Search

CN-121973438-A - 3D printing method and system based on visual detection

CN121973438ACN 121973438 ACN121973438 ACN 121973438ACN-121973438-A

Abstract

The application belongs to the technical field of 3D printing, and particularly relates to a 3D printing method and system based on visual detection, wherein the method comprises the steps of obtaining physical image information and generating printing information; the method comprises the steps of preprocessing printing information to obtain preprocessed printing information, carrying out geometric path prediction according to the preprocessed printing information to generate a predicted geometric path, carrying out printing according to the predicted geometric path and calibrating based on visual detection in real time, carrying out thermodynamic deformation prediction on the printing information to obtain a thermodynamic deformation path, predicting extrusion speed of a nozzle material to obtain a material speed path, carrying out weight fusion on the thermodynamic deformation path and the material speed path to obtain a path adjustment coefficient, generating the predicted geometric path according to the path adjustment coefficient, carrying out advanced prediction on a wire drawing phenomenon generated in the movement of the nozzle, and carrying out intervention.

Inventors

  • LIU JIURU
  • LIU JIA

Assignees

  • 西北工业大学

Dates

Publication Date
20260505
Application Date
20251229

Claims (10)

  1. 1. A 3D printing method based on visual inspection, the method comprising the steps of: acquiring real object image information and generating printing information; Preprocessing the printing information to obtain preprocessed printing information; Carrying out geometric path prediction according to the preprocessed printing information to generate a predicted geometric path; Printing is performed according to the predicted geometric path, and real-time calibration is performed based on visual detection.
  2. 2. The 3D printing method based on visual inspection according to claim 1, wherein the acquiring the physical image information, generating the printing information, comprises: capturing image information of a real object by one or more structure light scanners in 360 degrees; and carrying out three-dimensional modeling according to the acquired image information.
  3. 3. The 3D printing method based on visual inspection according to claim 2, wherein the preprocessing step of the printing information is: Converting discrete point cloud information in the printing information into a watertight grid model, and specifically, converting a grid model by solving a poisson equation; optimizing the quality of the grid model, specifically, preprocessing, including: Grid cleaning, namely deleting floating noise points and outliers; Hole filling, namely complementing surface holes caused by shielding or reconstruction failure; The mesh is simplified, namely the number of triangle patches is reduced on the premise of keeping the shape characteristics; smoothing, namely smoothing the serrated or uneven surface, and improving the aesthetic degree of the model; And (3) error repair, namely checking and repairing the model errors, and carrying out manual auxiliary repair when serious errors are involved.
  4. 4. A 3D printing method based on visual inspection according to claim 3, wherein the geometric path prediction based on the preprocessed printing information is specifically: carrying out thermodynamic deformation prediction on the printing information to obtain a thermodynamic deformation path; predicting the extrusion speed of the nozzle material to obtain a material speed path; carrying out weight fusion on the thermodynamic deformation path and the material speed path to obtain a path adjustment coefficient; and generating a predicted geometric path according to the path adjustment coefficient.
  5. 5. The 3D printing method based on visual inspection according to claim 1 or 4, wherein the printing according to the predicted geometrical path and the real-time calibration based on visual inspection are specifically: performing three-dimensional geometric printing according to the predicted geometric path, and preferentially printing the same material in the same material area and then replacing the material; in the printing process, the detection equipment is adopted for real-time visual detection and calibration.
  6. 6. A3D printing system based on visual detection comprises an image information acquisition module, a printing information preprocessing module, a geometric path prediction module and a visual detection calibration module; The image information acquisition module is used for acquiring the image information of the real object and generating printing information; The printing information preprocessing module is used for preprocessing the printing information and acquiring the preprocessed printing information; The geometric path prediction module is used for carrying out geometric path prediction according to the preprocessed printing information and generating a predicted geometric path; and the visual detection calibration module is used for printing according to the predicted geometric path and calibrating in real time based on visual detection.
  7. 7. The 3D printing system based on visual inspection according to claim 6, wherein the image information acquisition module comprises a captured image information unit and a three-dimensional modeling unit; A captured image information unit for capturing image information of the object 360 degrees using one or more structured light scanners; and the three-dimensional modeling unit is used for carrying out three-dimensional modeling according to the acquired image information.
  8. 8. The 3D printing system based on visual inspection according to claim 7, wherein the geometric path prediction module comprises a thermodynamic path prediction unit, a material extrusion speed prediction unit, a fusion unit and a geometric path generation unit; The thermodynamic path prediction unit is used for carrying out thermodynamic deformation prediction on the printing information; the material extrusion speed prediction unit is used for predicting the extrusion speed of the nozzle material; The fusion unit is used for carrying out weight fusion on the thermodynamic deformation path and the material speed path; And the geometric path generating unit is used for generating a predicted geometric path according to the path adjustment coefficient. In the embodiment of the application, the visual detection and calibration module comprises a printing unit and a visual detection unit; the printing unit is used for performing three-dimensional geometric printing according to the predicted geometric path; And the visual detection unit is used for performing real-time visual detection calibration by adopting detection equipment in the printing process.
  9. 9. An electronic device, comprising: The apparatus of any one of claims 1 to 5, wherein the apparatus comprises one or more processors, memory, and one or more computer programs stored in the memory, the one or more computer programs comprising instructions that, when executed by the apparatus, cause the apparatus to perform the steps of the visual detection-based 3D printing method of any one of claims 1 to 5.
  10. 10. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program which, when run on a computer, causes the computer to perform the steps of the vision detection based 3D printing method according to any one of claims 1 to 5.

Description

3D printing method and system based on visual detection Technical Field The application belongs to the technical field of 3D printing, and particularly relates to a 3D printing method and system based on visual detection. Background The 3D printing technology has been widely used in fields of industrial manufacturing, medical treatment, education, creative design, etc. due to its characteristics of high efficiency, flexibility and customizable. Along with the continuous expansion of application scenes, the precision control and the real-time quality monitoring of the printing process become key problems for restricting the further development of the technology. Especially in printing tasks with complex structures or high precision requirements, minor deviations in the printing process (such as material extrusion anomalies, layer-to-layer misalignment, temperature fluctuations, etc.) may lead to significant deviations of the final product from the design model. Therefore, how to realize real-time visual monitoring and real-time error correction of the printing process has become a core research direction for improving the reliability and the qualification rate of the finished products of 3D printing. At present, the quality control of the 3D printing process mainly depends on two technologies, namely, local parameter monitoring based on sensors (such as laser ranging and infrared temperature sensing) for detecting basic parameters such as flatness of a printing platform and temperature of a spray head, and a later detection technology, namely, comparing a finished product with an original model through three-dimensional scanning or manual measurement after printing is finished. However, the prior art still has the following significant drawbacks: 1. dynamic deviations from the design model cannot be identified and fed back in real time during printing. For example, when the material is not extruded uniformly or the path planning is abnormal, the system can often find the deviation of the result after the printing is finished, and the time and the material are wasted. 2. Microscopic defects such as wiredrawing phenomenon generated in the movement of the spray head, material accumulation at corners caused by speed change, weak interlayer bonding and the like are often perceived after accumulation is a significant problem, and the capability of foreknowledge and intervention is lacking. Disclosure of Invention In order to overcome the problems in the prior art, the application provides a 3D printing method and system based on visual detection, which adopts the following technical scheme: In a first aspect, the present application provides a 3D printing method based on visual inspection, including: acquiring real object image information and generating printing information; Preprocessing the printing information to obtain preprocessed printing information; Carrying out geometric path prediction according to the preprocessed printing information to generate a predicted geometric path; Printing is performed according to the predicted geometric path, and real-time calibration is performed based on visual detection. Further, the obtaining the physical image information and generating the printing information include: capturing image information of a real object by one or more structure light scanners in 360 degrees; and carrying out three-dimensional modeling according to the acquired image information. Further, the step of preprocessing the print information includes: Converting discrete point cloud information in the printing information into a watertight grid model, and specifically, converting a grid model by solving a poisson equation; optimizing the quality of the grid model, specifically, preprocessing, including: Grid cleaning, namely deleting floating noise points and outliers; Hole filling, namely complementing surface holes caused by shielding or reconstruction failure; The mesh is simplified, namely the number of triangle patches is reduced on the premise of keeping the shape characteristics; smoothing, namely smoothing the serrated or uneven surface, and improving the aesthetic degree of the model; And (3) error repair, namely checking and repairing the model errors, and carrying out manual auxiliary repair when serious errors are involved. Further, the geometric path prediction according to the preprocessed printing information specifically includes: carrying out thermodynamic deformation prediction on the printing information to obtain a thermodynamic deformation path; predicting the extrusion speed of the nozzle material to obtain a material speed path; carrying out weight fusion on the thermodynamic deformation path and the material speed path to obtain a path adjustment coefficient; and generating a predicted geometric path according to the path adjustment coefficient. Further, the printing is performed according to the predicted geometric path, and the real-time calibration based on the visual det