System and method for detecting corrosion in reinforced concrete

Abstract

A system for detecting corrosion in reinforced concrete is provided. The system comprises a multi-spectral camera for capturing image data, the image data being captured at a plurality of pixels of the multi-spectral camera and comprising, for each pixel, a spectral signature being a measurement of reflectance across a predetermined range of wavelengths. The system further comprises a processor, for processing the image data received from the multi-spectral camera. The processor is configured to receive from the multi-spectral camera image data of a surface of the reinforced concrete. The processor is further configured to determine whether corrosion is present in the reinforced concrete.

Description

[0001] System and method for detecting corrosion in reinforced concrete

[0002] Technical field

[0003] The present invention relates to a system and method for inspection of concrete, in particular to identify corrosion or rust in reinforced concrete. The reinforced concrete may form part of a concrete structure. The present method and system make use of spectral imaging in order to identify corrosion.

[0004] Background

[0005] Inspecting large concrete structures such as large bridges can be challenging. Completion of structural inspection may be very time consuming and expensive. In many cases, access to some elements of the structure is difficult, due to the height and position of those elements, the surrounding terrain and the environmental conditions.

[0006] One way of inspecting concrete structures is by use of remote imaging, such as cameras. One challenge is how to detect the onset of corrosion through the presence of corrosion products on the surface of the structure. This is at least because the concrete surfaces may include many different types of painting, stains, or other colouration.

[0007] Spectral imaging devices are known to be used for detecting a deteriorated portion of concrete on an inspection target surface of a structure. For example, JP4116926B2 discloses a spectral imaging device that utilises a spectral imaging camera which uses a plurality of wavelength bands and discriminates the surface to be inspected into a concrete region and a non-concrete region, based on the spectral intensity of each location. For the concrete regions, this device illuminates the surface with light and detects and measures intensity of reflected light at different wavelengths to detect degradation of concrete. Degradation is based on sulphate and chloride degradation. Chinese patent

[0008] CN107576619B relies on spectral libraries of concrete raw materials and comparisons in an iterative processing manner that is complicated and requires significant processing power.

[0009] There is therefore the need for a method of detecting the presence of corrosion products on a concrete surface, in a simple manner so that detection can be fast and easy. Summary

[0010] A system for detecting corrosion in reinforced concrete is provided. The system comprises a multi-spectral camera for capturing image data, the image data being captured at a plurality of pixels of the multi-spectral camera and comprising, for each pixel, a spectral signature being a measurement of reflectance across a predetermined range of wavelengths. The system further comprises a processor, for processing the image data received from the multi-spectral camera. The processor is configured to receive from the multi-spectral camera image data of a surface of the reinforced concrete. The processor is further configured to determine whether corrosion is present in the reinforced concrete.

[0011] In one example, corrosion is determined as being present if the spectral signature captured at one or more pixels meets all of the following conditions:

[0012] RB - RA > ki,

[0013] RB - Rc > k2, and

[0014] Rc - RA > k3.

[0015] Each of RA, RB and Rc are reflectance values at different wavelengths, wi, w2and w3respectively, within the predetermined range of wavelengths, wherein wi < w2< w3, wherein ki >= 0, k2>= 0, and k3>= 0.

[0016] In another example, corrosion is determined as being present if the spectral signature captured at one or more pixels complies with at least one criterion of a group of criteria.

[0017] The group of criteria comprises: a reflectance at a longest wavelength of the predetermined range of wavelengths is greater than a reflectance at a shortest wavelength of the predetermined range of wavelengths; and a peak in the spectral signature is identified within the predetermined range of wavelengths (preferably between 750 nm and 790 nm).

[0018] In another example, corrosion is determined as being present if the spectral signature captured at one or more pixels complies with each criterion of the group of criteria. Advantageously, by requiring that a spectral signature meet each of the criteria in the group of criteria in order to signify corrosion, instances of false positives may be reduced. The two criteria described above (increasing reflectance across the range and a peak within the range) have been found to be particularly effective at reliably classifying whether staining on the surface of the reinforced concrete is caused by corrosion products or not. Moreover, these criteria are easy to assess (compared to some prior art methods, which involve complex comparisons with libraries of spectra) and therefore assessment of the proposed criteria may be performed quickly and in a computationally efficient manner.

[0019] The system may be for detecting degradation of a reinforced concrete structure. The degradation may be due to corrosion of steel embedded in the reinforced concrete. The system may capture an image of a portion of the structure and detect corrosion within that portion. The portion may be referred to as a “sample”. Since only the surface of the sample is visible, the image data relates to the surface. Corrosion occurring within the reinforced concrete, beneath the surface, can cause staining on the surface. The system may detect staining that is characteristic of corrosion in the image data.

[0020] The system may be configured to capture one or more images, each image comprising a plurality of pixels and corresponding a spectral signature. The “image data” may be data comprised in a single image or may be data from a plurality of images.

[0021] Advantageously, the proposed methods utilise spectral analysis in a manner that is simple and fast to determine the likelihood of the presence of corrosion products on the surface of a reinforced concrete structure.

[0022] These methods may also be used for detecting degradation in other types of concrete. However, these methods are particularly effective for detecting degradation of reinforced concrete because corrosion of the embedded iron or steel in reinforced concrete creates characteristic patches of discolouration on the surface of the reinforced concrete. These patches are caused by iron oxide or “rust” produced when the embedded iron or steel corrodes. The proposed methods are able to identify surface patches caused by rust and differentiate these from other types of surface discolouration. A peak in the spectral intensity is a peak in the reflectance. In other words, a reflectance at a wavelength between the longest wavelength and the shortest wavelength is greater than a reflectance at either the longest wavelength or the shortest wavelength.

[0023] Identification of a peak may comprise determining that a reflectance at an expected peak position, at a wavelength between the longest wavelength and the shortest wavelength, is greater than a reflectance at either the longest wavelength or the shortest wavelength.

[0024] The reflectance at the expected peak position may be at least 3% higher than the reflectance at the shortest wavelength and at least 1% higher than the reflectance at the longest wavelength.

[0025] The expected peak position may be between 750 nm and 790 nm. More preferably, the expected peak position may be between 750 nm and 760 nm. The expected peak position may be 750 nm, 760 nm, 770 nm, 780 nm or 790 nm.

[0026] Alternatively, identification of a peak may comprise analysing the spectral intensity to identify a turning point and / or local maximum.

[0027] The processor may be further configured to assign a flag for storage with the image data in a computer memory, and / or for output to a user interface, the flag identifying whether corrosion is determined to be present in the image data.

[0028] The processor may be further configured to assign an indicator associated with each pixel of the plurality of pixels, the indicator for storage with the image data in a computer memory and / or for output to a user interface, the indicator identifying whether the spectral signature captured at the said pixel complies with at least one criterion of the group of criteria.

[0029] The processor may be further configured to map each pixel to a location at the surface of the reinforced concrete.

[0030] The shortest wavelength of the predetermined range of wavelengths may be 600 nm or greater. The shortest wavelength of the predetermined range of wavelengths may be 650 nm or less. The shortest wavelength of the predetermined range of wavelengths may be between 600 nm and 750 nm. The shortest wavelength of the predetermined range of wavelengths may be between 600 nm and 650 nm. For example, the shortest wavelength of the predetermined range of wavelengths may be 600 nm, 610 nm, 620 nm, 630 nm, 640 nm or 650 nm.

[0031] The longest wavelength of the predetermined range of wavelengths may be 950 nm or less. The longest wavelength of the predetermined range of wavelengths may be 900 nm or greater. The longest wavelength of the predetermined range of wavelengths may be between 790 nm and 950 nm. The longest wavelength of the predetermined range of wavelengths may be between 950 nm and 900 nm. For example, the longest wavelength of the predetermined range of wavelengths may be 900 nm, 910 nm, 920 nm, 930 nm 940 nm or 950 nm.

[0032] The processor may be further configured to, prior to determining whether corrosion is present in the reinforced concrete, filter or smooth the spectral signature captured at each of the plurality of pixels.

[0033] The processor may be further configured to determine that corrosion is present in the reinforced concrete if the spectral signature captured at one or more pixels meets all of the following conditions:

[0034] RB - RA > 3,

[0035] RB - Rc > 1 , and Rc - RA > 0.

[0036] In other words, ki = 3, k2 = 1 , and ks = 0 in some examples.

[0037] In one example, RA is the reflectance at 650 nm, RBis the reflectance at 750 nm, and Rc is the reflectance at 900 nm. In other words, wi = 650 nm W2 = 750 nm, and W3 = 900 nm.

[0038] In another example, RA is the reflectance at 650 nm, RBis the reflectance at 790 nm and Rc is the reflectance at 950 nm. In other words, wi = 650 nm W2 = 790 nm, and W3 = 950 nm.

[0039] The group of criteria may further comprise a criterion that the spectral signature does not exceed a threshold reflectance. In other words, the reflectance measured across the predetermined range of wavelengths is less than a threshold reflectance. In some examples, the threshold reflectance may be 50%. In other examples, the threshold reflectance may be 40% or 60%.

[0040] The system may further comprise a moveable carrier upon which the multi-spectral camera is mounted, the moveable carrier for positioning the multi-spectral camera relative to the surface of the reinforced concrete.

[0041] The system may be configured to control the moveable carrier to position the multi-spectral camera.

[0042] The system may be configured to capture a plurality of images of the surface of the reinforced concrete. The system may be configured to control the moveable carrier to reposition the multi-spectral camera between capturing each of the plurality of images.

[0043] The moveable carrier may be an unmanned aerial vehicle or drone.

[0044] A method for detecting corrosion in reinforced concrete is also provided. The method comprises receiving, from a multi-spectral camera, image data of a surface of the reinforced concrete, wherein the multi-spectral camera captures the image data at a plurality of pixels, the image data comprising, for each pixel, a spectral signature being a measurement of reflectance across a predetermined range of wavelengths. The method further comprises determining whether corrosion is present in the reinforced concrete.

[0045] In one example, corrosion is determined as being present if the spectral signature captured at one or more pixels meets all of the following conditions:

[0046] RB - RA > ki,

[0047] RB - Rc > k2, and

[0048] Rc - RA > k3.

[0049] Each of RA, RB and Rc are reflectance values at different wavelengths, wi, w2and w3respectively, within the predetermined range of wavelengths, wherein wi < w2< w3, wherein ki >= 0, k2>= 0, and k3>= 0. In another example, corrosion is determined as being present if the spectral signature captured at one or more pixels complies with at least one criterion of a group of criteria.

[0050] The group of criteria comprises: a reflectance at a longest wavelength of the predetermined range of wavelengths is greater than a reflectance at a shortest wavelength of the predetermined range of wavelengths; and a peak in the spectral signature is identified within the predetermined range of wavelengths (preferably between 750 nm and 790 nm).

[0051] In another example, corrosion is determined as being present if the spectral signature captured at one or more pixels complies with each criterion of the group of criteria.

[0052] The method may further comprise assigning a flag for storing with the image data in a computer memory and / or for outputting to a user interface, the flag identifying whether corrosion is determined to be present in the image data.

[0053] The method may further comprise assigning an indicator associated with each pixel of the plurality of pixels, the indicator for storing with the image data in a computer memory and / or for outputting to a user interface, the indicator identifying whether the spectral signature captured at the said pixel complies with at least one criterion of the group of criteria.

[0054] The method may further comprise mapping each pixel to a location at the surface of the reinforced concrete.

[0055] The method may further comprise, prior to determining whether corrosion is present in the reinforced concrete, filtering or smoothing the spectral signature captured at each of the plurality of pixels.

[0056] The method may further comprise determining that corrosion is present in the reinforced concrete if the spectral signature captured at one or more pixel meets all of the following conditions:

[0057] RB - RA > 3,

[0058] RB - Rc > 1 , and

[0059] Rc - RA > 0. In other words, ki = 3, k2 = 1 , and ks = 0 in some examples.

[0060] In one example, RA is the reflectance at 650 nm, RBis the reflectance at 750 nm, and Rc is the reflectance at 900 nm. In other words, wi = 650 nm W2 = 750 nm, and W3 = 900 nm.

[0061] In another example, RA is the reflectance at 650 nm, RBis the reflectance at 790 nm and Rc is the reflectance at 950 nm. In other words, wi = 650 nm W2 = 790 nm, and W3 = 950 nm.

[0062] In another example, a method for detecting corrosion in reinforced concrete is provided. The method comprises receiving from a multi-spectral camera image data of a surface of the reinforced concrete, wherein the multi-spectral camera captures the image data at a plurality of pixels, the image data comprising, for each pixel, a spectral signature being a measurement of reflectance across a predetermined range of wavelengths. The method further comprises determining whether corrosion is present in the reinforced concrete, wherein corrosion is determined as being present if the spectral signature captured at one or more pixels complies with all of the following conditions:

[0063] RB - RA > 3;

[0064] RB- Rc > 1 ; and

[0065] Rc - RA > 0.

[0066] Where RA is the reflectance at a first wavelength within the predetermined range of wavelengths (e.g., 600 nm or 650 nm), RBis the reflectance at a second wavelength within the predetermined range of wavelengths (e.g., 750 nm or 790 nm), and Rc is the reflectance at a third wavelength within the predetermined range of wavelengths (e.g., 900 nm or 950 nm).

[0067] In another example, a system for detecting corrosion in reinforced concrete is provided. The system comprises a multi-spectral camera configured to capture image data of a surface of the reinforced concrete, wherein the multi-spectral camera captures the image data at a plurality of pixels, the image data comprising, for each pixel, a spectral signature being a measurement of reflectance across a predetermined range of wavelengths. The system further comprises a processor configured to determine whether corrosion is present in the reinforced concrete, wherein corrosion is determined as being present if the spectral signature captured at one or more pixels complies with all of the following conditions:

[0068] RB - RA > 3; RB - Rc > 1 ; and Rc - RA > 0.

[0069] Where RA is the reflectance at a first wavelength within the predetermined range of wavelengths (e.g., 600 nm or 650 nm), RBis the reflectance at a second wavelength within the predetermined range of wavelengths longer than the first wavelength (e.g., 750 nm or 790 nm), and Rc is the reflectance at a third wavelength within the predetermined range of wavelengths longer than the second wavelength (e.g., 900 nm or 950 nm).

[0070] In yet a further example, a system for detecting the presence of corrosion products on a concrete surface of a structure is provided. The system comprises: i. an image acquisition means for acquiring a hyperspectral image of a surface of a structure under inspection; ii. a positioning means for positioning the image acquisition means; ill. a processing means for processing said image; iv. a communication means for communicating information to a remote computer; wherein the processing of said image comprises the steps of: a. generating a spectral curve of the image; b. determining the reflectance values for three or more wavelengths within a waveband of interest; c. applying a transfer function through numerical processing of said reflectance values at said three or more wavelengths; d. checking if all three conditions below hold true

[0071] RB - RA >ki RB — Rc >k2Rc — RA >k3and if all three conditions hold true then set a “Rust Index” to TRUE; wherein, said waveband of interest is between 650nm and 950nm.

[0072] In some examples, ki = 3, k2= 1 , and k3= 0.

[0073] Said surface under inspection may be segmented into a grid forming a plurality of pixels, wherein a spectral curve is generated for each pixel and a “Rust Index” computed for each pixel, and wherein based on the values of the “Rust Index” for each pixel, an area is delineated in which area it is determined that there are corrosion products present.

[0074] The processing of said image, prior to computation of “Rust Index”, may further comprise, applying appropriate filtering to said image.

[0075] The methods and systems described may further comprise additional steps or features described in relation to other embodiments and examples. For example, the methods may further comprise assigning a flag for storing with the image data in a computer memory and / or for outputting to a user interface, assigning an indicator associated with each pixel of the plurality of pixels, mapping each pixel to a location at the surface of the reinforced concrete, and / or filtering or smoothing the spectral signature captured at each of the plurality of pixels.

[0076] Brief description of the figures

[0077] The present invention is described with reference to a number of non-limiting examples. A number of examples are illustrated in the accompanying drawings.

[0078] Figure 1 is a schematic representation of a typical large structure that is surveyed.

[0079] Figure 2 is a representation of the main subsystems of a system for detecting the presence of corrosion products on a concrete surface of a structure.

[0080] Figure 3A is a photograph of a test sample of concrete having patches of corrosion products.

[0081] Figure 3B is a photograph of a test sample of concrete having patches of corrosion products, with a pixel grid superimposed.

[0082] Figure 4 illustrates an example of a typical spectral curve for a surface area having corrosion products and a surface area not having corrosion products.

[0083] Figure 5A is an illustration of the key characteristics of a spectral curve for a surface area having corrosion products. Figure 5B is closeup view of figure 5A, focusing on the region of the spectral curve that falls within the waveband of interest.

[0084] Detailed description of specific examples

[0085] The present application relates to inspection of large concrete structures using spectral imaging. In a specific example, a system and a method for detecting corrosion in reinforced concrete by determining the presence of corrosion products on a concrete surface is provided. The system comprises five main subsystems:

[0086] 1 ) positioning means,

[0087] 2) image capturing means,

[0088] 3) processing means,

[0089] 4) information transmission means, and

[0090] 5) remote computing system.

[0091] Once the image capturing means is positioned at the appropriate location, an image is captured of the target surface by the image-capturing means. The image is processed by the processing means and the processed information is transmitted to the remote computing device.

[0092] The positioning means may be an aerial vehicle on which the image capturing means is mounted. The aerial vehicle may be a remotely operated unmanned aerial vehicle or “drone”.

[0093] The image capturing means may be a multi-spectral camera. The proposed methods utilise a multi-spectral camera to obtain one or more multispectral images of one or more concrete surfaces. Each multispectral image comprises a plurality of pixels. Each multispectral image comprises, for each pixel, a spectral signature or “spectral curve”. The spectral signature being a measurement of reflectance across a predetermined range of wavelengths.

[0094] The processing means may be a processor. The processor processes the spectral curves to detect specific conditions within a predetermined range of wavelengths, referred to as a “spectral band”. In some examples, the spectral band is from 650 nm to 950 nm. The information transmission means may be a transceiver.

[0095] For each pixel (for each spectral signature), a “Rust Index” may be determined using simple arithmetic operations. The “Rust Index” is indicative of the presence of corrosion products on the concrete surface. The large surface can then be mapped with a grid, each grid representing a pixel having an “Rust Index” value, and hence generate a grid with pixels indicating presence or not of corrosion products at each pixel. Advantageously, this method provides a simple way to determine whether corrosion is present in reinforced concrete and, if so, where on the surface the corrosion products are visible.

[0096] Figure 1 is a schematic representation of a typical large structure that is surveyed (1). Structures that need surveying (1) are typically large-scale structures such as bridges. The structure under inspection (1) is seen as comprising a plurality of surface segments (10) each of which needs to be inspected for the presence of corrosion products. Corrosion products are typically iron oxides, which have leached to the surface after the reinforcing iron or steel members of the structure have undergone corrosion. An image acquisition means (3) such as a multi-spectral camera is used to obtain images of each of the plurality of surface segments under inspection (10). To do so, the image acquisition means (3) needs to be repositioned at an appropriate distance and angle from a target surface segment (10). This is done with the assistance of a positioning means (2). In one preferred embodiment the positioning means (2) is a drone or aerial vehicle (e.g., an unmanned aerial vehicle, UAV). The image acquisition means (3) may be a multispectral camera.

[0097] Figure 2 is a representation of the main subsystems of a system according to a specific example. The system is for detecting corrosion in reinforced concrete by detecting the presence of corrosion products on a concrete surface of a structure.

[0098] The system comprises of five main subsystems: a) positioning means (20), b) image capturing means (30), c) processing means (6), d) information transmission means (61), and e) remote computing system (62). The positioning means (20) in a preferred embodiment is an aerial vehicle or “drone”. The use of a drone provides flexibility in approaching the target surface at the desired distance and angle. Through programmed navigation, it is also possible to automate the process. In some examples, the drone may be configured to automatically capture a plurality of images of different surface segments of the reinforced concrete, by adjusting the position of the image capturing means between capturing each image.

[0099] The image capturing means (30) is a multi-spectral camera. In some examples, the multi- spectral camera has at least three spectral channels. The multi-spectral camera is configured to capture raw image data at a plurality of pixels. The raw image data comprising, for each pixel, a spectral signature, which comprises a measurement of reflectance across a predetermined range of wavelengths;

[0100] In some examples, the processing means (6) comprises a digital signal processing chip, memory, and numerical processing. The processing means (6) processes a spectral curve of each of the pixels of the target surface segment (10), determines for each pixel if there is presence of corrosion products, and then designates that pixel accordingly (e.g., by assigning an indicator associated with each pixel to identify whether the spectral signature is consistent with staining caused by corrosion products).

[0101] The raw image data and / or processed information comprising pixel designation may be transmitted to a remote computer (62) via information transmission means (61 ), such as a wireless transceiver. Wireless transmission may take place via a variety of means and protocols. A remote computing system (62) may be a smart handheld device, a remote server or any other computer.

[0102] Figure 3A is a photograph of a test sample of concrete having patches of corrosion products. Corrosion products, predominantly iron oxides, tend to leach to the surface and can be visible to the naked eye. In the sample shown in figure 3A, areas of the surface that are clear of the presence of corrosion products (4) are visible. Two patches of surface area having corrosion products present (5a, 5b) are also visible. The patches where corrosion products are present (5a, 5b) typically have an irregular shape. For simplicity of reference, these patches will be referred to as “rust patches”. When the pixel size under examination is small enough, the irregular shape of the rust patches is not a concern since each pixel will be denoted as a Rust or No-Rust and in this manner the area of a rust patch will be delineated effectively. This is illustrated in figure 3B.

[0103] Figure 3B is a photograph of a surface of a test sample of concrete under examination. Patches of corrosion products are visible, with a pixel grid (11) superimposed. The image comprises a plurality of pixels (12). The pixel grid (11 ) is arranged according to optimal pixel size (12). The spectral curve from each pixel (12) is analysed to determine if in that pixel (12) is predominantly an area of presence of corrosion products or not and make a binary determination of Rust, No-Rust to characterise each pixel. Delineation (13) of an area having presence of corrosion products (corrosion patch 5a) is performed according to the characterisation of each pixel.

[0104] The images of the concrete surface illustrated in Figures 3A and 3B are provided for illustrative purposes only. The pixels of the images illustrated in 3A and 3B do not correspond with the pixels captured by the multi-spectral camera. The pixels captured by the multi-spectral camera in one example are illustrated by the pixel grid (11 ) and example pixel (12) in Figure 3B.

[0105] Figure 4 is an illustrates an example of a typical spectral curve for a surface area having corrosion products (a “RUST” pixel) and a surface area not having corrosion products (a “No RUST” pixel). The spectral curve of a surface pixel not having corrosion products (72) generally has higher reflectance than the spectral curve of a surface pixel having corrosion products present (71). The spectral curve is analysed with a special interest in a specific wavelength band (70). In the example illustrated in Figure 4, the predetermined range of wavelengths forming the specific wavelength band (70) is between 650nm and 900nm. In other examples, the predetermined range of wavelengths forming the specific wavelength band (70) is between 650nm and 950nm. In further examples, the predetermined range of wavelengths forming the specific wavelength band is between 600nm and 900nm. In still further examples, the predetermined range of wavelengths forming the specific wavelength band is between 600nm and 950nm.

[0106] The segment of the spectral curve within the waveband of interest (71a, 72a) is examined and key characteristics of the spectral curve are processed. A first characteristic is that a spectral curve of a surface not having corrosion products (72) tends to have a negative slope (72a) within the waveband of interest, while the spectral curve of a surface having corrosion products (71 ) tends to have a positive slope (71 a) until reaching a maxima within the waveband of interest.

[0107] To determine characteristics of the spectral curve within the waveband of interest, reflectance values may be evaluated at key wavelengths. Some key wavelengths are illustrated in Figure 4 (650nm, 750nm, and 900nm).

[0108] Figure 5A is an illustration of the key characteristics of a spectral curve for a surface area having corrosion products. Figure 5B is closeup view of Figure 5A, focusing on the region of the spectral curve that falls within the waveband of interest. It can be seen that the slope (85) within the waveband of interest is positive. Key wavelengths are denoted as wi, w2, w3, W4. At least wi, w2and w3fall within what is referred to as the “waveband of interest”, which is within 650nm and 900nm in some examples. Lines of intersection (81 , 82, 83, 84) are illustrated at wavelength wi (81 ), wavelength w2(82), wavelength w3(83) and wavelength w4(84). Lines 81 , 82, 83 and 84 intersect the spectral curve at points A, B, C and D respectively. At each of these points the reflectance value is denoted as RA, RB, RC and RD respectively. These reflectance values correspond with key wavelengths wi, w2, w3and w4, respectively.

[0109] In some examples, for each pixel under examination, reflectance values RA, RB, RC and RD are processed through a transfer function that returns a binary value that denotes the presence or absence of corrosion products. The transfer function is called Rustjndex and characterises each pixel as a “Rust” pixel (Rustjndex = TRUE) or “No Rust” pixel (Rust ndex = FALSE).

[0110] In one example, the transfer function assesses the spectral curve according to the following criteria: the spectral signature has a positive slope; and a peak is identified in the spectral signature, within the predetermined range of wavelengths.

[0111] The transfer function may determine that the spectral signature has a positive slope by determining that the reflectance at the longest wavelength of the predetermined range of wavelengths is greater than the reflectance at the shortest wavelength. Alternatively, the transfer function may evaluate the slope at a particular point on the spectral curve (e.g., at the first key wavelength wi). Evaluation of the slope may be performed after smoothing.

[0112] Preferably, the transfer function may return Rustjndex = TRUE when all of the criteria are met. Alternatively, the transfer function may return Rustjndex = TRUE when any one or more of the criteria are met.

[0113] In another example, the transfer function comprises subtractions. According to this example, Rust ndex = TRUE, when all of the following conditions are met:

[0114] 1) RB- RA> i;

[0115] 2) RB- Rc> 2; and

[0116] 3) Rc- RA>k3.

[0117] In one example ki=3, k2=1 , k3=1.

[0118] In another example ki=0, k2=0, k3=0.

[0119] In another example, the transfer function comprises divisions. According to a specific example, Rustjndex = TRUE, when

[0120] 1 ) Rs / RA>kn;

[0121] 2) RB / Rc>ki2; and

[0122] 3) Rc / RA>ki3.

[0123] In one example kn=0, ki2=0, ki3=0. In another example, kn=1.12, ki2=1.05, ki3=1.02.

[0124] In yet another example, the transfer function comprises a combination of subtractions and divisions.

[0125] In the above examples, the transfer function is configured to return a “Rust Index” with a Boolean value of either TRUE or FALSE. In alternative examples, the transfer function may return an integer or float value indicating a likelihood of corrosion staining being present in the corresponding pixel. In this case, such an integer or float value could be converted to a Boolean value by applying a threshold. For example, values above a threshold value may be converted to a TRUE value, while values below the threshold value may be converted to a FALSE value.

[0126] Any of the methods described herein may be implemented as a computer program. The computer program may be configured to control the positioning means (e.g., a guidance system), image capturing means, image processing means and / or information transmission means.

[0127] A remote computer system may also be provided, configured to operate in accordance with certain methods disclosed herein. For example, whilst image processing operations may preferably be performed by a processor onboard the aerial vehicle, image processing operations may alternatively be performed by a remote computer.

[0128] The remote computer may be a cloud computing resource, such as a virtual machine.

[0129] The information transmission means may comprise at least one communication interface, particularly comprising one or both of a transmitter and receiver.

[0130] Although specific embodiments have been described, the skilled person will understand that various modifications and variations are possible. For example, whilst the disclosure is described in relation to existing technological components and nomenclature, it will be understood that changes to the components (and / or nomenclature) are possible, but the present disclosure may still be applicable in this case. Also, combinations of any specific features shown with reference to one embodiment or with reference to multiple embodiments are also provided, even if that combination has not been explicitly detailed herein.

[0131] Where this application refers to a processor or remote computer, for instance, this may actually be a pair of processors, or remote computers (primary and failover), for redundancy.

[0132] The examples may be carried out on any suitable data processing device, such as a personal computer, laptop, mobile telephone, server, virtual machine, and the like. The above description of the systems and methods has been simplified for purposes of discussion, and is intended to provide a specific example to illustrate the invention. Different types of systems and methods may be used, as will be appreciated by the skilled person. It will be appreciated that the boundaries between logic blocks are merely illustrative and that alternative embodiments may merge logic blocks or elements, or may impose an alternate decomposition of functionality upon various logic blocks or elements.

[0133] It will be appreciated that the above-mentioned functionality may be implemented as one or more corresponding modules as hardware and / or software. For example, the above- mentioned functionality may be implemented as one or more software components for execution by a processor of the system. Alternatively, the above-mentioned functionality may be implemented as hardware, such as on one or more field-programmable-gate-arrays (FPGAs), and / or one or more application-specific-integrated-circuits (ASICs), and / or one or more digital-signal-processors (DSPs), and / or other hardware arrangements. Method steps implemented in flowcharts contained herein, or as described above, may each be implemented by corresponding respective modules. Moreover, multiple method steps implemented in flowcharts contained herein, or as described above, may be implemented together by a single module.

[0134] Examples may be implemented by computer software or a “computer program”. A storage medium and a transmission medium carrying the computer software are also provided. The computer software may comprise one or more instructions, or code, that, when executed by a computer, causes the methods described to be performed. Computer software may be a sequence of instructions designed for execution on a computer system, and may include a subroutine, a function, a procedure, a module, an object method, an object implementation, an executable application, an applet, a servlet, source code, object code, a shared library, a dynamic linked library, and / or other sequences of instructions designed for execution on a computer system. The storage medium may be a magnetic disc (such as a hard drive or a floppy disc), an optical disc (such as a CD-ROM, a DVD-ROM or a Blu-ray disc), or a memory (such as a ROM, a RAM, EEPROM, EPROM, Flash memory or a portable / removable memory device), etc. The transmission medium may be a communications signal, a data broadcast, a communications link between two or more computers, etc.

[0135] Each feature disclosed in this specification, unless stated otherwise, may be replaced by alternative features serving the same, equivalent or similar purpose. Thus, unless stated otherwise, each feature disclosed is one example only of a generic series of equivalent or similar features.

[0136] As used herein, including in the claims, unless the context indicates otherwise, singular forms of the terms herein are to be construed as including the plural form and vice versa. For instance, unless the context indicates otherwise, a singular reference herein including in the claims, such as "a" or "an" (such as a multispectral camera) means "one or more” (for instance one or more multispectral cameras). Throughout the description and claims of this disclosure, the words "comprise", "including", "having" and "contain" and variations of the words, for example "comprising" and "comprises" or similar, mean "including", and are not intended to (and do not) exclude other components.

[0137] The use of any and all examples, or exemplary language ("for instance", "such as", "for example" and like language) provided herein, is intended merely to better illustrate the invention and does not indicate a limitation on the scope of the invention unless otherwise claimed. No language in the specification should be construed as indicating any nonclaimed element as essential to the practice of the invention.

[0138] Any steps described in this specification may be performed in any order or simultaneously unless stated or the context requires otherwise. Moreover, where a step is described as being performed after a step, this does not preclude intervening steps being performed.

[0139] All of the aspects and / or features disclosed in this specification may be combined in any combination, except combinations where at least some of such features and / or steps are mutually exclusive. As described herein, there may be particular combinations of aspects that are of further benefit, such the aspects of determining a set of compensation parameters and applying a set of compensation parameters to measurements. In particular, the preferred features of the invention are applicable to all aspects of the invention and may be used in any combination. Likewise, features described in non- essential combinations may be used separately (not in combination).

[0140] A method of manufacturing and / or operating any of the devices disclosed herein is also provided. The method may comprise steps of providing each of the features disclosed and / or configuring or using the respective feature for its stated function.

Claims

CLAIMS:1 . A system for detecting corrosion in reinforced concrete, the system comprising: a multi-spectral camera for capturing image data, the image data being captured at a plurality of pixels of the multi-spectral camera and comprising, for each pixel, a spectral signature being a measurement of reflectance across a predetermined range of wavelengths, wherein a shortest wavelength of the predetermined range of wavelengths is between 600 nm and 750 nm, wherein a longest wavelength of the predetermined range of wavelengths is between 790 nm and 950 nm; a processor, for processing the image data received from the multi-spectral camera, wherein the processor is configured to: receive from the multi-spectral camera image data of a surface of the reinforced concrete; determine whether corrosion is present in the reinforced concrete, wherein either: a) corrosion is determined as being present if the spectral signature captured at one or more pixels meets all of the following conditions:RB - RA > ki,RB - Rc > k2, andRc - RA > ks, wherein each of RA, RB and Rc are reflectance values at different wavelengths, w1 , w2 and w3 respectively, within the predetermined range of wavelengths, wherein w1 < w2 < w3, wherein ki >= 0, k2>= 0, and ks >= 0; or b) corrosion is determined as being present if the spectral signature captured at one or more pixels complies with each criterion of a group of criteria, the group of criteria comprising: a reflectance at the longest wavelength of the predetermined range of wavelengths is greater than a reflectance at the shortest wavelength of the predetermined range of wavelengths; and a peak in the spectral signature is identified between 750 nm and 790 nm.

2. The system of claim 1 , wherein the processor is further configured to assign a flag for storage with the image data in a computer memory, and / or for output to a user interface, the flag identifying whether corrosion is determined to be present in the image data.

3. The system of claim 1 or claim 2, wherein the processor is further configured to assign an indicator associated with each pixel of the plurality of pixels, the indicator for storage with the image data in a computer memory and / or for output to a user interface, the indicator identifying whether the spectral signature captured at the said pixel complies with at least one criterion of the group of criteria.

4. The system of any preceding claim, wherein the processor is further configured to map each pixel to a location at the surface of the reinforced concrete.

5. The system of any preceding claim, wherein: the shortest wavelength of the predetermined range of wavelengths is between 600 nm and 650 nm; and / or the longest wavelength of the predetermined range of wavelengths is between 900 nm and 950 nm.

6. The system of any preceding claim, wherein the processor is further configured to, prior to determining whether corrosion is present in the reinforced concrete, filter or smooth the spectral signature captured at each of the plurality of pixels.

7. The system of any preceding claim, wherein: ki = 3 k2= 1 k3= 0 wherein RAis the reflectance at 650 nm, RBis the reflectance at 750 nm, and wherein Rc is the reflectance at 900 nm.

8. The system of any preceding claim, wherein the group of criteria further comprises a criterion that the spectral signature does not exceed a threshold reflectance.

9. The system of any preceding claim, further comprising a moveable carrier upon which the multi-spectral camera is mounted, the moveable carrier for positioning the multi- spectral camera relative to the surface of the reinforced concrete.

10. The system of claim 9, wherein the moveable carrier is an unmanned aerial vehicle.

11. A method for detecting corrosion in reinforced concrete, the method comprising: receiving from a multi-spectral camera image data of a surface of the reinforced concrete, wherein the multi-spectral camera captures the image data at a plurality of pixels, the image data comprising, for each pixel, a spectral signature being a measurement of reflectance across a predetermined range of wavelengths, wherein a shortest wavelength of the predetermined range of wavelengths is between 600 nm and 750 nm, wherein a longest wavelength of the predetermined range of wavelengths is between 790 nm and 950 nm; determining whether corrosion is present in the reinforced concrete, wherein either: a) corrosion is determined as being present if the spectral signature captured at one or more pixels meets all of the following conditions:RB - RA > ki, RB - Rc > k2, and Rc - RA > ks, wherein each of RA, RB and Rc are reflectance values at different wavelengths, w1 , w2 and w3 respectively, within the predetermined range of wavelengths, wherein w1 < w2 < w3, wherein ki >= 0, k2>= 0, and ks >= 0; or b) corrosion is determined as being present if the spectral signature captured at one or more pixels complies with at each of a group of criteria, the group of criteria comprising: a reflectance at the longest wavelength of the predetermined range of wavelengths is greater than a reflectance at the shortest wavelength of the predetermined range of wavelengths; and a peak in the spectral signature is identified between 750 nm and790 nm.

12. The method of claim 11 , further comprising assigning a flag for storing with the image data in a computer memory and / or for outputting to a user interface, the flag identifying whether corrosion is determined to be present in the image data.

13. The method of claim 11 or claim 12, further comprising assigning an indicator associated with each pixel of the plurality of pixels, the indicator for storing with the image data in a computer memory and / or for outputting to a user interface, the indicator identifying whether the spectral signature captured at the said pixel complies with at least one criterion of the group of criteria.

14. The method of any one of claims 11 to 13, further comprising mapping each pixel to a location at the surface of the reinforced concrete.

15. The method of any one of claims 11 to 14, further comprising, prior to determining whether corrosion is present in the reinforced concrete, filtering or smoothing the spectral signature captured at each of the plurality of pixels.

16. The method of any one of claims 11 to 15, wherein : ki = 3 k2= 1 k3= 0 wherein RAis the reflectance at 650 nm, RBis the reflectance at 750 nm, and wherein Rc is the reflectance at 900 nm.

Images