Method and device for detecting asbestos fibres in a building material and / or in rubble fractions, and computer program product

Abstract

The invention describes a device and a method (100) for detecting asbestos fibres (12) in a building material (11) and / or in rubble fractions (11), wherein the method (110) comprises: irradiating a building material (11) and / or rubble fractions (11) with unpolarized white light (λ0); subsequently splitting reflected light (λref) into polarized light beams (λref, pol) in at least two polarization directions; and subsequently detecting the polarized light (λref, pol) reflected by the building material and / or by the rubble fractions and divided in at least two directions, evaluating the detected reflected polarized light (λref, pol) in order to detect a possible colour difference in the at least two polarization directions, wherein asbestos fibres (12) are present in the building material (11) and / or in the rubble fractions (11) when the colour difference is detected.

Description

METHOD AND DEVICE FOR THE DETECTION OF ASBESTOS FIBERS IN A BUILDING MATERIAL AND / OR IN CONSTRUCTION WASTE FRACTIONS, AS WELL AS COMPUTER PROGRAM PRODUCT Description A method for detecting asbestos fibers in a building material and / or in construction waste fractions and a device for detecting asbestos fibers in a building material and / or in construction waste fractions are described. In light of the German Substitute Building Materials Ordinance [1] and the European Union's call for its member states to significantly increase the use of recycled materials in the construction industry [2], the use of sensor-based bulk material sorting for construction and demolition waste has been in high demand in recent years. Around 600 million tons of mineral raw materials are used annually in the construction sector in Germany alone [3]. Currently, primary raw materials are predominantly used in the production of new building materials, as only 81 million tons of construction and demolition waste are recycled for construction applications each year. A large proportion of demolition material is landfilled or used as fill material in road construction [4]. In the area of ​​construction waste processing, material generated after demolition must be tested for hazardous substances, as these require separate handling. The most common hazardous substance found in construction waste is asbestos, a material primarily used in the 1960s for its insulating and stabilizing properties, or added to other building materials in the form of asbestos fibers. If asbestos fibers are released into the air, for example during demolition, they pose a risk to human health due to their ability to penetrate deep into the lungs and can lead to lung cancer (so-called "asbestosis"). The early and unambiguous detection of asbestos in construction waste is therefore a necessary step to determine the subsequent treatment steps for the waste and to prevent asbestos fibers from entering recycled building materials. For conventional asbestos products such as sprayed asbestos, asbestos cement, cardboard, and cords, analytical methods are now highly standardized and uniform, ensuring reliable results. Asbestos is analyzed using scanning electron microscopes (VDI 3866 Sheet 5 "Determination of asbestos in technical products - scanning electron microscopy method").

[2004] ) or specially equipped light microscopes (VDI 3866 Sheet 4 “Determination of asbestos in technical products - Phase contrast microscopy method”

[2002] ) detected in material samples. Fingernail-sized pieces are sufficient for this; however, larger pieces make it easier for samplers and laboratories to carry out sufficiently accurate analyses, so that materials with little or no suspicion of asbestos content can be separated from those where there is a high suspicion. These methods are reliable from 1% asbestos content upwards; however, at concentrations below 1%, only chance findings can result unless the samples are prepared using special digestion procedures. Using analytical methods that make it possible to reliably detect asbestos contents below 1%, contamination is found in many applications with very large affected areas, especially in all polymer- or organically bound materials such as floor coverings, adhesives, fillers and corrosion protection coatings. For this purpose, it is necessary to incinerate the binder component in the oven before the actual analysis in the laboratory, and then to dissolve the lime and gypsum components with dilute hydrochloric acid. Both components mask the asbestos in the sample. Given the implications of these findings, these analytical methods must be incorporated into the routine of all laboratories. Sampling strategies must be significantly intensified, and existing reports and inventories must be reviewed where necessary. (Source: Pollutants in Indoor Spaces and on Buildings, pp. 215-216, German Association for Pollutant Remediation, 2nd edition, 2014) When detecting asbestos fibers using microscopy, polarization filters are often employed. These allow the detection of not only the shape but also the specific optical properties of asbestos fibers. Through manual operation and visual evaluation, the effect of birefringence can be observed. Another method for detecting asbestos fibers, which is not yet established due to cost, involves hyperspectral sensors in the short-wave infrared (SWIR) range. This method is less precise but is suitable for use by non-specialists and for high material throughput. Two approaches exist, depending on the type of sensor. Hyperspectral imaging sensors in the short-wave infrared (SWIR) range enable the identification of asbestos fibers. Studies have demonstrated this using line-scanning HSI sensors. A two-dimensional image is generated by the samples moving orthogonally beneath the sensor. The spectra of the individual pixels form a characteristic spectral fingerprint that is processed using statistical analysis methods. Previous studies have varied considerably in the distance between the sensor and the sample, and thus in the effective spatial resolution of the objects under investigation. In recent articles, the trend is moving away from microscopic laboratory investigations towards larger-area detection of multiple samples [5]. Hyperspectral point sensors consist of only a photodiode or a single pixel. This makes them more cost-effective and smaller in size. They are therefore often referred to as handheld spectrometers. However, depending on their image acquisition method, there are limitations in the covered spectral range and the quality of the acquired spectral information. Furthermore, only a single measurement at one point on the sample is possible. Previous approaches use an HSI point sensor as a handheld device for asbestos identification. The measurement is performed close to the sample. The previous approach uses an expensive sensor which, despite its handheld nature, has a bulky and unwieldy design. Due to advances in handheld spectrometers, this approach can no longer be considered state of the art [6]. One object of the present invention is to provide a method and a device with which large quantities of construction demolition waste can be automatically tested for asbestos, in particular directly on site at the construction site or at a construction waste processor. This problem is solved by the subject matter of the attached independent claims. The present invention describes a novel method for detecting asbestos fibers using an imaging sensor in a single setup. It utilizes a crystal-optical property of asbestos fibers, the so-called pleochroism effect. Compared to the prior art, this method has the advantage that It is possible to test large quantities of construction waste for asbestos in a short time using cost-effective sensors. The proposed method for detecting asbestos fibers in a building material and / or in construction waste fractions involves irradiating the building material and / or construction waste fractions with unpolarized white light. This reflects light rays, which are then to be detected. When detecting the reflected light rays, the reflected light is split (A re f) into polarized light rays (A re f, poi) in at least two polarization directions; and subsequent detection of the polarized light reflected by the building material and / or the construction waste fractions, split in at least two directions. After detecting the polarized reflected light, it is evaluated to identify any color difference in the at least two polarization directions. If a color difference is detected, asbestos fibers are present in the building material and / or in the construction waste fractions. The failure to detect an asbestos fiber, which corresponds to the failure to detect a color difference, using the proposed method does not guarantee that the construction waste is free of asbestos. However, the detection of a color difference guarantees that asbestos is present in the construction waste. Another aspect of the present invention relates to a device for detecting asbestos fibers in a building material and / or in construction waste fractions. The device comprises a light source for irradiating a building material and / or construction waste fractions with unpolarized white light. The device also comprises a splitter for splitting the reflected light into polarized light beams in at least two polarization directions; and a sensor for detecting polarized light reflected from the building material and / or the construction waste fractions in at least two polarization directions. The device further comprises an evaluation device for evaluating the detected reflected polarized light in order to detect any color difference in the at least two polarization directions, wherein the detection of such a color difference indicates the presence of asbestos fibers in the building material and / or the construction waste fractions. The light source and sensor can be, for example, a polarization camera. The evaluation device can be a computer, to which the captured signals—that is, the reflected, polarized, and detected light rays—are transferred. This allows the computer to run an algorithm with the Run the transmitted signals to determine a color difference, provided that asbestos is present in the construction waste. The proposed device detects asbestos fibers that are freely present and can interact with polarized light. Fibers still embedded in a concrete matrix remain invisible to the system consisting of the device and the construction debris being imaged and are not registered as asbestos. However, after mechanical treatment (e.g., in a jaw crusher), the likelihood increases that individual fibers will be exposed and thus detectable, as disclosed by the method described herein. Furthermore, a computer program product comprising instructions which, when loaded and run on a computer, cause the computer to perform a procedure described herein, in particular the evaluation of the detected reflected polarized light, is described. Furthermore, a computer-readable medium is described on which the computer program product is stored. The method and apparatus described herein utilize an effect called pleochroism. Pleochroism, from the Greek word for "multicolored," is the phenomenon where minerals exhibit different colors when viewed from different crystallographic directions under polarized light. This property is due to the anisotropic nature of minerals, which causes them to absorb light differently along different crystallographic axes. Pleochroism is frequently observed in minerals that show a significant difference in light absorption along various crystallographic directions. Pleochroism is typically observed using a polarizing microscope, where the mineral is placed between crossed polarizers and the stage is rotated to different orientations to observe color changes. Rotating the stage can cause the mineral to display a range of colors, from no color (extinction) to one or more distinct colors. The number of colors and the intensity of the pleochroism can provide important clues for mineral identification, as different minerals exhibit unique pleochroic properties. Asbestos, from the Greek word meaning "imperishable," is the collective term for naturally occurring, fibrous silicate minerals with fiber diameters down to 2 micrometers (1 micrometer equals one thousandth of a millimeter). Asbestos is chemically very resistant, insensitive to heat, and non-combustible. It exhibits high elasticity and tensile strength and, due to its bonding properties with other materials, can be easily processed into products. Because of its special properties, asbestos has been used in a wide variety of products since around 1930. These include panels for building construction, brake and clutch linings for vehicles, seals, and molding compounds for high thermal or chemical loads. Further advantageous embodiments of the present invention are the subject of dependent patent claims. Preferred embodiments of the present teaching are described below in connection with the accompanying figures. It is understood that the described embodiments do not limit the scope of the teaching described herein. The figures show: Fig. 1 shows a flowchart of the process for detecting asbestos fibers in a building material and / or in construction waste fractions; Fig. 2 shows a device for detecting asbestos fibers in a building material and / or in construction waste fractions; Fig. 3 shows a more detailed flowchart of the process for detecting asbestos fibers in a building material and / or in construction waste fractions; and Fig. 4 is a false-color image showing a result after carrying out the method for detecting asbestos fibers in a building material and / or in construction waste fractions. The principle of the teaching disclosed herein will be further clarified below with reference to possible embodiments, whereby the detailed description of individual embodiments does not constitute a limitation of the teaching described herein. Individual aspects of the invention described herein are described below in Figures 1 to 4. In the present application, identical reference numerals refer to identical or equivalent elements, and it is not necessary for all reference numerals to be repeated in all drawings. The established method for detecting asbestos fibers is performed in the laboratory using microscopes. This evaluation relies on manual visual assessment by specialists. The optical properties used in this method form the basis of a new procedure for detecting asbestos fibers macroscopically, i.e., by examining a larger area of ​​the sample, using imaging sensors. Sensor data analysis is automated using intelligent algorithms. This enables low-volume, in-line operation. Fig. 1 shows a flowchart of the proposed method 100 for the detection of asbestos fibers 12 in a building material 11 and / or in construction waste fractions 11. In a first step 110, the method 100 comprises irradiating a building material 11 and / or construction waste fractions 11 with unpolarized white light Ao. The irradiation with unpolarized light can be carried out with the aid of a source 20, which may be a polarization camera or an external light source emitting unpolarized light. The unpolarized light strikes the building material 11 and / or the construction waste fractions 11 and is reflected by it / them. In a second step 120, the method first comprises splitting the reflected light into polarized light beams A. re f, poiin at least two polarization directions. Preferably, the reflected light is split into four polarization directions. Subsequently, in a third step 130, the polarized light A, reflected from the building material and / or from the construction waste fractions and split into at least two directions, is detected. re f, po i. The detection of the polarized light reflected from the building material and / or from the construction waste fractions, split in at least two directions A re f, poi This is done by means of a sensor. The sensor can be a polarization camera or another device that detects the polarized light A reflected from the building material and / or the construction waste fractions, split in at least two directions. re f, poi is suitable. In a fourth step 140, the procedure includes an evaluation of the detected reflected polarized light A.re f, po i, in order to detect a possible color difference in the at least two polarization directions, wherein, upon detection of the color difference, exposed asbestos fibers 12 in the building material 11 and / or in the Construction waste fractions 11 are present. The evaluation 140 is carried out with the help of an evaluation device 50, such as a computer or a smartphone. Fig. 2 shows a device for detecting asbestos fibers in a building material and / or in construction waste fractions, which is designed to carry out the method 100 described herein. The basis for detection, particularly automated detection, is the pleochroism effect, or birefringence, present in asbestos fibers. In this effect, the color of anisotropic crystals varies depending on the direction of vibration; that is, different parts of the light spectrum are absorbed to varying degrees depending on the direction of transmission. This phenomenon, in which a mineral exhibits a perceptible color change and / or different color intensities, is called pleochroism. This direction-dependent absorption can range from weak to very strong. The mineral chrysotile, a representative of asbestos, exhibits, for example, weak pleochroism ranging from pale greenish-yellow to pale greenish. Crucial for pleochroism is the dependence of absorption on wavelength, polarization, and propagation direction. When unpolarized white light is shone onto a surface, the resulting color impression, known as surface color, is an additive mixture of the two polarized colors. This effect can be demonstrated by splitting reflected light rays from the object under investigation into polarized light rays, which then exhibit a color difference. This is achieved by separating the rays, for example, using a crystal, such as a calcite crystal in a dichroscope, or by employing polarizing filters, such as in a polarizing microscope. The splitting of the reflected light A re f in polarized light rays A re f, poiThis can be done using a crystal or at least two polarizing filters. The crystal or polarizing filters are integrated into a polarizing camera. Alternatively, a calcite crystal in a dichroscope or a polarizing microscope could be used to polarize the reflected light rays. re f in reflected, polarized light rays A re f, P to divide oi. Detection is performed automatically, in particular by means of an imaging sensor 40, and / or evaluation is performed automatically, in particular based on an algorithmic evaluation method. After detection of the reflected polarized light A re f, po i, which is captured as an electronic signal, the electronic signal can be passed to the evaluation device on which the algorithmic evaluation procedure can be executed. Detecting the reflected polarized light A re f,poi at a distance d to the building material 11 and / or to the construction waste fractions 11 of 20 cm or more, preferably 30 cm or more. Method 100 can therefore be carried out on a macroscopic level, i.e., at a possible distance d of the sample to the imaging sensor 40 of >20 cm. The described pleochroism effect forms the basis of the proposed method 100, i.e., detection and evaluation, particularly automated detection, using an imaging sensor on a macroscopic level. A polarization camera can be used for detection, with which four images with four different polarization directions can be simultaneously acquired. Preferably, the detection of the reflected polarized light A is carried out. re f, Poi using a polarization camera which simultaneously captures at least two, preferably four, images, each with a different polarization direction. A polarization camera can, for example, be taken to a construction site where the rubble is located. There, it can be photographed at a distance d > 20 cm, so that the reflected polarized light A re f, poi as an electronic signal. The polarization camera can transmit the electronic signal to the evaluation device, such as a smartphone or a laptop, i.e., a computer, for evaluation of the electronic signal. Fig. 3 shows a more detailed flowchart of process steps 130 and 140 of process 100 for the detection of asbestos fibers 12 in a building material 11 and / or in construction waste fractions 11. In step 130, the reflected, polarized light A is detected. re f, Poi by taking at least two, preferably four, images with different polarizing filter positions, i.e., in different polarization directions. Each image includes polarized reflected light in one of the different polarization directions. Each pixel of a single captured image comprises four RGGB subpixels, with each RGGB subpixel containing four subsubpixels. Each subsubpixel has a polarization filter to adjust the reflected light A reThe system captures images in at least two, and preferably four, polarization directions. The smallest unit, the subsubpixels, contain the polarization filters. Thus, a maximum of four polarization directions can be captured. From each pixel, four RGB color images with four different polarization directions can be "calculated." Evaluation is then performed using these four images, although only a minimum of two images are required. Each individual image then has an RGB vector associated with each "pixel." Each image for each subsubpixel comprises an RGB vector. In other words, each pixel, consisting of four RGB subpixels, is itself subdivided into four subsubpixels. Each of these four smallest pixel units has a polarization filter that can perform a rotation of 0°, 90°, 45°, and 135°. The result of a single capture is therefore RGB images with four different polarization directions.These recordings serve as input for algorithmic evaluation, particularly automated evaluation. The goal of image analysis is to detect changes in color value. The following processing chain is used for this purpose: Preferably, evaluating the potential color difference involves performing a color space transformation of the captured images from an RGB space to an HSV space (see 141 in Fig. 3). The RGB images with different polarization directions, converted as described above, form the input (also referred to as input above) to the algorithm in step 141. For this purpose, the algorithm uses at least two RGB images for further evaluation. Therefore, we refer to (transformed from the sensor data) RGB images whose pixels consist of RGB vectors. This means that the term "pixel-wise" is used hereafter, referring to the transformed RGB images at the algorithm's input. Consequently, the term "pixel-wise" was used in Fig. 3. Following step 141, noise reduction is performed in step 142, in particular using a median filter and / or a Gaussian filter. Simultaneously or sequentially, and especially starting from the transformed RGB images at the algorithm's input, a pixel-wise calculation of a difference and / or variance (see 143 in Fig. 3) and a sub-subpixel-wise calculation of the distances between the RGB vectors are performed. among themselves (see 144 in Fig. 3). The results from steps 143 and 144 are used in step 145 to perform a pixel-by-pixel comparison of the calculated difference and / or the calculated variance and / or the calculated distances, each with a threshold value. The evaluation in step 140 is based on the acquired subsubpixels or the RGB images with different polarization directions transformed from the subsubpixels. Steps 130, 141 to 145 are performed in the order shown in Fig. 3. The distance measure used is the Euclidean distance, which is a scalar value. The threshold value is therefore also a scalar value. Preferably, the method, in particular starting from the transformed RGB images at the input of the algorithm, comprises pixel-wise post-processing of the calculated difference and / or the calculated variance and / or the calculated, wherein the post-processing comprises evaluating a grayscale of the pixels, in particular of the transformed RGB images at the input of the algorithm. Grayscale analysis involves determining the number of adjacent pixels or sub-subpixels that differ from a defined threshold. Two pixels or sub-subpixels are considered adjacent if they are next to each other. Adjacent pixels then form a cluster, the size of which (number of pixels) can be determined. The threshold could be zero, meaning that non-zero values ​​(greater than or less than zero) differ from the defined threshold. However, it is also possible to define a different threshold, one that takes an offset into account, for example. The threshold could be set at 100 or another value. In post-processing, the number of adjacent pixels exceeding the threshold is recorded. If the number of contiguous pixels exceeds a defined value, this indicates a positive detection of asbestos-12. Preferably, the method 100 comprises representing the calculated difference and / or the calculated variance and / or the calculated distances of each pixel, in particular subsubpixel, in a mask in which the possible color difference or possible color differences is / are marked as the contiguous number of neighboring pixels, in particular subsubpixels. The mask is given in the form of an n x M matrix, where n is the number of columns and m is the number of rows. A cluster corresponds to the number of neighboring pixels, in particular subsubpixels. The result is therefore a A mask is created in which areas in the images exhibiting pleochroic properties are marked. The mask can be visualized as a false-color image (see Fig. 4). The detection of asbestos fibers 12 via the pleochroism effect is ultimately carried out by examining the created mask or by measuring the size of contiguous clusters within the mask. Method 100 comprises visualizing the calculated difference and / or the calculated variance and / or the calculated distances of each pixel, in particular each sub-subpixel, in a false-color image 400 to demonstrate the presence of asbestos fibers 12 in the building material 11 and / or in the construction waste fractions 11. Fig. 4 shows a false-color image 400 in which a result after carrying out method 100 for the detection of asbestos fibers 12 in a building material 11 and / or in construction waste fractions 11 can be seen. In Fig. 4, the detected asbestos fibers 12 are shown against a dark background. As previously described, Fig. 2 shows a device for detecting asbestos fibers in a building material and / or in construction waste fractions, which is configured to carry out the method 100 described herein. The device 10 for detecting asbestos fibers 12 in a building material 11 and / or in construction waste fractions 11 comprises a source 20 for irradiating a building material 11 and / or construction waste fractions 11 with unpolarized white light Ao. The source 20 can be an external light source or can be a source integrated into the polarization camera. In Fig. 2, the source 20 can be a polarization camera. The device 10 further comprises a splitting device 30 for splitting the reflected light A to be detected. re f in polarized light rays A re f, poiin at least two polarization directions; and a sensor 40 for detecting polarized light A reflected from the building material 11 and / or from the construction waste fractions 11 re f, poiin at least two polarization directions. The dividing device 30 and the sensor 40 can be two separate devices. In particular, the sensor 40 is a polarization camera for detection. In particular, an integrated solution of the dividing device 30 and the sensor in the source 20 (as shown in Fig. 2), as in a polarization camera, can be used. The polarization camera is portable and can be easily moved to a location where the building material 11 and / or the construction waste fractions 11 to be photographed are present. The polarization camera images of the building material 11 and / or the construction waste fractions 11 can then be used to carry out the method 100 to divide the building material 11 and / or the construction waste fractions 11. Asbestos 12 to be tested. The device 10 is equipped with an evaluation device 50 for evaluating the detected reflected polarized light A. re f,po i, connectable or connected to detect a possible color difference in the at least two polarization directions, wherein, upon detection of the color difference, asbestos fibers 12 are present in the building material 11 and / or in the construction waste fractions 11. It is conceivable that the dividing device 30 is used to divide the reflected light A re f in polarized light rays A re f, P oi is a crystal or is given by at least two polarizing filters. Polarizing filters are, for example, integrated into the polarizing camera. The sensor 40 for detection is designed to detect the reflected polarized light A ref, to automatically detect poi, wherein the sensor 40 is an imaging sensor such as the polarization camera, which simultaneously captures at least two, preferably four, images, each with a different polarization direction. The sensor 40 images the building material 11 and / or the construction waste fractions 11 to be imaged as a grayscale in pixels. Each pixel of a single captured image of the sensor 40 comprises four RGGB subpixels, with each RGGB subpixel having four subsubpixels, each subsubpixel having a polarization filter to reduce the reflected light A re f in one polarization direction of at least two, in particular four, directions. In order to carry out the method 100 with the device 10, two images of the building material 11 and / or the construction waste fractions 11 as grayscales in pixels are sufficient, each image representing a different polarization direction of the reflected light. The proposed method 100, in particular for evaluating the recorded images of the light reflected by the building material 11 and / or the construction waste fractions 11, can be implemented as a computer program. The computer program comprises instructions which, when loaded and run on a computer, cause the computer to execute the method 100, in particular the evaluation of the detected reflected polarized light A. re f, po i, as described herein. The proposed method 100 can be configured as a computer-readable medium on which the computer program product is stored. In contrast to existing solutions, the approach described herein enables, for the first time, the testing of large quantities of construction and demolition waste for asbestos, e.g., directly on-site at the construction site or at a construction waste processing facility. Unlike microscopic analyses in a laboratory, no sample preparation is required to detect asbestos. The method described here can therefore also serve as an incoming inspection for construction and demolition waste. Using readily available camera technology, materials can be tested for asbestos content in a non-contact inspection system. This allows for the testing of larger samples, up to and including 100% testing. Furthermore, the approach described herein is cost-effective compared to the technologies mentioned in the introduction. Suitable polarization cameras are currently available for a price in the upper three-figure euro range. In contrast to an examination using a microscope in the laboratory, the method 100 enables a particularly automated detection of asbestos fibers 12 at a greater distance from the sample and therefore a greater area coverage in an in-line operation. To achieve a true circular economy in the construction sector, the recovery of materials from construction and demolition waste for the production of new, high-quality recycled (RC) products is essential. In addition to the generally existing requirements for sorting materials into pure fractions, the detection of hazardous substances is crucial. Put simply, current national and international regulations and laws stipulate that if there is sufficient suspicion of contamination with hazardous substances, particularly asbestos, the material must be landfilled and may not be reused. Due to rapidly increasing landfill costs and resource scarcity (e.g., sand), it is important to keep the proportion of material requiring landfilling as low as possible and, conversely, to maximize the amount of material that can be used in RC products. Although some aspects have been described in connection with a process, it is understood that these aspects also constitute a description of a corresponding apparatus, such that a block or component of the apparatus can also be understood as a corresponding process step or as a feature of a process step. For reasons of redundancy, a complete description of the present invention in the form of apparatus features is omitted here. In the preceding detailed description, various features were sometimes grouped together in examples to streamline the disclosure. This type of disclosure should not be interpreted as indicating that the claimed examples have more features than are expressly stated in each claim. Rather, as the following claims reflect, the subject matter may consist of fewer than all the features of a single disclosed example. Consequently, the following claims are hereby incorporated into the detailed description, with each claim potentially representing a separate, independent example.While each claim can stand as a separate example, it should be noted that, although dependent claims refer back to a specific combination with one or more other claims, other examples also include a combination of dependent claims with the subject matter of any other dependent claim, or a combination of any feature with other dependent or independent claims. Such combinations are included unless it is stated that a specific combination is not intended. Furthermore, it is intended that a combination of features of a claim with any other independent claim is also included, even if that claim is not directly dependent on the independent claim. Bibliography

[0001] https: / / www. buzer.de / s1. htm?q=Substitute Building Material Ordinanceq&f=1 [2] I ittps / / www. bmuv.de / qesetz / richtlinie-2008-98-eq-ueber-abfaelle-und-zur-aufhe- Dunq-bestimmter-richtli-nien [3] http: / / www.kreislaufwirtschaft-bau.de / Arge / Bericht-10.pdf [4] R. S. Paranhos, B. G. Cazacliu, C. H. Sampaio, C. O. Petter, R. O. Neto und F. Huchet. „A sorting method to value recycled concrete“. In: Journal of Cleaner Production 112 (2016), S. 2249-2258. [5] G. Bonifazi, G. Capobianco, und S. Serranti, „Hyperspectral Imaging and Hierarchical PLS-DA Ap-plied to Asbestos Recognition in Construction and Demolition Waste“, Applied Sciences, Bd. 9, Nr. 21, Art. Nr. 21, Jan. 2019, doi: 10.3390 / app9214587. [6] Mueller, Anette & Seidemann, Marko & Schnellert, T.. (2011). Quick detection of asbestos. AT re-covery. 1. 56-69.

Claims

Patent claims 1. Method (100) for detecting asbestos fibers (12) in a building material (11) and / or in construction waste fractions (11), wherein the method (110) comprises: Irradiation of a building material (11) and / or of construction waste fractions (11) with unpolarized white light (Ao); subsequent splitting of reflected light (A re f) into polarized light rays (A re f, po i) in at least two polarization directions; and subsequent Detection of polarized light reflected from the building material and / or the construction waste fractions, split in at least two directions (A re f, po i) Evaluating the detected reflected polarized light (A re f, poi) to detect a possible color difference in at least two polarization directions, wherein, upon detection of the color difference, asbestos fibers (12) are present in the building material (11) and / or in the construction waste fractions (11).

2. Method (100) according to claim 1, wherein splitting the detected reflected light (A re f) into polarized light rays (A re f, po i) with a crystal or with at least two polarizing filters.

3. Method according to any of the preceding claims, wherein the detection of the reflected polarized light (A re f, po i) automatically, in particular by means of an imaging sensor (40), and / or the evaluation is carried out automatically, in particular on the basis of an algorithmic evaluation procedure.

4. Method according to any of the preceding claims, wherein the detection of the reflected polarized light (A re f, poi) at a distance (d) from the building material (11) and / or from the construction waste fractions (11) of 20 cm or more, preferably of 30 cm or more.

5. Method according to any of the preceding claims, wherein the detection of the reflected polarized light (A re f, po i) using a polarization camera which simultaneously captures at least two, preferably four, images, each with a different polarization direction.

6. The method of claim 5, wherein each pixel of a single captured image comprises four RGGB subpixels, wherein each RGGB subpixel has four subsubpixels, and wherein each subsubpixel has a polarization filter to reflect the reflected light (Aref) in one polarization direction of at least two, in particular four, directions, in particular as reflected polarized light (A re f, po i) to record.

7. The method of claim 6, wherein each image comprises an RGB vector for each sub-subpixel.

8. Method according to claim 6 or 7, wherein the evaluation of the possible color difference comprises: Performing a color space transformation of the captured images from an RGB space to an HSV space; Performing noise reduction, in particular with a median filter and / or a Gaussian filter; pixel-wise, in particular sub-subpixel-wise, calculating a difference and / or variance; pixel-wise, in particular sub-subpixel-wise, calculating distances between the RGB vectors; pixel-wise, in particular sub-subpixel-wise, comparing the calculated difference and / or the calculated variance and / or the calculated distances with a threshold value.

9. The method of claim 8, wherein the method comprises pixel-wise, in particular sub-subpixel-wise, post-processing of the calculated difference and / or the calculated variance and / or the calculated distances, wherein the post-processing comprises evaluating a grayscale of the sub-subpixels.

10. The method according to claim 9, wherein the evaluation of the grayscale comprises determining a number of neighboring pixels, in particular subsubpixels, which differ from a defined threshold.

11. Method according to claim 10, wherein the method comprises: Representing the calculated difference and / or the calculated variance and / or the calculated distances of each pixel, especially subsubpixel, in a mask in which the possible color difference or the possible color differences are marked as the contiguous number of neighboring pixels, especially subsubpixels.

12. Method according to any one of claims 7 to 11, wherein the method comprises: visualizing the calculated difference and / or the calculated variance and / or the calculated distances of each pixel, in particular subsubpixels, in a false color image (400) to show the presence of asbestos fibers (12) in the building material (11) and / or in the construction waste fractions (11).

13. Device (10) for detecting asbestos fibers (12) in a building material (11) and / or in construction waste fractions (11), the device comprising: a source (20) for irradiating a building material (11) and / or construction waste fractions (11) with unpolarized white light (Ao); a splitting device (30) for splitting the detected reflected light (Aref) into polarized light beams (A re f, poi) in at least two polarization directions; and a sensor (40) for detecting polarized light (A) reflected from the building material (11) and / or from the construction waste fractions (11). re f, po i) in at least two polarization directions; and an evaluation device (50) for evaluating the detected reflected polarized light (A re f, po i) to detect a possible color difference in at least two polarization directions, wherein, upon detection of the color difference, asbestos fibers (12) are present in the building material (11) and / or in the construction waste fractions (11).

14. Device according to claim 13, wherein the source (20) for irradiation and the sensor (40) for detection are two separate devices or a single device The device is, in particular the sensor (40) for detection, a polarization camera.

15. Device according to claim 13 or 14, wherein the dividing device (30) is for dividing the reflected light (A re f) into polarized light rays (A re f, po i) is a crystal or at least two polarizing filters.

16. Device according to any one of the preceding claims 13 to 15, wherein the sensor (40) is configured to automatically detect the reflected polarized light (Aref, poi), wherein the sensor (40) is an imaging sensor such as a polarization camera which simultaneously captures at least two, preferably four images with a different polarization direction.

17. Device according to claim 16, wherein each pixel of a single captured image comprises four RGGB subpixels, wherein each RGGB subpixel comprises four subsubpixels, wherein each subsubpixel has a polarization filter to capture the reflected light (Aref) in one polarization direction of at least two, in particular four, directions.

18. Computer program product, comprising instructions which, when loaded and run on a computer, cause the computer to perform a procedure, in particular the evaluation of the detected reflected polarized light (A re f, po i) according to any one of claims 1 to 10.

19. Computer-readable medium on which the computer program product according to claim 18 is stored.

Images