Analysis of factors affecting the accuracy of optical scanning measurement and their countermeasures

1 Introduction

Reverse engineering (or reverse engineering) is a rapidly developing emerging technology. Physical reverse engineering technology is used to redesign products based on existing physical products, or to digitally convert some product models (such as clay models) that cannot be directly expressed and designed by digital means but can only be expressed in physical form, so as to implement digital design and manufacturing of these products. For example, when performing aerodynamic design on the appearance of aircraft, missiles and other aircraft, the physical model obtained entirely based on experimental optimization cannot be directly modeled and expressed using the existing CAD/CAM system. Therefore, in order to realize the digital design and manufacturing of these aircraft, it is necessary to use the point cloud data acquisition technology in physical reverse engineering to convert the physical model into a CAD model. At present, physical reverse engineering technology has been widely used in new product design, product maintenance, product online testing and other aspects.

In physical reverse engineering, in order to realize the digitization of objects, the corresponding measuring or scanning equipment must be used to measure or scan the three-dimensional physical model of the product to obtain the spatial topological discrete point data point cloud of the physical model. Therefore, point cloud data collection is the first task that must be completed in physical reverse engineering. Among various physical measurement technologies, the optical scanning point cloud data collection technology that has emerged in recent years has many advantages such as high measurement efficiency, good data integrity, wide application range, and wide data collection range (from a few millimeters to tens of meters).

In recent years, many domestic universities and research institutes have conducted a lot of research and development on physical reverse technology, and have made breakthroughs in some key technologies. Through independent development and technical cooperation, the development and application of contact measurement technology and equipment in China have become increasingly mature. However, for non-contact measurement of optical scanning point cloud data acquisition technology, there are still few public reports on mature products developed in China and their successful applications in the automotive, mold and other industries. In order to promote the research and development of optical scanning point cloud data acquisition and processing technology and provide external technical services, our school has introduced mobile optical scanning equipment produced by a foreign company. During the use of the equipment, the problem of excessive measurement errors is often caused by operating errors, improper adjustments and other reasons. In order to thoroughly digest the imported technology from abroad, the author conducted research and analysis on this in combination with work practice, and proposed corresponding solutions.

2 Main manifestations of measurement errors

(1) The collected data is missing or the data density does not meet the requirements. Using such incomplete data for point cloud fitting will result in large errors and it is difficult to achieve the required measurement accuracy.

(2) The data collection results for the same surface are presented as multi-layer point clouds. This situation often occurs when the object being measured is a large workpiece or a transparent object.

(3) Single-frame data collection is inaccurate, affecting the overall measurement accuracy.

(4) The cumulative error is too large, resulting in obvious deviations in the measurement results.

(5) Errors in point cloud stitching lead to large measurement errors.

(6) There are too many coarse points (noise) in the measurement results.

3. Error Cause Analysis and Countermeasures to Improve Accuracy

Based on actual work experience, through test analysis, the author summarizes the main reasons for large measurement errors as follows: improper calibration, improper use of rulers; improper selection of probe lens combination; improper measurement sequence; improper selection of measurement strategy; improper placement of workpiece surface marking points; improper operation during measurement; improper pretreatment of the workpiece surface; improper post-processing; improper selection of measurement environment, etc. The following are analyzed separately:

(1) Improper calibration and improper use of scale

The scanning measuring head (probe) is composed of a light source, a CCD camera and a corresponding lens group. Before collecting point cloud data, the probe needs to be initialized first. The main contents include:

① Select different lens combinations according to the size of the object being measured, the number of surface features and their complexity;

② Determine the intensity of the main light source according to the measurement site conditions, the surface morphology and surface treatment of the object being measured;

③ Calibrate the selected lens combination according to the system standard workflow, so that the calibration accuracy value is ≤0.020;

If the above work is not performed before measurement, and the previously calibrated probe is used directly for measurement, the measurement accuracy may not be guaranteed because the lens combination, light source intensity, and calibration accuracy do not meet the measurement requirements, resulting in large errors.

During measurement, if the probe is impacted or collided due to operational errors, the probe should be checked in time. If damaged, it should be repaired; if not damaged, the probe must be recalibrated; even if no operational errors occur during measurement, if the measurement time is long, the probe must be quickly calibrated regularly to check its accuracy.

The ruler is an essential tool for locating the marking points on the entire workpiece using a digital camera when collecting data on large workpieces. The standard size on the ruler used should be consistent with the size value displayed when processing using photos. [page]

(2) Improper selection of probe lens combination

When collecting point cloud data on the surface of large workpieces, a lens combination with a larger measurement range should be selected to achieve rapid collection of overall data; for areas with more and smaller features, it is best to use a lens combination with a smaller measurement range to perform prominent measurements of local small features to obtain better measurement results.

For large workpieces, if a lens combination with a smaller measurement range is selected for measurement, more identification points for point cloud stitching are required on the workpiece surface, which will prolong the workpiece preprocessing time and increase the measurement time span. The error caused by the change of ambient temperature over time will be reflected in the measurement results and will affect the overall measurement efficiency. If a digital camera is used to locate the identification points on the entire workpiece and the stitching is automatically performed during measurement, stitching errors are prone to occur due to the large number of identification points and the increased probability of the identification points having the same relationship. If a digital camera is not used to locate the identification points on the entire workpiece, and the adjacent single point clouds are stitched using common identification points, the stitching cumulative error will be too large due to the large number of stitching times.

On the contrary, for small workpieces, if a lens combination with a larger measuring range is selected for measurement, the small features on the workpiece cannot be accurately reflected, and the measurement results cannot meet the required accuracy. It is necessary to replace the appropriate lens combination, re-calibrate and measure.

(3) Improper measurement order

The measurement sequence refers to the superposition sequence between adjacent single-frame measurement results during measurement. Taking the measurement of a slender workpiece as an example, the rectangular area shown in 1,2,3,4,5 in the figure is the measurement range of the current calibration probe. When measuring the workpiece in the radial arrangement shown in the figure, the middle position of the workpiece is measured first, and after the measurement of the first frame in the middle is completed, the second frame is measured, and then the second frame is spliced ​​with the first frame using three common marking points. At this time, a splicing error will be generated. Similarly, when the third frame is spliced ​​with the second frame, a splicing error will also be generated. Assuming that all splicing errors are of the same size, the cumulative error generated between 1,2,3 is 28, and the cumulative error generated between 1,4,5 is also 28. The measurement results show that the cumulative error between 1,2,3 does not form a superposition relationship with the cumulative error between 1,4,5, so the total cumulative error is still 28. If the workpiece is measured in the order shown in Figure 2, the maximum cumulative error is 48. Therefore, when measuring, the measurement sequence of "based on the center and arranged in a radial shape" should be adopted as much as possible to reduce the cumulative error.

(4) Improper selection of measurement strategy

During measurement, the workpieces to be measured should be classified into large workpieces, medium-sized workpieces, small-sized multi-feature workpieces, inner cavity workpieces, etc. Different measurement strategies should be adopted for each type of workpiece.

When measuring large workpieces, you can first use a digital camera to locate the marking points overall, and then select a lens combination with a large single-frame measurement range for measurement; if the workpiece is too large, you can measure it twice and then use a common reference point to stitch it together; if there are many local small features in a large-sized workpiece, you can use a lens combination with a smaller measurement range to perform local measurements after the measurement is basically completed. In order to facilitate automatic stitching of small-range measurements, the density of reference points should be increased when pre-processing the local area.

When measuring medium and small workpieces, attention should be paid to adopting the correct measurement sequence to reduce cumulative errors. In fact, the measurement strategy for large-sized workpieces can also be used for medium and small workpieces, but it must be equipped with a digital camera and related software for overall positioning of the reference point.

When measuring the inner surface of a workpiece, in order to overcome the depth of field limitation of the optical scanning measurement equipment, some technical means can be used to convert the inner cavity measurement into the outer shape measurement. For example, silicone can be injected into the inner cavity of the workpiece, and then taken out after it solidifies to measure its outer shape.

(5) Improper placement of marking points on the workpiece surface

Regardless of whether you are measuring a large workpiece or a medium or small workpiece, you will encounter the problem of placing marking points on the workpiece surface.

Data collection of large workpieces generally requires the use of rulers and digital cameras. Measurement can be carried out in two steps: the first step is to use large digital points to construct the overall marker points used for the splicing of single-frame measurement point clouds. In order to ensure that the marker point cloud for single-frame measurement is obtained through normal calculation, the arrangement rules shown in Figure 1 must be followed; the second step is to use the marker point cloud as a reference system. The system will compare the marker points in each single-frame point cloud measured with the marker points in the existing reference point cloud. If the two match, they will be automatically spliced, and there is no need for overlap between two adjacent single-frame point clouds. Measurement can also be carried out, but the premise is that each single-frame point cloud must contain at least three marker points. At this time, the marker points need to be appropriately pasted to the measured area, otherwise it will cause measurement difficulties or reduce measurement accuracy.

The pasting of marker points for medium-sized workpieces is different from that for large workpieces. Since the automatic (or manual) stitching of two adjacent point clouds needs to be completed based on the common marker points of adjacent single point clouds, the pasting density of marker points for medium-sized workpieces should be greater than that for large workpieces, otherwise it will be difficult to stitch two adjacent point clouds.

Since the placement of marking points on small workpieces will mask the features on the workpiece to varying degrees, the surface of the workpiece should be affixed with as few or no marking points as possible to obtain more complete scanning data.

In addition, the marking points should generally be pasted on a relatively flat position on the workpiece to reduce the difficulty of filling in the point cloud at the marking points and the corresponding measurement errors.

(6) Improper operation during measurement

During the measurement process, pay attention to the following operating points:

① Adjust the probe position so that the measured part is within the measurement range of both probes at the same time;

② Adjust the light intensity of the main light source and the clarity of the marking point and the workpiece surface respectively, so that the marking point and the workpiece surface of the measured part are as clear as possible;

③During the measurement process, the probe should be impacted or collided as much as possible. If this happens accidentally, the probe should be checked and recalibrated in time to maintain the accuracy of subsequent measurements. Otherwise, the measurement results will show that the measured component data is missing, or even the measurement cannot be continued. [page]

(7) Improper pretreatment of the workpiece surface

Before starting the measurement, the workpiece surface needs to be properly pre-processed. If the workpiece shape is very simple and the workpiece size is small, data collection can be completed through a single-frame measurement, and it is only necessary to make the workpiece surface form diffuse reflection under the irradiation of the main light source. However, under normal circumstances, it is difficult to complete the data collection of a complete workpiece through a single-frame scanning measurement, and it is also difficult for a general workpiece surface to form a diffuse reflection that meets the measurement requirements under the irradiation of the main light source. Therefore, some reference points must be preset on the workpiece surface, and the common reference points must be used to splice the measurement results of each time, and the workpiece surface must be evenly sprayed with a colorant to form a more ideal diffuse reflection on the workpiece surface.

Improper surface pretreatment of the workpiece mainly refers to:

① Some parts of the workpiece surface reflect too much or absorb too much light, which cannot form diffuse reflection suitable for scanning requirements, resulting in the inability to form an effective point cloud. The measurement results show that the data of this part is missing;

② Lack of sufficient reference points makes it impossible to stitch. Even if a point cloud can be formed, it is only a scattered point cloud instead of a whole point cloud;

③ The consistency of the reference points on the workpiece surface is too strong and lacks characteristics, which makes it impossible for the system to effectively identify the stitching position of a single point cloud, which easily leads to stitching errors and makes it difficult to form the overall point cloud of the measured workpiece.

Improper pretreatment of the workpiece surface to be measured also includes non-measurement factors such as failure to correct the part of the workpiece surface that cannot correctly reflect the design intent, failure to repair the workpiece surface in time after being damaged during measurement, improper placement of the workpiece (such as the workpiece being stressed), etc. In addition, when silicone injection is performed on the workpiece cavity (such as the engine airway) to form a model, insufficient injection volume or too many bubbles in the silicone will also cause the formed model to fail to correctly reflect the actual shape of the workpiece cavity.

(8) Improper post-processing

In optical scanning measurement, not all the measured data is point cloud data. The measurement process is actually the process of forming an image of the workpiece. To obtain point cloud data, the image data must be post-processed using the ATOS system. For the results generated by stitching a single point cloud, all single data must first be aligned using several common identification points to reduce cumulative errors; then the aligned point cloud is recalculated to convert the image data into point cloud data. At this point, the point cloud data may still have problems such as uneven density and many coarse error points. It can be further processed by triangulation (Polygonize) to finally obtain point cloud data of better quality.

Of course, post-processing includes more than just the above. In actual measurement, the data points obtained by scanning are not necessarily limited to the physical model being measured. Some random points in the measurement environment that do not belong to the model are also measured at the same time. Therefore, these unnecessary points must be removed during post-processing to reduce the possibility of errors when constructing a 3D CAD model based on point cloud data.

In addition, post-processing also includes simplification of point clouds. In a physical inverse point cloud, the accuracy requirements for each part of the workpiece are not exactly the same. Therefore, some less important parts can be simplified by reducing the point cloud density; some more important parts can be simplified by increasing their point cloud density. This not only ensures the accuracy requirements of the 3D model construction, but also greatly improves the modeling efficiency.

Of course, post-processing includes more than just the above. In actual measurement, the data points obtained by scanning are not necessarily limited to the physical model being measured. Some random points in the measurement environment that do not belong to the model are also measured at the same time. Therefore, these unnecessary points must be removed during post-processing to reduce the possibility of errors when constructing a 3D CAD model based on point cloud data.

In addition, post-processing also includes simplification of point clouds. In a physical inverse point cloud, the accuracy requirements for each part of the workpiece are not exactly the same. Therefore, some less important parts can be simplified by reducing the point cloud density; some more important parts can be simplified by increasing their point cloud density. This not only ensures the accuracy requirements of the 3D model construction, but also greatly improves the modeling efficiency.

4 Conclusion

In recent years, physical reverse engineering technology is playing an increasingly important role in new product design, product modification design, mold manufacturing, etc., and its application in automobile manufacturing, aerospace, machine tools, national defense, electronics, molds and other fields is becoming more and more extensive. However, the current research on related technologies in China is still lagging behind, and related technical equipment still mainly relies on imports. Therefore, it is imperative to research and develop physical reverse engineering technology and equipment with independent intellectual property rights in my country and realize commercial application as soon as possible. This paper analyzes the factors affecting the accuracy in the actual application of optical scanning point cloud data acquisition system, and puts forward corresponding countermeasures, hoping to help improve the application level of related measuring equipment, and also provide some practical experience for reference for the research of physical reverse engineering technology, so as to promote the continuous improvement of the development and application level of reverse engineering technology in my country.