Description

Process for detecting the three-dimensional structure of an object and apparatus for implementing the process

Technical Field

The present invention relates to a process for detecting the three- dimensional structure of an object which is fed along a movement path, as well as an apparatus able to implement the process. In particular, the present invention is intended for application in the wood processing sector, where the geometric structure, for example of logs, has to be detected, then saved in a computer for subsequent use.

Background Art At present, to do this, the log is fed along a path until it comes to a detection zone where, by means of a laser triangulation scanner, the position in space of each point of the outer surface of the log is identified. Since detection occurs as the log is fed forward (usually positioned with its longitudinal axis parallel with the direction of feed), it is carried out in a plurality of steps one after another during which the various sections of the log are detected one after another as they gradually arrive in the detection zone.

Consequently, to obtain the complete structure of the log, at the end the data detected must be reprocessed so as to virtually place the various sections detected alongside one another, thus reconstructing the complete log. However, this known technology has a significant disadvantage.

If between the start and the end of detection of the various section of which it consists, the log moves relative to the conveyor which is feeding it forward, after the movement the detection system sees the various sections as if they were

moved relative to the previous ones. As a result, at the moment of composition of the overall geometric structure a log is obtained in which every movement has been transformed into an alteration of the structure of the log.

Other methods have also been proposed over the years for detecting the geometric structure of objects.

For example, methods were proposed in which the entire geometric structure of the log is detected simultaneously by a plurality of laser triangulation devices. This solution, although not affected by any movements by the object, in reality cannot be applied for objects that are medium or large in size, where it would be necessary to use a very large number of detection devices, leading to a practically unsustainable cost.

Alternatively, it was proposed that the log should be held stationary and the detection zone should be moved along the log. However, although this solution solves the basic problem of possible log movements, it requires every log to remain stationary for a long time, with a consequent collapse in system productivity.

Disclosure of the Invention

In this situation, the technical purpose which forms the basis of the present invention is to provide a process for detecting the three-dimensional structure of an object which overcomes the above-mentioned disadvantages.

In particular, the technical purpose of the present invention is to provide a process for detecting the three-dimensional structure of an object which is not affected by any movements by the object during detection. The technical purpose of the present invention is also to provide a process for detecting the three-dimensional structure of an object which guarantees high productivity and a low cost.

Another technical purpose of the present invention is to provide an apparatus which allows implementation of the process disclosed. The technical purpose specified and the aims indicated are substantially

achieved by a process for detecting the three-dimensional structure of an object and by a relative apparatus as described in the claims herein.

Brief Description of the Drawings Further features and advantages of the present invention are more apparent in the detailed description below, with reference to several preferred, non-limiting embodiments of a process for detecting the three-dimensional structure of an object and of a relative apparatus, described with reference to the accompanying drawings, in which: Figure 1 is a perspective view of a log and a system for detecting its geometric shape able to implement the process in accordance with the present invention;

Figure 2 illustrates a section of an object which can be detected at a first detection zone; Figure 3 shows the section of Figure 2 as it would be detected in a second detection zone after an object lateral translation; and

Figure 4 shows the section of Figure 2 as it would be detected in a second detection zone after an object rotation on the conveyor.

Detailed Description of the Preferred Embodiments of the Invention

In accordance with the present invention, the process for detecting the three- dimensional structure of an object comprises first the known operating steps of: feeding the object 1 along a movement path. This may be achieved, for example, by placing the object 1 on a conveyor 2; at a first detection zone 3 positioned along the movement path, detecting the geometric shape 4 of at least one section of the object 1. hi particular, at least one section of the object 1 is detected, which is transversal to the direction of extension of the movement path at the first detection zone 3; repeating the detection step for all of the transversal sections of the object 1 which reach the first detection zone 3. hi this way, a plurality of partial geometric

shapes 4 of the object 1 can be obtained, corresponding to the sections positioned one after another which gradually reach the first detection zone 3; and, finally combining, preferably electronically, all of the partial geometric shapes 4 previously detected, to obtain the overall geometric shape of the object 1. As is known, any object 1 which can move normally has six degrees of freedom, which, identifying a Cartesian reference system, may be identified as three possibilities for translation parallel with the three axes, and three possibilities for rotation about axes parallel with the Cartesian ones.

In the case of detecting the geometric shape 4 of an object 1, it may therefore to useful to establish a Cartesian reference system of the type illustrated in Figure 1, in which the axis X corresponds with the direction of extension of the movement path, the axis Y with a direction (horizontal) perpendicular to it, and the axis Z with a direction (vertical) perpendicular to the horizontal plane in which the axis X lies. Each movement by the object 1 may therefore be reconstructed as a set of translations and rotations relative to the above-mentioned three Cartesian axes.

Therefore, if the object 1 is fed forward correctly, its movement will follow that of the movement means (conveyor 2) and will be describable only as a translation parallel with the axis X. However, as the object 1 is fed forward along the movement path, the object

1 may also move relative to the conveyor 2.

By way of example, such a movement may occur in the time interval separating detection of two zones of the object 1 positioned one after another. Consequently, at the first detection zone 3 the two sections positioned one after another would be detected as if moved relative to one another. This situation is schematically illustrated in Figures 2 to 4, representing the case of how the geometric shape 4 of the section of an object 1 could be detected if it were to remain centred on the conveyor 2 (the dashed line represents the central plane of the conveyor 2) as indicated in Figure 2, if the section were to perform a lateral translation relative to the conveyor 2 as indicated in Figure 3 or if it were to

perform an anti-clockwise rotation rolling on the conveyor 2 as illustrated in Figure 4 (rototranslation).

To overcome this problem, the process disclosed involves a new detection step during which, at least at a second detection zone 5 positioned along the path, downstream of the first detection zone 3, the geometric shape 4 of the same transversal sections of the object 1 detected in the first detection zone 3 is detected again.

For each section, the geometric shape 4 detected in the first detection zone 3 is compared with the geometric shape 4 detected in the second detection zone 5, to identify, based on said comparison, whether or not the object 1 moved during the time which elapsed for the passage of each section from the first detection zone 3 to the second detection zone 5.

Again with reference to Figures 2 to 4, by way of example it may be imagined that a predetermined first section of the object 1 at the first detection zone 3 is detected as having the geometric shape 4 indicated in Figure 2. If, when it arrives in the second detection zone 5, the new reading is again represented by Figure 2, it may be concluded that in passing from the first detection zone 3 to the second detection zone 5 either the object 1 did not move, or its overall movement was null. In any event, since at the moment when the first section examined is detected in the second detection zone 5, a second section is detected in the first detection zone 3, it can be said that the arrangement relative to one another of the readings taken in the first detection zone 3, which represent the above-mentioned first and second sections of the object 1, corresponds to the actual shape of the object 1. One may be sure that the object 1 is in the same position relative to the conveyor means.

If, in contrast, in the second detection zone the first section were to be detected as indicated in Figure 3, it could be inferred that during the passage of the first section from the first detection zone to the second detection zone 5 the object 1 performed one or more movements which can be defined, as a whole, as a lateral translation (parallel with the axis Y in a reference system such as that of Figure 1).

However, the second section at that moment detected in the first detection zone 3 is also affected by said object 1 translation. Therefore, it can be said that the representations detected in the first detection zone 3, which respectively represent the first section and the second section, are affected by the translation which happened in the meantime. Thus, they describe the actual position of the two sections relative to one another without a translation factor equal to that detected by the comparison between the first section detected in the first detection zone 3 and the first section detected in the second detection zone 5.

Similar reasoning may be applied if in the second detection zone 5 the first section were to be detected as illustrated in Figure 4. However, in this case the movement by the object 1 is not a simple translation, but the combination of a lateral translation and a rotation about an axis perpendicular with the sheet (with reference to a Cartesian system as illustrated in Figure 1, the translation parallel with the axis Y, and the rotation about an axis parallel with the axis X). Once any movements by the object 1 have been identified, the present invention allows this information to be considered during the step of combination of the partial geometric shapes 4, with the aim of correcting any detection errors caused by object 1 movement. For this purpose, it is sufficient that for each section detected an equation is saved which describes its movement caused by the movement of the object 1 being detected, relative to at least one other reference section. Li this case, such equations are used in the step of combining the geometric shapes 4 to correct the readings on the basis of the movements detected. For example, starting with the end section saved it is possible to gradually add the subsequent ones, if necessary applying the transformation equations which allow the elimination of errors caused by object 1 movement.

In the simplest embodiment, in which in the first and second detection zones 3, 5 only a section of the object 1 perpendicular to the movement path is detected, the only movements effectively detectable are those which keep the section in the same plane in which it is lying (therefore, with reference to the Cartesian system of Figure 1, translations parallel with the axis Y and rotations about axes parallel

with the axis X).

However, it should be noticed that for it to be advantageously applied, the present invention must preferably be intended to detect objects which have no axial symmetry, such as logs. If not, it would be impossible to establish with certainty any rotations by the objects on themselves.

In the more complex embodiments there may be further devices for increasing the precision of the process.

For example, there may be another detection step, to be implemented at least at a third detection zone 6 positioned along the movement path, downstream of the second detection zone 5, during which each section is detected for a third time.

The new reading is then compared with the geometric shape 4 detected at least in the first detection zone or the second detection zone 5, to provide any new information about movements by the object 1.

Consequently, when identifying any movements by the object 1 as each section passed from the first detection zone 3 to the second 5 and to the third detection zone 6, what is detected in the third detection zone 6 is also considered.

Depending on requirements, there may also be several third detection zones 6 one after another along the path.

Figure 1, relative to a log, shows the case of one first detection zone 3, one second detection zone 5 and one third detection zone 6. In each of them a section of the object 1 is detected in a vertical plane perpendicular to the feed path.

In other embodiments, in each detection zone a section of the object 1 may be detected according to a plane (for example vertical) set at an angle to the feed path. It is also possible that in each detection zone a plurality of separate transversal sections of the object 1 is detected. In this case, in all of the detection zones, the position relative to one another of the planes in which the sections to be detected lie must be the same. If not, the readings could not be compared with one another. For example, for each detection zone 3, 5, 6 two sections may be detected

which lie in two planes set at an angle relative to one another (and, preferably, relative to the plane perpendicular to the feed path), hi this way, it is simpler to detect any object 1 rotations about an axis parallel with the axis Z of Figure 1.

In general, in all of the embodiments, preferably the detection steps involve, for each section to be detected, illuminating the object 1 surface to create an illuminated zone and identifying the position in space of each illuminated point by means of the laser triangulation technique.

As already indicated, the present invention also relates to an apparatus for detecting the three-dimensional geometric structure of an object 1 able to implement the process described above.

First, it comprises a conveyor 2 able to feed an object 1 along a movement path.

First detection means 7 are positioned at a first detection zone 3 along the path, and are designed to detect the geometric shape 4 of at least one section of the obj ect 1 , transversal to the movement path.

Similarly, there are second detection means 8 positioned at a second detection zone 5, along the path downstream of the first detection zone 3, for again detecting the geometric shape 4 of the sections of the object 1 detected at the first detection zone 3. Processing means (not illustrated) are connected to the first detection means

7 and the second detection means 8 to acquire the geometric shapes 4 they detect and combine them in an object 1 overall geometric shape. Moreover, the processing means are programmed to identify any object 1 movements relative to the conveyor 2 during feed and to use said information to correct the data detected as indicated above.

The apparatus may also comprise third detection means 9 positioned at a third detection zone 6, along the path and downstream of the second detection zone 5, for again detecting the geometric shape 4 of the sections of the object 1 detected at the first detection zone 3 and the second detection zone 5. As described above, in this case the processing means will also take into

consideration the information received from the third detection means 9.

Finally, in general, the detection means consist of laser triangulation devices and are able to acquire sections of the object 1 perpendicular to or simply at an angle to the movement path. The present invention brings important advantages.

First, the present invention allows the detection of the three-dimensional structure of objects in a safe way and irrespective of any unwanted movements by the object.

It also allows productivity to be kept at current levels with an operating cost that is only slightly higher.

However, in general, the present invention is relatively easy to produce and even the cost linked to implementation of the invention is not very high.

The invention described above may be modified and adapted in several ways without thereby departing from the scope of the inventive concept. All details of the invention may be substituted by other technically equivalent elements and, in practice, all of the materials used, as well as the shapes and dimensions of the various components, may be any according to requirements.