Online density measurement method and system

Abstract

The invention discloses an online density measurement method and system. The system is composed of a machine vision unit, an assembly line, a volume calculation unit, a weighing unit, a density calculation unit and a measurement processing unit. The density measurement method comprises the following steps: 1) the machine vision unit calculates the contour 3D point cloud of the surface of a measured object on an assembly line by using a 3D camera, and performs detection imaging on the internal structure of the measured object by using an X-ray camera to generate internal 2D image information; 2) a volume calculation unit calculates the volume of the measured object according to the 3D point cloud of the surface profile; 3) a weighing unit weighs the measured object on the assembly line to obtain the mass of the measured object; 4) a density calculation unit calculates the density of the measured object; and 5) comparing the density and the internal 2D image of the measured object with a detection standard value by the measurement processing unit, and processing an unqualified detected object. The device has the advantages that the density of the measured object can be measured in real time in an online and non-contact mode, the measured object cannot be damaged, and the product quality of the measured object on a production line can be improved.

Description

Technical Field

[0001] This patent belongs to the fields of intelligent manufacturing, machine vision, and measuring instruments. Background Technology

[0002] Current methods for measuring the density of objects are all contact methods. This involves first weighing the object and its waterproof covering material, then wrapping the object in the waterproof material and placing it in a container filled with solvent. The volume of the object is obtained by measuring the change in solvent level, and finally, the density is calculated by dividing the weight by the volume. Examples include density measurement methods and methods for measuring the density of calcined products of radioactive waste liquids (patent application number...).

[0003] 202110684206.0. A method for measuring the density of an object (CN201410503100.6A) involves placing the object at the temperature required for density measurement, measuring the object's mass, applying negative pressure twice to the environment surrounding the object, and measuring the temperature, volume, and pressure of the gas during the two negative pressure measurements. The density value of the object is then calculated using a formula.

[0004] Current methods are all contact-based measurements, which require complex human-machine interaction, can damage the object being measured, and cannot be used to measure products in production on the production line in real time. Summary of the Invention

[0005] To address the aforementioned problems, this invention discloses an online density measurement method and system. The system comprises a machine vision unit, an assembly line, a volume calculation unit, a weighing unit, a density calculation unit, and a measurement processing unit. The density measurement method is as follows: 1) The machine vision unit uses a 3D camera to calculate the 3D point cloud of the surface contour of the object being measured on the assembly line, and uses an X-ray camera to detect and image the internal structure of the object, generating an internal 2D image; 2) The volume calculation unit calculates the volume of the object being measured based on the 3D point cloud of the surface contour; 3) The weighing unit weighs the object being measured on the assembly line to obtain its mass; 4) The density calculation unit calculates the density of the object being measured; 5) The measurement processing unit compares the density and internal 2D image of the object being measured with a detection standard value, and processes unqualified objects. The advantage of this invention is that it can measure the density of the object being measured in real time online and non-contactly, improving the product quality of the measured object on the production line.

[0006] To achieve the above objectives, the present invention provides the following two methods:

[0007] 1. A non-contact 3D camera is used to scan and image the 3D point cloud of the measured object's outline, thereby calculating the object's volume. Simultaneously, an online real-time electronic scale is used to obtain the object's mass, and the packing density is calculated using the following formula:

[0008] Density = mass / volume

[0009] 2. Using a non-contact X-ray camera, an internal 2D image of the object being measured is detected, and its X-ray absorption rate is then calculated. A template object with known density, belonging to the same product category as the object being measured, is used beforehand. Its internal 2D image and X-ray absorption rate are obtained using an X-ray camera and used as template parameters. The internal 2D image and absorbance of the object being measured are compared with the template parameters to map the density of the object being measured.

[0010] The machine vision unit is the main device for non-contact optical measurement. This invention provides two methods for obtaining the contour and internal structure of a measured object: First, a 3D camera using visible light laser line scanning or a 3D structured light array camera obtains 3D point cloud information of the contour, and then calculates the volume of the measured object. Second, a non-visible light X-ray camera or gamma-ray camera probes the interior of the measured object, obtaining its internal structure and X-ray absorption rate, while simultaneously detecting impurities, cracks, and the uniformity of molecular crystallization, serving as indicators of density uniformity. Preferably, a machine learning method is used to train a model using the internal 2D image and X-ray absorption rate of a template object with known density. The density calculation unit then performs inference based on the model on the measured object online, thereby obtaining the density of the measured object.

[0011] In this invention, multiple 3D cameras are calibrated in a unified coordinate system. Based on the 3D point cloud of the object's contour and the height function z(x,y) within this unified coordinate system, the volume is calculated by integrating the data over the x[x1,xn] and y[y1,ym] intervals containing the object. In the specific computer software implementation, the embodiments of this patent use a quadratic summation method to calculate the volume integral.

[0012] Preferably, the two-dimensional coordinate system of the 3D camera is calibrated with the two-dimensional coordinate system of the non-visible X-ray camera, and the system can determine the uniformity of the internal density of the measured object.

[0013] The weighing unit of this invention is an online weighing electronic scale that can measure the weight of an object in motion. The accuracy of the electronic scale is derived from a density standard. Attached Figure Description

[0014] Figure 1 A schematic diagram illustrating the principle of online 3D camera scanning and weighing for density calculation using this invention.

[0015] Figure 2 Schematic diagram of the principle of measuring density using the X-ray camera of this invention.

[0016] Figure 3 For applying the first embodiment of the present invention Detailed Implementation

[0017] Figure 1 A schematic diagram illustrating the principle of online 3D camera scanning and weighing for density calculation using this invention.

[0018] 101. 3D cameras acquire images and 3D point clouds of the measured object's contour online. In this embodiment, multiple 3D cameras acquire the contour of the measured object. The multiple 3D cameras are connected to a computer, and computer software stitches together the point clouds from the multiple 3D cameras to form a complete contour 3D point cloud.

[0019] 102. Calculate the volume of the object being measured. After obtaining the 3D point cloud and the height function z(x,y), the computer calculates the volume according to the volume calculation formula.

[0020] 103. The mass is obtained through online weighing and then transmitted to the computer.

[0021] 104. Computer calculation of the density of the measured object = mass / volume.

[0022] Figure 2 Schematic diagram of the principle of measuring density using the X-ray camera of this invention.

[0023] The 201X-ray camera can detect the 2D image and absorptivity inside the measured object online. In this embodiment, the X-ray camera consists of an X-ray emitter and an X-ray detector, and the detector is connected to a computer.

[0024] 202. Based on the template model, reason and judge the measured object. The template model is stored in advance in the computer, and reason and judge the 2D image and absorption rate of the measured object based on the template model.

[0025] 203. Calculate the density of the measured object based on the density of the template object, and calculate the density of the measured object based on the known density of the template object.

[0026] 204. The density of the template object is obtained by using a densitometer. A standard template object is selected in advance, and its density is measured and obtained by a densitometer.

[0027] A 205X-ray camera was used to obtain the internal 2D image and absorption rate of the template object, and the template model was trained.

[0028] Figure 3 For applying the first embodiment of the present invention

[0029] This embodiment is an online device for measuring the density of ceramic tiles. A template ceramic tile is selected in advance and its density is measured using a densitometer. An X-ray camera is used to obtain the internal 2D image and X-ray absorption rate of the template ceramic tile. Based on these data, a template model is trained using machine learning software on a computer.

[0030] The 301 online weighing device is a production line type electronic scale that is connected to a computer.

[0031] 302 The tile being measured;

[0032] The 303 machine vision unit has a field of view consistent with the width of the tile being measured. The machine vision unit connects to a computer and consists of two components: 1) two 3D line scan cameras; 2) an X-ray emitter and an X-ray detector.

[0033] The computer infers and judges the internal 2D image and X-ray absorption rate of the tile being tested and the template model, and maps the density of the tile being tested based on the density of the known template tile.

[0034] 304 stainless steel production line, conveying ceramic tiles.

Claims

1. An online density measurement method and system, the system comprising: a machine vision unit, an assembly line, a volume calculation unit, a weighing unit, a density calculation unit, and a measurement processing unit; the density measurement method is as follows: 1) The machine vision unit uses a 3D camera to calculate the 3D point cloud of the surface contour of the object being measured on the assembly line, and uses an X-ray camera to detect and image the internal structure of the object being measured to generate internal 2D image information; 2) The volume calculation unit calculates the volume of the object being measured based on the 3D point cloud of the surface contour; 3) The weighing unit weighs the object being measured on the assembly line to obtain its mass; 4) The density calculation unit calculates the density of the object being measured; 5) The measurement processing unit compares the density and internal 2D image of the object being measured with the detection standard value, and processes unqualified objects being measured.

2. The method according to claim 1, characterized in that, 3D machine vision unit imaging of the surface contour of a measured object on an assembly line refers to imaging the surface contour of the measured object by setting multiple 3D cameras at multiple different angles, with the multiple 3D cameras calibrated in a unified coordinate system, and stitching the point clouds generated by the multiple 3D cameras into a complete 3D point cloud information of the measured object contour.

3. The method and system according to claims 1 and 2, characterized in that, The volume calculation unit calculates the volume of the measured object based on the 3D point cloud coordinates of the object's outline in the table. Using a unified coordinate system, based on the coordinates of n 3D point clouds of the object's outline, the volume calculation unit obtains the height function z(x,y) of the measured object using laser triangulation. It then integrates this function within the x and y intervals [x1,xn] and [y1,ym] of the measured object to obtain its volume.

4. The method according to claim 1, characterized in that, The mass of an object measured on a production line is obtained by weighing it using an online electronic scale in conjunction with the production line.

5. The method according to claim 1, characterized in that, The density calculation unit calculates the density of the measured object, including at least the following methods: 1) Based on the volume obtained by the 3D camera, the density is obtained by dividing the mass by the volume; 2) Based on the X-ray camera, the X-ray absorptivity data is trained and established in advance using the 2D image information of multiple standard template objects with known density, and the X-ray absorptivity data related to the 2D image information of the measured object obtained online is compared with the standard template to obtain the density of the measured object.

6. The method according to claim 1, characterized in that, The measurement and processing unit uses a pre-determined density standard value to measure and judge the density of the measured object, and uses a machine learning-trained model to infer and judge the internal 2D image of the detected object. The processing method for unqualified detected objects includes: 1) starting the ejection device on the production line to eject the unqualified measured object from the production line; 2) starting the alarm device.

7. The method according to claim 2, characterized in that, A 3D camera images the surface contour of a measured object. This is an imaging method that works in conjunction with the movement of an assembly line. The method includes at least: 1) a 3D line scan camera that captures images of the measured object as it moves on the assembly line. 2) 3D area array camera, which keeps the object being measured still and acquires its image.

8. The method and system according to claims 1 and 4, characterized in that, The 3D machine vision unit and the weighing unit work in parallel under system control. While the object being measured moves along the conveyor belt on the weighing unit, the 3D machine vision unit images the object being measured.

9. The system according to claim 1, characterized in that, The assembly line is a mechanism that carries the movement of the object being measured, and it is equipped with sensors that monitor the position of the object being measured, so as to promptly notify the 3D machine vision unit and the weighing unit to start working.

10. The method and system according to claims 1 to 9, characterized in that, The 3D machine vision unit, production line, volume calculation unit, weighing unit, density calculation unit, and measurement processing unit are controlled by a unified computer system and PLC. The computer system is responsible for the calculation work of the 3D machine vision unit, volume calculation unit, density calculation unit, and measurement processing unit, while the PLC controls the start and stop of the production line, weighing unit, 3D machine vision unit, and measurement processing unit.

Images