MX2013015395A

MX2013015395A - Method for verifying two-dimensional or three-dimensional geometries in an object from the obtention of a point cloud.

Info

Publication number: MX2013015395A
Application number: MX2013015395A
Authority: MX
Inventors: Salvador Eduardo Ramirez Brambila; Luis Raul Hernandez Rodriguez Gil
Original assignee: Demaq Technologies S A De C V
Priority date: 2013-12-19
Filing date: 2013-12-19
Publication date: 2015-06-18

Abstract

The present invention refers to a method for verifying the two-dimensional or three-dimensional geometries in an object from the obtention of a point cloud, where an information processor that uses mathematical algorithms is used for identifying the equation corresponding to the geometry of the points acquired with an optical device, in order to establish its variations and parameters.

Description

METHOD FOR THE VERIFICATION OF TWO-DIMENSIONAL OR THREE-DIMENSIONAL GEOMETRICS OF AN OBJECT FROM THE OBTAINING OF A CLOUD OF POINTS DESCRIPTION FIELD OF THE INVENTION The present invention refers to a method for the verification of two-dimensional or three-dimensional geometries of an object from the obtaining of a point cloud, where by means of an information processor that uses mathematical algorithms the equation corresponding to the geometry of the points taken with an optical device.

BACKGROUND Currently there are known methods and systems for identifying geometries and detecting variations when inspecting objects or pieces of manufacture, for example, document US6665080B1 refers to a method for determining the deviations in the measured geometrical dimensions and / or the position of a starting object. of defined and desired values of said dimensions and / or geometric position of the object. Before determining the deviations, the measurement values of the dimensions and / or the position of the object are adapted to the values desired, taking into account the values of the defined tolerance of said desired values of the dimensions and / or the position of the object.

The patent application document US2005043849A1 refers to the positioning errors of a multi-axis numerical control machine and a flexible attachment of the work piece that are related through a mathematical model of statistical total error in the position of the machine tool (for example, a drill for drilling holes), and the maximum allowed quantities assigned for each of the individual factors that contribute to the total error. The mathematical model allows all errors or tolerances of the fixture attachment and the machine to be realistically distributed in order to maintain the accuracy of what results in the production of parts within acceptable limits. The relationship between the machine and the accessory is marked by a probe mounted on the spindle of the machine that plumbs the clamping device and measures the positions of the movable clamping members of the clamping fixture. The operation and alignment of the probe are checked with the probe against a fixed monument. The overall positioning accuracy of the machine is inspected throughout a work space of the machine using a laser measuring instrument. A master coordinate system in relation to the forms of the machine is created inside the laser instrument by calculating a linear adjustment of the acquired position of the machine coordinates as the machine moves along of each of the two perpendicular axes of the machine, and the master coordinate system is used for the accuracy and precision of the accessory by the global controls, thus eliminating the need for a reference system based on the background.

The patent application document US2013170709A1 discloses an automatic method and system for the visual inspection of the railway infrastructure, the invention refers to a system of visual inspection and method for the maintenance of the infrastructures, in particular the railway infrastructures. It is a system capable of operating in real time, fully automatically, for the automatic detection of the presence / absence of the characterization of the members of the infrastructure itself, for example the fastening locks of the rails to the sleepers CN102052078B discloses a real-time guidance system of a melting coating machine with multiple data sensors comprising a general laser measurement sensor, a laser target, a first prism, a second prism, a rear view prism and a computer , where the rear view prism is used to detect whether the position of the general laser measurement sensor is changed or not in the measurement process the laser target is used to measure a position angle of the coating machine in real time, the respectively, the first prism and the second prism are respectively arranged at both ends of the coating machine; the general laser measurement sensor measures the coordinates of the first prism, the second prism and the laser target prism under a geodetic coordinate system by the lasers it emits; and the computer makes the calculations to obtain a multigroup of the position angles by combining the position angle data measured by the laser target and performs the processing of the fusion in order to obtain the optimized position angles of the Coating machine for the application of the real-time guidance of the coating machine.

However, none of the state of the art documents discloses a method for obtaining two-dimensional or three-dimensional geometries from point characteristics of a cloud of points by means of an optical device, where this information is sent to a processor of information that assesses geometry, its properties, dimensions and location, can be two-dimensional or three-dimensional to make the comparison against the result of the algorithm including tolerance, which is defined by the user, depending on everything that is preferred to evaluate in the inspection, to know if the required specifications are met or not, such as identifying an object outside the acceptable range.

As can be seen, the difference with the state of the art is that the present invention is an identifier method of geometric characteristics of several Two-dimensional elements in a three-dimensional space, the ultimate goal is to measure something.

The method consists of detecting differences in some of the axes to obtain key points, those key points (presences), evaluate them with respect to other key points and that evaluation is used to obtain the geometric characteristics (center, radius, or long, if is a rectangle, side, width, if they are two planes, the angle between them), since you have these characteristics, you can compare them with a design specification with a certain tolerance, so that if the tolerance is met, you can take a decision (passes - does not pass), is very flexible in how to show it (remove the piece, stop the line, signal with light / sound).

It is indicated from the beginning the points to be evaluated or read of each scan or shot of the scanner, to evaluate everything in a line.

At the beginning you read all the points of the shot, from each shot you take certain key points, which are the ones that go into the evaluation of geometries.

The scanning position, that is to say where the line is required to be set, is programmed to obtain the key points, which are depending on the type of geometry to be evaluated (circle is at least 3), in this way it says that there are differences in a certain axis. If you define the key points, you are given the information of a circle (a geometry) to evaluate, that is what you start program, with the position and the geometry that should match, this identifies what kind of geometries and where they are in the piece, before evaluating the piece. Based on the key points, the presence is evaluated.

The characteristics or geometrical properties of the geometry to be evaluated, once there is presence, with the model of the geometry, is identified with which geometric configuration can coincide, trying to adjust the points in the characteristic equation of the circle, so that automatically it obtains the radius, the center, of the circle (geometry), which are the basic characteristics of a circle, and with these, depending on what is required (area, perimeter, etc.) it can also be programmed to obtain it, said geometry can be two-dimensional or three-dimensional.

Once the properties have been defined to identify the geometry and its location, depending on the design specification that is being evaluated, the comparison is made against the result of the algorithm including tolerance, all this is defined by the user, depending on everything it is preferred to evaluate, to identify if they are fulfilled or not.

The machines of measurement by coordinates by contact, do something similar, but they are by contact, they are tactile, not laser, not optical, it is a Cartesian robot, it handles the three axes by linear coordinates. There have been equations that are made for these coordinate measuring machines using a laser, but the support system is different, because it is not a robot, but in this machine everything moves in three dimensions X, Y and Z, the movement It is not as free as in a robot, so within the advantages of this method are the increase in degrees of freedom or trajectories and that the point cloud is not taken by contact.

BRIEF DESCRIPTION OF THE FIGURES Figure 1 illustrates the step where the object to be inspected or measured is positioned in a known place with respect to a reference or universal coordinate system.

Figure 2 shows the step where the geometry or geometries are defined and therefore the minimum points to obtain in each geometry.

Figure 3 illustrates how the zones to be analyzed are defined in the object to be inspected or measured where the points that describe the geometries to be measured are located.

Figure 4 shows how the minimum necessary points that describe the geometries are located by identifying the edges of the object to be inspected in the area to be analyzed.

Figure 5 shows the defined geometry, once these points are considered as a reduced cloud.

Figure 6 shows how to obtain the characteristic parameters of said geometry, call center, diameter, dimensions, etc., in terms of the coordinates with respect to the universal coordinate system or the reference.

Figure 7 is a perspective view of an example in which two areas of a piece such as a bottle are taken defining its dimensions.

DETAILED DESCRIPTION OF THE INVENTION The method of the present invention employs a system that includes a series of apparatuses, such as a transportation means for a measuring instrument, an inspection zone, the manufacturing part and an information processing device.

We start with an optical device to acquire point clouds in three dimensions or range images, either passive or active, the optical device for the acquisition of point clouds can be mobile or fixed, we also start from a support medium that is mobile and can hold the object to be inspected or measured within the field of work of both, both the optical device for acquisition of point clouds and the support medium.

The optical device for acquisition of point clouds in three dimensions or range images, can be passive or active, this refers to the methods of acquisition of point clouds, passive is the one that does not require any type of energy projected on an object to return something to the device, if necessary requires certain lighting conditions. The assets are those that project some type of energy that by that projection returns the information that is the point cloud, the interferometry. The liabilities do not project any type of energy on the object and the assets do.

The device by its nature collects information of the object in three dimensions, either by means of coordinates x, y and z, in each point or characteristic of the object, which has been the point cloud or by means of an image with depth information, which has been the image of rank.

In an embodiment, the optical device is held by the support means, that is, the device is mobile, the object is fixed to obtain the characteristics of the object. The optical device acquires the point clouds that are related to a universal coordinate system or a reference no matter where it is, as long as the geometric relationships between the characteristics of the object are maintained, for which the object is fixed.

A mathematical algorithm is used that processes the information obtained from specific points of the point cloud, these specific points are elements that characterize geometries, which are obtained from the identification of the edges or points of change in the trend of the coordinates of the rest of the points, in this case a minimum of two points are identified, depending on the type of geometry, example, for a line at least 2 points, for a circle at least 3 points, and thus, a plurality of points depending on the points necessary to characterize the geometry to be detected.

Although a single section is taken, you can obtain several geometries of one, then one area, the measurement area is larger if you want to obtain several points, in a single flash you get the specific area, but the line , that line is made repeatedly to become one of area, to obtain several lines, as if they were several slices.

The method consists of the following steps: 1. It has the specified dimensions or obtained from the base model, design or reference pattern, call design specification, CAD model or a previously measured reference object, specifying the respective tolerance to be considered. 2. The object to be inspected or measured must be positioned in a known place with respect to a reference or universal coordinate system as shown in Figure 1, where the number 2 within the circle indicates this step.

The geometry (s) and hence the minimum points to obtain in each geometry are defined, as well as the parameters that are to be measured of said or said geometries, as shown in figure 2, where the number 3 within the circle indicates this He passed.

The zones to be analyzed are defined in the object to be inspected or measured where the points describing the geometries to be measured are located to send the instruction to the support medium and to the optical device for acquiring the point cloud in three dimensions or images of the range or zones to be inspected in the object, as shown in figure 3, where the number 4 within the circle indicates this step.

The position of the areas to be inspected in the support medium is programmed and the capture of the point cloud or the range image in the optical device is programmed in the programmed position in the support medium, synchronizing the execution of each one of them. they having a relation between each position of the support medium, that is, the coordinates, and the values obtained by the acquisition device of the point cloud in three dimensions or range images.

The coordinates of each point cloud capture or range image made by the optical acquisition device are transform to unify the coordinate system of the optical device to the universal coordinate system that uses the medium of support in order to generate a single point cloud with known reference.

The point cloud obtained by the zones to be analyzed in the object is divided according to the location of the minimum necessary points that describe the defined geometries.

The minimum necessary points that describe the geometries are located by identifying the edges of the object to be inspected in the area to be analyzed, or points of change in the trend of the coordinates of the rest of the points, in this case, the limits of a trend that is based on the difference in some of the axes and in advance it is defined if the point to be searched for is in the maximum or minimum value of said difference in that respective axis, that is, the points of change suppose the existence of a difference in any of the coordinate axes, where it is known beforehand if the point describing the geometry is at the maximum or minimum value of said difference in that axis to characterize the defined geometry, as shown in the figure 4, where the number 8 in the circle indicates this step.

To characterize the defined geometry, once you have these points as a reduced cloud, adapt to the characteristic equations of the geometry you want to approximate, as shown in Figure 5, where the number 9 within the circle indicates this step.

The characteristic parameters of said geometry, call center, diameter, dimensions, etc., are obtained in terms of the coordinates with respect to the universal coordinate system or the reference, as shown in Figure 6, where the number 10 within the Circle indicates this step.

If you have several geometries or several parameters of these geometries, you can obtain the geometric and dimensional relationship or measurements between them based on their coordinates.

The measurements obtained, of one or several geometries, can be compared against the dimensions specified or obtained from the base model, design, or reference pattern considering the tolerance.

If the result of the comparison is within the defined tolerance, it is taken as an acceptable measurement or as an object within the acceptable range. 14. If the result of the comparison is outside the defined tolerance, it is considered as an unacceptable measurement or as an object outside the acceptable range.

The present invention corresponds to a method that starts from scanning the critical sections of a piece and not the whole piece itself, because it would take more time to form and obtain all the coordinates of the point cloud, so it must first be determined the specific sections of the piece to be inspected, is indicated to the scanner or optical device to which section or sections of the piece is directed to take a cloud of points, for example a sheet of a ring binder that must bring perforations of a certain diameter and at a precise distance for the rings to pass, it is irrelevant to take samples of the whole sheet when in fact the critical sections are the perforations, then the inspection is programmed specifically in the perforations and a reference point is taken from the corners of the sheet to know that the diameter or perimeter of that hole is within the specified or that the distance between the centers of the s perforations is within a predetermined range, another example can be a bottle like the one shown in figure 7 where the scanning is done only in the base and in the mouth of the bottle and can be in sections of each of these where the scanner sends a line that captures one or two points and these points are captured by an information processor that compares these points in the equation of a circle and on the basis of this determines if the points found correspond to the equation of a circle and also determining the diameter or perimeter to compare them with the predetermined values, likewise this method also allows to identify geometries in large pieces that require to be very precise as for example a turbine.

Claims

CLAIMS Having described the invention as above, the content of the following claims is claimed as property:

1. A method for the verification of two-dimensional or three-dimensional geometries of an object from the obtaining of a cloud of points which comprises: The object or piece to be inspected or measured is positioned in a predetermined place with respect to a reference or universal coordinate system; The geometry or geometries are defined and therefore the minimum points to obtain in each geometry, as well as the parameters that are desired to measure said or said geometries; The critical zones to be analyzed are identified or defined in the object to be inspected where the points describing the geometries to be measured are located; the instruction is sent by means of a controller to a support means and to an optical device for acquiring the point cloud in three dimensions, synchronizing the execution of each one of them having a relation between each position of the support means, that is to say the coordinates, and the values obtained by the acquisition device of the point cloud in three dimensions; the coordinates of each point cloud capture made by the optical acquisition device are transformed by means of an information processor to unify the coordinate system of the optical device to the universal coordinate system that uses the support means in order to generate a single point cloud with known reference; the point cloud obtained by the zones to be analyzed in the object is sectioned according to the location of the minimum necessary points that describe the defined geometries; The minimum necessary points that describe the geometries are located identifying the edges in the object to be inspected of the area to analyze, or points of change in the trend of the coordinates of the rest of the points; the limits of a trend that are based on the difference in some of the coordinate axes are identified; by means of an information processor that uses mathematical algorithms it is identified with which equation that characterizes a certain geometry coincide the points taken by the optical device; the characteristic parameters of said geometry are obtained by means of the information processor, call center, diameter, dimensions, etc., in terms of the coordinates with respect to the universal coordinate system or the reference; the measurements obtained from one or more geometries are compared by means of the information processor against the specified or predetermined dimensions of the piece considering tolerance.

2. A method for the verification of two-dimensional or three-dimensional geometries of an object from the obtaining of a cloud of points according to claim 1 where if you have several geometries or several parameters of these geometries, you can obtain the geometric relationship and dimensional measurements between them based on their coordinates.

3. A method for the verification of two-dimensional or three-dimensional geometries of an object from the obtaining of a cloud of points according to claim 1 wherein if the result of the comparison is within the defined tolerance it is taken as an acceptable measurement or as an object within the acceptable range.

4. A method for the verification of two-dimensional or three-dimensional geometries of an object from the obtaining of a cloud of points according to claim 1 wherein if the comparison result is outside the defined tolerance it is considered as an unacceptable measurement or as an object outside the acceptable range.