

10. The curvature detection device according to claim 7, characterized in that: The detection mechanism includes: An image processing unit, used for processing the image information of the end of the rod to be detected collected by the collection mechanism, so as to extract the moving track of the rod to be detected; A detection and judgment unit, used for analyzing the motion trajectory of the bar to be detected obtained by the image processing unit and obtaining a curvature detection result of the bar to be detected, wherein the curvature detection result includes qualified bars and defective bars; A result statistics unit, used for performing result statistics according to the curvature detection result obtained by the detection and judgment unit; The control unit is used to control the picking mechanism to classify and collect defective bars and qualified bars among the bars that have completed the inspection according to the curvature inspection result.

11. The curvature detection device according to any one of claims 1 to 6, characterized in that: The unloading mechanism comprises: Mounting frame; A lower hopper is rotatably connected to the mounting frame, the lower hopper is used to accommodate the bar to be tested, the lower hopper is at least partially a conical structure, the small end of the lower hopper is arranged toward the conveying platform, and the small end of the lower hopper is provided with a lowering port; A first driving assembly, connected to the lower hopper, the first driving assembly being capable of driving the lower hopper to swing relative to the mounting frame; A receiving plate, wherein a receiving groove is provided on one side of the receiving plate facing the feeding port; The second driving component is arranged on the mounting frame and connected to the receiving plate. The second driving component can drive the receiving plate to move in the direction close to the discharge port, and is used for the accommodating groove to be arranged opposite to the discharge port, so that the rod to be inspected falls into the accommodating groove through the discharge port.

12. A method for detecting curvature, characterized in that: The following steps are involved: Acquire multiple image information of the bar to be detected, wherein the multiple image information is image information of the end of the bar to be detected at different positions collected when the bar to be detected is driven by the conveying component to roll along the conveying platform; Performing image processing on multiple image information to obtain multiple characteristic image points; Connect multiple feature image points to obtain the motion trajectory of the bar to be detected; The bending test result of the bar to be tested is determined according to the maximum and minimum values ​​of the motion trajectory.

13. The curvature detection method according to claim 12, characterized in that: Acquiring multiple image information of the bar to be inspected includes the following steps: Obtain video information of the bar to be inspected within a preset time; The video frame of the video information is extracted to obtain a plurality of image information.

14. The curvature detection method according to claim 12, characterized in that: The image processing method for performing image processing on multiple image information includes at least one of a background difference method, an optical flow method, a frame difference method, and a machine learning target detection algorithm.

15. The curvature detection method according to claim 12, characterized in that: Performing image processing on multiple image information includes the following steps: Performing image processing on multiple image information and preset background image information to obtain multiple characteristic image points; The preset background image information is an image when there is no bar to be detected on the conveying platform.

16. The curvature detection method according to claim 15, characterized in that: Before obtaining multiple image information or by calling a database, obtain preset background image information.

17. The curvature detection method according to claim 15, characterized in that: Obtaining multiple feature image points includes the following steps: When the image information is a point image, the coordinates corresponding to the point image are determined as feature image points; When the feature image point is a regional image, the contour of the regional image is obtained, the minimum circumscribed rectangle is determined according to the contour of the regional image, and the coordinates corresponding to the center point of the minimum circumscribed rectangle are determined as the feature image point.

18. The curvature detection method according to claim 12, characterized in that: Determining the bend test result of the bar to be tested includes the following steps: According to the difference between the maximum value and the minimum value of the motion trajectory, the curvature of the bar to be tested is determined; Compare the curvature of the bar to be tested with the preset curvature; Determine whether the curvature of the bar to be tested is less than or equal to a preset curvature; If yes, determine that the bending test result of the bar to be tested is a qualified bar; If not, it is determined that the bending test result of the bar to be tested is a defective bar.

19. The curvature detection method according to claim 18, characterized in that: After determining the bending test result of the bar to be tested, the following steps are included: According to the curvature test results of the bars to be tested, the control picking mechanism will classify and collect the defective bars and qualified bars among the bars that have completed the test.

Images