ES2802900A1 - Method for cleaning chewing gum residue or its derivatives (Machine-translation by Google Translate, not legally binding) - Google Patents

Abstract

Method for cleaning chewing gum residue or its derivatives. The present invention refers to a method of cleaning chewing gum residues on floors or vertical walls. It is characterized by a stage for obtaining the image of the residue to be cleaned, a stage for determining the surface to be cleaned and the action of a laser beam on it. (Machine-translation by Google Translate, not legally binding)

Description

DESCRIPTION

METHOD FOR CLEANING UP CHEWING GUM OR DERIVATIVE RESIDUES

FIELD OF THE INVENTION

The floors and facades in urban environments often show easily visible alterations due to the human factor. Among them are polymeric residues, better known as gum or chewing gum. In particular, these wastes are difficult to remove and that is why their disposal is a problem in city cleaning.

Currently, cleaning methods are used that achieve the removal of the residue with a high percentage of success but that are highly aggressive with the surface in question and also with the environment. This invention will address the cleaning of chewing gums adhered to floors and facades of urban environments.

BACKGROUND OF THE INVENTION

Chewing gum and its composition

Chewing gum is a non-swallow flavored chewable tablet with a gum-like texture. It is composed of (this composition was extracted from the ingredients indicated in a gum pack):

- Gum base composed of E-306, a non-harmful antioxidant of vegetable origin.

- Sweeteners: sorbitol, xylitol, mannitol, maltitol, aspartame, acesulfame K and sucralose. - Humectant E-422, a stabilizer of chemical or synthetic origin that is not harmful.

- Acidity correctors such as citric and malic acid.

- Emulsifier: sunflower lecithin.

Gum base is an insoluble and non-nutritive substance that allows chewing gum to be chewed for hours without significant changes.

This gum is manufactured from food-grade polymers, plasticizers, texture modifiers and emulsifiers, among other ingredients that give chewing gum its properties. unique. To have this quality, the gum base must meet international requirements such as PDA 21 CFR 172.615 and the Food Chemical Codex (FCC) specifications.

Methods currently used for disposal

Currently there are three techniques for cleaning this waste. These are shown below:

• Cleaning with pressurized water: consists of projecting a jet of hot water at high pressure on the surface to be treated using a lance connected to a hose (see Figure 1). For this method, it is essential to have a hydraulic pump to obtain the working pressure and a water tank, which is its main disadvantage in addition to the amount of water generated in the environment. Sometimes, such as in crowded and crowded areas, the system needs a vacuum cleaner to remove the water sprayed on the streets.

Often this method is also combined with the addition of certain chemical compounds to the water or with the previous application of said compounds on the areas to be treated in order to carry out more satisfactory cleaning and promote disinfection. The results of the application of this technique are not entirely unfavorable, since it manages to eliminate the gum but it also tends to produce some erosion, cause detachments or leave residual marks on the treated surface, as can be seen in Figure 2.

• Steam cleaning: it is based on applying water steam at high temperature and pressure (180 ° C and 350 to 500 bar) to the area of interest by means of a pressure gun or by means of a device that moves while maintaining contact with the surface to be treated. It is necessary to have a water tank and equipment for heating it, however, the consumption of this resource is greatly reduced and therefore so does the amount of wastewater generated. In Figures 3 and 4 you can see examples of the application of this method.

The results by this method are very similar to those obtained by applying water under pressure. The practically total elimination of the gum is appreciated but residual marks can also be observed on the treated surface.

• Cleaning using solvents: it involves applying a mixture of chemical solvents on the surface in question and then rubbing the area with a brush. This method is generally carried out using an integrated brush gun connected to a portable backpack worn by the operator, where the cleaning solution is stored.

Sometimes these chemicals are also applied with brushes, leaving a few minutes to act afterwards and the residue being removed with a spatula or with pressurized water. Other times, the use of freezer sprays is chosen in order to harden the gum, and then remove it with a plastic spatula.

The main advantages of this technique are its portability and easy handling for the operator. On the other hand, cleaning using chemical compounds can be too aggressive or abrasive on certain surfaces.

BRIEF DESCRIPTION OF THE INVENTION

For all of the foregoing, a set of advantages can be highlighted that involves cleaning surfaces by using laser equipment compared to existing cleaning methods. They are as follows:

- Absence of mechanical actions, as opposed to cleaning methods that use spatulas to remove adhering residues

- It has no inertia, therefore it can be moved easily, high precision is achieved and high speed can be reached

- There is no wear or corrosion on the tool, as it does end up in the nozzles of the guns used when applying methods that use pressurized water or water with solvents

- Possibility of energy concentration in a defined area, which implies a reduced thermally affected area (which can be zero) and low distortion of the material

- There is no erosion of the substrate as in the case of the water jet with abrasive

- Corrosion or chemical degradation of the treated surface does not occur as occurs when using chemical solvents

This invention proposes the use of a laser for the elimination of chewing gum on the surface of floors or vertical walls. In particular, the application of the laser for the cleaning of chewing gum will be studied in pavements or facades made from natural rocks such as granite, marble, slate or similar materials. As mentioned above, it is possible to use lasers for cleaning surfaces, but there are no precedents for the application of lasers for cleaning this type of chewing gum residue. That is why this method has an absolutely novel and original character, exploring an unprecedented application of lasers. Therefore, this work will contribute to the development of solutions for the maintenance of urban environments and progress in laser applications, and more specifically to the development of cleaning tasks using the CO ² laser.

Therefore, the present invention will pursue two fundamental objectives: first, the experimental determination of the optimal conditions for the total elimination of the plastic residue from the stone, minimizing the damage caused to it. Second, it is intended to carry out the design of a portable equipment incorporating a high-power laser, an artificial vision system and the necessary components for the correct operation of the system.

The present invention defines a method of cleaning chewing gum residues on floors or vertical walls according to claim 1. The dependent claims define preferred embodiments of the invention.

BRIEF DESCRIPTION OF THE FIGURES

Figure 1. Hydro-cleaning with trailer in urban environment. Source: MPA cleanings.

Figure 2. Appearance of the flooring after a treatment with pressurized water. Source: Novatecnic advanced urban services.

Figure 3. Steam cleaning with a machine in direct contact. Source: Information newspaper.

Figure 4. Steam cleaning with a gun. Source: Teleprensa digital newspaper. Figure 5. Operator carrying the solvent cleaning system and additional brush. Source: Conantec.

Figure 6. Operator removing gum with spatula Source: SR Radio portal.

Figure 7. Gum on Rosa Porriño stone with a sawn finish.

Figure 8. Chicle-stone work limits.

Figure 9. Irradiance distribution in the plane transverse to the beam.

Figure 10. Simple representation of the projection of the beam on the surface to be treated. Figure 11. Experimental measurement of the diameter of the CO ² laser beam used.

Figure 12. Chewing gum tests No. 1

Figure 13. Tests on chewing gum No. 2.

Figure 14. Chewing gum # 23, test 69, B = 140 mm / s.

Figure 15. Chewing gum # 23, test 70, B = 150 mm / s.

Figure 16. Chewing gum # 24, tests 71 and 72, B = [160; 170] mm / s respectively.

Figure 17. Chewing gum # 16, trial 57, A = 1 pass.

Figure 18. Chewing gum # 17, A = 10 passes.

Figure 19. Tests 1 and 2 on stone.

Figure 20. Tests 3, 4, 5 and 6 on stone.

Figure 21. Tests 7, 8, 9, 10 and 11 on stone.

Figure 22. Tests on stone at 140 mm / s and [21.1; 22.4; 23.7; 25.0] W respectively. Figure 23. Tests on stone at 150 mm / s.

Figure 24. Tests on stone at [400; 420; 440] mm / s.

Figure 25. Chicle work zones as a function of power-speed torque.

Figure 26. Chicle work zones as a function of the power-n0 pair of passes.

Figure 27. Chicle work zones as a function of the number of passes-speed pair.

Figure 28. Chicle work zones at maximum power as a function of speed-n0 torque of passes.

Figure 29. Areas of involvement of the Rosa Porrino stone when irradiated with laser.

Figure 30. Intersection of the total gum removal zone (A) and the zone in which the stone is not damaged (B).

Figure 31. Work area in black outline and representation of tests as an example of evolution.

Figure 32. Chewing gum n ° 42, A = [28-33] passes, B = [710-720] mm / s and C = 25 W.

Figure 33. Chewing gum n ° 43, A = [28-35] passes, B = [730-740] mm / s and C = 25 W.

Figure 34. Flow diagram of the image processing stages: original image, grayscale image, binarization, image processing, final image, and laser delivery.

Figure 35. Original image of two chewing gums on the stone.

Figure 36. Grayscale image next to the binary image obtained by setting a threshold of t = 0.55.

Figure 37. Grayscale image together with the binary image obtained by setting a threshold of t = 0.7.

Figure 38. Grayscale image next to the binary image obtained by setting a threshold of t = 0.85.

Figure 39 Eroded image with circular nucleus of radius 9.

Figure 40 Eroded image with circular nucleus of radius 20.

Figure 41. Image after the elimination of the small areas with a number less than 19,000 pixels.

Figure 42. Image after the dilation process with a circular nucleus of radius 20.

Figure 43. Comparison between the original RGB image and the processed binary image.

DETAILED DESCRIPTION OF THE INVENTION

Pre-tests

First, Rosa Porriño stone samples were taken and chewing gums were progressively adhered to the rough face of the pieces. This finish is known as serrated or serrated and provides a smooth but somewhat rough, porous and matte surface, with a slightly blurred appearance. It is achieved by cutting granite blocks using steel or diamond strips, diamond wire or diamond blades. In figure 7 you can see the chewing gum on the Rosa Porriño stone with a sawn finish.

These were left outdoors for about a month until the first tests began. Logically, the healing time is not the same as the gum would spend on the streets of an urban environment, however as a first approach or study of laser gum cleaning it is acceptable.

Then, bearing in mind that one of the main objectives of this work is to find the optimal parameters for cleaning this polymer without causing significant damage to the underlying stone, the problem was posed from the beginning as illustrated in figure 8, which shows gum-stone working limits.

Therefore, what is interesting is to look for irradiance and energy thresholds per unit length for both the stone and the chewing gum, in such a way that, if the threshold / is exceeded for the gum but not the the stone, it is possible to operate in a range of values that allows the elimination of the residue without producing visible imperfections in the stone.

The concepts involved are:

• Irradiance: varies with the distance to the lens and is the magnitude used to describe the power of the laser beam per unit area of the beam at that distance.

where Pm represents the average power of the laser beam and the denominator of the equation is the area that the beam sweeps over the part, 0 ^h being the diameter of the beam in the sample.

* Energy per unit of length:

The irradiance distribution in a plane transverse to the laser beam obeys a single peak Gaussian profile, which corresponds to the TEM 00 transverse mode. This distribution is illustrated in figure 9.

For the simplification of the calculations, the mean value of the irradiance will be taken and it will be assumed to be uniform throughout the entire exposure time of the beam on the surface.

Average, uniform

where r is the exposure time: r = -, where v is the speed of the beam.

Thus, the energy per unit length that represents the energy deposited by the beam per unit length in the direction of advance of the ray (x-axis) is:

Figure 10 shows a diagram of how the laser beam would be swept along the x axis at a tracking speed v. The width of the channel is given by the diameter of the beam in the sample, 0H.

The C 02 laser software used allows the configuration of various parameters. Thus, apparently previously, it seems reasonable that the work parameters taken for all the tests are the following:

• Power of the laser beam: the laser software used allows it to be varied with a range of [50 - 1000] parts per thousand of the 25 W output power. It represents the energy deposited on the work surface per unit of time. Hereinafter the author will refer to the power of the laser in watts.

• Tracking speed: with a range of [20 - 1500] mm / s. Determine the interaction time between the beam and the material to be treated.

• Number of passes: it is to be expected that just an interaction of a pair of power-speed values will not be sufficient to completely remove the gum from the surface.

In addition, the following were fixed as constants: the working distance, at 29.5 cm -30 cm, which in this way coincides in the working plane the focal plane of the optical system that integrates the laser head and the type of substrate on which do the tests (sawn surface of Rosa Porrino stone).

Experimental procedure

43 chewing gums and two Rosa Porriño stone samples were used and 173 tests were carried out, of which 114 were on chewing gums and 59 on granite directly.

The tests were carried out by irradiating the laser beam in a direction perpendicular to the stone samples, in a continuous regime and focusing the beam on the surface of the chewing gums.

In order to be able to determine the separation value between the fill lines, the diameter of the laser beam was measured experimentally. For this, a line was drawn in the LaserCAD environment and irradiated with the beam on photographic paper. Next, the optical microscope was used to visualize the laser trace together with a ruler with minimum divisions of 0.5 mm. This is illustrated in figure 11.

As can be seen, the diameter of! beam is about half a millimeter. This is why the distance between lines was set to 0.5 mm (FÍII300), since the overlap between beam traces is practically zero.

An experiment design was also determined, which was the following: A (number of passes) and B (tracking speed) would be set and C (power) would be varied to obtain a minimum of three values of this.

In addition to this, it was also necessary to perform experiments on bare stone to determine the damage threshold of the Rosa Porriño granite substrate.

Analysis of samples

In this chapter an analysis is made of the influence that the variation of the three chosen parameters (power, speed and number of passes) has on both the chewing gum and the stone. All the essays are arranged by rows in each of the photographs.

Influence on chewing gum

Influence of power variation

Figures 12 and 13 show the first experiments carried out. For all of them, A = 1 pass and B = 100 mm / s.

Figure 12 shows the tests on gum # 1 and Figure 13 shows the tests on gum # 2.

Table 1. Power values set for each test performed on chewing gum n ° 1 and n °

2.

As can be seen in the previous figures, an increase in the radiant power implies a greater elimination of the plastic residue due to the fact that there is greater heating on the surface thereof, which facilitates the vaporization of the gum.

Influence of speed variation

The following four tests were performed with A = 6 passes and C = 17.1 W.

Figure 14. Chewing gum # 23, test 69, B = 140 mm / s.

Figure 15. Chewing gum # 23, test 70, B = 150 mm / s.

A slightly greater amount of gum remaining after the treatment is observed in test 70 than that remaining in test 69. This result is logical because the higher the speed of the pass, the lower the energy per unit length than the laser provides the residue. The two trials in Figure 16 show another example.

Figure 16. Chewing gum # 24, tests 71 and 72, B = [160; 170] mm / s respectively.

Influence of the variation of the number of passes

The values set for tests 57 and 62 were: B = 80 mm / s and C = 19.7 W.

Figure 17. Chewing gum # 16, trial 57, A = 1 pass.

Figure 18. Chewing gum # 17, A = 10 passes.

It can be verified by comparing test 57 with test 62, that an increase in the number of passes translates into greater removal of the polymeric residue from the substrate. The increasing absorption of heat by the gum that occurs with each new pass causes more of this gum to vaporize.

Influence on the stone

Carrying out tests directly on the stone is also of great importance, because it is essential to know the limits it presents in terms of incident beam power and beam tracking speed. The experiments were analyzed as follows:

Analysis of the influence of radiant power

For each speed value shown in Table 1, seven tests were performed with the power values indicated therein. For all of them, the number of passes was equal to one. In most of the tests on chewing gum, the value of 17.1 W was used, and ended up choosing, for power, then it seemed logical to start from that value to carry out the tests on the stone.

Table 2. Power and speed values used for tests on stone.

Figure 19 shows tests 1 and 2 on stone. Figure 20 shows tests 3, 4, 5 and 6 on stone. Figure 21 shows tests 7, 8, 9, 10 and 11 on stone.

The photographs in Figures 19-21 are shown by way of evolution. Tests 1 and 2 on stone correspond to power values of 21.1 W and 25.0 W respectively, both with speed values of 20 mm / s. In Figure 20, test 11 was performed at 23.7 W and 130 mm / s, where less damage is observed.

It was found that it was from 140 mm / s when it was difficult to distinguish the area of the irradiated stone from the non-irradiated one. This is illustrated in Figure 22, showing tests on stone at 140 mm / s and [21.1; 22.4; 23.7; 25.0] W respectively.

Analysis of the influence of passing speed

10 tests were carried out taking the fixed power (at the value taken in the first block of tests on the chewing gum) and increasing the speed from 140 mm / s, since with that value the best result was obtained visually in the previous analysis.

Figure 23 shows tests on stone at 150 mm / s. Figure 24 shows tests on stone at [400; 420; 440] mm / s.

Figures 23 and 24 illustrate the difference between, for example, irradiating at a rate of 150 mm / s, as shown by test 16, and irradiating at [400; 420; 440] mm / s as shown in Figure 24, which shows the intact stone.

Results

After having carried out all the necessary tests on the chewing gum, to achieve the elimination of the polymeric residue without damaging the stone and on the granite, to obtain reference values that would not cause detachment or alterations in its surface, an analysis was carried out exhaustive of the data obtained.

As the varied parameters were power, speed and number of passes, all of them were represented by graphs as follows:

• Chewing gum tests varying the power, speed and number of passes. • Chewing gum tests varying the speed and number of passes at maximum and fixed power (productivity regime)

• Tests on stone varying speed and power at fixed passes

In this section only the most significant graphs are presented that allow us to explain the conclusive results of the experiments carried out.

First, attention will be focused on seeing only the behavior of the gum after being irradiated with laser, then that of the stone alone and finally a work area will be defined taking into account the two previous analyzes.

Relating only to chewing gum

• Chewing gum tests varying the power, speed and number of passes (The average time of this group of tests is around one minute)

In the first place, we began by contrasting the values of the three parameters chosen from each test against the real result that it produced on each gum. Afterwards, different regions were created according to the effect produced on each gum in each of the three graphs made (power-speed, power-passes and passes-speed). As a result, different zones were obtained in the area of each graph, where each represents an effect that is achieved on the gum.

The work areas shown below have been obtained by experimental observation. As the parameters studied are three (power, speed and number of passes) and in each of the graphs only two parameters can be taken into account, it is necessary to create three graphical representations to provide complete results. These will be illustrated below.

- Power-speed

Table 3. Summary of the effects produced on the gum in each of the four zones of the power-speed graph.

Figure 25 shows the gum work zones as a function of power-speed torque.

- Power-passes

Table 4. Summary of the effects produced on the gum in each of the four areas of the power-number of passes graph.

Figure 26 shows the gum work zones as a function of the power-number of passes torque.

- Pass-speed

Table 5. Summary of the effects produced on the chewing gum in each of the four areas of the n ° of passes-speed graph.

Figure 27 shows the chewing gum work zones as a function of the torque number of passes speed.

• Tests on chewing gum varying the speed and the number of passes at maximum and fixed power (productivity regime)

In this second block of tests, the power was set at 25 W (maximum power provided by the laser used) and the speed was varied until reaching the maximum value that ensured total cleaning without damaging the stone in a reasonable number of passes. It is important to note that with the values of the starting parameters, that is, with those established in the previous test block, acceptable cleaning results are already obtained: the gum is completely removed but the stone is not always intact. . Therefore, the objective is to try to cause minor damage to the stone and carry out this removal as quickly as possible.

Table 6. Summary of the effects produced on chewing gum under the productivity regime

Figure 28 shows the working areas on gum at maximum power as a function of the speed-n0 torque of passes.

The time invested in cleaning was approximately 50 s, which implies a certain advance compared to the minute reached in the first test block.

Relating only to stone

^• Trials on stone varying speed and power at fixed passes

In this section the graph that summarizes the conclusions derived from the analysis of the tests carried out on the exposed stone will be presented. For the 59 experiments on granite, the number of passes was set to one. This was so because the stone is clean from the beginning. What happens is that it would be possible to damage the stone even if some parameter values are set that do not damage the stone in one pass. This is due to the progressive heating that the stone undergoes due to the accumulation of successive passes on the same point of the substrate.

Thus, the graph in Figure 29 provides only a first approximation when analyzing how laser radiation affects the Rosa Porriño stone.

Table 7. Summary of the effects produced on bare granite.

Figure 29 shows the affected areas of the Rosa Porriño stone when irradiated with laser.

Relating to the stone-gum set. Work zone.

This section tries to group the results obtained previously so that they serve as a guide from which to start when carrying out future experiments.

At the beginning of the chapter, the cleaning problem was posed in such a way that the objective was to find irradiance and energy thresholds per unit length for both the stone and the gum.

It is seen that after all the analysis of the tests, it is more practical and revealing to observe the results in a graph that represents the parameters of the laser that have been modified; In this case, the power-speed graph has been chosen because the number of passes must be a parameter established by the operator by observing the work surface, since this is closely related to the thickness of each gum in question and therefore the value that it can take varies in a very wide range.

Therefore, it seems logical to think that the work area will be defined by the intersection of two regions: the area where the gum was completely removed and the area where the stone was not damaged. This would be the case if the number of passes set for all tests on chewing gum and stone had been the same, but it must be taken into account that the tests on stone were carried out with only 1 pass. Therefore, at the intersection of both green areas, there are trials in which it is possible to eliminate the gum but also damage the stone (see Figure 30).

Figure 30 shows the intersection of the zone of total gum removal (A) and the zone in which the stone is not damaged (B).

Thus, it is necessary to limit the work area only to tests that have fulfilled both objectives. To do this, the experiments contained in the area surrounded by white in figure 30 were reviewed and it was verified from which the damage on the granite was null.

Figure 31 shows the resulting work area surrounded in black and it can be seen that from 320 mm / s an optimal result is achieved. It also includes two photographs of two tests, one carried out at 140 mm / s, 17.1 W and 6 passes, where the wear of the stone is observed; and the other done at the same power, 440 mm / s and 25 passes in which the granite is perceived in perfect condition.

Summary of Results

Next, the images of the tests in which a total removal of the polymeric residue was achieved without causing any damage to the Rosa Porriño stone will be shown. underlying. A table will also be presented summarizing the optimal values of the chosen laser parameters.

In Figure 32 the tests of gum no.42 are shown, A = [28-33] passes, B = [710-720] mm / s and C = 25 W. In figure 33 the tests of gum no. 43, A = [28-35] passes, B = [730-740] mm / s and C = 25 W.

Table 8. Summary of optimal laser parameter values.

With regard to the application of this cleaning technique, it would be convenient to set the power and speed values as indicated in the previous table and establish the number of passes at a starting value equal to 10; in this way, the operator can subsequently adjust the number of passes according to the result observed after the first treatment. This will ensure a finer adjustment and therefore a more satisfactory result.

Image capture and treatment

The inclusion of an artificial vision system in this cleaning application with key lasers and essential to ensure an optimal result because it allows to visualize in real time, through a camera, the surface that is susceptible to be treated with laser. This facilitates the correct positioning of the portable system by the operator. If this system is not integrated, both the areas contaminated by plastic waste and the areas of the surface that are clean from the beginning will be irradiated, causing even greater damage to the substrate in addition to wasting electrical and optical energy.

Thus, this section describes the procedure followed to process the Images that the camera will take. Afterwards, it will end with the presentation of a proof of concept in a real laser.

In order for the image of a real object to be processed, it must be obtained in digital format. This first step (digltalization) is performed by the image capture device or camera, which divides the image into as many fragments arranged by columns and rows, called pixels, as indicated by its resolution. Each of these pixels is represented by a number of bits set by the technology of each camera. In particular, the analog-to-digital converter (CAD) of the selected camera is configured to provide 12 bits per pixel of the image. The quality of the image taken will be the better the greater the number of pixels in an image or the greater the number of bits used to define each pixel. Once the image has been acquired, the process of processing it by computer can begin.

In this work, the processing was carried out by means of the numerical calculation software MATLAB (version R2018b). The flow chart in Figure 34 shows the particular phases that make up the process followed.

As can be seen in Figure 34, the final objective of this process is to be able to send the laser control software an image that contains only two basic and sufficient information: an area that defines the background (the substrate, what is not wanted irradiate) and another area (or areas) that enclose the waste to be treated (target area).

First, an image of the chewing gums was taken on the Rosa Porrino substrate and it was used as a sample in the treatment (see Figure 35).

It was necessary to convert the previous image (RGB image), which is in color, to a grayscale image that would allow its subsequent treatment. In a grayscale image each pixel is represented with a bit value in the range of 2 to 8 bits or more. Thus, with the MATLAB rgb2gray command, a grayscale image is obtained where each pixel of the Image is represented by 8 bits, then each one of them (pixels) can take a gray value of between 28 = 256 tones.

This image was then binarized. This process consists of comparing the level of gray that each pixel of the image presents with a predefined threshold value (t), and depending on whether that level is greater or less than the threshold, all the pixels are classified into two segments. Each grouping is assigned a logical value (0 or 1), where '0' usually corresponds to black and 'T' to white. Figures 36 to 38 show the grayscale image (left) and the binary image (right) together, for different threshold values, in particular t = 0.55 in Figure 36, t = 0.7 in the figure 37 and t = 0.85 in figure 38.

The ideal threshold value was obtained through successive tests as can be seen in the three previous figures. The threshold was set at t = 0.7, which was the one for which the best result was obtained.

Once the binary image was obtained, the target area and the bottom area were defined more clearly. For this, a morphological filtering was carried out that consisted of the application of two basic operators in the following order: erosion and dilation. Both need to have a defined nucleus and its dimension, which can have the shape of a circle, square and others. The core is the pattern that defines to which zone (object or background) each evaluated point of the image belongs.

Erosion consists of eliminating the perturbations of an image that are contained in the dimension of the nucleus that is applied on the image. The following images show the results obtained after the erosion of the binary image. In both, the core taken was circular, but in Figure 39 it was applied with radius 9 and in Figure 40 the radius was 20. The radius was set at 9 for the best result shown.

Before the dilation operation, it was necessary to remove all the white elements that appear in Figure 39 that are not part of the two chewing gums. Although they do not belong to the waste, the software classified them as 'object'. To eliminate these zones, a command was executed that allows removing all the connected object zones that contain less than a parameterizable number of pixels. For the image analyzed in this process, this value turned out to be 19,000 pixels. The result can be seen in figure 41.

What remains is to fill in the gaps that are observed that have been inside and on the edges of the gum. This can be done by a dilation operation, which is the reverse of erosion. It consists of increasing the size of the object in question by applying a nucleus, in this case also circular and with radius 20. Thus, it selects the points for which the translation of the nucleus touches some point on the object. In figure 42 the image is observed after this treatment and with it the processing of the original image ends. A comparative figure between the starting image and the final image obtained is also shown (see Figure 43).

It is observed that the result of the treatment in the upper gum is not entirely optimal. This may be due to a lack of lighting as well as a lack of refinement in the image processing. The objective of this project is not to improve or advance image processing techniques but to carry out a fairly valid preliminary treatment that justifies the use of an artificial vision system as an important part in laser gum cleaning and to demonstrate that it is possible to send said image treated to the software that controls the beam of any laser.

In addition to the above, the following components are necessary to ensure optimal working conditions:

- Vapor evacuation system: to eliminate the gases that are produced when vaporizing the gum residues with the laser.

- Artificial vision system: consisting mainly of a camera (and its lens) that will be used to take images in real time of the surface to be treated. Subsequently, these photographs will be properly processed and sent to the laser control software, which will move the motors that govern the galvanic mirrors of the scanning head, deflecting the beam in such a way that it only irradiates the regions that occupy the gum or gum in question. .

- Alternative identification system for the waste to be cleaned: Additionally, an identification system for the waste of chewing gum to be cleaned may be used by means of a portable spectroscopy system coupled to the artificial vision system, in such a way that it automatically identifies whether the residue to be cleaned is adequate or not.

- Optical targeting system. It can be done through lenses or mirrors or both.

Regarding the system to evacuate the vapors produced, two options were considered, based on opposite operations:

• Elimination by means of an air jet: this alternative entails the need to include an air compressor that can provide an air stream over the area to be treated.

• Aspirator and gas extraction nozzle: this option consists of eliminating the vapors by sucking them up and filtering them through an active carbon filter to eliminate pollutants from the gas stream.

Claims (12)

1 - Method for cleaning chewing gum residues on floors or vertical walls, characterized in that it comprises a stage of obtaining the image of the residue to be cleaned by means of an artificial vision system, a stage of determining the surface to be cleaned , and an application step on the area covered by the residue of a laser beam, in such a way that the chewing gum residue is eliminated without altering the pavement or wall on which it is adhered. 2. - Method according to claim 1, characterized in that the step of determining the surface to be cleaned comprises: converting the image obtained to a grayscale image, in the case that the image obtained is a color image; binarize the grayscale image; apply an erosion operator on the binarized image; eliminate areas of the image that contain less than a parameterizable number of pixels; and apply a dilation operator. 3. - Method according to any of the preceding claims, characterized in that the application stage on the area covered by the residue of a laser beam is carried out with a CO ² laser. 4 - Method according to any of the preceding claims, characterized in that the power of the laser beam is in the range of 1.25 to 25 W. 5 - Method according to the preceding claim, characterized in that the power of the laser beam is 25 W. 6. - Method according to any of the preceding claims, characterized in that the laser beam tracking speed is in the range of 20 to 1500 mm / s. 7. - Method according to the preceding claim, characterized in that the laser beam tracking speed is in the range of 730 to 740 mm / s. 8. - Method according to any of the preceding claims, characterized in that a number of passes of the laser beam greater than or equal to 10 is carried out. 9. - Method according to the preceding claim, characterized in that the number of passes of the laser beam is between 28 and 35. 10. Method according to any of the preceding claims, characterized in that it comprises using a vapor evacuation system to eliminate the gases that are produced when vaporizing the chewing gum residues with the laser. 11. Method according to the preceding claim, characterized in that the removal of gases is carried out by means of an air jet provided by an air compressor. 12.- Method according to claim 10, characterized in that the removal of gases is carried out by suctioning them with an aspirator and a gas extraction nozzle and their subsequent filtering through an active carbon filter to remove contaminants from the gaseous stream. 13 - Method according to any of the preceding claims, characterized in that the artificial vision system includes a camera with a lens to take images in real time of the surface to be treated and that the application of the laser beam is carried out by means of a head scanning comprising motor-driven galvanic mirrors; where the method comprises processing the images taken by the camera and sending the images to a laser control software, to move the motors that govern the galvanic mirrors of the scanning head, deflecting the beam in such a way that it only irradiates the occupied regions by chewing gum. 14 - Method according to any of the preceding claims, characterized in that the focusing is done by means of lenses and / or mirrors.