Device and method for measuring density of solid product

Abstract

The invention discloses a solid product density measuring device which comprises a robot system and a visual system used for collecting point cloud on the surface of a product, the visual system comprises a 3D data collecting unit arranged on the robot system, and the robot system is arranged to be used for controlling the product and the 3D data collecting unit to move among different positions. According to the solid product density measurement device, the accuracy and the automation level of solid product density measurement can be improved. The invention further discloses a solid product density measuring method.

Description

Technical Field

[0001] The present invention belongs to the technical field of automatic detection devices, and in particular, relates to a device and method for measuring the density of a solid product. Background Art

[0002] Product density testing is a key indicator for evaluating the physical properties of products. It plays a vital role in many industries, including but not limited to petroleum, chemicals, building materials, light industry and medical equipment manufacturing, international trade, national defense security, and scientific research. This parameter plays an indispensable role in ensuring product quality, optimizing production processes, and promoting the development of new products. By accurately measuring the density of products, companies can better understand the purity and chemical stability of materials and whether they meet relevant standard requirements, thereby providing strong data support for production and market circulation.

[0003] Density is defined as the amount of mass contained in a unit volume of a substance. Its mathematical expression is ρ = m / V, where ρ represents density, m represents the mass of the object, and V is the volume of space occupied by the object. Based on this basic formula, density measurement can be achieved through direct or indirect methods, that is, first determine the mass and volume information of the sample to be tested, and then calculate its specific density value based on this.

[0004] When it comes to volume measurement of solid products, it is particularly important to choose the right technical means, which mainly depends on the specific morphological characteristics (such as geometric shape), size, type of material, and the requirements for the accuracy of the measurement results of the object being measured. For those solid objects with regular shapes and easy to describe (such as cubes, cuboids or cylindrical structures), the existing technology can usually directly and accurately measure the length of each dimension and quickly obtain the total volume data based on the corresponding mathematical model. However, when faced with objects with complex and changeable shapes that are difficult to quantify using conventional methods, a more flexible and effective alternative is needed, such as the displacement method. The basic principle of this technology is to completely immerse the target object in a container filled with a known volume of liquid (usually water), observe and record the changes in the liquid level caused by the placement of the object, and then calculate the actual liquid volume displaced by the object, that is, the volume of the object itself. It is worth noting that the prerequisite for adopting such an indirect measurement strategy is that the sample being measured will neither dissolve in the selected medium nor produce any chemical reaction with it that may affect the accuracy of the final result.

[0005] A Chinese patent with application number 202420164996.9 discloses a fully automatic density meter, including a weighing sensor, which is connected to a driving mechanism for driving the weighing sensor to rise and fall, the weighing sensor and the driving mechanism are electrically connected to a control host respectively, a tray is suspended below the weighing sensor by a suspension rope, and a liquid storage tank with a top opening is provided below the tray; the weighing sensor can be driven to rise and fall by the driving mechanism, and the weighing sensor can drive the tray to rise and fall by the suspension rope, so that the solid to be measured in the tray automatically descends and is immersed in the liquid in the liquid storage tank for weighing after being weighed in the air. The user does not need to measure the weight of the solid to be measured in the air once and then use a clamp such as tweezers or pliers to transfer the solid to be measured and immerse it in the liquid to measure the weight once, thereby making it more convenient to measure the density of the solid to be measured and reducing the probability of measurement errors.

[0006] It is desirable to provide an improved solid product density measuring device, particularly with respect to a shielding member that improves the appearance shielding effect of the corner adjustment area. Summary of the invention

[0007] The present invention aims to solve at least one of the technical problems existing in the prior art. To this end, the present invention provides a solid product density measuring device, the purpose of which is to improve the accuracy and automation level of solid product density measurement.

[0008] In order to achieve the above-mentioned purpose, the technical solution adopted by the present invention is: a solid-state product density measuring device, including a robot system and a visual system for collecting product surface point clouds, the visual system includes a 3D data acquisition unit arranged on the robot system, and the robot system is configured to control the product and the 3D data acquisition unit to move between different positions.

[0009] The 3D data acquisition unit is a planar array 3D camera.

[0010] The robot system is provided with a first clamp unit for clamping products, and the 3D data acquisition unit is located at one side of the first clamp unit.

[0011] The solid product density measuring device further comprises a centering fixture, and the product centering fixture is configured to clamp the product and realize the centering of the product.

[0012] The solid product density measuring device also includes a weighing and detecting platform, which is configured to measure product quality.

[0013] A detection placement mechanism is arranged on the upper part of the weighing detection platform.

[0014] The solid product density measuring device further comprises a second clamp unit, which is configured to clamp the product during the loading and unloading process.

[0015] The present invention also provides a solid product density measurement method, which uses the solid product density measurement device and includes the steps of:

[0016] S1. The product reaches the detection position and the detection is triggered;

[0017] S2, the robot carries the 3D data acquisition unit to the detection position to photograph the product;

[0018] S3, locate the product position and posture;

[0019] S4, robot grabs products;

[0020] S5. The robot places the product on the weighing and testing platform for weighing;

[0021] S6. The robot carries the 3D data acquisition unit and moves to the weighing and testing platform to take pictures of the product;

[0022] S7. Locate the product position and posture;

[0023] S8, robot grabs products;

[0024] S9, the robot places the product into the centering fixture and clamps it for centering;

[0025] S10, the robot takes the first photo of the product with the 3D data acquisition unit according to the pre-set teaching position, and collects point cloud data of the product surface at multiple positions and postures;

[0026] S11. After the robot grabs the product, it turns the product over and places it on the transfer table;

[0027] S12, the robot clamps the product and places it in the centering fixture for clamping and centering;

[0028] S13, the robot takes a second photo of the product with the 3D data acquisition unit according to the pre-set teaching position, and collects point cloud data of the product surface at multiple positions and postures;

[0029] S14. Splicing surface point cloud data of products with multiple positions and postures;

[0030] S15. Compute product height, diameter and actual volume based on the point cloud of the spliced ​​product surface;

[0031] S16, calculating the density of the product according to the actual volume and mass of the product, and outputting the calculation result for display;

[0032] S17. The robot grabs the product and unloads it.

[0033] In step S3 and step S7, when the visual system locates the product position, a coarse positioning algorithm based on 2D image and 3D point cloud fusion and a fine positioning algorithm based on ICP point cloud registration are used.

[0034] When the coarse positioning algorithm based on the fusion of 2D images and 3D point clouds is used to locate the position of the product, the point cloud blocks containing the product point clouds are first segmented by clustering using the Euclidean clustering algorithm of the point cloud, and then the point cloud blocks containing the product point clouds are projected onto the plane through plane projection, and then mapped to the deep 2D image through the positional relationship of the point cloud on the plane, each pixel in the 2D image represents the projection depth, and finally the edge is extracted in the deep 2D image, and the position of the product in the projection plane is obtained by template matching, and then the position of the product located in the plane is reversely projected to obtain the actual position of the product.

[0035] The solid product density measuring device of the present invention can improve the accuracy and automation level of solid product density measurement. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] This specification includes the following drawings, which show the following contents:

[0037] Figure 1 It is a structural schematic diagram of a solid product density measuring device of the present invention;

[0038] Figure 2 is a flow chart of a method for measuring density of a solid product of the present invention;

[0039] The following are marked in the figure: 1. Robot system; 2. Control system; 3. Weighing and detection platform; 4. 3D data acquisition unit and first fixture unit; 5. Centering fixture; 6. Second fixture unit; 7. Loading and unloading trays. DETAILED DESCRIPTION

[0040] The specific implementation methods of the present invention are further explained in detail below by describing the embodiments with reference to the accompanying drawings, with the aim of helping those skilled in the art to have a more complete, accurate and in-depth understanding of the concept and technical solution of the present invention and facilitating its implementation.

[0041] It should be noted that, in the following embodiments, the “first” and “second” mentioned do not represent an absolute distinction in structure and / or function, nor do they represent a sequential order of execution, but are merely for the convenience of description.

[0042] like Figure 1As shown, the present invention provides a solid product density measuring device, including a control system 2, a robot system 1 and a visual system for collecting product surface point clouds, the visual system includes a 3D data acquisition unit arranged on the robot system, and the robot system 1 is configured to control the product and the 3D data acquisition unit to move between different positions.

[0043] Specifically, the present invention aims to solve the problem of high-precision volume measurement of solid complex-shaped products. The 3D surface structured light sensor is used to scan the product surface point cloud at multiple positions and angles, and the complete product surface point cloud is spliced ​​together. The algorithm is used to quickly and accurately measure the volume of the product, thereby improving the accuracy and automation level of product density measurement.

[0044] In an embodiment of the present invention, the 3D data acquisition unit is an array 3D camera, which mainly uses the 3D camera to collect the three-dimensional point cloud of the product surface, calculates its volume by reconstructing the three-dimensional model of the product, and uses a weighing sensor to measure its mass, thereby calculating the density of the product.

[0045] like Figure 1 As shown, the robot system 1 is provided with a first fixture unit for gripping products, and the 3D data acquisition unit is located on one side of the first fixture unit. The robot system includes a robot and a robot base, which is used to load the 3D data acquisition unit and move it to different positions; at the same time, a first fixture unit is installed on one side of the 3D data acquisition unit, and the first fixture unit loads and unloads products before and after the product inspection is completed.

[0046] Before measuring the product density, the position and posture relationship between the 3D data acquisition unit and the robot needs to be calibrated in advance. It is also necessary to set the product model to be detected and teach the robot's position and posture for each model of product to meet the needs of complete product photography.

[0047] like Figure 1 As shown, the control system 2 includes electrical components such as a host computer, a PLC, and a control cabinet. The control system is connected to the robot, the first fixture unit, and the 3D data acquisition unit.

[0048] like Figure 1 As shown, the solid product density measuring device of the present invention further comprises a centering fixture 5, and the product centering fixture 5 is configured to clamp the product and realize the centering of the product.

[0049] like Figure 1 As shown, the solid product density measuring device of the present invention further includes a weighing detection platform 3, which is configured to measure product quality, and a weighing sensor is disposed on the weighing detection platform 3. A detection placement mechanism is disposed on the upper portion of the weighing detection platform 3 to prevent the robot from directly contacting the weighing sensor when picking up and placing materials, thereby avoiding damage to the weighing sensor.

[0050] like Figure 1 As shown, the solid product density measuring device of the present invention further comprises a second clamp unit 6, and the second clamp unit 6 is configured to clamp the product during the loading and unloading process of the product.

[0051] like Figure 1 As shown, the solid product density measuring device of the present invention further comprises a loading and unloading tray 7, and the loading and unloading tray 7 is used for storing products.

[0052] like Figure 1 As shown, the present invention also provides a solid product density measurement method, which adopts the solid product density measurement device of the above structure and includes the following steps:

[0053] S1. The product reaches the detection position and the detection is triggered;

[0054] S2. The robot carries the 3D data acquisition unit to the inspection location to photograph the product;

[0055] S3, locate the product position and posture;

[0056] S4, robot grabs products;

[0057] S5. The robot places the product on the weighing and testing platform for weighing;

[0058] S6. The robot carries the 3D data acquisition unit and moves to the weighing and testing platform to take pictures of the product;

[0059] S7. Locate the product position and posture;

[0060] S8, robot grabs products;

[0061] S9, the robot places the product into the centering fixture and clamps it for centering;

[0062] S10, the robot takes the first photo of the product with the 3D data acquisition unit according to the pre-set teaching position, and collects point cloud data of the product surface at multiple positions and postures;

[0063] S11. After the robot grabs the product, it turns the product over and places it on the transfer table;

[0064] S12, the robot clamps the product and places it in the centering fixture for clamping and centering;

[0065] S13, the robot takes a second photo of the product with the 3D data acquisition unit according to the pre-set teaching position, and collects point cloud data of the product surface at multiple positions and postures;

[0066] S14. Splicing surface point cloud data of products with multiple positions and postures;

[0067] S15. Compute product height, diameter and actual volume based on the point cloud of the spliced ​​product surface;

[0068] S16, calculating the density of the product according to the actual volume and mass of the product, and outputting the calculation result for display;

[0069] S17. The robot grabs the product and unloads it.

[0070] In the above step S1, the product whose density needs to be measured is placed in the loading and unloading tray to trigger the detection.

[0071] In the above step S2, the robot carries the 3D data acquisition unit and moves to the detection position. The 3D data acquisition unit takes a preliminary photo of the product to obtain preliminary three-dimensional information of the product.

[0072] In the above steps S3 and S7, the visual system uses a coarse positioning algorithm based on 2D image and 3D point cloud fusion and a fine positioning algorithm based on ICP point cloud registration when locating the product position.

[0073] In the above step S5, the robot moves the product to the weighing and testing platform, weighs the product, and obtains the quality of the product.

[0074] In the above step S6, the robot carries the 3D data acquisition unit to photograph the product again to obtain three-dimensional information of the product under weighing.

[0075] In the above step S9, the robot moves the product to the centering fixture to ensure that the product is in the correct position and orientation for more accurate 3D data acquisition.

[0076] In the above step S10, the robot carries a 3D data acquisition unit to shoot the product from multiple angles according to the preset trajectory and position, and collects comprehensive product surface point cloud data.

[0077] In the above step S11 , the robot turns the product over so as to photograph the other side of the product.

[0078] In the above step S12, the flipped product is centered again to prepare for a second shot.

[0079] In the above step S13, the robot carries the 3D data acquisition unit to take multi-angle photos of the product again according to the preset trajectory and position, and collects comprehensive product surface point cloud data.

[0080] In the above step S14, the point cloud data obtained by taking pictures in step S10 and step S13 are spliced ​​to form a complete three-dimensional model of the product, thereby completing the reconstruction process of the three-dimensional model of the product.

[0081] In the solid product density measurement method of the embodiment of the present invention, a 3D visual positioning algorithm is adopted. The 3D visual positioning algorithm is mainly composed of two parts: 1 is a coarse positioning algorithm based on the fusion of 2D image and 3D point cloud, and 2 is a fine positioning algorithm based on point cloud registration of iterative closest point (ICP).

[0082] When using a coarse positioning algorithm based on the fusion of 2D images and 3D point clouds to locate the product position, the Euclidean clustering algorithm of the point cloud is first used to segment the point cloud blocks containing the product point cloud by clustering, and then the point cloud blocks containing the product point cloud are projected onto the plane through plane projection, and then mapped to the deep 2D image through the positional relationship of the point cloud on the plane. Each pixel in the 2D image represents the projection depth, and finally the edge is extracted in the deep 2D image, and the position of the product in the projection plane is obtained by template matching, and the position of the product located in the plane is reversely projected to obtain the actual position of the product.

[0083] Since there will be a loss in position accuracy when mapping the position relationship of the point cloud on the plane to the deep 2D image, the positioning accuracy cannot meet the grasping requirements, and the ICP registration algorithm is required to further accurately locate the product position.

[0084] In the above step S15, a volume measurement algorithm based on the product surface point cloud is used to calculate the actual volume of the product. The volume measurement algorithm based on the product surface point cloud is mainly performed by plane segmentation, calculating the area inside the point cloud after segmentation, and then multiplying it by the segmentation thickness to obtain the volume of the segmented point cloud. Finally, the volumes of all segmented point clouds are accumulated to obtain the actual volume of the entire product.

[0085] Regarding the calculation of the area inside the point cloud after segmentation, the segmented point cloud is first projected onto the segmentation plane, and then the projected point cloud is refined and segmented to obtain multiple groups of closed point cloud curves. Bezier curve fitting is performed on each group of closed point cloud curves to obtain the fitting curve, and then the area enclosed by each fitting curve is calculated. By analyzing the enclosing relationship of the fitting curve, it can be determined whether the area enclosed by the fitting curve belongs to the inside or outside of the product. The area of ​​the point cloud belonging to the inside of the product after segmentation is obtained by adding up the areas enclosed by all the curves belonging to the inside of the product and then subtracting the areas enclosed by all the curves belonging to the outside of the product.

[0086] By plane segmenting the product surface point cloud and calculating the volume of each group of segmented point clouds, the product volume can be quickly and accurately measured. This method not only improves the accuracy of measurement, but also enhances the flexibility and adaptability of measurement.

[0087] The solid-state product density measurement device and method of the embodiment of the present invention adopts non-contact 3D scanning technology to capture and scan the 3D point cloud of the product surface, and uses the product surface point cloud to quickly and accurately calculate the product volume, thereby obtaining the product density, thereby achieving fast and accurate non-contact automatic measurement of solid-state products.

[0088] The rough positioning algorithm based on the fusion of 2D images and 3D point clouds and the fine positioning algorithm based on ICP point cloud registration can realize fast and accurate positioning of products, providing guarantee for the grabbing and handling of products. This fusion technology not only improves the accuracy and efficiency of positioning, but also provides a solid foundation for subsequent precise positioning. By plane segmenting the point cloud on the surface of the product and calculating the volume of each group of segmented point clouds, the volume of the product can be measured quickly and accurately.

[0089] The solid product density measurement device and method of the embodiment of the present invention avoids the errors or damage that may be caused by contact in the traditional measurement method. This is especially important for fragile, delicate or surface-sensitive products. The automated measurement process greatly improves the measurement efficiency, reduces the reliance on manual operation, and reduces the risk of human error.

[0090] The present invention is described above by way of example in conjunction with the accompanying drawings. Obviously, the specific implementation of the present invention is not limited to the above-mentioned methods. As long as various non-substantial improvements are made using the method concept and technical solution of the present invention; or the above concept and technical solution of the present invention are directly applied to other occasions without improvement, they are all within the protection scope of the present invention.

Claims

1. A solid product density measuring device, characterized in that: It includes a robot system and a vision system for collecting surface point clouds of products. The vision system includes a 3D data collection unit arranged on the robot system. The robot system is configured to control the product and the 3D data collection unit to move between different positions.

2. The solid product density measuring device according to claim 1, characterized in that: The 3D data acquisition unit is a planar array 3D camera.

3. The solid product density measuring device according to claim 1, characterized in that: The robot system is provided with a first clamp unit for clamping products, and the 3D data acquisition unit is located at one side of the first clamp unit.

4. The solid product density measuring device according to any one of claims 1 to 3, characterized in that: Also included is a centering fixture, which is configured to clamp the product and achieve centering of the product.

5. The solid product density measuring device according to any one of claims 1 to 3, characterized in that: It also includes a weighing and testing platform, which is configured to measure product quality.

6. The solid product density measuring device according to claim 5, characterized in that: A detection placement mechanism is arranged on the upper part of the weighing detection platform.

7. The solid product density measuring device according to any one of claims 1 to 3, characterized in that: It also includes a second clamp unit, which is configured to clamp the product during the product loading and unloading process.

8. A method for measuring the density of a solid product, characterized in that: The solid product density measuring device according to any one of claims 1 to 7 is used, and comprises the steps of: S1. The product reaches the detection position and the detection is triggered; S2, the robot carries the 3D data acquisition unit to the detection position to photograph the product; S3, locate the product position and posture; S4, robot grabs products; S5. The robot places the product on the weighing and testing platform for weighing; S6. The robot carries the 3D data acquisition unit and moves to the weighing and testing platform to take pictures of the product; S7. Locate the product position and posture; S8, robot grabs products; S9, the robot places the product into the centering fixture and clamps it for centering; S10, the robot takes the first photo of the product with the 3D data acquisition unit according to the pre-set teaching position, and collects point cloud data of the product surface at multiple positions and postures; S11. After the robot grabs the product, it turns the product over and places it on the transfer table; S12, the robot clamps the product and places it in the centering fixture for clamping and centering; S13, the robot takes a second photo of the product with the 3D data acquisition unit according to the pre-set teaching position, and collects point cloud data of the product surface at multiple positions and postures; S14. Splicing surface point cloud data of products with multiple positions and postures; S15. Compute product height, diameter and actual volume based on the point cloud of the spliced ​​product surface; S16, calculating the density of the product according to the actual volume and mass of the product, and outputting the calculation result for display; S17. The robot grabs the product and unloads it.

9. The solid product density measuring device according to claim 8, characterized in that: In step S3 and step S7, when the visual system locates the product position, a coarse positioning algorithm based on 2D image and 3D point cloud fusion and a fine positioning algorithm based on ICP point cloud registration are used.

10. The solid product density measuring device according to claim 9, characterized in that: When the coarse positioning algorithm based on the fusion of 2D images and 3D point clouds is used to locate the position of the product, the point cloud blocks containing the product point clouds are first segmented by clustering using the Euclidean clustering algorithm of the point cloud, and then the point cloud blocks containing the product point clouds are projected onto the plane through plane projection, and then mapped to the deep 2D image through the positional relationship of the point cloud on the plane, each pixel in the 2D image represents the projection depth, and finally the edge is extracted in the deep 2D image, and the position of the product in the projection plane is obtained by template matching, and then the position of the product located in the plane is reversely projected to obtain the actual position of the product.

Images