Cigarette rod end face detection method, device and system, and control processing apparatus

Abstract

The present disclosure relates to a cigarette tobacco end face detection method, device and system, and a control processing device. The cigarette tobacco end face detection device comprises a rack, wherein the rack is provided with a detection station; a conveying and clamping mechanism configured to convey and clamp a first cigarette and a second cigarette to the detection station, so that a first tobacco end face of the first cigarette and a second tobacco end face of the second cigarette are arranged opposite to each other and in the same detection area; an end face positioning mechanism arranged at the detection station and configured to reflect the first tobacco end face and the second tobacco end face to the same camera field of view; a light measurement assembly configured to acquire tobacco end face images and end face height information; and a control processing device configured to process the tobacco end face images and the end face height information and determine a detection result. The present disclosure can realize single-camera double-end-face synchronous acquisition and has end face height measurement capability.

Description

Technical Field

[0001] This disclosure relates to the field of tobacco product quality testing technology, and in particular to a method, apparatus and system for detecting the end face of tobacco shreds in cigarettes, as well as control and processing equipment. Background Technology

[0002] After cigarette manufacturing is completed, it is usually necessary to inspect for defects in appearance, circumferential dimensions, draw resistance, combustion, and other items to ensure the quality of the finished product. In actual production, after the cigarette is cut to form the tobacco end face, technical problems such as missing tobacco strands, voids, loose tobacco strands, dents / protrusions, tears, and adhesion contamination may occur at the end face. Summary of the Invention

[0003] Research has revealed that end-face defects such as missing fibers, voids, loose fibers, depressions / protrusions, tears, and adhesion contamination can affect smoking consistency and appearance quality. Furthermore, the end face is located at the axial end of the cigarette, where the space is narrow and dust is abundant, making it difficult for related technologies to obtain a stable and clear end-face image from the side.

[0004] In view of at least one of the above technical problems, this disclosure provides a method, device and system for detecting the end face of cigarette tobacco, and a control and processing device that can realize simultaneous acquisition of dual end faces by a single camera and has the ability to measure the end face height.

[0005] According to one aspect of this disclosure, a device for detecting the end face of tobacco shreds in a cigarette is provided, comprising:

[0006] A frame, wherein a testing station is provided on the frame;

[0007] The conveying and clamping mechanism is configured to convey and clamp the first cigarette and the second cigarette to the testing station, such that the first tobacco end face of the first cigarette and the second tobacco end face of the second cigarette face each other and are in the same testing area.

[0008] The end face positioning mechanism set at the detection station is configured to reflect the first tobacco end face and the second tobacco end face to the same camera field of view.

[0009] The optical measurement component is configured to acquire images of the tobacco end face and end face height information; and

[0010] The control processing device is configured to process the tobacco end face image and end face height information and determine the detection result.

[0011] In some embodiments of this disclosure, the end face positioning mechanism is further configured to form an optical path reference for end face imaging and measurement.

[0012] In some embodiments of this disclosure, the end face positioning mechanism includes:

[0013] The V-shaped reflector is configured to reflect the first tobacco end face and the second tobacco end face into the same camera field of view.

[0014] In some embodiments of this disclosure, the V-shaped reflective component has a first reflective surface and a second reflective surface, which are respectively configured to reflect the images of the first tobacco end face and the second tobacco end face located on both sides of the detection station to the same camera field of view.

[0015] In some embodiments of this disclosure, the optical measurement component includes:

[0016] camera;

[0017] A surface light source is configured to provide uniform illumination to the end face of the tobacco shreds;

[0018] The first structured light emitting end and the second structured light emitting end are arranged corresponding to the first reflective surface and the second reflective surface, respectively, and are configured to project structured light onto the first tobacco end face and the second tobacco end face.

[0019] In some embodiments of this disclosure, the end face positioning mechanism includes:

[0020] A transparent protective component is disposed between the detection station and the optical measurement component and is configured to isolate tobacco dust.

[0021] In some embodiments of this disclosure, the transparent protective component is a glass protective plate or a transparent acrylic protective plate.

[0022] In some embodiments of this disclosure, the transparent protective component is provided with at least one of an anti-reflective coating and an anti-reflective coating on the side facing the camera.

[0023] In some embodiments of this disclosure, the transparent protective component and the frame together form a semi-enclosed end-face detection cavity.

[0024] In some embodiments of this disclosure, the end face detection cavity is provided with an air curtain assembly or a purge assembly, which is configured to purge the inner surface of the transparent protective assembly.

[0025] In some embodiments of this disclosure, the conveying clamping mechanism is provided on the frame along the production line direction.

[0026] In some embodiments of this disclosure, the conveying and clamping mechanism includes a first clamping unit and a second clamping unit positioned opposite each other, configured to clamp the first cigarette and the second cigarette and feed the end faces of the first tobacco shreds and the end faces of the second tobacco shreds toward each other into the detection station.

[0027] In some embodiments of this disclosure, the first clamping unit clamps the first cigarette and positions the first tobacco end face of the first cigarette toward the second clamping unit, and the second clamping unit clamps the second cigarette and positions the second tobacco end face of the second cigarette toward the first clamping unit.

[0028] In some embodiments of this disclosure, the structured light emitting end is a line laser emitter or a stripe projector.

[0029] In some embodiments of this disclosure, the control processing device is configured to perform triangulation based on the positional offset of structured light in the end face image to obtain at least one of the following: the depth of the concavity, the height of the protrusion, and the inclination of the end face relative to the reference surface of the cigarette paper port.

[0030] In some embodiments of this disclosure, the control processing device is configured to segment the end face image and determine at least one defect type among end face missing wire, end face void, end face loose wire, end face depression, end face protrusion, end face tear, and end face contamination; and output a rejection signal to the actuator to reject unqualified cigarettes.

[0031] In some embodiments of this disclosure, the first reflecting surface is a plane mirror.

[0032] In some embodiments of this disclosure, the second reflecting surface is a plane mirror.

[0033] In some embodiments of this disclosure, the included angle between the first reflective surface and the second reflective surface ranges from 80° to 100°.

[0034] In some embodiments of this disclosure, the angles between the first reflective surface and the second reflective surface and the axis of the cigarette branch range from 40° to 55°.

[0035] In some embodiments of this disclosure, the angle between the optical axis of the camera and the normal to the plane containing the transparent protective component ranges from 35° to 45°.

[0036] In some embodiments of this disclosure, the distance from the inner surface of the transparent protective component to the end face of the tobacco shreds ranges from 50mm to 70mm.

[0037] According to another aspect of this disclosure, a method for detecting the end face of tobacco shreds in a cigarette is provided, comprising:

[0038] The conveying and clamping mechanism of the cigarette tobacco end face detection device as described in any of the above embodiments is controlled to convey and clamp the first cigarette and the second cigarette to the detection station, so that the first tobacco end face of the first cigarette and the second tobacco end face of the second cigarette are facing each other and are in the same detection area.

[0039] The end face positioning mechanism of the cigarette tobacco end face detection device as described in any of the above embodiments is controlled to reflect the first tobacco end face and the second tobacco end face to the same camera field of view;

[0040] Controlling the photometric component of the tobacco end-face detection device as described in any of the above embodiments to acquire images of the tobacco end-face and end-face height information; and

[0041] The image of the tobacco end face and the end face height information are processed to determine the detection result.

[0042] In some embodiments of this disclosure, obtaining the tobacco end-face image and end-face height information includes:

[0043] Turn on the surface light source and two sets of structured light emitters, and control the camera to synchronously acquire end-face images including the first tobacco end face and the second tobacco end face via the V-shaped reflector.

[0044] In some embodiments of this disclosure, the activation of the surface light source and the two sets of structured light emitters, and the control of the camera to synchronously acquire end-face images including the first tobacco end-face and the second tobacco end-face via a V-shaped reflector assembly, include:

[0045] A controlled surface light source provides uniform diffuse illumination to obtain end-face texture images;

[0046] Controlling the first and second structured light emitting ends to project structured light stripes onto corresponding end faces and acquiring structured light images includes at least one of the following steps:

[0047] Using a time-division multiplexing method, the first structured light transmitter is lit up first to acquire the first structured light image, and then the second structured light transmitter is lit up to acquire the second structured light image.

[0048] By employing a multi-wavelength multiplexing method, the first structured light emitting end and the second structured light emitting end are controlled to use different center wavelengths, and the first structured light image and the second structured light image are separated through different color channels of the camera in a single exposure.

[0049] By employing spatial multiplexing, the stripes on both ends of the image are located in a first region of interest and a second region of interest that do not overlap, allowing the first structured light image and the second structured light image to be extracted in the same frame.

[0050] In some embodiments of this disclosure, processing the tobacco end-face image and end-face height information and determining the detection result includes:

[0051] The end face image is divided into regions and geometrically corrected to obtain a first end face image and a second end face image, respectively;

[0052] Based on the first end face image and the second end face image, the end face circumferential boundary, tobacco filling area and defect area are extracted, and the judgment result of at least one defect type among the end face missing tobacco, end face void, end face loose tobacco, end face depression, end face protrusion, end face tear and end face contamination is determined.

[0053] The position of structured light stripes is detected in the end face image and calibrated and converted to determine the end face concavity depth, protrusion height, end face flatness index, and end face tilt index.

[0054] The judgment result, the end face indentation depth, protrusion height, end face flatness index and end face tilt index are compared with a predetermined threshold or standard model to determine the test result. The test result includes at least one of the following: the tobacco end face is qualified or unqualified, defect type and defect level.

[0055] In some embodiments of this disclosure, the extraction of the end face circumferential boundary, tobacco filling area, and defect area based on the first end face image and the second end face image includes:

[0056] A candidate edge point set is obtained through edge detection;

[0057] The end face circumferential parameters are determined by using at least one of the Hough circle transform and least squares circle fitting.

[0058] Based on the circumferential parameters, an end face mask and a cigarette paper port ring mask are generated to locate and quantify end face tears, port abnormalities, and end face contamination.

[0059] In some embodiments of this disclosure, the determination result for at least one defect type among end face missing wire, end face void, end face loose wire, end face depression, end face bulge, end face tear, and end face contamination includes:

[0060] The tobacco filling area is divided within the end face mask to obtain the tobacco mask;

[0061] Calculate the tobacco filling ratio;

[0062] If the tobacco filling rate is lower than the predetermined filling rate threshold or if there are consecutive missing sectors in the radial partition, it is determined that the end face is missing tobacco.

[0063] In some embodiments of this disclosure, the determination result for at least one defect type among end face missing wire, end face void, end face loose wire, end face depression, end face bulge, end face tear, and end face contamination includes:

[0064] The tobacco filling area is divided within the end face mask to obtain the tobacco mask;

[0065] Connectivity analysis is performed on the complement region of the tobacco mask;

[0066] In the case of a connected region with a void area greater than the area threshold and a shape that satisfies the circularity or aspect ratio constraints, the end face void is determined, and the void area, location, and number are output.

[0067] In some embodiments of this disclosure, the method for detecting the end face of the tobacco in the cigarette stick further includes:

[0068] The alignment delay is calculated based on the conveying speed, the distance from the detection point to the rejection point, and the system cycle.

[0069] Write the detection results into the queue;

[0070] Based on the alignment delay, a rejection signal is output at the corresponding time.

[0071] In some embodiments of this disclosure, the step of detecting the position of structured light stripes in the end face image and performing calibration conversion to determine the end face concavity depth, protrusion height, end face flatness index, and end face tilt index includes:

[0072] Based on the camera intrinsic parameter matrix and distortion parameters, the pixel coordinates of the center point of the structured light stripe are back-projected into the ray direction in the camera coordinate system;

[0073] The intersection of the ray direction and the plane equation of the structured light plane yields a three-dimensional point;

[0074] Calculate the height using the plane equation relative to the reference plane at the cigarette paper end;

[0075] The end face concavity depth, protrusion height, end face flatness index, and end face tilt index are calculated based on the height.

[0076] According to another aspect of this disclosure, a control processing apparatus is provided, comprising:

[0077] The control module is configured to control the conveying and clamping mechanism of the cigarette tobacco end-face detection device as described in any of the above embodiments to convey and clamp the first cigarette and the second cigarette to the detection station, so that the first tobacco end face of the first cigarette and the second tobacco end face of the second cigarette face face each other and are in the same detection area; control the end-face positioning mechanism of the cigarette tobacco end-face detection device as described in any of the above embodiments to reflect the first tobacco end face and the second tobacco end face to the same camera field of view; control the photometric component of the cigarette tobacco end-face detection device as described in any of the above embodiments to acquire tobacco end-face images and end-face height information; and

[0078] The processing module is configured to process the tobacco end face image and end face height information and determine the detection result.

[0079] According to another aspect of this disclosure, a control processing apparatus is provided, comprising:

[0080] The memory is configured to store instructions; and

[0081] A processor coupled to the memory is configured to execute the cigarette tobacco end face detection method as described in any of the above embodiments, based on instructions stored in the memory.

[0082] According to another aspect of this disclosure, a cigarette tobacco end face detection system is provided, including a cigarette tobacco end face detection device as described in any of the above embodiments and a host computer, wherein the host computer is configured to receive the detection results output by the control processing device and drive a sorting mechanism or a rejection mechanism to reject unqualified cigarettes.

[0083] According to another aspect of this disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions that, when executed by a processor, implement the cigarette tobacco end face detection method as described in any of the above embodiments.

[0084] According to another aspect of this disclosure, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, it implements the cigarette tobacco end face detection method as described in any of the above embodiments.

[0085] This invention enables simultaneous acquisition of data from two end faces using a single camera and provides the capability to measure the height of the end faces. Attached Figure Description

[0086] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this disclosure. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0087] Figure 1 These are schematic diagrams of some embodiments of the cigarette tobacco end face detection device disclosed herein.

[0088] Figure 2 This is a schematic diagram of the end face positioning mechanism and V-shaped reflective assembly in some embodiments of this disclosure.

[0089] Figure 3 This is a schematic diagram showing the installation relationship of the optical measurement components in some embodiments of this disclosure.

[0090] Figure 4 These are schematic diagrams of some embodiments of the method for detecting the end face of tobacco shreds in cigarettes disclosed herein.

[0091] Figure 5 This is a schematic diagram of some embodiments of the control and processing device disclosed herein.

[0092] Figure 6 This is a schematic diagram of the structure of some other embodiments of the control and processing device disclosed herein. Detailed Implementation

[0093] The technical solutions of the embodiments of this disclosure will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this disclosure, and not all embodiments. The following description of at least one exemplary embodiment is merely illustrative and is in no way intended to limit this disclosure or its application or use. All other embodiments obtained by those skilled in the art based on the embodiments of this disclosure without creative effort are within the scope of protection of this disclosure.

[0094] Unless otherwise specifically stated, the relative arrangement, numerical expressions, and values ​​of the components and steps set forth in these embodiments do not limit the scope of this disclosure.

[0095] At the same time, it should be understood that, for ease of description, the dimensions of the various parts shown in the accompanying drawings are not drawn according to actual scale.

[0096] Techniques, methods, and equipment known to those skilled in the art may not be discussed in detail, but where appropriate, such techniques, methods, and equipment should be considered part of the specification.

[0097] In all examples shown and discussed herein, any specific values ​​should be interpreted as merely exemplary and not as limitations. Therefore, other examples of exemplary embodiments may have different values.

[0098] It should be noted that similar labels and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be discussed further in subsequent figures.

[0099] The study also found that related technologies often employ dual or multiple cameras to simultaneously detect the ends of cut tobacco shreds, resulting in high equipment costs, complex calibration and maintenance, and the easy contamination of the lens and optical window by tobacco shred debris near the ends, further reducing detection stability.

[0100] In view of at least one of the above technical problems, this disclosure provides a method, apparatus and system for detecting the end face of tobacco in cigarettes, and a control and processing device. The present disclosure will be described below through specific embodiments.

[0101] Figure 1 These are schematic diagrams of some embodiments of the cigarette tobacco end-face detection device disclosed herein. Figure 1As shown, the tobacco end face detection device disclosed herein may include a frame 1, a conveying and clamping mechanism 2, an end face positioning mechanism, an optical measurement component 6, and a control and processing device 7.

[0102] The frame 1 is provided with a testing station 3.

[0103] The conveying and clamping mechanism 2 is configured to convey and clamp the first cigarette and the second cigarette to the testing station, such that the first tobacco end face of the first cigarette and the second tobacco end face of the second cigarette face each other and are in the same testing area.

[0104] In some embodiments of this disclosure, such as Figure 1 As shown, the conveying and clamping mechanism 2 is arranged on the frame along the production line direction. The conveying and clamping mechanism is used to convey and clamp the cigarette sticks, so that the tobacco end face of the cigarette stick is located at the inspection station.

[0105] In some embodiments of this disclosure, such as Figure 1 As shown, the conveying and clamping mechanism includes a first clamping unit 21 and a second clamping unit 22 facing each other, configured to clamp the first cigarette and the second cigarette and send the first tobacco end face 91 and the second tobacco end face 92 toward each other into the detection station 3.

[0106] In some embodiments of this disclosure, the first clamping unit 21 clamps the first cigarette and positions the first tobacco end face 91 of the first cigarette toward the second clamping unit 22, and the second clamping unit 22 clamps the second cigarette and positions the second tobacco end face 92 of the second cigarette toward the first clamping unit 21, thereby placing the first tobacco end face and the second tobacco end face in the same detection area.

[0107] The end face positioning mechanism located at the detection station 3 is configured to reflect the first tobacco end face 91 and the second tobacco end face 92 into the field of view of the same camera 61.

[0108] In some embodiments of this disclosure, the end face positioning mechanism may also be configured to form an optical path reference for end face imaging and measurement.

[0109] Figure 2 This is a schematic diagram of the end-face positioning mechanism and V-shaped reflective assembly in some embodiments of this disclosure. For example... Figure 2 and Figure 1 As shown, the end face positioning mechanism may include a V-shaped reflective component 5.

[0110] The V-shaped reflector 5 is configured to reflect the first tobacco end face 91 and the second tobacco end face 92 into the same camera field of view.

[0111] In some embodiments of this disclosure, such as Figure 2 As shown, the V-shaped reflective component 5 has a first reflective surface 51 and a second reflective surface 52, which are respectively configured to reflect the images of the first tobacco end face 91 and the second tobacco end face 92 located on both sides of the detection station 3 to the same camera field of view.

[0112] In some embodiments of this disclosure, such as Figure 2 As shown, a V-shaped structure is formed between the first reflective surface 51 and the second reflective surface 52, which is used to guide the reflected images of the first tobacco end face and the second tobacco end face to the same camera field of view and to image them on the same image plane.

[0113] In some embodiments of this disclosure, the first reflecting surface may be a plane mirror.

[0114] In some embodiments of this disclosure, the second reflecting surface may be a plane mirror.

[0115] In some embodiments of this disclosure, the included angle between the first reflective surface and the second reflective surface can range from 80° to 100°, preferably 90°.

[0116] In some embodiments of this disclosure, the angles between the first reflective surface and the second reflective surface and the axis of the cigarette can range from 40° to 55°.

[0117] The embodiments disclosed above can reflect both end faces to the same camera field of view through a V-shaped reflector component, thereby achieving synchronous detection of both end faces with a single camera, reducing hardware costs and calibration difficulty.

[0118] In some embodiments of this disclosure, as shown in 1, the end face positioning mechanism may further include a transparent protective component 4.

[0119] In some embodiments of this disclosure, as shown in 1, the end-face positioning mechanism may include a transparent protective component 4 and a V-shaped reflective component 5. The V-shaped reflective component 5 is disposed on the side of the transparent protective component 4 near the cigarette. The transparent protective component 4 is used to isolate tobacco dust, and the V-shaped reflective component 5 is used to reflect both end faces to the same camera field of view, realizing single-camera dual-end-face imaging.

[0120] The transparent protective component 4 is disposed between the detection station 3 and the optical measurement component 6 and is configured to isolate tobacco dust.

[0121] In some embodiments of this disclosure, the transparent protective component 4 may be a glass protective plate or a transparent acrylic protective plate.

[0122] In some embodiments of this disclosure, the transparent protective component 4 is preferably a glass protective plate.

[0123] In some embodiments of this disclosure, the transparent protective component 4 is provided with at least one of an anti-reflective coating and an anti-reflective coating on the side facing the camera 61.

[0124] In some embodiments of this disclosure, the transparent protective component 4 can form a semi-enclosed end-face detection cavity together with the frame 1.

[0125] In some embodiments of this disclosure, the distance from the inner surface of the transparent protective component 4 to the end face of the tobacco is preferably in the range of 50mm-70mm, so as to balance imaging resolution and installation space.

[0126] In some embodiments of this disclosure, the end face detection cavity is provided with an air curtain assembly or a blowing assembly, which is configured to blow the inner surface of the transparent protective assembly to reduce the impact of tobacco dust adhesion on imaging.

[0127] The embodiments disclosed above can isolate tobacco dust through transparent protective components and air curtain components or blowing structures, thereby reducing lens contamination and improving long-term operational stability.

[0128] In some embodiments of this disclosure, such as Figure 1 As shown, a photometric component 6 is installed on the frame 1, which is configured to acquire images of the tobacco end face and end face height information.

[0129] Figure 3 This is a schematic diagram showing the installation relationship of the optical measurement components in some embodiments of this disclosure. For example... Figure 3 and Figure 1 As shown, the optical measurement component 6 may include a camera 61, a surface light source 62, a first structured light emitting end 63, and a second structured light emitting end 64.

[0130] In some embodiments of this disclosure, the camera 61 is preferably an industrial camera and is equipped with a fixed-focus lens or a telecentric lens.

[0131] In some embodiments of this disclosure, the angle between the optical axis of the camera 61 and the normal to the plane containing the transparent protective component 4 ranges from 35° to 45°.

[0132] In some embodiments of this disclosure, the angle between the optical axis of the camera 61 and the normal to the plane containing the transparent protective component 4 is preferably about 40°, so as to reduce specular reflection interference while satisfying the field of view coverage.

[0133] A surface light source 62 is configured to provide uniform illumination to the end face of the tobacco shreds. The surface light source can be configured to provide uniform illumination for end face texture and defect segmentation.

[0134] In some embodiments of this disclosure, the light source 62 may be a diffuse surface light source or a ring surface light source, arranged around or coaxially with the camera 61, to provide uniform illumination to the end face.

[0135] The surface light source of the above embodiments of this disclosure can provide uniform illumination, which facilitates stable segmentation of defects such as missing wires, voids, loose wires and edge breaks on the end face.

[0136] The first structured light emitting end 63 and the second structured light emitting end 64 are arranged corresponding to the first reflective surface 51 and the second reflective surface 52, respectively, and are configured to project structured light onto the first tobacco end face 91 and the second tobacco end face 92.

[0137] In some embodiments of this disclosure, the structured light emitting end can be a line laser emitter or a stripe projector to form structured light stripes on the end face.

[0138] The photometric component of the above embodiments of this disclosure includes a camera, a surface light source, and two sets of structured light emitters. The two sets of structured light emitters in this embodiment project stripes or lines of laser light onto two end faces respectively, for acquiring height information such as concavities and convexities of the end faces.

[0139] The two sets of structured light emitting ends in the above embodiments of this disclosure can provide end face height information and can simultaneously complete the measurement of end face concavity, convexity and flatness, thereby improving the detection dimensions and accuracy.

[0140] In some embodiments of this disclosure, such as Figure 1 As shown, a control and processing device 7 is installed inside the frame 1, which is configured to process the end face image and end face height information of the tobacco shreds, determine and output the detection results.

[0141] In some embodiments of this disclosure, the control processing module 7 is electrically or communicatively connected to the camera 61, the surface light source 62, and the structured light emitters 63 and 64. The control processing module 7 can be configured to acquire and process end-face images and structured light information.

[0142] In some embodiments of this disclosure, such as Figure 1 As shown, the control processing module 7 can also be connected to the rejection actuator 8 to output a rejection signal for non-conforming products.

[0143] In some embodiments of this disclosure, the control processing device 7 may be configured to perform triangulation based on the positional offset of structured light in the end face image to obtain at least one of the following: the depth of the concavity of the tobacco end face relative to the reference surface of the cigarette paper port, the height of the protrusion, and the inclination of the end face.

[0144] In some embodiments of this disclosure, the control processing device 7 may be configured to segment the end face image and determine at least one defect type among end face missing wire, end face void, end face loose wire, end face depression, end face protrusion, end face tear, and end face contamination; and output a rejection signal to the actuator to reject unqualified cigarettes.

[0145] The control and processing device of the above embodiments of this disclosure can perform region division, distortion correction, end face boundary positioning and defect segmentation on the acquired end face image, and output the end face flatness and defect type in combination with structured light triangulation measurement.

[0146] The above-described embodiments of this disclosure have a compact structure, making them easy to place at cutting or sorting stations, thereby enabling high-speed online detection and rejection of defective products.

[0147] The cigarette tobacco end-face inspection device of the above embodiments of this disclosure includes a frame, a conveying and clamping mechanism, an end-face positioning mechanism, an optical measurement component, and a control and processing module. The conveying and clamping mechanism is used to position the tobacco end-face of the cigarette to be tested to the inspection station; the end-face positioning mechanism includes a V-shaped reflective component disposed between the two end faces and a transparent protective component disposed adjacent to the V-shaped reflective component; the optical measurement component includes a camera, a surface light source, and two sets of structured light emitting ends. The imaging optical path of the camera reflects the images of the tobacco end faces on the left and right sides to the same camera imaging surface through the V-shaped reflective component, so as to realize the synchronous acquisition of the two end faces by a single camera; the structured light emitting ends project stripe / line lasers onto the tobacco end faces to obtain end-face height and concavity information, and the surface light source provides uniform illumination for end-face defect segmentation; the control and processing module outputs defect judgment results such as missing tobacco strands, voids, loose tobacco strands, concavity / protrusion, and edge damage based on the acquired images. The embodiments disclosed above, through V-shaped reflection and a transparent protective structure that is dustproof and dirtproof, reduce the number of cameras and maintenance frequency while ensuring the accuracy and robustness of end-face detection, making them suitable for high-speed online detection in cigarette production lines.

[0148] Figure 4 These are schematic diagrams illustrating some embodiments of the cigarette tobacco end-face detection method of this disclosure. Preferably, this embodiment can be executed by the cigarette tobacco end-face detection device, the cigarette tobacco end-face detection system, or the control and processing equipment of this disclosure. Figure 4 As shown, Figure 4 The method of the embodiment may include at least one of steps 100 to 400.

[0149] In step 100, the conveying and clamping mechanism of the cigarette tobacco end face detection device as described in any of the above embodiments is controlled to convey and clamp the first cigarette and the second cigarette to the detection station, so that the first tobacco end face of the first cigarette and the second tobacco end face of the second cigarette are facing each other and are in the same detection area.

[0150] In step 200, the end face positioning mechanism of the tobacco end face detection device as described in any of the above embodiments is controlled to reflect the first tobacco end face and the second tobacco end face to the same camera field of view.

[0151] In step 300, the photometric component of the tobacco end face detection device as described in any of the above embodiments is controlled to acquire the tobacco end face image and end face height information.

[0152] In some embodiments of this disclosure, steps 100 to 300 may include: clamping positioning and triggering.

[0153] In some embodiments of this disclosure, steps 100 to 300 (the clamping, positioning, and triggering steps) may include: the conveying clamping mechanism 2 clamps the first cigarette and the second cigarette by the first clamping unit 21 and the second clamping unit 22 respectively and sends them into the detection station 3, so that the end face 91 of the first tobacco shred and the end face 92 of the second tobacco shred are arranged facing each other and simultaneously within the same field of view of the camera 61; the control processing device 7 receives the arrival signal of the encoder / photoelectric sensor or calculates the triggering time according to the conveying speed, and triggers the camera 61 to complete the exposure.

[0154] To ensure the stability of high-speed online detection, the embodiments disclosed above preferably employ hardware triggering, and write the trigger timestamp into the cache for subsequent alignment with the removal execution mechanism 8.

[0155] In some embodiments of this disclosure, step 300 may include: turning on the surface light source and two sets of structured light emitters, and controlling the camera to synchronously acquire end-face images including the first tobacco end-face and the second tobacco end-face via a V-shaped reflector.

[0156] In some embodiments of this disclosure, step 300 may include: synchronous acquisition.

[0157] In some embodiments of this disclosure, step 300 may include at least one of steps 310 to 320.

[0158] In step 310, the control surface light source provides uniform diffuse illumination to obtain an end-face texture image.

[0159] In some embodiments of this disclosure, step 310 may include: during the triggered exposure, controlling the processing device 7 to control the surface light source 62 to provide uniform diffuse illumination to obtain the end face texture image I_tex.

[0160] In step 320, the first structured light emitting end 63 and the second structured light emitting end 64 are controlled to project structured light stripes onto the corresponding end faces and acquire structured light images I_sl.

[0161] In some embodiments of this disclosure, the surface light source image I_tex in step 310 and the structured light image I_sl in step 320 are acquired in a time-division manner. When acquiring I_sl, the surface light source 62 is turned off and only the structured light emitting ends 63 and 64 are lit to improve the stripe signal-to-noise ratio and reduce the interference of end face texture on stripe extraction.

[0162] In some embodiments of this disclosure, the first structured light emitter 63 and the second structured light emitter 64 employ time-division multiplexing control or multi-wavelength multiplexing control with different center wavelengths, enabling the camera 61 to acquire structured light stripes of the first tobacco end face 91 and the second tobacco end face 92 in the same frame or adjacent frames, and obtain I_sl1 and I_sl2 through color channel separation or ROI partitioning separation.

[0163] In some embodiments of this disclosure, step 320 may include at least one of steps 321 to 323.

[0164] In step 321, time-division multiplexing is used to first illuminate the first structured light transmitter 63 to acquire the first structured light image I_sl1, and then illuminate the second structured light transmitter 64 to acquire the second structured light image I_sl2.

[0165] In step 322, a multi-wavelength multiplexing method is used to control the first structured light emitter 63 and the second structured light emitter 64 to use different center wavelengths (e.g., red / green), and the first structured light image and the second structured light image are separated through different color channels of the camera 61 in one exposure.

[0166] In step 323, spatial multiplexing is used so that the stripes on both ends are located in the first region of interest (ROI1) and the second region of interest (ROI2) in the image, which do not overlap. The first structured light image and the second structured light image can be extracted in the same frame.

[0167] The above embodiments of this disclosure can all be used in conjunction with the acquisition of the surface light source 62 to achieve joint detection of "two-dimensional texture + three-dimensional structured light".

[0168] In step 400, the image of the tobacco end face and the end face height information are processed and the detection result is determined.

[0169] In some embodiments of this disclosure, step 400 may include at least one of steps 410 to 440.

[0170] In step 410, the end face image is divided into regions and geometrically corrected to obtain a first end face image and a second end face image, respectively.

[0171] In some embodiments of this disclosure, step 410 may include at least one of steps 411 to 413.

[0172] In step 411, the end face region is divided.

[0173] In some embodiments of this disclosure, step 411 may include: based on the imaging positions of the two reflecting surfaces (first reflecting surface 51 and second reflecting surface 52) of the V-shaped reflecting component 5 in the camera's field of view, pre-calibrating the partition boundaries of the two end faces, or online detecting the position of the edge / boundary line of the V-shaped reflecting component 5, dividing the end face texture image I_tex into the first end face ROI1 and the second end face ROI2, and obtaining I1_raw and I2_raw, where ROI is the region of interest; similarly, dividing the structured light image I_sl into I_sl1_raw and I_sl2_raw.

[0174] In step 412, camera calibration and distortion correction are performed.

[0175] In some embodiments of this disclosure, step 412 may include: performing intrinsic parameter calibration on the camera 61 to obtain the camera intrinsic parameter matrix K and distortion parameters D (radial distortion k1, k2, k3 and tangential distortion p1, p2), and calculating the normalized coordinates x_n corresponding to the pixel coordinates (u, v) as follows: x_n = undistort( K^{-1}·[u, v, 1]^T, D ), where undistort(·) represents performing inverse mapping on the radial / tangential distortion to obtain the corrected normalized coordinates.

[0176] In step 413, homography transformation and perspective correction are performed.

[0177] In some embodiments of this disclosure, step 413 may include: since the end face exhibits perspective distortion after being folded 5 times by the V-shaped reflector, the homography matrices H1 and H2 of the two end face views are calibrated for ROI1 and ROI2 respectively, and I1_raw and I2_raw are mapped to the end face frontal coordinate system to obtain the first end face image I1 and the second end face image I2: I1(x,y)=I1_raw( H1·[x,y,1]^T ), I2(x,y)=I2_raw( H2·[x,y,1]^T ); similarly, I_sl1 and I_sl2 are obtained from the structured light image. The H1 / H2 can be obtained by acquiring the virtual views of the two end faces at the detection station 3 on the standard calibration plate and calculating the four-point perspective transformation, or by solving the multi-point least squares problem.

[0178] In some embodiments of this disclosure, in steps 412 and 413, distortion correction and homography transformation are performed on the two end face regions according to formulas (1) and (2), respectively:

[0179] I1= warp(undistort(I1_raw,K,D),H1) (1)

[0180] I2 = warp(undistort(I2_raw,K,D),H2) (2)

[0181] In formulas (1) and (2), H1 and H2 are obtained by four-point perspective transformation of the standard calibration plate or by multi-point least squares solution.

[0182] In step 420, the end face circumferential boundary, tobacco filling area and defect area are extracted based on the first end face image and the second end face image. The determination result of at least one defect type among the following is determined and output: end face missing tobacco, end face void, end face loose tobacco, end face depression, end face protrusion, end face tear and end face contamination.

[0183] In some embodiments of this disclosure, step 420, which involves extracting the end face circumferential boundary, tobacco filling area, and defect area based on the first end face image and the second end face image, may include at least one of steps 421 to 423.

[0184] In step 421, preprocessing is performed.

[0185] In some embodiments of this disclosure, step 421 may include: graying and denoising I1 and I2 (median filtering / bilateral filtering), and performing brightness normalization (such as CLAHE or background fitting subtraction) to suppress the slow-changing illumination caused by the non-uniformity of the surface light source 62 and residual contamination on the surface of the transparent protective component 4.

[0186] In step 422, the end face circumference and the cigarette paper port reference are extracted.

[0187] In some embodiments of this disclosure, step 422 may include: obtaining a set of candidate edge points E through edge detection; determining the end face circumferential parameters (c,r) using at least one of Hough circle transform and least squares circle fitting; and generating an end face mask M_end and a cigarette paper port ring mask M_ring based on the circumferential parameters to locate and quantify end face tears, port anomalies, and end face contamination.

[0188] In some embodiments of this disclosure, step 422 may include: performing edge detection (Canny / Scharr) on the preprocessed end face image to obtain a candidate edge point set E; obtaining the end face circumference parameters (c,r) using Hough circle transform or least squares circle fitting, where c is the center and r is the radius; generating an end face mask M_end with (c,r) and constructing a ring mask M_ring (e.g., a ring with r-Δr≤ρ≤r+Δr) near the cigarette paper port for subsequent determination of tears or port anomalies.

[0189] In some embodiments of this disclosure, step 422 may include: obtaining the end face circumference parameters (c,r) by edge detection and circle fitting, obtaining the tobacco mask M_tobacco by adaptive threshold segmentation and morphological operation within the end face mask M_end, and further outputting the defect locations and areas of end face missing shreds, end face voids, end face contamination and end face breaks through connected component analysis.

[0190] In step 423, the tobacco filling area is segmented.

[0191] In some embodiments of this disclosure, step 423 may include: within M_end, segmenting the tobacco region and background using a global threshold (Otsu's maximum inter-class variance method) or an adaptive threshold (local mean / Gaussian) to obtain a binary mask M_tobacco; further, morphological opening and closing operations, hole filling, and connected component filtering may be used to remove noise. For scenes with large variations in lighting / texture, a lightweight semantic segmentation network (such as U-Net or a cropped version of DeepLab) may be used to segment the end-face texture to obtain M_tobacco, thereby improving robustness.

[0192] In some embodiments of this disclosure, step 420, which involves determining the result of at least one defect type among end face missing wire, end face void, end face loose wire, end face depression, end face protrusion, end face tear, and end face contamination, may include at least one of steps 424 to 427.

[0193] In step 424, the end face is determined to be missing a wire.

[0194] In some embodiments of this disclosure, step 424 may include: dividing the tobacco filling area within the end face mask M_end to obtain a tobacco mask; calculating the tobacco filling rate η; and determining that the end face is missing tobacco when the tobacco filling rate η is lower than a predetermined filling rate threshold η_min or when there are consecutive missing sectors in the radial partition.

[0195] In some embodiments of this disclosure, step 424 may include: calculating the tobacco filling rate η = area(M_tobacco) / area(M_end) within M_end; and determining end face missing tobacco when η is lower than the threshold η_min or when there are consecutive missing sectors in the radial partition.

[0196] In step 425, end face void determination is performed.

[0197] In some embodiments of this disclosure, step 425 may include: segmenting the tobacco filling area within the end face mask to obtain a tobacco mask M_tobacco; performing connected component analysis on the complement region (M_end\M_tobacco) of the tobacco mask M_tobacco; and determining end face voids in the case where the area of ​​a void connected component is greater than the area threshold A_void and the shape satisfies the circularity or aspect ratio constraints, and outputting the void area, location, and number.

[0198] In some embodiments of this disclosure, steps 424 and 425 may include: dividing the tobacco filling area within the end-face mask M_end to obtain the tobacco mask M_tobacco, and calculating the filling rate η=area(M_tobacco) / area(M_end) and characteristic quantities such as the area of ​​the void connected domain A_void, the number N_void, and the maximum equivalent diameter D_voidmax. When η is lower than a first threshold or A_void, N_void, and D_voidmax are higher than the corresponding thresholds, it is determined to be an end-face missing tobacco or an end-face void defect.

[0199] In step 426, the loose fibers or fuzzy fibers are determined.

[0200] In some embodiments of this disclosure, step 426 may include: constructing a boundary neighborhood M_edge (e.g., r-Δe≤ρ≤r) near the circumferential boundary of the end face, and statistically analyzing high-frequency texture indices (variance, energy, entropy) or the mean gradient magnitude Ḡ within M_edge; when the texture roughness exceeds a threshold and there is a long and thin connected region (which can be obtained by skeletonization), it is judged as loose fibers / fuzz on the end face, and the length L_fiber and the number N_fiber of loose fibers are output.

[0201] In some embodiments of this disclosure, step 426 may include: calculating local texture features (including but not limited to grayscale variance, gradient energy, LBP histogram or entropy value) within the end face mask M_end and extracting the filament skeleton length L_fiber or the number of filaments N_fiber near the cigarette paper port ring mask M_ring to determine the end face loose filament defect.

[0202] In step 427, contamination is determined.

[0203] In some embodiments of this disclosure, step 427 may include: calculating color / grayscale outliers in the end face region, for example: calculating the color difference ΔE or grayscale deviation ΔI with the standard template in HSV or Lab color space, where HSV stands for Hue (hue, saturation, color purity) Value (brightness), and Lab color space uses L to represent brightness, a to represent red-green axis chromaticity, and b to represent yellow-blue axis chromaticity; when ΔE or ΔI exceeds the threshold and the area of ​​the connected region exceeds the threshold A_stain, it is judged as end face contamination.

[0204] In some embodiments of this disclosure, step 427 may include: the end face detection cavity is provided with an air curtain assembly or a purge assembly, the air curtain assembly or purge assembly forms an air curtain flowing along the inner surface of the transparent protective assembly 4, and the control processing device 7 monitors the degree of contamination based on the change in the surface reflection intensity or imaging contrast of the transparent protective assembly 4, and outputs a cleaning / maintenance prompt or automatically adjusts the exposure parameters when the threshold is exceeded.

[0205] The defect types in steps 424 to 427 of the above embodiments of this disclosure can be output individually or combined according to defect level.

[0206] In step 430, the position of structured light stripes is detected in the end face image and calibrated and converted to determine and output the end face concavity depth, protrusion height, end face flatness index and end face tilt index.

[0207] In some embodiments of this disclosure, step 430 may include: obtaining a three-dimensional point X by intersecting the structured light stripe center point (u,v) with the ray-light plane, and obtaining the end face plane normal vector n_z by performing plane fitting on the three-dimensional point, the end face tilt θ=arccos((n0·n_z) / (|n0||n_z|)), and the end face flatness index includes Flat_PV=max(h)-min(h) and Flat_RMS=√mean((h-mean(h))^2).

[0208] In some embodiments of this disclosure, step 430 may include at least one of steps 431 to 436.

[0209] In step 431, stripe extraction and subpixel localization are performed.

[0210] In some embodiments of this disclosure, step 431 may include: performing bandpass filtering or directional filtering on the stripe regions in the structured light images I_sl1 and I_sl2 to enhance the stripes; and obtaining the sub-pixel coordinates (u,v) of the stripe centerline using Steger centerline extraction, gray-scale centroid method, or Gaussian fitting. A set of stripe center points P={(u_i,v_i)} can be obtained for each row or column.

[0211] In some embodiments of this disclosure, the extraction of the center point of the structured light stripe is performed using the Steger line center extraction algorithm, the gray-scale centroid method, or a sub-pixel localization algorithm that fits the stripe intensity profile with Gaussian / parabolic curves. The extracted stripe point set is then fitted with RANSAC straight lines / curves to remove outliers.

[0212] In step 432, pixel to space ray.

[0213] In some embodiments of this disclosure, step 432 may include: back-projecting the pixel coordinates of the center point of the structured light stripe into the ray direction in the camera coordinate system based on the camera intrinsic parameter matrix and distortion parameters.

[0214] In some embodiments of this disclosure, step 432 may include: using the camera intrinsic parameter K and the distortion parameter D, mapping the fringe center point (u,v) to the normalized ray direction r in the camera coordinate system: r = normalize( undistort(K^{-1}[u,v,1]^T, D) ), where r is the camera ray direction.

[0215] In step 433, the light plane model is used for triangulation.

[0216] In some embodiments of this disclosure, step 433 may include: finding the intersection point of the ray direction with the plane equation of the structured light plane π to obtain a three-dimensional point X; and calculating the height relative to the plane equation of the cigarette paper port reference plane.

[0217] In some embodiments of this disclosure, step 433 may include: establishing the optical plane equation π for the first structured light emitting end 63 and the second structured light emitting end 64 respectively: n^TX + d = 0, where, as Figure 2As shown, π is the structured light plane, n is the light plane normal vector of π, and d is the plane offset of π. The parameters of the structured light plane π are (n,d). When the origin of the camera coordinate system is located at the camera optical center, the spatial point (three-dimensional point) X corresponding to the fringe point can be obtained by finding the intersection of the ray and the light plane, as shown in formula (3). In formula (3), n and d are obtained by system calibration (for example, placing a standard plane plate at different known heights to collect fringe and fitting the light plane with least squares). If a displacement model is used, the offset of the fringe in the image Δu = u - u0 (u0 is the fringe position at the reference plane) can be mapped to the height h: h = f(Δu), where h is the height relative to the reference plane, and f(·) can be a polynomial, piecewise linear or lookup table (LUT) model. This model is also obtained by calibration, and Δu is the offset of the fringe relative to the reference position.

[0218] X = (-d / (n^T r)) · r (3)

[0219] In some embodiments of this disclosure, the parameters (n,d) of the structured light plane π are obtained by collecting structured light stripes at different known heights using a standard plane calibration plate and performing least-squares fitting.

[0220] In step 434, the reference surface is determined and the concavity / convexity is calculated.

[0221] In some embodiments of this disclosure, step 434 may include: defining the cigarette paper port reference plane as the cigarette paper port reference plane O: n0^TX + d0 = 0. For example... Figure 2 As shown, the parameters of the reference plane π0 at the cigarette paper port are (n0, d0); π0 can be obtained by plane fitting of the three-dimensional points of structured light sampled near the circumference of the end face (cigarette paper port), or by mechanical reference + calibration curing. For each three-dimensional point X_i, calculate its signed distance to π0 (along the n0 direction) h_i = (n0^T X_i + d0) / ||n0||, then the end face concavity depth H_concave can be determined according to formula (4), and the end face protrusion height H_protrude can be determined according to formula (5).

[0222] H_concave = max(0, -min_i(h_i)) (4)

[0223] H_protrude = max(0, max_i(h_i)) (5)

[0224] In some embodiments of this disclosure, the cigarette paper port reference plane π0 is obtained by plane fitting of three-dimensional points on the end face circumferential boundary or the cigarette paper port ring area, or by pre-calibration and temperature drift compensation in online detection.

[0225] In step 435, flatness and tilt are determined.

[0226] In some embodiments of this disclosure, steps 434 and 435 may include: calculating the end face recess depth, protrusion height, end face flatness index, and end face tilt index based on the height.

[0227] In some embodiments of this disclosure, step 435 may include: using {h_i} as the end face height field, the end face flatness index may be PV and RMS indices: Flat_PV = max_i(h_i) - min_i(h_i), Flat_RMS = sqrt(mean_i( (h_i - mean(h))^2 ) ); simultaneously, a plane fitting can be performed on the three-dimensional point set {X_i} to obtain the end face fitting plane πz, whose normal vector is n_z; the end face inclination θ is defined as the angle between the reference normals n0 and n_z: θ = arccos( (n0·n_z) / (|n0||n_z|) ). The above H_concave, H_protrude, Flat_PV, Flat_RMS, and θ may be output for the first end face and the second end face, respectively.

[0228] In some embodiments of this disclosure, steps 432 to 435 may include: projecting the pixel coordinates (u,v) of the center point of the structured light stripe into a ray direction r in the camera coordinate system using the camera intrinsic parameter matrix K and the distortion parameter D, and intersecting the ray direction r with the plane equation n^TX + d = 0 of the structured light plane π to obtain a three-dimensional point X, and then calculating the height h relative to the plane equation n0^TX + d0 = 0 of the cigarette paper port reference plane π0, where h = (n0^TX + d0) / ||n0||, and calculating the end face concave depth H_concave, protrusion height H_protrude, and end face flatness indices Flat_PV and Flat_RMS from the height h.

[0229] In step 436, the dual end faces or dual light sources are separated.

[0230] In some embodiments of this disclosure, step 436 may include: when time-division multiplexing is used to acquire I_sl1 and I_sl2, they can be respectively mapped to ROI1 / ROI2 according to the timestamp; when multi-wavelength multiplexing is used, red / green stripes can be separated in the RGB channel; when there is stripe crossing or scattering interference, stripe association can be performed by structured light coding (phase / Gray code) or direction consistency constraints.

[0231] In step 440, the output is determined and discarded.

[0232] In some embodiments of this disclosure, step 440 may include at least one of steps 441 to 442.

[0233] In step 441, the detection results are determined and output.

[0234] In some embodiments of this disclosure, step 441 may include: comparing the judgment result (defect type or feature quantity) obtained in step 420 and the three-dimensional index (end face indentation depth, protrusion height, end face flatness index and end face tilt index) obtained in step 430 with a predetermined threshold or standard model (statistical model) to determine and output the detection result, wherein the detection result includes at least one of the following: the tobacco end face qualification or non-qualification judgment result, defect type and defect level.

[0235] In some embodiments of this disclosure, step 441 may include: inputting the two-dimensional defect features (defect type or feature quantity) obtained in step 420 and the three-dimensional indicators (end face indentation depth, protrusion height, end face flatness index and end face tilt index) obtained in step 430 into a rule base or machine learning classifier (including but not limited to SVM, random forest or lightweight neural network) to obtain the defect type and defect level, and setting differentiated thresholds for different defect types to reduce the false rejection rate.

[0236] In step 442, the output is discarded.

[0237] In some embodiments of this disclosure, step 442 may include: calculating the alignment delay N≈round(L / (v_c·T)) based on the conveying speed v_c, the distance L from the detection point to the rejection point, and the system period T; writing the detection result into a FIFO queue; and outputting a rejection signal at the corresponding time according to the alignment delay.

[0238] In some embodiments of this disclosure, the step of outputting a rejection signal at a corresponding time based on the alignment delay may include: outputting a rejection signal to the rejection execution mechanism 8 at the corresponding time to achieve online rejection.

[0239] In some embodiments of this disclosure, the steps of writing the detection result into a FIFO queue and outputting a rejection signal at the corresponding time according to the alignment delay may include: writing the detection result into a FIFO queue to output a rejection signal at the corresponding time, and using timestamp interpolation for compensation when the encoder pulse and the camera trigger are out of sync.

[0240] In some embodiments of this disclosure, the method for detecting the end face of cigarette tobacco shreds further includes: when the production line switches specifications or the ambient temperature changes, self-calibrating using a preset standard cigarette or a defect-free cigarette, updating the structured light reference fringe position u0, exposure parameters, and threshold model to maintain detection accuracy.

[0241] The above embodiments of this disclosure provide a device and method for detecting the end face of cigarette sticks and tobacco shreds in a cigarette production line, belonging to the field of online detection and quality control based on optical measurement technology.

[0242] The above embodiments of this disclosure provide a cigarette end-face detection solution that is compact, has strong dust resistance, can achieve simultaneous acquisition of dual end-face data by a single camera, and has the ability to measure end-face height.

[0243] In view of the technical problems existing in related technologies, such as the large influence of dust on end face imaging, insufficient robustness of end face defect identification, and high cost and complex maintenance of multi-camera solutions, the above-mentioned embodiments of this disclosure propose a cigarette end face detection device and detection method to achieve simultaneous detection of the two end faces after cutting, and to take into account both end face defect identification and end face concavity or convexity height measurement.

[0244] Figure 5 These are schematic diagrams illustrating some embodiments of the control processing apparatus of this disclosure. For example... Figure 5 As shown, this disclosure controls the processing equipment (e.g. Figure 1 The control processing device 7) in the embodiment may include at least one of the control module 51 and the processing module 52.

[0245] Control module 51 is configured to control the conveying and clamping mechanism of the cigarette tobacco end face detection device as described in any of the above embodiments to convey and clamp the first cigarette and the second cigarette to the detection station, so that the first tobacco end face of the first cigarette and the second tobacco end face of the second cigarette face face each other and are in the same detection area; control the end face positioning mechanism of the cigarette tobacco end face detection device as described in any of the above embodiments to reflect the first tobacco end face and the second tobacco end face to the same camera field of view; and control the photometric component of the cigarette tobacco end face detection device as described in any of the above embodiments to acquire tobacco end face images and end face height information.

[0246] In some embodiments of this disclosure, when the control module 51 acquires the end face image and end face height information of the tobacco shreds, it can be configured to turn on the surface light source and two sets of structured light emitters, and control the camera to synchronously acquire end face images including the first tobacco shred end face and the second tobacco shred end face via the V-shaped reflector component.

[0247] In some embodiments of this disclosure, when the control module 51 turns on the surface light source and two sets of structured light emitting ends, and controls the camera to synchronously acquire end-face images including the first tobacco end face and the second tobacco end face via the V-shaped reflection component, it can be configured to control the surface light source to provide uniform diffuse illumination to obtain end-face texture images; and control the first structured light emitting end and the second structured light emitting end to project structured light stripes onto the corresponding end faces and acquire structured light images.

[0248] In some embodiments of this disclosure, when the control module 51 controls the first structured light emitting end and the second structured light emitting end to project structured light stripes onto corresponding end faces and acquire structured light images, it can be configured to perform at least one of the following operations: using time-division multiplexing, first lighting up the first structured light emitting end to acquire the first structured light image, and then lighting up the second structured light emitting end to acquire the second structured light image; using multi-wavelength multiplexing, controlling the first structured light emitting end and the second structured light emitting end to use different center wavelengths respectively, and separating the first structured light image and the second structured light image through different color channels of the camera in one exposure; using spatial multiplexing, the stripes on both end faces are located in a first region of interest and a second region of interest that do not overlap in the image, and the first structured light image and the second structured light image can be extracted in the same frame respectively.

[0249] The processing module 52 is configured to process the tobacco end face image and end face height information and determine the detection result.

[0250] In some embodiments of this disclosure, the processing module 52 can be configured to perform region division and geometric correction on the end face image to obtain a first end face image and a second end face image respectively; extract the end face circumferential boundary, tobacco filling area and defect area based on the first end face image and the second end face image, and determine the judgment result of at least one defect type among end face missing tobacco, end face void, end face loose tobacco, end face depression, end face protrusion, end face tear and end face contamination; detect the position of structured light stripes in the end face image and perform calibration conversion to determine the end face depression depth, protrusion height, end face flatness index and end face tilt index; compare the judgment result, the end face depression depth, protrusion height, end face flatness index and end face tilt index with a predetermined threshold or standard model to determine the detection result, wherein the detection result includes at least one of the following: tobacco end face qualified or unqualified judgment result, defect type and defect level.

[0251] In some embodiments of this disclosure, when the processing module 52 extracts the end face circumferential boundary, tobacco filling area, and defect area based on the first end face image and the second end face image, it can be configured to obtain a candidate edge point set through edge detection; determine the end face circumferential parameters by using at least one of Hough circle transform and least squares circle fitting; and generate an end face mask and a cigarette paper port ring mask based on the circumferential parameters to locate and quantify end face tears, port abnormalities, and end face contamination.

[0252] In some embodiments of this disclosure, when the processing module 52 determines the determination result of at least one defect type among end face missing shreds, end face voids, end face loose shreds, end face depressions, end face protrusions, end face tears, and end face contamination, it can be configured to divide the tobacco filling area within the end face mask to obtain a tobacco mask; calculate the tobacco filling rate; and determine end face missing shreds if the tobacco filling rate is lower than a predetermined filling rate threshold or if there are continuous sector missing areas in the radial partition.

[0253] In some embodiments of this disclosure, when the processing module 52 determines the result of at least one defect type among end face missing shreds, end face voids, end face loose shreds, end face depressions, end face protrusions, end face tears, and end face contamination, it can be configured to segment the tobacco filling area within the end face mask to obtain a tobacco mask; perform connected component analysis on the complement region of the tobacco mask; and determine end face voids when there are connected components with voids whose area is greater than the area threshold and whose shape satisfies circularity or aspect ratio constraints, and output the void area, location, and number.

[0254] In some embodiments of this disclosure, the control processing device may also be configured to calculate the alignment delay based on the conveying speed, the distance from the detection point to the rejection point, and the system cycle; write the detection result into a queue; and output a rejection signal at the corresponding time according to the alignment delay.

[0255] In some embodiments of this disclosure, when the processing module 52 detects the position of the structured light stripes in the end face image and performs calibration calculations to determine the end face concavity depth, protrusion height, end face flatness index, and end face tilt index, it can be configured to back-project the pixel coordinates of the center point of the structured light stripes into the ray direction in the camera coordinate system based on the camera intrinsic parameter matrix and distortion parameters; intersect the ray direction with the plane equation of the structured light plane to obtain a three-dimensional point; calculate the height relative to the plane equation of the cigarette paper port reference plane; and calculate the end face concavity depth, protrusion height, end face flatness index, and end face tilt index based on the height.

[0256] In some embodiments of this disclosure, the control processing device may also be configured to perform the cigarette end face detection method described in any of the above embodiments of this disclosure.

[0257] Figure 6 This is a schematic diagram of the structure of some other embodiments of the control processing device disclosed herein. For example... Figure 6 As shown, the control processing device includes a memory 61 and a processor 62.

[0258] The memory 61 is used to store instructions, and the processor 62 is coupled to the memory 61. The processor 62 is configured to execute instructions stored in the memory to implement the tobacco end face detection method of any of the above embodiments of this disclosure.

[0259] like Figure 6 As shown, the control processing device also includes a communication interface 63 for exchanging information with other devices. Additionally, the control processing device includes a bus 64, through which the processor 62, communication interface 63, and memory 61 communicate with each other.

[0260] Memory 61 may include high-speed RAM, and may also include non-volatile memory, such as at least one disk drive. Memory 61 may also be a memory array. Memory 61 may also be divided into blocks, and the blocks may be combined into virtual volumes according to certain rules.

[0261] Furthermore, processor 62 may be a central processing unit (CPU), an application-specific integrated circuit (ASIC), or one or more integrated circuits configured to implement embodiments of the present disclosure.

[0262] According to another aspect of this disclosure, a cigarette tobacco end face detection system is provided, comprising any of the embodiments described above (e.g., Figures 1 to 3 The cigarette tobacco end face detection device and host computer described in any embodiment, wherein the host computer is configured to receive the detection results output by the control and processing equipment and drive the sorting mechanism or rejection mechanism to reject unqualified cigarettes.

[0263] According to another aspect of this disclosure, a computer program product is provided, comprising a computer program, wherein when the computer program is executed by a processor, it implements the cigarette tobacco end face detection method as described in any of the above embodiments.

[0264] According to another aspect of this disclosure, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores computer instructions that, when executed by a processor, implement the cigarette tobacco end face detection method as described in any of the above embodiments.

[0265] In some embodiments of this disclosure, the computer-readable storage medium may be a non-transitory computer-readable storage medium.

[0266] Those skilled in the art will understand that embodiments of this disclosure can be provided as methods, apparatus, or computer program products. Therefore, this disclosure can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, this disclosure can take the form of a computer program product embodied on one or more computer-usable non-transitory storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0267] This disclosure is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this disclosure. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, create a machine for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0268] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0269] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0270] The control processing device, control module, and processing module described above can be implemented as a general-purpose processor, programmable logic controller (PLC), digital signal processor (DSP), application-specific integrated circuit (ASIC), field-programmable gate array (FPGA), or other programmable logic device, discrete gate or transistor logic device, discrete hardware component, or any suitable combination thereof for performing the functions described in this application.

[0271] This concludes the detailed description of the present disclosure. To avoid obscuring the concept of the disclosure, some details known in the art have not been described. Those skilled in the art will fully understand how to implement the technical solutions disclosed herein based on the above description.

[0272] Those skilled in the art will understand that all or part of the steps of the above embodiments can be implemented by hardware, or by a program instructing the relevant hardware to implement them. The program can be stored in a non-transitory computer-readable storage medium, such as a read-only memory, a disk, or an optical disk.

[0273] The description in this disclosure is provided for illustrative and descriptive purposes only and is not intended to be exhaustive or to limit the disclosure to its forms. Many modifications and variations will be apparent to those skilled in the art. The embodiments were chosen and described in order to better illustrate the principles and practical application of this disclosure and to enable those skilled in the art to understand this disclosure and to design various embodiments with various modifications suitable for a particular purpose.

Claims

1. A device for detecting the end face of tobacco shreds in a cigarette stick, comprising: A frame, wherein a testing station is provided on the frame; The conveying and clamping mechanism is configured to convey and clamp the first cigarette and the second cigarette to the testing station, such that the first tobacco end face of the first cigarette and the second tobacco end face of the second cigarette face each other and are in the same testing area. The end face positioning mechanism set at the detection station is configured to reflect the first tobacco end face and the second tobacco end face to the same camera field of view. The optical measurement component is configured to acquire images of the tobacco end face and end face height information; and The control processing device is configured to process the tobacco end face image and end face height information and determine the detection result.

2. The tobacco end face detection device according to claim 1, wherein: The end face positioning mechanism is also configured to form an optical path reference for end face imaging and measurement; The end face positioning mechanism includes: The V-shaped reflector is configured to reflect the first tobacco end face and the second tobacco end face to the same camera field of view; The V-shaped reflective component has a first reflective surface and a second reflective surface, which are respectively configured to reflect the images of the first tobacco end face and the second tobacco end face located on both sides of the detection station to the same camera field of view.

3. The tobacco end face detection device according to claim 2, wherein, The optical measurement component includes: camera; A surface light source is configured to provide uniform illumination to the end face of the tobacco shreds; The first structured light emitting end and the second structured light emitting end are arranged corresponding to the first reflective surface and the second reflective surface, respectively, and are configured to project structured light onto the first tobacco end face and the second tobacco end face.

4. The tobacco end face detection device according to claim 3, wherein, The end face positioning mechanism includes: A transparent protective component is disposed between the detection station and the optical measurement component and is configured to isolate tobacco dust.

5. The tobacco end face detection device according to claim 4, wherein: The transparent protective component is a glass protective plate or a transparent acrylic protective plate; The transparent protective component has at least one of an anti-reflective coating and an anti-reflective coating on the side facing the camera. The transparent protective component and the frame together form a semi-enclosed end-face detection cavity.

6. The tobacco end face detection device according to claim 5, wherein, The end face detection cavity is provided with an air curtain assembly or a purging assembly, which is configured to purify the inner surface of the transparent protective assembly.

7. The tobacco end face detection device according to any one of claims 1 to 6, wherein: The conveying clamping mechanism is provided on the frame along the production line direction; The conveying and clamping mechanism includes a first clamping unit and a second clamping unit that are positioned opposite each other on the left and right, and is configured to clamp the first cigarette and the second cigarette and send the end face of the first tobacco shred and the end face of the second tobacco shred facing each other into the detection station; The first clamping unit clamps the first cigarette and positions the first tobacco end face of the first cigarette toward the second clamping unit, and the second clamping unit clamps the second cigarette and positions the second tobacco end face of the second cigarette toward the first clamping unit.

8. The tobacco end face detection device according to any one of claims 3 to 6, wherein: The structured light emitting end is a line laser emitter or a stripe projector; The control processing device is configured to perform triangulation based on the positional offset of the structured light in the end face image to obtain at least one of the following: the depth of the concavity, the height of the protrusion, and the inclination of the end face relative to the reference surface of the cigarette paper port.

9. The tobacco end face detection device according to any one of claims 1 to 6, wherein: The control and processing device is configured to segment the end face image and determine at least one defect type among end face missing wire, end face void, end face loose wire, end face depression, end face protrusion, end face tear, and end face contamination; and output a rejection signal to the actuator to reject unqualified cigarettes.

10. The tobacco end face detection device according to any one of claims 2 to 6, wherein: The first reflecting surface is a plane mirror; The second reflecting surface is a plane mirror; The angle between the first reflective surface and the second reflective surface ranges from 80° to 100°. The angles between the first reflective surface and the second reflective surface and the axis of the cigarette branch range from 40° to 55°.

11. The tobacco end face detection device according to any one of claims 4 to 6, wherein: The angle between the optical axis of the camera and the normal to the plane containing the transparent protective component ranges from 35° to 45°. The distance from the inner surface of the transparent protective component to the end face of the tobacco shreds ranges from 50mm to 70mm.

12. A method for detecting the end face of tobacco shreds in a cigarette, comprising: The conveying and clamping mechanism of the cigarette tobacco end face detection device as described in any one of claims 1 to 11 conveys and clamps the first cigarette and the second cigarette to the detection station, so that the first tobacco end face of the first cigarette and the second tobacco end face of the second cigarette face face each other and are in the same detection area. The end face positioning mechanism of the cigarette tobacco end face detection device as described in any one of claims 1 to 11 is controlled to reflect the first tobacco end face and the second tobacco end face to the same camera field of view; The photometric component of the tobacco end-face detection device as described in any one of claims 1 to 11 is controlled to acquire images of the tobacco end-face and end-face height information; and The image of the tobacco end face and the end face height information are processed to determine the detection result.

13. The method for detecting the end face of tobacco shreds in a cigarette according to claim 12, wherein, The acquisition of the tobacco end face image and end face height information includes: Turn on the surface light source and two sets of structured light emitters, and control the camera to synchronously acquire end-face images including the first tobacco end face and the second tobacco end face via the V-shaped reflector.

14. The method for detecting the end face of tobacco shreds in a cigarette according to claim 13, wherein, The open surface light source and two sets of structured light emitters control the camera to synchronously acquire end-face images including the first tobacco end face and the second tobacco end face via a V-shaped reflector assembly, including: A controlled surface light source provides uniform diffuse illumination to obtain end-face texture images; Controlling the first and second structured light emitting ends to project structured light stripes onto corresponding end faces and acquiring structured light images includes at least one of the following steps: Using a time-division multiplexing method, the first structured light transmitter is lit up first to acquire the first structured light image, and then the second structured light transmitter is lit up to acquire the second structured light image. By employing a multi-wavelength multiplexing method, the first structured light emitting end and the second structured light emitting end are controlled to use different center wavelengths, and the first structured light image and the second structured light image are separated through different color channels of the camera in a single exposure. By employing spatial multiplexing, the stripes on both ends of the image are located in a first region of interest and a second region of interest that do not overlap, allowing the first structured light image and the second structured light image to be extracted in the same frame.

15. The method for detecting the end face of tobacco shreds in a cigarette according to any one of claims 12 to 14, wherein, The process of processing the tobacco end-face image and end-face height information and determining the detection result includes: The end face image is divided into regions and geometrically corrected to obtain a first end face image and a second end face image, respectively; Based on the first end face image and the second end face image, the end face circumferential boundary, tobacco filling area and defect area are extracted, and the judgment result of at least one defect type among the end face missing tobacco, end face void, end face loose tobacco, end face depression, end face protrusion, end face tear and end face contamination is determined. The position of structured light stripes is detected in the end face image and calibrated and converted to determine the end face concavity depth, protrusion height, end face flatness index, and end face tilt index. The judgment result, the end face indentation depth, protrusion height, end face flatness index and end face tilt index are compared with a predetermined threshold or standard model to determine the test result. The test result includes at least one of the following: the tobacco end face is qualified or unqualified, defect type and defect level.

16. The method for detecting the end face of tobacco shreds in a cigarette according to claim 15, wherein, The extraction of the end face circumferential boundary, tobacco filling area, and defect area based on the first end face image and the second end face image includes: A candidate edge point set is obtained through edge detection; The end face circumferential parameters are determined by using at least one of the Hough circle transform and least squares circle fitting. Based on the circumferential parameters, an end face mask and a cigarette paper port ring mask are generated to locate and quantify end face tears, port abnormalities, and end face contamination.

17. The method for detecting the end face of tobacco shreds in a cigarette according to claim 16, wherein, The determination result for identifying at least one defect type among end face missing wire, end face void, end face loose wire, end face depression, end face bulge, end face tear, and end face contamination includes: The tobacco filling area is divided within the end face mask to obtain the tobacco mask; Calculate the tobacco filling ratio; If the tobacco filling rate is lower than the predetermined filling rate threshold or if there are consecutive missing sectors in the radial partition, it is determined that the end face is missing tobacco.

18. The method for detecting the end face of tobacco shreds in a cigarette according to claim 16, wherein, The determination result for identifying at least one defect type among end face missing wire, end face void, end face loose wire, end face depression, end face bulge, end face tear, and end face contamination includes: The tobacco filling area is divided within the end face mask to obtain the tobacco mask; Connectivity analysis is performed on the complement region of the tobacco mask; In the case of a connected region with a void area greater than the area threshold and a shape that satisfies the circularity or aspect ratio constraints, the end face void is determined, and the void area, location, and number are output.

19. The method for detecting the end face of tobacco shreds in a cigarette according to any one of claims 12 to 14, further comprising: The alignment delay is calculated based on the conveying speed, the distance from the detection point to the rejection point, and the system cycle. Write the detection results into the queue; Based on the alignment delay, a rejection signal is output at the corresponding time.

20. The method for detecting the end face of tobacco shreds in a cigarette according to claim 15, wherein, The step of detecting the position of structured light stripes in the end face image and performing calibration conversion to determine the end face concavity depth, protrusion height, end face flatness index, and end face tilt index includes: Based on the camera intrinsic parameter matrix and distortion parameters, the pixel coordinates of the center point of the structured light stripe are back-projected into the ray direction in the camera coordinate system; The intersection of the ray direction and the plane equation of the structured light plane yields a three-dimensional point; Calculate the height using the plane equation relative to the reference plane at the cigarette paper end; The end face concavity depth, protrusion height, end face flatness index, and end face tilt index are calculated based on the height.

21. A control processing device, comprising: The control module is configured to control the conveying and clamping mechanism of the cigarette tobacco end face detection device as described in any one of claims 1 to 11 to convey and clamp the first cigarette and the second cigarette to the detection station, such that the first tobacco end face of the first cigarette and the second tobacco end face of the second cigarette face face each other and are in the same detection area; control the end face positioning mechanism of the cigarette tobacco end face detection device as described in any one of claims 1 to 11 to reflect the first tobacco end face and the second tobacco end face to the same camera field of view; and control the photometric component of the cigarette tobacco end face detection device as described in any one of claims 1 to 11 to acquire tobacco end face images and end face height information. as well as The processing module is configured to process the tobacco end face image and end face height information and determine the detection result.

22. A control processing device, comprising: The memory is configured to store instructions; as well as A processor coupled to the memory, the processor being configured to execute the cigarette tobacco end face detection method as described in any one of claims 12 to 20 based on instructions stored in the memory.

23. A cigarette tobacco end face detection system, comprising a cigarette tobacco end face detection device as described in any one of claims 1 to 11 and a host computer, wherein, The host computer is configured to receive the detection results output by the control and processing equipment and drive the sorting mechanism or rejection mechanism to reject unqualified cigarettes.

24. A computer-readable storage medium, wherein, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the cigarette tobacco end face detection method as described in any one of claims 12 to 20.

25. A computer program product comprising a computer program, wherein, When the computer program is executed by the processor, it implements the cigarette tobacco end face detection method as described in any one of claims 12 to 20.

Images