AU2020259802A1 - A system and method for asbestos identification - Google Patents

Abstract

Disclosed is an asbestos identification apparatus, comprising an optical component having a magnification factor, for obtaining a magnified image of a sample; an image acquisition means to acquire an image data from the magnified image; and an image processing module which is adapted to receive the image data, to determine whether an asbestos signature is present in the image data.

Description

A SYSTEM AND METHOD FOR ASBESTOS IDENTIFICATION

This application claims priority from Australian Application No.s 2019901334 and 2019901335 both filed on 17 April 2019, the contents of which are to be taken as incorporated herein by this reference.

TECHNICAL FIELD

This disclosure relates to the identification of asbestos, in particular the identification of asbestos in situ as it is typically found as part of a secondary matrix such as in building materials.

BACKGROUND ART

In the area of construction or home building or renovation, asbestos needs to be identified on a regular basis, as it has the potential to cause significant harm to workers or inhabitants through asbestos related diseases. This is a significant problem in homes built in Australia before the mid 1980s when the use of asbestos based products, for example cement sheet was common in building parts such as flues, roof eaves etc. When houses are undergoing renovations or demolition there is particularly a risk to workers and reliable asbestos information is needed.

Typically asbestos identification requires people specialised in the field with specialist equipment and skills, and includes substantial equipment or time requirements. For asbestos in particular three techniques are generally used for identification: light microscopy by a skilled operator, x-ray crystallography and NIR spectroscopy. It is generally not practical to use this equipment in the field, as they are expensive and often not portable. In the case of microscopy techniques only someone extremely skilled in the field can identify differences between Asbestos fibres and non- Asbestos materials. Such information is not typically accessible to people who work on sites for example, tradies, builders etc.

It is to be understood that, if any prior art is referred to herein, such reference does not constitute an admission that the prior art forms a part of the common general knowledge in the art, in Australia or any other country. SUMMARY

In a first aspect, the present invention provides an asbestos identification apparatus, comprising

an optical component having a magnification factor, for obtaining a magnified image of a sample;

an image acquisition means to acquire an image data from the magnified image;

an image processing module which is adapted to receive the image data, to determine whether an asbestos signature is present in the image data.

The apparatus can have a light source which is adapted to illuminate the sample, at the time of image acquisition.

The light source can have a setting which is chosen to enhance a detection accuracy and/or possibility of the asbestos signature.

The light source can include one or more lighting components. The lighting components can include components which generate lights of different wavelengths.

One of the lighting components can be infrared or UV light.

The image processing module can have an image detection module which is trained to detect the asbestos signature.

The image processing module can further comprise a pre-processing module, which pre-processes the image data, and a pre-processed data is provided to the image detection module.

The pre-processing module can include one or more pre-processing functions, which are chosen in accordance with the asbestos signature, so as to enhance a possibility of detecting whether the asbestos signature is present in the image data.

The signature can be a morphological signature, a spectral signature, or a combination thereof.

The optical component can be a microscope. The apparatus can have a controller which includes said image processing module. The controller can be a processing unit of a computing device.

The computing device can be a mobile device.

The optical component having the magnification factor can be retrofitted to said mobile device. For example the optical component can be clipped onto the mobile device.

In a second aspect, the invention provides a method of training an image recognition system having a machine learning algorithm, including providing a plurality of training images to the image recognition system, the plurality of images being acquired from samples known to contain asbestos, the plurality of training images having a magnification factor.

The method can include pre-processing the plurality of training images before inputting the training images to the image recognition system.

The training images can include one or more sets of images, each set being taken of samples known to contain a different sub-type of asbestos.

The training images can include at least one set of training images taken of samples known to contain a substance known to resemble asbestos

The method can further include providing to the image recognition system, one or more sets of training images, each further provided set being magnified images of samples known to contain a different material which resembles asbestos.

The method includes providing illumination at the time of image acquisition. The illumination can be consistently provided for each set of training images.

The method includes providing a plurality of sets of training images, each being acquired with a different illumination setting.

The method includes tagging each of the one or more training images to identify portion or portions therein containing the asbestos signature. In a third aspect, the invention provides a method of asbestos identification or detection, including acquiring an image data of a sample using an apparatus mentioned in the first aspect above.

The method can include providing the image data to an image detection system which is trained using the method mentioned in the second aspect above.

In a fourth aspect, the invention provides an application for asbestos identification or detection, including an image processing module, the image processing module including an image detection program which is trained using the method mentioned in the second aspect above.

The application can include a control module for a device for controlling an image acquisition device.

The application can include a control module for a light source.

The application can include a user interface module for user to input control commands.

The application can include an executable program which when executed is adapted to cause a display of or associated with a computing device on which the mobile application resides, to display an output.

The output can be an image which is processed by the image processing module, further annotated to indicate location or locations of identified asbestos. In a fifth aspect, the invention provides a computer programme, comprising instructions for controlling a computing device to implement the application mentioned in the fourth aspect mentioned above.

In a sixth aspect, the invention provides a computer readable medium, providing a computer programme mentioned above. BRIEF DESCRIPTION OF THE DRAWINGS

Embodiments will now be described by way of example only, with reference to the accompanying drawings in which

Figure 1 is a schematic representation of a device for asbestos identification;

Figure 2 is an image of a sample which is a cement sheet building material including asbestos fibres, the images showing the edges of the sample;

Figure 3 is a schematic representation of a training process for the image recognition program and the image recognition;

Figure 4(1) a training image of chrysotile asbestos;

Figure 4(2) is the pre-processed image where pre-processing has been performed to accentuate the morphological structure of the chrysotile asbestos fibres, where the structures are tagged;

Figure 5(1) a training image of organic fibres;

Figure 5(2) is the pre-processed image where pre-processing has been performed to accentuate the morphological structure of the organic fibres, where the structures are tagged;

Figure 6(1) is a training image for training the image recognition module, where crocidolite asbestos fibres are tagged;

Figure 6(2) is a training image for trained image recognition module, where chrysotile asbestos fibres are tagged;

Figure 7 (1) is an output image where portions of the image determined to show suspected asbestos is marked in the square;

Figure 7 (2) is an output image where portions of the image determined to likely not show asbestos is marked in the square; and

Figure 8 is a schematic representation of a portable microscope which can be clipped onto a mobile device. DETAILED DESCRIPTION

In the following detailed description, reference is made to accompanying drawings which form a part of the detailed description. The illustrative embodiments described in the detailed description, depicted in the drawings and defined in the claims, are not intended to be limiting. Other embodiments may be utilised and other changes may be made without departing from the spirit or scope of the subject matter presented. It will be readily understood that the aspects of the present disclosure, as generally described herein and illustrated in the drawings can be arranged, substituted, combined, separated and designed in a wide variety of different configurations, all of which are contemplated in this disclosure.

The invention discussed herein enables asbestos identification using visual images.

Asbestos at both a macroscopic and microscopic scale materials have visual properties that indicate information about their structure. The information includes a unique morphology (characteristic) or other visual properties (particularly, interaction with lights of different colours including infrared light) which provide visual properties of this information.

The present invention provides a device and method for asbestos identification, where the images are taken in such a way that enables an enhancement of the signature unique to asbestos. Training images thus taken are used in machine learning, where the signature of asbestos is quantifiably identified from the visual images. This contrasts with the conventional identification method, where asbestos identification is not reliably performed by the human eye, and identification usually involves an expensive or cumbersome equipment.

Referring to Figure 1, an asbestos identification system 100, in accordance with one embodiment of the current invention, includes a camera 102 for acquiring an image of the sample (e.g. cement sheet) containing a substance to be identified. In the example shown in Figure 1, the camera 102 is external to the computing device. However, it can be built into the computing device in another embodiment. A magnifying component 112 is coupled to the camera 102. That is, the camera 102 receives the output of the magnifying component 112. The camera 102 is therefore acquiring a magnified image of the sample.

The system 100 preferably includes a light source for illuminating the sample, at the time of image acquisition by the camera 102. It can include a combination of lighting components with different wavelengths, e.g. infrared, coloured lights and white lights. The light source will be set to operate at a particular setting to optimise the image acquisition for the recognition of the asbestos signature.

In a particular embodiment, a microscope, which provides both the magnifying component and the light source, is used.

The detection system 100 includes a controller 108, which can be or reside in the processor of a computing device such as a computer, a mobile phone, or a tablet. The computing device is a portable device in embodiments intended to allow on- the-spot asbestos identification in the field (e.g. in the worksite). The controller 108 controls the operation of the camera and/or the illumination and magnification components, and includes the processing module which performs the asbestos identification.

The detection system 100 can be provided as a mobile application. It can be embodied as an executable program which is adapted to run on the processor of a portable or mobile device such as a laptop, smart phone or a tablet. In an embodiment of this type, the controller 108 includes an application module 110, which when activated, will launch a user interface with which a user interacts to operate the system 100. The user interface may be displayed on a display 114 that is associated with or that is a part of, the portable or mobile device on which the controller 108 is installed.

In an embodiment, the controller 108 will include a camera control module and/or a light source control module, which includes a data field which stores the light setting to be used for asbestos identification. It is preferred that the light setting is pre-programmed, so that the illumination is automatically controlled by the controller 108, and its setting does not require the user’s manipulation.

The controller 108 includes an image processing module 106 for processing image data from the camera 102. In some example the controller 108 also sends control signals to the camera 102 to operate it, either automatically or on a trigger action by the user (e.g. activating a switch, interacting with a touch screen, voice command, etc).

In some examples, the controller 108 is in data and/or electrical communication to the light source 104 to control the operation of the light source. For example, the controller 108 supplies power to the light source 104, and the power supply is switched by the controller to turn the light source, or one or more light source components, on and off. In other examples, the light source 104 is manually operated, to switch the whole light source, or various lighting components in the light source, on or off.

The image processing module 106 will include an image recognition (detection) program or module 116 to determine whether the imaged sample contains asbestos. Before the data is run through the image recognition module 116, it may be pre- processed (such as but not limited to, filtered, colour, contrast or brightness adjusted, transformed by comparison to images from different light sources) by a pre-processing module 118, to visually accentuate the asbestos signature.

The algorithm for the image recognition module 116 is obtained by using a large set of images to train an image recognition program. The image recognition program comprises appropriate supervised machine learning programs such as, but not limited to, deep neural networks, to recognise the asbestos signature, being the structural feature and/or the spectral feature of asbestos.

The presentation of the materials depends on the matrix or the“Surrounding Environment”, i.e. how the substance is embedded into its surrounding materials (e.g. how the asbestos is embedded into the building material). The detection method provides information about asbestos, even when it is embedded in a complex matrix, as the case can be in the way asbestos was used in home building products.

The training of the image recognition algorithm involves acquiring at least a threshold number of training images, being images of samples known to contain asbestos, i.e. positive samples of asbestos. The training images will also include negative samples, being images of samples known not to contain asbestos, i.e. negative samples of asbestos. The positive and negative samples may each include finer classification, into sub-types of positive samples (for example, different Asbestos types) and negative samples for example different fibre types), so that the algorithm can learn to recognise the sub-types.

Specifically, for training for asbestos it is preferred that the sample and the training images contain at least some images from the edge portions of the samples. In some cases where detection of fibrous materials is being trained, the fibres can be revealed by examining the edges of the samples in which the fibrous materials are embedded. An example is shown in Figure 2, where the imaged sample is a slab material. The image shows the edges of the material, as indicated by the arrows.

Figure 3 depicts a conceptual schematic for training the detection algorithm. The training process 120 used to train an image recognition module 116 includes acquiring a training data set 122. It will be appreciated that the larger the data set, the better the algorithm can be trained. In example implementations done during development, between 3000 and 10000 training images were acquired. It will be appreciated that re-training of the system may be performed at any time, as more training images or examples become available.

The training images are acquired with the coupled optical device/lighting - being the magnifying device 112 coupled with the camera 102. This ensures the training images are acquired with a consistent built-in magnification factor compared to the actual sample. In an example implementation, the magnifying device 112 is a small size microscope with a 60X magnification. The illumination by the light source has a particular setting, or one of a plurality of predefined light settings. The light setting includes a setting for the colour of the light (i.e. wavelength) and/or intensity of the light. The training images can include images acquired with different illumination settings. This helps to train the algorithm to recognise different subtypes of asbestos or negative samples. The different light settings also account for situations where different settings may be used to acquire the image to be tested. However, it is preferred that the light setting for the acquisition of the training images be the same as the light setting for the acquisition of the test images.

In an embodiment, the training images includes a plurality of training images for each sub-type of asbestos - e.g. different sets of images of“asbestos”,“chrysotile asbestos”,“crocidolite asbestos”,“amosite asbestos”, are included. This enables the detection algorithm to return a“finer grain” result or of a finer classification than a simple asbestos/not asbestos identification.

Negative training images are also used. Emphasis may be placed on negative examples, being materials having signatures which resemble the signature of the asbestos fibres. For example, training images can be acquired of“organic fibres” and “manufactured glass fibres”, which resemble asbestos fibres but are not asbestos.

The training images are then tagged, to identify areas which show the asbestos fibres 124. The images, or the tagged or identified areas, are labelled. Depending on the information available, the labelling can provide different levels of information. On the most general level, the image is labelled“asbestos” or“not asbestos”. Or the labelling can provide finer detail to identify the asbestos fibres in different phases, and other fibres of similar appearance- e.g.“organic fibres”, “manufactured glass fibre”; “asbestos”, “chrysotile asbestos”, “crocidolite asbestos”,“amosite asbestos”, etc.

The classified or labelled training images are fed into an image recognition program 116, which includes a learning component to execute a learning process 126 using the training images. The image recognition program or module will associate each “label” or“class” with the structures that are highlighted or tagged in the training images of that class or with that label. The trained image recognition program 116 is included in the image processing module 106, to process the acquired image data 130 (or a pre-processed version thereof) during use, and identify whether the acquired image data show the signatures of asbestos.

Figures 4 to 6 show examples of the above described process. Figure 4(1) shows a training image of chrysotile asbestos. Figure 4(2) is the pre-processed image where pre-processing has been performed to accentuate the morphological structure of the chrysotile asbestos fibres, where the identified structures are tagged by outlining in a rectangular box. Figures 5(1) and 5(2) are similar images, but of organic (not asbestos fibres). Although not shown, the same process can be applied to e.g. crocidolite asbestos fibres. Figures 6(1) and 6(2) are examples of training images supplied to the trained image recognition module 118, where crocidolite asbestos and chrysotile asbestos fibres, respectively, are tagged in the rectangular outlines and labelled.

In one example, these training images are processed by different image recognition software programs, each can be customised to recognise a different characteristic in the asbestos. Each program is trained appropriately using sufficient examples of images containing the characteristic of interest. In one embodiment, the image recognition programs can each be trained to detect one type of signature. Thus, for a system which is able to identify a number different asbestos sub-types, the test image (or a pre-processed version thereof) will be inputted to a corresponding number programs for the identification of the number of sub-types. Alternatively, one program can be trained to identify the signatures of two or more sub-types.

The image recognition module 116, with the trained algorithm, will be used to determine whether the test images include any of the signatures relating to asbestos, or a negative example resembling asbestos.

Variations and modifications may be made to the parts previously described without departing from the spirit or ambit of the disclosure. In a particular embodiment, the system can be retrofitted to, e.g. a user’s existing mobile device having a built-in camera, by loading the program modules needed to the mobile device, and by fitting the magnifying device and light source to the mobile device. For example, a portable microscope can be permanently or removably retrofitted to the mobile device. It can, for instance, be clipped onto a mobile phone (removable retrofit) on which the mobile application for the detection system is installed. Figure 7 schematically depicts the embodiment where a portable microscope 150 having gap 152 which allows the portable microscope 150 to be fitted onto a mobile device (not shown). The attachment can be achieved by a friction fit, a biased clamping or clipping joint, or by another mechanism. When the microscope 150 is attached to the mobile device, the camera lens of the mobile device will line up with the output field of the microscope 150. The microscope 150 includes a light source 154 which as mentioned above, may include different lighting components. The light source 154 is adapted to illuminate the sample 156 of which the test image is being taken. It is preferred that the test images are taken under a consistent light source, to ensure a consistent performance of the detection system.

Figures 8(1) and 8(2) depict example output images. Figure 8 (1) is an output image where portions of the image determined to show suspected asbestos is marked in the square. Figure 8 (2) is an output image where a portion of the image, where asbestos is suspected but determined by the program to likely not show asbestos, is marked in the square.

As alluded above, the various processing functions and control interfaces can be embodied as modules which are provided as executable codes, which are adapted to be installed on the processing unit of the control device.

In the claims which follow and in the preceding description of the invention, except where the context requires otherwise due to express language or necessary implication, the word “comprise” or variations such as “comprises” or “comprising” is used in an inclusive sense, i.e. to specify the presence of the stated features but not to preclude the presence or addition of further features in various embodiments of the invention.

Claims (31)

THE CLAIMS DEFINING THE INVENTION ARE AS FOLLOWS:

1. An asbestos identification apparatus, comprising

an optical component having a magnification factor, for obtaining a magnified image of a sample;

an image acquisition means to acquire an image data from the magnified image;

an image processing module which is adapted to receive the image data, to determine whether an asbestos signature is present in the image data.

2. The apparatus of claim 1, further comprising a light source which is adapted to illuminate the sample, at the time of image acquisition.

3. The apparatus of claim 2, wherein the light source has a setting which is chosen to enhance a detection accuracy and/or possibility of the asbestos signature.

4. The apparatus of claim 2 or claim 3, wherein the light source includes one or more lighting components.

5. The apparatus of claim 4, wherein the lighting components include components which generate lights of different wavelengths.

6. The apparatus of any one of the preceding claims, wherein the image processing module comprises an image detection module which is trained to detect the asbestos signature.

7. The apparatus of claim 6, wherein the image processing module comprises a pre-processing module, which pre-processes the image data, and a pre- processed data is provided to the image detection module.

8. The apparatus of claim 7, wherein the pre-processing module includes one or more pre-processing functions, which are chosen in accordance with the asbestos signature, so as to enhance a possibility of detecting whether the signature is present in the image data.

9. The apparatus of any one of the preceding claims, wherein said signature is a morphological signature, a spectral signature, or a combination thereof.

10. The apparatus of any one of the preceding claims, wherein said optical component is a microscope.

11. The apparatus of any one of the preceding claims, comprising a controller which includes said image processing module.

12. The apparatus of claim 11, wherein said controller is a processing unit of a computing device.

13. The apparatus of claim 12, wherein said computing device is a mobile device.

14. The apparatus of claim 13, wherein said optical component having the magnification factor is retrofitted to said mobile device.

15. A method of training an image recognition system with a machine learning algorithm, including providing a plurality of training images to the image recognition system, the plurality of images being acquired from samples known to contain asbestos, the plurality of training images having a magnification factor.

16. The method of claim 15, including pre-processing the training images before inputting the training images to the image recognition system.

17. The method of claim 15 or claim 16, wherein the training images includes one or more sets of images, each set being taken of samples known to contain a different sub-type of asbestos.

18. The method of any one of claims 15 to 17, further including providing to the image recognition system one or more sets of training images, each set being magnified images of samples known to contain a different material which resembles asbestos.

19. The method of any one of claims 15 to 18, including providing illumination at the time of image acquisition.

20. The method of claim 19, wherein illumination is consistently provided for each set of training images.

21. The method of claim 20, wherein at least some of the training images are taken under a different illumination setting than the other training images.

22. The method of any one of claims 15 to 21, including tagging each training image to identify portion or portions therein containing the asbestos signature.

23. A method of asbestos identification or detection, including acquiring a test image data of a sample using an apparatus as claimed in any one of claims 1 to 14.

24. A method as claimed in claim 23, including providing the image data to an image detection algorithm which is trained using the method of any one of claims 15 to 22.

25. An application for asbestos identification or detection, including an image processing module, the image processing module including an image detection program which is trained using the method claimed in any one of claims 15 to 22.

26. The application of claim 25, including a control module for a device for controlling an image acquisition device.

27. The application of claim 25 or claim 26, including a control module for a light source.

28. The application of any one of claims 25 to 27, including a user interface module for user to input control commands.

29. The application of any one of claims 25 to 28, including an executable program which when executed is adapted to cause a display of or associated with a computing device on which the mobile application resides, to display an output.

30. The application of claim 29, the output being an image which is processed by the image processing module, further annotated to indicate location or locations of identified asbestos.

31. A computer program, comprising instructions for controlling a computing device to implement the application as claimed in any one of claims 25 to 30.

32, A computer readable medium, providing a computer programme in accordance with claim 31.