Material thickness uniformity adjusting method and device based on double depth cameras

Abstract

The embodiment of the invention discloses a material thickness uniformity adjusting method and device based on double depth cameras. The method comprises the following steps: acquiring a depth image set of tobacco materials on a material conveying belt collected by double depth cameras; determining a point cloud data set of the tobacco material based on the depth image set and the internal and external parameters of the double depth cameras; performing point cloud registration and point cloud fusion on the point cloud data set, and constructing a three-dimensional material model of the tobacco material; based on the three-dimensional material model and a pre-constructed scraper physical model, establishing a target function by taking material thickness uniformity as an optimization target; and performing maximum value solving on the target function based on a gradient method, determining leveling information of the scraper, and controlling the scraper to perform thickness uniformity adjustment on the tobacco material based on the leveling information. Through the technical scheme of the embodiment of the invention, the thickness uniformity of the tobacco material can be accurately and conveniently adjusted, and the adjustment efficiency and accuracy of the thickness uniformity of the tobacco material are improved.

Description

Technical Field

[0001] The embodiments of the present invention relate to the technical field of tobacco processing in a cigarette processing technology, and in particular to a method and device for adjusting material thickness uniformity based on a dual-depth camera. Background Art

[0002] In the tobacco shred process, the existing material flow control generally has two feeding methods to control the flow: quantitative feeding device and vibrating plate feeding device. There are two types of material flow control devices that are widely used.

[0003] The first is a quantitative feeding device, which refers to a material flow control system that uses a feeder, a metering tube, a controlled electronic belt scale or a feeder, a metering tube, and a horizontal conveyor belt. Tobacco sheets or shreds are fed by a steep-angled elevator and dropped from a hopper into a metering tube or dosing tube. The dosing tube is vertically connected to the hopper. Transparent windows on either side of the tube wall are equipped with dual-position grating sensors for upper and lower limits, controlling the temporary storage of the appropriate amount of tobacco material within the tube. A press roller is installed between the lower outlet of the dosing tube and the electronic belt scale. The belt runs forward, driving the tobacco material into the processing process. The press roller and the electronic belt scale interact to control the relatively stable and quantitative flow of tobacco material.

[0004] However, the surface of the tobacco material at the outlet of the metering tube is corrugated and uneven, the dynamic adjustment accuracy of the electronic scale is low, and the feedback lag is large, which affects the instantaneous flow uniformity of the material at the entrance of processes such as adding materials, drying, and adding flavors, resulting in uneven processing intensity in subsequent processes.

[0005] The second is the vibrating feeding plate, which mainly relies on the cooperation of pulse electromagnets and annular spring sheets. The vibration plate with a certain amplitude makes the tobacco material in the hopper vibrate in the vertical direction, and the annular spring sheet drives the material to perform torsional vibration around it, achieving a quantitative feeding effect. The feeding effect can be adjusted and controlled by changing the excitation frequency.

[0006] However, when the material is only moved forward by vibrating the vibrating disk and the annular spring sheet with a certain amplitude, instantaneous flow interruption often occurs, requiring manual intervention to replenish the material to achieve a stable feed flow, which also affects the uniformity of the conveyed material. Summary of the Invention

[0007] The embodiments of the present invention provide a method and device for adjusting material thickness uniformity based on a dual-depth camera, so as to accurately and conveniently adjust the thickness uniformity of tobacco materials, thereby improving the adjustment efficiency and accuracy of the thickness uniformity of tobacco materials.

[0008] In a first aspect, an embodiment of the present invention provides a method for adjusting material thickness uniformity based on a dual-depth camera, comprising:

[0009] Obtaining a depth image set of tobacco materials on a material conveyor belt captured by a dual depth camera;

[0010] Determining a point cloud dataset of the tobacco material based on the depth image set and the internal and external parameters of the dual depth camera; wherein the internal parameters are determined by a checkerboard calibration method; and the external parameters are determined by a calibration plate calibration method;

[0011] Performing point cloud registration and point cloud fusion on the point cloud data set to construct a three-dimensional material model of the tobacco material;

[0012] Based on the three-dimensional material model and the pre-built scraper physical model, an objective function is established with material thickness uniformity as the optimization goal;

[0013] The objective function is maximized based on the gradient method to determine the leveling information of the scraper, and the scraper is controlled based on the leveling information to adjust the thickness uniformity of the tobacco material; the leveling information includes: leveling height, leveling speed and stroke.

[0014] Optionally, the method further includes: before obtaining the depth image set of the tobacco material on the material conveyor belt captured by the dual-depth camera, obtaining a material in place signal sent by the limit electric eye, and generating an image acquisition signal for waking up the dual-depth camera based on the material in place signal; and controlling the dual-depth camera to capture the depth image set of the tobacco material on the material conveyor belt based on the image acquisition signal.

[0015] Optionally, the method further includes: before determining the point cloud dataset of the tobacco material, performing image preprocessing on each depth image in the depth image set based on a preset image preprocessing method to obtain a preprocessed depth image set.

[0016] Optionally, the method also includes: determining a point cloud dataset of the tobacco material in the camera coordinate system based on the depth image set and the internal parameters of the dual-depth camera; and determining a point cloud dataset of the tobacco material in the world coordinate system based on the point cloud dataset of the tobacco material in the camera coordinate system and the external parameters of the dual-depth camera.

[0017] Optionally, the method further includes: performing rough alignment on each point cloud data based on the centroid of each point cloud data in the point cloud data set to obtain a roughly aligned point cloud data set; using any point cloud data in the roughly aligned point cloud data set as source point cloud data, and using the remaining point cloud data as selected point cloud data; determining the selected pixel point corresponding to each source pixel point in the source point cloud data in the selected point cloud data through nearest neighbor search; performing point cloud fusion on the source point cloud data and the selected point cloud data based on the established association relationship between the source pixel point and the selected pixel point, to construct a three-dimensional material model of the tobacco material.

[0018] Optionally, the method also includes: the tobacco material is a roughly scraped tobacco material; before obtaining the depth image set of the tobacco material on the material conveyor belt acquired by the dual-depth camera, a depth image set of the unscraped tobacco material on the material conveyor belt acquired by the dual-depth camera is acquired; based on the depth image set and the internal and external parameters of the dual-depth camera, a point cloud data set of the unscraped tobacco material is determined; point cloud registration and point cloud fusion are performed on the point cloud data set to construct a three-dimensional material model of the unscraped tobacco material; regional extreme value detection is performed on the three-dimensional material model to determine each local maximum point and local minimum point in the three-dimensional material model; the depth values corresponding to the local maximum point and the local minimum point are averaged to determine the rough scraping height of the scraper, and the scraper is controlled based on the rough scraping height to perform rough scraping processing on the unscraped tobacco material.

[0019] Optionally, the method also includes: the dual depth camera includes: a first depth camera and a second depth camera; the first depth camera and the second depth camera are arranged at a cross-perspective layout at a preset angle; the first depth camera is a depth camera arranged on the left side of the positive direction of conveying materials along the material conveyor belt; the second depth camera is a depth camera arranged on the right side of the positive direction of conveying materials along the material conveyor belt; the depth image set includes: a first depth image and a second depth image; the first depth image is a depth image captured by the first depth camera at a perspective slightly to the left of the tobacco material; the second depth image is a depth image captured by the second depth camera at a perspective slightly to the right of the tobacco material.

[0020] In a second aspect, an embodiment of the present invention further provides a device for adjusting material thickness uniformity based on a dual-depth camera, the device comprising:

[0021] A first depth image set acquisition module is used to acquire a depth image set of the tobacco material on the material conveyor belt captured by the dual depth camera;

[0022] a first point cloud data set determination module, configured to determine a point cloud data set of the tobacco material based on the depth image set and internal and external parameters of the dual depth camera; the internal parameters are determined by a checkerboard calibration method; and the external parameters are determined by a calibration plate calibration method;

[0023] a first three-dimensional material model construction module, configured to perform point cloud registration and point cloud fusion on the point cloud dataset to construct a three-dimensional material model of the tobacco material;

[0024] An objective function establishment module is used to establish an objective function based on the three-dimensional material model and the pre-built scraper physical model, with the material thickness uniformity as the optimization target;

[0025] The thickness uniformity adjustment module is used to solve the maximum value of the objective function based on the gradient method, determine the leveling information of the scraper, and control the scraper to adjust the thickness uniformity of the tobacco material based on the leveling information; the leveling information includes: leveling height, leveling speed and stroke.

[0026] In a third aspect, an embodiment of the present invention further provides an electronic device, comprising:

[0027] one or more processors;

[0028] a memory for storing one or more programs;

[0029] When the one or more programs are executed by the one or more processors, the one or more processors implement the material thickness uniformity adjustment method based on the dual-depth camera as provided in any embodiment of the present invention.

[0030] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements a material thickness uniformity adjustment method based on a dual-depth camera as provided in any embodiment of the present invention.

[0031] In a fifth aspect, an embodiment of the present invention provides a computer program product, including a computer program, which, when executed by a processor, implements a material thickness uniformity adjustment method based on a dual-depth camera as provided in any embodiment of the present invention.

[0032] The technical solution of the embodiment of the present invention is to obtain a depth image set of tobacco materials on a material conveyor belt collected by a dual-depth camera, thereby using the dual-depth camera to achieve multi-perspective collaborative perception, and obtain depth images of tobacco materials from multiple angles, that is, multi-source data; based on the depth image set and the internal and external parameters of the dual-depth camera, a point cloud data set of the tobacco material is determined; the internal parameters are determined by a checkerboard calibration method; the external parameters are determined by a calibration plate calibration method; point cloud registration and point cloud fusion are performed on the point cloud data set to construct a three-dimensional material model of the tobacco material. , thereby realizing multi-source data fusion so as to make feedback adjustments on this basis; based on the three-dimensional material model and the pre-built scraper physical model, an objective function is established with the material thickness uniformity as the optimization target; the objective function is solved for the maximum value based on the gradient method, the scraper leveling information is determined, and the scraper is controlled to adjust the thickness uniformity of the tobacco material based on the leveling information, thereby realizing accurate and convenient adjustment of the thickness uniformity of the tobacco material, and improving the adjustment efficiency and accuracy of the thickness uniformity of the tobacco material; the leveling information includes: leveling height, leveling speed and stroke.

[0033] It should be understood that the content described in this section is not intended to identify the key or important features of the embodiments of the present invention, nor is it intended to limit the scope of the present invention. Other features of the present invention will become readily understood through the following description. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0035] Figure 1 This is a flow chart of a method for adjusting material thickness uniformity based on a dual-depth camera provided in Example 1 of the present invention;

[0036] Figure 2 This is a flow chart of a method for adjusting material thickness uniformity based on a dual-depth camera provided in Example 2 of the present invention;

[0037] Figure 3 This is a system example diagram of a method for adjusting material thickness uniformity based on a dual-depth camera according to a second embodiment of the present invention;

[0038] Figure 4 1 is a schematic structural diagram of a device for adjusting material thickness uniformity based on a dual-depth camera according to a third embodiment of the present invention;

[0039] Figure 5 3 is a schematic structural diagram of an electronic device for implementing a method for adjusting material thickness uniformity based on a dual-depth camera according to an embodiment of the present invention. DETAILED DESCRIPTION

[0040] In order to enable those skilled in the art to better understand the solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the embodiments described are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts should fall within the scope of protection of the present invention.

[0041] It should be noted that the terms "first", "second", etc. in the description and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that the numbers used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0042] Example 1

[0043] Figure 1 A flowchart of a material thickness uniformity adjustment method based on a dual-depth camera is provided for the first embodiment of the present invention. This embodiment is applicable to the case of adjusting the thickness uniformity of tobacco materials. The method can be performed by a material thickness uniformity adjustment device based on a dual-depth camera. The material thickness uniformity adjustment device based on a dual-depth camera can be implemented in the form of hardware and / or software. The material thickness uniformity adjustment device based on a dual-depth camera can be configured in an electronic device. Figure 1 As shown, the method includes:

[0044] S110 : Acquire a depth image set of the tobacco material on the material conveyor belt captured by the dual depth cameras.

[0045] The dual depth camera includes: a first depth camera and a second depth camera. The first depth camera and the second depth camera are arranged at a cross-viewing angle at a preset angle. The preset angle can be, but is not limited to, 45°. The first depth camera is a depth camera arranged on the left side of the positive direction of the material conveying along the material conveyor belt. The second depth camera is a depth camera arranged on the right side of the positive direction of the material conveying along the material conveyor belt. The depth image set includes: a first depth image and a second depth image. The first depth image is a depth image captured by the first depth camera at a perspective slightly to the left of the tobacco material. The second depth image is a depth image captured by the second depth camera at a perspective slightly to the right of the tobacco material. The material conveyor belt is used to convey all materials falling from the outlet of the metering tube. The tobacco material may refer to part of the material conveyed on the material conveyor belt. For example, the tobacco material may be, but is not limited to, the material within a specific area pre-set on the material conveyor belt. In an embodiment of the present invention, the tobacco material may be, but is not limited to, material that has not been leveled or material that has been leveled once.

[0046] Specifically, a first depth image of the tobacco material on the material conveyor belt captured by the first depth camera is obtained, and a second depth image of the tobacco material on the material conveyor belt captured by the second depth camera at the same time is obtained.

[0047] For example, the dual-depth camera's capture range covers the entire conveyor belt, preventing the wide conveyor belt from incomplete field of view. Circular polarized fill light is used to capture depth images from the dual-depth camera, suppressing reflections from tobacco leaves of varying formulations, grades, and colors. This effectively captures two slightly offset depth images of the same tobacco material on the conveyor belt.

[0048] Based on the above technical solution, the method also includes: before obtaining the depth image set of the tobacco material on the material conveyor belt captured by the dual-depth camera, obtaining the material in place signal sent by the limit electric eye, and generating an image acquisition signal for waking up the dual-depth camera based on the material in place signal; based on the image acquisition signal, controlling the dual-depth camera to capture the depth image set of the tobacco material on the material conveyor belt.

[0049] The limit sensors can include an upper limit sensor and a lower limit sensor. The upper and lower limit sensors can be understood as grating sensors at the upper and lower limit positions. The upper limit sensor senses whether material has entered the metering tube. The lower limit sensor senses whether material is about to fall onto the material conveyor belt. The material arrival signal can be a trigger signal when the material reaches the lower limit position. The image acquisition signal can be used to trigger the depth camera to capture a depth image of the tobacco material.

[0050] Specifically, taking tobacco material as an example, which has not been scraped, the feed conveyor conveys the material at a constant speed to the metering tube. When the lower limit sensor detects the material, it sends a material arrival signal to the controller (such as a PLC). Based on the received material arrival signal, the controller generates an image acquisition signal for waking up the dual-depth camera and uses the image acquisition signal to wake up the dual-depth camera. After being awakened, the dual-depth camera collects a depth image set of the tobacco material on the material conveyor according to the image acquisition signal.

[0051] For example, taking tobacco material as an example of material that has been scraped once, when the controller controls the scraper to complete the material scraping operation, it sends an image acquisition signal to the dual-depth camera, so that the dual-depth camera collects a depth image set of the tobacco material on the material conveyor belt based on the image acquisition signal.

[0052] Based on the above technical solution, the method also includes: before determining the point cloud data set of the tobacco material, each depth image in the depth image set is preprocessed based on a preset image preprocessing method to obtain a preprocessed depth image set.

[0053] The preset image preprocessing method may refer to a preset method for removing noise. For example, the preset image preprocessing method may be, but is not limited to, bilateral filtering.

[0054] Specifically, bilateral filtering denoising is performed on the first depth image or the second depth image in the depth image set to reduce noise interference of sensor noise and material occlusion boundary artifacts.

[0055] For example, the image preprocessing operation is expressed as follows:

[0056]

[0057] Where i and j are used to represent pixel positions; Δx=ix, Δy=jy are pixel offsets; d p and d q is the pixel value; σ s Control space smoothness; σ r Controls the depth similarity tolerance.

[0058] S120 : Determine a point cloud dataset of the tobacco material based on the depth image set and the internal and external parameters of the dual depth camera.

[0059] Among them, the internal parameters are determined by the checkerboard calibration method. The external parameters are determined by the calibration plate calibration method. The internal parameters can refer to the inherent parameters that describe the internal optical characteristics of the camera and are directly related to the physical structure of the camera. The internal parameters can be used to establish the mapping relationship between the pixel coordinate system and the camera coordinate system. The internal parameters include: focal length (f x ,f y ), principal point coordinates (c x ,c y) and distortion coefficients (k1, k2, p1, p2, k3). The focal length is the optical focal length per pixel. The focal length can be used to determine the imaging scale. The principal point coordinates are the intersection of the optical axis and the imaging plane. The principal point coordinates can be used to reflect the sensor center offset. The distortion coefficients include radial distortion (k1, k2, k3) and tangential distortion (p1, p2). The distortion coefficients can be used to correct lens deformation errors. Extrinsic parameters can refer to parameters that describe the position and posture of the camera in the world coordinate system. Extrinsic parameters can be used to transform a three-dimensional point from the world coordinate system to the camera coordinate system. Extrinsic parameters include the rotation matrix (R) and the translation vector (t). The rotation matrix is a 3×3 matrix. The rotation matrix can be used to represent the rotation of the camera coordinate system relative to the world coordinate system. The translation vector is a 3×1 vector. The translation vector can be used to represent the position of the camera origin in the world coordinate system. The point cloud dataset includes: first point cloud data corresponding to the first depth image and second point cloud data corresponding to the second depth image. Point cloud data can refer to a data structure used to describe objects or scenes in three-dimensional space. Point cloud data can be composed of a large number of spatial coordinate points. Point cloud data can be used to represent tobacco materials in the form of spatial coordinate points.

[0060] Specifically, for the first depth image in the depth image set that is in the pixel coordinate system, the intrinsic parameters of the first depth camera are used to perform coordinate transformation on each pixel point in the first depth image to determine the first point cloud data of the first depth image in the camera coordinate system, and the extrinsic parameters of the first depth camera are used to perform coordinate transformation on each spatial coordinate point in the first point cloud data to determine the first point cloud data of the first depth image in the world coordinate system. For the second depth image in the depth image set that is in the pixel coordinate system, the intrinsic parameters of the second depth camera are used to perform coordinate transformation on each pixel point in the second depth image to determine the second point cloud data of the second depth image in the camera coordinate system, and the extrinsic parameters of the second depth camera are used to perform coordinate transformation on each spatial coordinate point in the second point cloud data to determine the second point cloud data of the second depth image in the world coordinate system.

[0061] Based on the above technical solution, "determining the point cloud dataset of the tobacco material based on the depth image set and the internal and external parameters of the dual-depth camera" may include: determining the point cloud dataset of the tobacco material in the camera coordinate system based on the depth image set and the internal parameters of the dual-depth camera; determining the point cloud dataset of the tobacco material in the world coordinate system based on the point cloud dataset of the tobacco material in the camera coordinate system and the external parameters of the dual-depth camera.

[0062] Specifically, the coordinates (u, v) of each pixel in the depth image correspond to the coordinates (x, y, z) of the three-dimensional point cloud. The conversion formula is:

[0063]

[0064] Where z is the depth value of the (u,v) position in the depth image. If you need to convert the point cloud to the world coordinate system, you need to combine the external parameters as follows:

[0065]

[0066] S130 , performing point cloud registration and point cloud fusion on the point cloud dataset to construct a three-dimensional material model of the tobacco material.

[0067] The three-dimensional material model may be a three-dimensional model constructed based on the physical object of the tobacco material. In the embodiment of the present invention, the three-dimensional material model is similar to the three-dimensional model of a mountain range, both of which have shape features such as peaks, valleys and ridges.

[0068] Specifically, the centroid of the first point cloud data and the centroid of the second point cloud data in the point cloud data set are determined, point clouds are registered based on the centroids of the two, and point clouds are fused based on the registration results to construct a three-dimensional material model of the tobacco material.

[0069] S140. Based on the three-dimensional material model and the pre-built scraper physical model, an objective function is established with material thickness uniformity as the optimization target.

[0070] Specifically, the objective function can be expressed as:

[0071]

[0072] Where D is the surface area of tobacco material, A is the area of the region; H(v,s,x,y) is the thickness distribution after scraper treatment; is the average thickness.

[0073] S150 , solving the maximum value of the objective function based on the gradient method, determining the flatness information of the scraper, and controlling the scraper to adjust the thickness uniformity of the tobacco material based on the flatness information.

[0074] Leveling information includes leveling height, leveling speed, and travel. Leveling height is the height set by the scraper during leveling. Leveling speed is the speed at which the scraper moves during leveling. Travel is the distance the scraper moves during leveling.

[0075] Specifically, the leveling speed and stroke are randomly initialized, and the gradient method is used to solve the maximum value of the objective function to obtain the leveling speed v and stroke s for the best material thickness uniformity. The material thickness uniformity is defined as the negative variance of the thickness distribution after treatment, and the optimization goal is to maximize the uniformity. The thickness of the material model after scraper leveling (equivalent to the leveling height): Assume that the scraper action is equivalent to h(x,y) smoothing filtering of the surface convexity of the three-dimensional model, and the filter kernel function g(v,s,x,y) is related to v and s. For example, the Gaussian kernel function:

[0076] H(v,s,x,y)=h(x,y)*g(v,s;x,y)

[0077] Among them, the relationship between the variance σ of the filter kernel and the scraper parameter is:

[0078]

[0079] The specific form of the kernel function is:

[0080]

[0081] The optimization goal is to maximize the objective function J(v,s), thereby guiding the adaptive leveling scraper to complete material leveling.

[0082] The technical solution of the embodiment of the present invention is to obtain a depth image set of tobacco materials on a material conveyor belt collected by a dual-depth camera, thereby using the dual-depth camera to achieve multi-perspective collaborative perception, and obtain depth images of tobacco materials from multiple angles, that is, multi-source data; based on the depth image set and the internal and external parameters of the dual-depth camera, a point cloud data set of the tobacco material is determined; the internal parameters are determined by a checkerboard calibration method; the external parameters are determined by a calibration plate calibration method; point cloud registration and point cloud fusion are performed on the point cloud data set to construct a three-dimensional material model of the tobacco material, thereby realizing multi-source data fusion for feedback adjustment on this basis; based on the three-dimensional material model and a pre-constructed scraper physical model, an objective function is established with material thickness uniformity as the optimization target; the objective function is solved for the maximum value based on the gradient method to determine the flatness information of the scraper, and the scraper is controlled to adjust the thickness uniformity of the tobacco material based on the flatness information, thereby achieving accurate and convenient adjustment of the thickness uniformity of the tobacco material, thereby improving the adjustment efficiency and accuracy of the thickness uniformity of the tobacco material; the flatness information includes: flattening height, flattening speed and stroke.

[0083] On the basis of the above technical solution, the tobacco material in S110 is a roughly scraped tobacco material, and the method also includes: before obtaining the depth image set of the tobacco material on the material conveyor belt captured by the dual-depth camera, obtaining the depth image set of the unscraped tobacco material on the material conveyor belt captured by the dual-depth camera; determining the point cloud data set of the unscraped tobacco material based on the depth image set and the internal and external parameters of the dual-depth camera; performing point cloud registration and point cloud fusion on the point cloud data set to construct a three-dimensional material model of the unscraped tobacco material; performing regional extreme value detection on the three-dimensional material model to determine each local maximum point and local minimum point in the three-dimensional material model; performing mean processing on the depth values corresponding to the local maximum point and the local minimum point, determining the rough scraping height of the scraper, and controlling the scraper to perform rough scraping processing on the unscraped tobacco material based on the rough scraping height.

[0084] Roughly scraped tobacco material can refer to material that has been scraped once. Unscraped tobacco material refers to material that has not been scraped. The local area in regional extreme value detection is a linear array detection area divided along the direction of travel of the material conveyor belt. Linear array extreme value points include: local maximum points and local minimum points.

[0085] Specifically, before acquiring a depth image set of the tobacco material on the material conveyor belt captured by the dual-depth camera, a depth image set of the unscraped tobacco material on the material conveyor belt captured by the dual-depth camera is acquired; based on the depth image set and the internal and external parameters of the dual-depth camera, a point cloud dataset of the unscraped tobacco material is determined; point cloud registration and point cloud fusion are performed on the point cloud dataset to construct a three-dimensional material model of the unscraped tobacco material. The depth gradient of each line array is calculated, and the gradient mutation points are selected according to a preset threshold value as the line array extreme points. For example, the preset threshold value can be a depth gradient greater than 15 mm / m or less than -15 mm / m. The depth values corresponding to the local maximum point and the local minimum point are averaged to determine the rough scraping height of the scraper, and the scraper is controlled to perform rough scraping on the unscraped tobacco material based on the rough scraping height. This allows rough scraping to be performed before fine scraping, avoiding the situation where a specific thickness or optimal thickness uniformity cannot be achieved in one scraping process, and provides a material basis for accurate scraping for the second scraping process, further improving the adjustment efficiency and accuracy of the thickness uniformity of the tobacco material.

[0086] Example 2

[0087] Figure 2 This is a flowchart of a method for adjusting material thickness uniformity based on a dual-depth camera provided in the second embodiment of the present invention. This embodiment, based on the above embodiment, describes in detail the process of constructing a three-dimensional material model of tobacco material. The explanations of the terms that are the same or corresponding to the above embodiments are not repeated here. Figure 2 As shown, the method includes:

[0088] S210: Acquire a depth image set of the tobacco material on the material conveyor belt captured by the dual depth cameras.

[0089] S220 : Determine a point cloud dataset of the tobacco material based on the depth image set and the internal and external parameters of the dual depth camera.

[0090] The internal parameters are determined by the checkerboard calibration method, and the external parameters are determined by the calibration plate calibration method.

[0091] S230 , performing coarse registration on each point cloud data set based on the centroid of each point cloud data set in the point cloud data set to obtain a coarsely registered point cloud data set.

[0092] S240 , taking any point cloud data in the roughly registered point cloud data set as source point cloud data, and taking the remaining point cloud data as candidate point cloud data.

[0093] S250 , determining a candidate pixel point corresponding to each source pixel point in the source point cloud data in the candidate point cloud data by nearest neighbor search.

[0094] Among them, the source pixel point in the source point cloud data is represented as p i . In the selected point cloud data, the source pixel point p i The nearest candidate pixel is denoted as q j The method for determining the pixel points to be selected is as follows:

[0095] q j =argmin||p i -q||

[0096] S260 , performing point cloud fusion based on the source point cloud data with established association relationships between the source pixel points and the pixel points to be selected and the point cloud data to be selected, to construct a three-dimensional material model of the tobacco material.

[0097] Specifically, point cloud fusion is performed on the source point cloud data and the point cloud data to be selected based on the established correlation relationship between the source pixel points and the pixel points to be selected, and a material model to be selected of the tobacco material is obtained. The center of mass calculation is performed on the material model to be selected, and then the surface of the model is reconstructed to obtain a three-dimensional material model of the tobacco material. A double verification mechanism of the center of mass constraint is introduced after the nearest neighbor search, which further improves the accuracy of the model construction.

[0098] Among them, calculate the center of mass:

[0099]

[0100] Among them, p k ∈Voxel i .

[0101] Surface reconstruction:

[0102]

[0103] Where d(x) is the distance from the point to the surface, and u is the cutoff threshold.

[0104] S270. Based on the three-dimensional material model and the pre-built scraper physical model, an objective function is established with material thickness uniformity as the optimization goal.

[0105] S280: Solve the maximum value of the objective function based on the gradient method, determine the flatness information of the scraper, and control the scraper to adjust the thickness uniformity of the tobacco material based on the flatness information.

[0106] The leveling information includes: leveling height, leveling speed and stroke.

[0107] The technical solution of the embodiment of the present invention is to perform coarse registration of each point cloud data based on the centroid of each point cloud data in the point cloud data set to obtain a coarsely registered point cloud data set; the centroid method only needs to calculate the geometric center of the point cloud and translate and align it, and the time complexity is O(n), which is much faster than the ICP algorithm O(n). 2 ) complexity, the speed is increased by more than 10 times, and the centroid coordinates are less affected by outliers, which can further improve the registration accuracy when the point cloud contains noise. Any point cloud data in the point cloud data set after coarse registration is used as the source point cloud data, and the remaining point cloud data is used as the candidate point cloud data; the candidate pixel point corresponding to each source pixel point in the source point cloud data is determined in the candidate point cloud data through nearest neighbor search; based on the source point cloud data and the candidate point cloud data with the established correlation relationship between the source pixel point and the candidate pixel point, point cloud fusion is performed to construct a three-dimensional material model of the tobacco material, thereby using coarse registration and fine registration to achieve multi-scale feature retention, such as coarse registration retains macro features, and fine registration repairs micro details, further improving the accuracy of model construction.

[0108] On the basis of the above technical solution, the three-dimensional material model can also be a model of the surface concavity and convexity of the tobacco material. In this model, the expression of the concavity and convexity extreme points (equivalent to the maximum and minimum points) is:

[0109] T rough =∑θ(n i ,n j )

[0110]

[0111] Among them, n i is the local normal vector of the rectangular patch, n j is the average normal vector of the rectangular patch area, θ(n i ,n j ) is the absolute value of the angle between the two normal vectors, T rough It is the sum of the angles between the normal vector of the rectangular patch and the normal vectors of the surrounding patches. It is used to measure the flatness and distribution of convex and concave extreme points on the surface of the patch area, evaluate the point cloud density, and screen the material conveying areas with low point cloud density and dense concave and convex extreme points to obtain the 3D material model for subsequent operations. The expression of the evaluation index is:

[0112]

[0113] Among them, α and β are weight coefficients, D max 、D min is the maximum and minimum regional point cloud density, T max 、T minis the maximum and minimum area convexity, T i Indicates the concavity and convexity of the material in the rectangular area of the field of view.

[0114] Illustratively, an embodiment of the present invention further provides a system that applies a method for adjusting material thickness uniformity based on a dual-depth camera. Figure 3 A system example diagram of a material thickness uniformity adjustment method based on a dual-depth camera is given. Figure 3 The system includes: a feeding conveyor belt, a quantitative tube, an upper limit electric eye, a lower limit electric eye, a thinning roller, a material conveyor belt, a height-adjustable automatic scraper, a first depth camera, a second depth camera and a controller.

[0115] Both the first and second depth cameras are Intel RealSense D435i depth cameras. In this system, when material passes through a metered feeding tube, the upper and lower sensors detect the passage of material and generate feedback signals to activate the dual depth cameras. The two Intel RealSense D435i depth cameras are arranged with a 45-degree cross-view angle. After coordinate calibration, the two cameras simultaneously capture a depth image. After image preprocessing, a fused 3D material model is generated. Based on this 3D material model, multiple extreme points in a linear array are detected and located, and their depth information is calculated. The average of the extreme depths in the linear array is used as the scraper's height for feedback control. An adaptive leveling scraper (an automatic, height-adjustable scraper) performs the initial leveling of the material. After the adaptive leveling scraper completes the initial leveling, the process repeats to generate a new 3D material model. Based on the physical scraper model and the new 3D material model, an objective function for optimizing material thickness uniformity is constructed. The objective function is maximized using a gradient method to determine the scraper's leveling height, speed, and stroke. These values are then used to control the scraper's thickness uniformity. The flattened and even tobacco material moves evenly under the action of the conveyor belt and is sent to the feeding barrel equipment for feeding processing.

[0116] The following is an embodiment of a material thickness uniformity adjustment device based on a dual-depth camera provided in an embodiment of the present invention. The device and the material thickness uniformity adjustment method based on a dual-depth camera in the above-mentioned embodiments belong to the same inventive concept. For details not fully described in the embodiment of the material thickness uniformity adjustment device based on a dual-depth camera, please refer to the embodiment of the material thickness uniformity adjustment method based on a dual-depth camera.

[0117] Example 3

[0118] Figure 4 This is a schematic diagram of the structure of a device for adjusting material thickness uniformity based on a dual-depth camera provided in the third embodiment of the present invention. Figure 4As shown, the device includes: a first depth image set acquisition module 410, a first point cloud data set determination module 420, a first three-dimensional material model construction module 430, an objective function establishment module 440 and a thickness uniformity adjustment module 450.

[0119] Among them, the first depth image set acquisition module 410 is used to obtain the depth image set of the tobacco material on the material conveyor belt collected by the dual depth camera; the first point cloud data set determination module 420 is used to determine the point cloud data set of the tobacco material based on the depth image set and the internal and external parameters of the dual depth camera; the internal parameters are determined by the checkerboard calibration method; the external parameters are determined by the calibration plate calibration method; the first three-dimensional material model construction module 430 is used to perform point cloud registration and point cloud fusion on the point cloud data set to construct a three-dimensional material model of the tobacco material; the objective function establishment module 440 is used to establish an objective function based on the three-dimensional material model and the pre-constructed scraper physical model, with the material thickness uniformity as the optimization target; the thickness uniformity adjustment module 450 is used to solve the maximum value of the objective function based on the gradient method, determine the scraper leveling information, and control the scraper to adjust the thickness uniformity of the tobacco material based on the leveling information; the leveling information includes: leveling height, leveling speed and stroke.

[0120] The technical solution of the embodiment of the present invention is to obtain a depth image set of tobacco materials on a material conveyor belt collected by a dual-depth camera, thereby using the dual-depth camera to achieve multi-perspective collaborative perception, and obtain depth images of tobacco materials from multiple angles, that is, multi-source data; based on the depth image set and the internal and external parameters of the dual-depth camera, a point cloud data set of the tobacco material is determined; the internal parameters are determined by a checkerboard calibration method; the external parameters are determined by a calibration plate calibration method; point cloud registration and point cloud fusion are performed on the point cloud data set to construct a three-dimensional material model of the tobacco material, thereby realizing multi-source data fusion for feedback adjustment on this basis; based on the three-dimensional material model and a pre-constructed scraper physical model, an objective function is established with material thickness uniformity as the optimization target; the objective function is solved for the maximum value based on the gradient method to determine the flatness information of the scraper, and the scraper is controlled to adjust the thickness uniformity of the tobacco material based on the flatness information, thereby achieving accurate and convenient adjustment of the thickness uniformity of the tobacco material, thereby improving the adjustment efficiency and accuracy of the thickness uniformity of the tobacco material; the flatness information includes: flattening height, flattening speed and stroke.

[0121] On the basis of the above technical solution, the device further includes:

[0122] An image acquisition signal generation module is used to obtain the material arrival signal sent by the limit electric eye and generate an image acquisition signal for waking up the dual-depth camera based on the material arrival signal;

[0123] The depth image set acquisition module is used to control the dual depth cameras to acquire the depth image set of the tobacco material on the material conveyor belt based on the image acquisition signal.

[0124] On the basis of the above technical solution, the device further includes:

[0125] The depth image set preprocessing module is used to perform image preprocessing on each depth image in the depth image set based on a preset image preprocessing method to obtain a preprocessed depth image set.

[0126] Based on the above technical solution, the first point cloud dataset determination module is specifically used to: determine the point cloud dataset of the tobacco material in the camera coordinate system based on the depth image set and the internal parameters of the dual-depth camera; determine the point cloud dataset of the tobacco material in the world coordinate system based on the point cloud dataset of the tobacco material in the camera coordinate system and the external parameters of the dual-depth camera.

[0127] Based on the above technical solution, the first three-dimensional material model construction module is specifically used to: perform rough alignment on each point cloud data based on the centroid of each point cloud data in the point cloud data set to obtain a roughly aligned point cloud data set; use any point cloud data in the roughly aligned point cloud data set as source point cloud data, and use the remaining point cloud data as selected point cloud data; determine the selected pixel point corresponding to each source pixel point in the source point cloud data in the selected point cloud data through nearest neighbor search; perform point cloud fusion on the source point cloud data and the selected point cloud data based on the established association relationship between the source pixel point and the selected pixel point to construct a three-dimensional material model of the tobacco material.

[0128] On the basis of the above technical solution, the tobacco material is a roughly scraped tobacco material;

[0129] The device also includes:

[0130] A second depth image set acquisition module is used to acquire a depth image set of the unscraped tobacco material on the material conveyor belt captured by the dual depth cameras;

[0131] A second point cloud data set determination module is used to determine a point cloud data set of the unscraped tobacco material based on the depth image set and the internal and external parameters of the dual depth camera;

[0132] The second 3D material model building module is used to perform point cloud registration and point cloud fusion on the point cloud data set to build a 3D material model of the unscraped tobacco material;

[0133] The extreme point determination module is used to perform regional extreme value detection on the three-dimensional material model and determine the local maximum and local minimum points in the three-dimensional material model;

[0134] The rough scraping processing module is used to perform mean processing on the depth values corresponding to the local maximum points and the local minimum points, determine the rough scraping height of the scraper, and control the scraper to perform rough scraping processing on the unscraped tobacco material based on the rough scraping height.

[0135] The material thickness uniformity adjustment device based on a dual-depth camera provided in an embodiment of the present invention can execute the material thickness uniformity adjustment method based on a dual-depth camera provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects for executing the material thickness uniformity adjustment method based on a dual-depth camera.

[0136] It is worth noting that in the above-mentioned embodiment of material thickness uniformity adjustment based on dual-depth cameras, the various units and modules included are only divided according to functional logic, but are not limited to the above-mentioned division, as long as the corresponding functions can be achieved; in addition, the specific names of the functional units are only for the convenience of distinguishing each other, and are not used to limit the scope of protection of the present invention.

[0137] Example 4

[0138] Figure 5 A schematic diagram of the structure of an electronic device 10 that can be used to implement an embodiment of the present invention is shown. The electronic device is intended to represent various forms of digital computers, such as laptop computers, desktop computers, workstations, personal digital assistants, servers, blade servers, mainframe computers, and other suitable computers. The electronic device can also represent various forms of mobile devices, such as personal digital processing, cellular phones, smart phones, wearable devices (such as helmets, glasses, watches, etc.) and other similar computing devices. The components shown herein, their connections and relationships, and their functions are merely examples and are not intended to limit the implementation of the present invention described and / or claimed herein.

[0139] like Figure 5 As shown, the electronic device 10 includes at least one processor 11 and a memory, such as a read-only memory (ROM) 12, a random access memory (RAM) 13, etc., which is communicatively connected to the at least one processor 11. The memory stores a computer program that can be executed by the at least one processor. The processor 11 can perform various appropriate actions and processes according to the computer program stored in the read-only memory (ROM) 12 or the computer program loaded from the storage unit 18 into the random access memory (RAM) 13. Various programs and data required for the operation of the electronic device 10 can also be stored in the RAM 13. The processor 11, ROM 12, and RAM 13 are connected to each other via a bus 14. An input / output (I / O) interface 15 is also connected to the bus 14.

[0140] Multiple components in the electronic device 10 are connected to the I / O interface 15, including an input unit 16, such as a keyboard, a mouse, etc.; an output unit 17, such as various types of displays, speakers, etc.; a storage unit 18, such as a magnetic disk, an optical disk, etc.; and a communication unit 19, such as a network card, a modem, a wireless communication transceiver, etc. The communication unit 19 allows the electronic device 10 to exchange information / data with other devices via a computer network such as the Internet and / or various telecommunication networks.

[0141] The processor 11 can be any general-purpose and / or specialized processing component with processing and computing capabilities. Some examples of the processor 11 include, but are not limited to, a central processing unit (CPU), a graphics processing unit (GPU), various specialized artificial intelligence (AI) computing chips, various processors running machine learning model algorithms, a digital signal processor (DSP), and any other suitable processor, controller, microcontroller, etc. The processor 11 executes the various methods and processes described above, such as the material thickness uniformity adjustment method based on the dual-depth camera.

[0142] In some embodiments, the material thickness uniformity adjustment method based on the dual-depth camera can be implemented as a computer program, which is tangibly contained in a computer-readable storage medium, such as the storage unit 18. In some embodiments, part or all of the computer program can be loaded and / or installed on the electronic device 10 via the ROM 12 and / or the communication unit 19. When the computer program is loaded into the RAM 13 and executed by the processor 11, one or more steps of the material thickness uniformity adjustment method based on the dual-depth camera described above can be performed. Alternatively, in other embodiments, the processor 11 can be configured to execute the material thickness uniformity adjustment method based on the dual-depth camera by any other appropriate means (for example, by means of firmware).

[0143] Various embodiments of the systems and techniques described herein can be implemented in digital electronic circuit systems, integrated circuit systems, field programmable gate arrays (FPGAs), application specific integrated circuits (ASICs), application specific standard products (ASSPs), system-on-chip systems (SOCs), programmable logic devices (CPLDs), computer hardware, firmware, software, and / or combinations thereof. These various embodiments can include being implemented in one or more computer programs that are executable and / or interpreted on a programmable system that includes at least one programmable processor, which can be a special purpose or general purpose programmable processor that can receive data and instructions from a storage system, at least one input device, and at least one output device, and transmit data and instructions to the storage system, the at least one input device, and the at least one output device.

[0144] Computer programs for implementing the methods of the present invention may be written in any combination of one or more programming languages. These computer programs may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing device, such that when the computer program is executed by the processor, the functions / operations specified in the flowcharts and / or block diagrams are implemented. The computer program may be executed entirely on the machine, partially on the machine, as a stand-alone software package, partially on the machine and partially on a remote machine, or entirely on a remote machine or server.

[0145] In the context of the present invention, computer-readable storage media can be tangible media that can contain or store a computer program for use with an instruction execution system, device or equipment or used in combination with an instruction execution system, device or equipment. Computer-readable storage media can include but are not limited to electronic, magnetic, optical, electromagnetic, infrared or semiconductor systems, devices or equipment, or any suitable combination of the foregoing. Alternatively, computer-readable storage media can be machine-readable signal media. More specific examples of machine-readable storage media can include electrical connections based on one or more lines, portable computer disks, hard disks, random access memories (RAM), read-only memories (ROM), erasable programmable read-only memories (EPROM or flash memory), optical fibers, portable compact disk read-only memories (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0146] To provide interaction with a user, the systems and techniques described herein can be implemented on an electronic device having: a display device (e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor) for displaying information to the user; and a keyboard and pointing device (e.g., a mouse or trackball) through which the user can provide input to the electronic device. Other types of devices can also be used to provide interaction with the user; for example, the feedback provided to the user can be any form of sensory feedback (e.g., visual feedback, auditory feedback, or tactile feedback); and input from the user can be received in any form (including acoustic input, voice input, or tactile input).

[0147] The systems and techniques described herein can be implemented in a computing system that includes back-end components (e.g., as a data server), or a computing system that includes middleware components (e.g., an application server), or a computing system that includes front-end components (e.g., a user computer with a graphical user interface or web browser through which a user can interact with implementations of the systems and techniques described herein), or a computing system that includes any combination of such back-end components, middleware components, or front-end components. The components of the system can be interconnected by any form or medium of digital data communication (e.g., a communication network). Examples of communication networks include: a local area network (LAN), a wide area network (WAN), a blockchain network, and the Internet.

[0148] A computing system may include clients and servers. The clients and servers are typically remote from each other and typically interact via a communication network. This client-server relationship arises through computer programs running on the respective computers, creating a client-server relationship. The server may be a cloud server, also known as a cloud computing server or cloud host. This server is a hosting product within the cloud computing service ecosystem that addresses the management difficulties and limited scalability of traditional physical hosting and VPS services.

[0149] An embodiment of the present invention further provides a computer program product, including a computer program, which, when executed by a processor, implements the material thickness uniformity adjustment method based on a dual-depth camera as provided in any embodiment of the present application.

[0150] During the implementation of the computer program product, computer program code for performing the operations of the present invention can be written in one or more programming languages or a combination thereof, including object-oriented programming languages such as Java, Smalltalk, C++, and conventional procedural programming languages such as "C" language or similar programming languages. The program code can be executed entirely on the user computer, partially on the user computer, as a separate software package, partially on the user computer and partially on a remote computer, or entirely on a remote computer or server. In the case of a remote computer, the remote computer can be connected to the user computer through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computer (for example, through the Internet using an Internet service provider). This program product and the material thickness uniformity adjustment method based on a dual-depth camera disclosed in each embodiment of the present application belong to the same inventive concept, and therefore will not be described in detail here.

[0151] It should be understood that the various forms of the processes shown above can be used to reorder, add, or delete steps. For example, the steps described in the present invention can be performed in parallel, sequentially, or in a different order, as long as the desired results of the technical solution of the present invention can be achieved. This is not limited herein.

[0152] The above specific embodiments do not limit the scope of protection of the present invention. Those skilled in the art will appreciate that various modifications, combinations, sub-combinations, and substitutions may be made based on design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention are intended to be included within the scope of protection of the present invention.

Claims

1. A method for adjusting material thickness uniformity based on a dual-depth camera, characterized in that: include: Obtaining a depth image set of tobacco materials on a material conveyor belt captured by a dual depth camera; Determining a point cloud dataset of the tobacco material based on the depth image set and the internal and external parameters of the dual depth camera; wherein the internal parameters are determined by a checkerboard calibration method; and the external parameters are determined by a calibration plate calibration method; Performing point cloud registration and point cloud fusion on the point cloud data set to construct a three-dimensional material model of the tobacco material; Based on the three-dimensional material model and the pre-built scraper physical model, an objective function is established with material thickness uniformity as the optimization goal; Solving the maximum value of the objective function based on a gradient method, determining the flatness information of the scraper, and controlling the scraper to adjust the thickness uniformity of the tobacco material based on the flatness information; The leveling information includes: leveling height, leveling speed and stroke.

2. The method according to claim 1, characterized in that Before acquiring the depth image set of the tobacco material on the material conveyor belt captured by the dual-depth camera, the method further includes: Obtaining a material arrival signal sent by the limit electric eye, and generating an image acquisition signal for waking up the dual-depth camera based on the material arrival signal; Based on the image acquisition signal, the dual depth cameras are controlled to acquire a depth image set of the tobacco material on the material conveyor belt.

3. The method according to claim 1, characterized in that Before determining the point cloud dataset of the tobacco material, the method further includes: Image preprocessing is performed on each depth image in the depth image set based on a preset image preprocessing method to obtain a preprocessed depth image set.

4. The method according to claim 1, wherein The step of determining a point cloud data set of the tobacco material based on the depth image set and the internal and external parameters of the dual depth camera comprises: Determining a point cloud dataset of the tobacco material in a camera coordinate system based on the depth image set and the internal parameters of the dual depth camera; Based on the point cloud dataset of the tobacco material in the camera coordinate system and the external parameters of the dual-depth camera, the point cloud dataset of the tobacco material in the world coordinate system is determined.

5. The method according to claim 1, wherein The performing point cloud registration and point cloud fusion on the point cloud data set to construct a three-dimensional material model of the tobacco material includes: Performing coarse registration on each point cloud data based on the centroid of each point cloud data in the point cloud data set to obtain a coarsely registered point cloud data set; Any point cloud data in the roughly registered point cloud data set is used as the source point cloud data, and the remaining point cloud data is used as the candidate point cloud data; Determine, in the candidate point cloud data, a candidate pixel point corresponding to each source pixel point in the source point cloud data by nearest neighbor search; Point cloud fusion is performed based on the source point cloud data with established association relationships between the source pixel points and the pixel points to be selected and the point cloud data to be selected, so as to construct a three-dimensional material model of the tobacco material.

6. The method according to claim 1, characterized in that The tobacco material is a roughly scraped tobacco material; Before acquiring the depth image set of the tobacco material on the material conveyor belt captured by the dual-depth camera, the method further includes: Obtaining a depth image set of unscraped tobacco material on a material conveyor belt captured by a dual-depth camera; Determining a point cloud dataset of the unscraped tobacco material based on the depth image set and the internal and external parameters of the dual depth camera; performing point cloud registration and point cloud fusion on the point cloud data set to construct a three-dimensional material model of the unscraped tobacco material; Performing regional extreme value detection on the three-dimensional material model to determine each local maximum point and local minimum point in the three-dimensional material model; The depth values corresponding to the local maximum point and the local minimum point are averaged to determine the rough scraping height of the scraper, and the scraper is controlled to perform rough scraping on the unscraped tobacco material based on the rough scraping height.

7. The method according to claim 1, characterized in that The dual depth camera includes: a first depth camera and a second depth camera; the first depth camera and the second depth camera are arranged with a cross-perspective layout at a preset angle; the first depth camera is a depth camera arranged on the left side of the positive direction of conveying materials along the material conveyor belt; the second depth camera is a depth camera arranged on the right side of the positive direction of conveying materials along the material conveyor belt; the depth image set includes: a first depth image and a second depth image; the first depth image is a depth image captured by the first depth camera at a perspective slightly to the left of the tobacco material; the second depth image is a depth image captured by the second depth camera at a perspective slightly to the right of the tobacco material.

8. A material thickness uniformity adjustment device based on a dual-depth camera, characterized in that: The device comprises: A first depth image set acquisition module is used to acquire a depth image set of the tobacco material on the material conveyor belt captured by the dual depth camera; a first point cloud data set determination module, configured to determine a point cloud data set of the tobacco material based on the depth image set and internal and external parameters of the dual depth camera; the internal parameters are determined by a checkerboard calibration method; and the external parameters are determined by a calibration plate calibration method; a first three-dimensional material model construction module, configured to perform point cloud registration and point cloud fusion on the point cloud dataset to construct a three-dimensional material model of the tobacco material; An objective function establishment module is used to establish an objective function based on the three-dimensional material model and the pre-built scraper physical model, with the material thickness uniformity as the optimization target; The thickness uniformity adjustment module is used to solve the maximum value of the objective function based on the gradient method, determine the leveling information of the scraper, and control the scraper to adjust the thickness uniformity of the tobacco material based on the leveling information; the leveling information includes: leveling height, leveling speed and stroke.

9. An electronic device, characterized in that: The electronic device comprises: one or more processors; a memory for storing one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the material thickness uniformity adjustment method based on the dual-depth camera as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the material thickness uniformity adjustment method based on a dual-depth camera as described in any one of claims 1 to 7 is implemented.

Images