Method of assessing visual impact of a 3D object on an environment

Abstract

The present invention relates to a computer-implemented method of assessing human visual impact of a three-dimensional (3D) digitally modelled object on an environment and calculating area of visual difference to be expressed as steradians.

Description

Method of assessing visual impact of a 3D object on an environmentTechnical Field

[0001] The present invention relates to a computer-implemented method of assessing human visual impact of a three-dimensional (3D) digitally modelled object on an environment.Background of Invention

[0002] Visual and landscape impact considerations are an important part of development approvals for projects, such as infrastructure projects, as they describe the impact of proposed projects on the visual amenity of the environment. Projects can affect a range of stakeholders in different ways, such as the general public, Traditional Owners, adjacent residents, commuters or others that may interact with that environment over time. The assessment of possible visual impact of objects to be built that form these projects is therefore an important factor in the planning and development process of the projects.

[0003] Current techniques for providing visual and landscape impact assessments use a combination of qualitative and quantitative approaches. Qualitative approaches are typically mainly useful for contextual analysis only and it will be appreciated that there are many problems associated with relying on these approaches.

[0004] Existing quantitative approaches have limitations too. They can be limited in their insight and / or are not robust in their application. For example, visualisation techniques have been widely used to provide landscape and visual impact assessments for proposed projects, such as wind farms. Given wind farms are often in environments with low levels of infrastructure development, wind turbine objects can have a high visual impact on the environment, especially for adjacent residencies and communities, depending on the viewing locations.

[0005] Existing visualisation techniques for quantitively assessing visual impact include manual techniques, such as making physical mock-ups of wind turbines and the environment, and then having an expert provide a subjective assessment. These techniques can be timeconsuming, costly, and lack the precision needed for consistent quantitative analysis.Furthermore, they can be limited in their ability to provide an assessment of the visual impact from multiple perspectives and scenarios, making them inefficient and sometimes inaccurate.

[0006] More recently, three-dimensional (3D) modelling techniques have been used for visualising the visual impact of proposed developments and projects on an environment. While 3D models enable stakeholders to explore proposed changes in a virtual environment, current techniques face significant technical challenges. Some of these challenges relate to the lack of integration between the 3D model of the environment and the objects to be built that are being modelled, such as wind turbines on a surrounding environment.

[0007] One existing modelling technique is a mapping technique generally used by planners and Geographical Information Systems (GIS) specialists called Zone of Theoretical Visibility. This technique produces a 2-dimensional (2D) map to evaluate whether proposed objects can be viewed from different locations in an environment. This technique typically describes whether objects can be viewed or not from different viewpoints, rather than a quantifying the extent of visual impact.

[0008] Another technique to quantify the extent to which the objects impact the surrounding environment is to use a grid counting method of rendered visualisations. This assessment is not particularly robust as the grid overlay of an equirectangular projection becomes increasingly distorted as you move from the horizon to the poles and transformation of the grid either vertically or horizontally can result in significantly different outcomes. Further, the rendering techniques typically used to produce the grid overlay do not accurately represent human perception and fail to account for field-of-view variations that influence how changes are perceived by observers.

[0009] That is, existing modelling techniques typically lack robust mechanisms to identify and quantify visual differences between the modelled original environment and the modelled objects in the environment. This makes it challenging to provide objective and consistent assessments of visual impact of the 3D modelled objects on the environment.

[0010] A reference herein to a patent document or other matter which is given as prior art is not to be taken as an admission that that document or matter was known or that theinformation it contains was part of the common general knowledge in Australia or elsewhere as at the priority date of any of the disclosure or claims herein. Such discussion of prior art in this specification is comprised to explain the context of the present invention in terms of the inventor's knowledge and experience.Summary of Invention

[0011] According to one aspect of the present invention there is provided a computer- implemented method of assessing human visual impact of a three-dimensional (3D) digitally modelled object on an environment, comprising: receiving a 3D model of an environment; receiving a 3D model of an object to be located in the environment; integrating the 3D model of the object into the 3D model of the environment to create an integrated model; receiving a selection of viewpoints of the environment for assessing human visual impact of the object on the environment; for a given viewpoint: rendering a first layer image of the integrated model from the given viewpoint of the environment using a spherical render with equirectangular projection; rendering a second layer image of the 3D model of the environment from the given viewpoint using a spherical render with equirectangular projection; computing a third image comprising pixels that differ between the first layer image and second layer image; calculating a projection coefficient weighting for each of the pixels that differ based on vertical angular displacement of the pixels relative to a horizon line to account for spherical distortion; calculating a resolution coefficient for the pixels based on resolution of the first layer image and the second layer image; assigning a binary outcome corresponding to whether the pixels differ between the first layer image and second layer image; and calculating area of visual difference by multiplying the binary outcome by the projection coefficient weightings for each the pixels and multiplying by the resolution coefficient to normalise the area of visual difference to be expressed as steradians.

[0012] The computer-implemented method is a computational technique that can be used for quantitively determining the proportional visual impact of 3D digitally modelled objects on an environment from different viewpoints, enabling new metrics to be calculated that provide significantly deeper insights than contemporary best practice methods. The calculations made by the method are significantly more accurate than existing methods as it accounts for distortion effects used in equirectangular projection.

[0013] The method evaluates the visual impact from spherical, i.e. 360-degree, viewpoints. This occurs by evaluating equirectangular projections of 360-degree, spherical renders from a viewpoint and correcting for distortion effects. The viewpoints of the environment correspond to a modelled digital camera at an average human eye-height.

[0014] In an embodiment, the method further comprises generating the 3D model of the environment from data comprising one or more of: survey data, LIDAR scans of the environment, photogrammetry data and 3D mesh data.

[0015] In an alternative embodiment, the method further comprises receiving one or more images of the environment to establish an equirectangular projected spherical image of the environment. The method may further comprise stitching these images to establish the equirectangular projected spherical image. For example, a digital camera may be rotated incrementally around a centre-point in the middle of the camera lens to obtain an equirectangular projected spherical image.

[0016] In an embodiment, the 3D model of an object to be located in the environment is generated from 3D mesh data of the object. Alternatively, the method further comprises receiving design input data and generating the 3D model of the object from this design input data. With reference to the above example, the object may be proposed infrastructure to be located in an environment, such as turbines of a wind farm. Variants of the object be 3D modelled can be made using extent modelling techniques.

[0017] In an embodiment, the step of computing the third image comprises determining a boundary extent of the object on the environment. Alternatively, the actual geometry of the object is used. The method may also comprise determining a range of motion of components of the object and updating the boundary extent of the object based on the range of motion. For example, the object is a wind turbine, and the range of motion is a circle representing the boundary extent of rotation of the wind turbine.

[0018] It will be appreciated that many modern architectural and engineering processes utilise 3D digital software to produce digital images, and this software can be utilised to generate the 3D models. In addition, early-stage concept plans can be used for this approach, such as "massing", where simplified geometry and assumed heights can beutilised. As a result, 3D digital development can be based on any inputs, whether 3D, 2D or stylised. Further, once the digital model foundations have been developed, additional modelling may be undertaken to increase detail. This is often referred to as a "level of detail" classification that is used to qualify the visual outputs produced.

[0019] The method evaluates the quantitative extent of the visual impact of the visual differences between the existing environment and the environment when the 3D digitally modelled object is located in the environment from different viewpoints. The location of the 3D digitally modelled object in the environment simulates the object being built. For example, the 3D object is a proposed windfarm. Embodiments of the method perform the evaluation using image processing techniques.

[0020] In an embodiment, the third image comprises rows of pixels. The projection coefficient is based on which row each of the pixels that differ are located to account for equirectangular projection.

[0021] In an embodiment, the method further comprises, for each of the rows of pixels, calculating a number of pixels that differ between the first layer image and second layer image, multiplying the number of pixels by the projection coefficient, and multiplying by the resolution coefficient to derive a value. The value is in steradians.

[0022] In an embodiment, the method further comprises adding the value for each of the rows of pixels to derive the area of visual difference.

[0023] In an embodiment, the projection coefficient is based on the cosine of the vertical displacement of each of the pixels that differ to account for equirectangular projection.

[0024] In an embodiment, the method further comprises providing an assessment of the human visual impact of the object on the environment from the area of visual difference based on one or more of: spherical visual impact, maximum field of view impact, minimum field of view impact, average field of view impact, and constrained fields of view impact.

[0025] This assessment of human visual impact is therefore based on metrics that are calculated from the area of visual difference. That is, the spherical visual impact, field of view impacts, and constrained fields of view impact are metrics that can provide significantlydeeper insights than contemporary best practice methods. The method thus provides a comprehensive quantitative approach for determine visual impact through a process that produces a series of metrics.

[0026] The spherical visual impact comprises a total amount of visual impact from a viewpoint. The constrained fields of view comprise limiting field of view impact across a vertical axis based on human field of interest.

[0027] In an embodiment, the method further comprises repeating for further viewpoints in the selection of viewpoints.

[0028] In an embodiment, the method further comprises generating an intensity map image indicative of the assessment of the human visual impact of the object on the environment across a grid of viewpoints of the selection of viewpoints, wherein colour of the intensity map indicates extent of the human visual impact of the object on the environment.

[0029] The method thus evaluates the spherical renders, corrects for distortion and computationally determines metrics, such as total impact, maximum, minimum and average impact for different viewpoints, for both constrained and unconstrained perspective ranges, to provide a quantitative assessment of the human visual impact of the object on the environment. The intensity map evaluates the assessment at across a grid of viewpoints to enables a heatmap of visual impact to be displayed. Different areas within a field-of-view will have different relative importance, which needs to be considered to provide a robust visual assessment.

[0030] The intensity map provides a number of advantages, such as providing a display of intensity of impact, rather than the binary option of whether it can be seen or not, and enabling the geometry of objects to be factored in, providing a greater insight into the visual impact

[0031] As mentioned, the intensity map can be applied to photomontage composites, rather than just survey data, enabling real world screening elements to be factored in.

[0032] Further, the intensity map is significantly more robust than, for example, boxcounting approaches which are do not account for the difference in box area based onvertical displacement due to equirectangular projection and are susceptible to manipulation through grid alignment. The method is significantly more precise too as it can be applied computationally on significantly higher resolutions, rather than the manual approach of 10- degree by 1-degree boxes. Also, the method produces better insights that are more flexible, as the method produces multiple different outcomes (maximum impact for constrained field of view, minimum impact for constrained field of view, total spherical viewpoint impact, average impact for constrained field of view, etc.) to be considered.

[0033] In an embodiment, the method further comprises: calculating a volume of visual difference by integrating the area of visual difference at sampled viewpoints over a geographic or planimetric area to be expressed as steradians metres-squared to produce a cumulative assessment of the human visual impact of the object on the environment.

[0034] Sampled viewpoints can be taken at higher densities at areas closer to the proposed infrastructure objects, or at areas of interest to increase the precision of the cumulative assessment.

[0035] In an embodiment, the method further comprises calculating a base case with no occlusion from given viewpoints by normalising the 3D model of the environment data to the same height, resulting in the viewpoint height, the object, such as infrastructure and potential occluding elements, all at the same vertical translation.

[0036] The surface of the viewpoint steradians values can similarly be integrated to calculate a volume represented the cumulative visual impact, to be expressed as steradian metres-squared.

[0037] To understand the sensitivity of an environment to occlusion of proposed objects, such as infrastructure, we calculate the approach based on the 3D model of the environment, e.g. survey or environment data, and then based on a normalised 3D model of the environment.

[0038] By dividing the cumulative visual assessment score using the survey or environment data by the cumulative visual assessment score using the normalised environment foundations, we obtain a visibility ratio. The complement of this is the Occlusion Ratio ( 1 - Visibility Ratio).

[0039] In an embodiment, the method further comprises providing a structured cumulative assessment by calculating the volume over select areas.

[0040] For example, the cumulative assessment of the human visual impact of the object on the environment can be overlayed with target areas in the environment to be able to calculate the cumulative impact of different layers, such as residential areas. This is particularly important for calculating impact over a different property boundary when, for example, compensation payments are involved.

[0041] According to another aspect of the present invention there is provided a computer-implemented method of assessing human visual impact of a three-dimensional (3D) digitally modelled object on an environment, comprising: receiving spherical photographs of an environment from a plurality of viewpoints; receiving a 3D model of an object to be located in the environment; compositing the 3D model of the object on the spherical photographs of the environment to create an integrated model; receiving a selection of viewpoints of the environment for assessing human visual impact of the object on the environment; for a given viewpoint: rendering a first layer image of the integrated model from the given viewpoint of the environment using a spherical render with equirectangular projection; rendering a second layer image of the 3D model of the environment from the given viewpoint using a spherical render with equirectangular projection; computing a third image comprising pixels that differ between the first layer image and second layer image; calculating a projection coefficient weighting for each of the pixels that differ based on vertical angular displacement of the pixels relative to a horizon line to account for spherical distortion; calculating a resolution coefficient for the pixels based on resolution of the first layer image and the second layer image; assigning a binary outcome corresponding to whether the pixels differ between the first layer image and second layer image; and calculating area of visual difference by multiplying the binary outcome by the projection coefficient weightings for each the pixels and multiplying by the resolution coefficient to normalise the area of visual difference to be expressed as steradians .

[0042] That is, the computer-implemented method is also able to be applied using photo-composites, where composited images can be evaluated. It will be appreciated thatobscuring and screening elements in the photo-composites can be factored into the evaluation of the area of visual difference.

[0043] According to another aspect of the present invention there is provided software for use with a computer comprising a processor and memory for storing the software, the software comprising a series of instructions executable by the processor to carry out the method as claimed in any one of the preceding claims.

[0044] According to another aspect of the present invention there is provided a system for assessing human visual impact of a three-dimensional (3D) digitally modelled object on an environment, the system comprising: a computing device comprising a memory and a processor, the computing device configured to implement a 3D modelling module and an image processing module, wherein the 3D modelling module is configured to: receive a 3D model of an environment; receive a 3D model of an object to be located in the environment; integrate the 3D model of the object into the 3D model of the environment to create an integrated model; and receive a selection of viewpoints of the environment for assessing human visual impact of the object on the environment, wherein the image processing module is configured, for a given viewpoint, to: render a first layer image of the integrated model from the given viewpoint of the environment using a spherical render with equirectangular projection; render a second layer image of the 3D model of the environment from the given viewpoint using a spherical render with equirectangular projection; compute a third image comprising pixels that differ between the first layer image and second layer image; calculate a projection coefficient weighting for each of the pixels that differ based on vertical angular displacement of the pixels relative to a horizon line to account for spherical distortion; calculate a resolution coefficient for the pixels based on resolution of the first layer image and the second layer image; assign a binary outcome corresponding to whether the pixels differ between the first layer image and second layer image; and calculate area of visual difference by multiplying the binary outcome by the projection coefficient weightings for each the pixels and multiplying by the resolution coefficient to normalise the area of visual difference to be expressed as steradians.

[0045] It will be appreciated by those persons skilled in the art that the area of visual difference could also be expressed as degrees-squared, square degrees, or radians-squared,which are non-standard units of solid angles that are used in the art. Further, the area of visual difference could be presented as a percentage of a sphere or a percentage of a human field of view.Brief Description of Drawings

[0046] Embodiments of the invention will now be described with reference to the accompanying drawings. It is to be understood that the embodiments are given by way of illustration only and the invention is not limited by this illustration. In the drawings:

[0047] Figure 1 is a flow chart of a computer-implemented method of assessing human visual impact of a three-dimensional (3D) digitally modelled object on an environment according to an embodiment of the present invention;

[0048] Figure 2 is a schematic of a system for assessing human visual impact of a three- dimensional (3D) digitally modelled object on an environment according to an embodiment of the present invention;

[0049] Figure 3 is a further flow chart of a computer-implemented method of assessing human visual impact of a three-dimensional (3D) digitally modelled object on an environment according to an embodiment of the present invention;

[0050] Figure 4 shows intensity map images of 3D digitally modelled objects on an environment from different views according to an embodiment of the present invention; and

[0051] Figure 5 shows an intensity map image and viewpoint images of 3D digitally modelled objects on an environment according to an embodiment of the present invention.Detailed Description

[0052] A summary of a computer-implemented method 10 of assessing human visual impact of a three-dimensional (3D) digitally modelled object on an environment is shown in Figure 1. The method comprising: receiving 12 a 3D model of an environment; receiving 14 a 3D model of an object to be located in the environment; integrating 16 the 3D model of the object into the 3D model of the environment to create an integrated model; receiving 18 a selection of viewpoints of the environment for assessing human visual impact of the objecton the environment; for a given viewpoint: rendering 18 a first layer image of the integrated model from the given viewpoint of the environment using a spherical render with equirectangular projection; rendering 20 a second layer image of the 3D model of the environment from the given viewpoint using a spherical render with equirectangular projection; computing 22 a third image comprising pixels that differ between the first layer image and second layer image; calculating 24 a projection coefficient weighting for each of the pixels that differ based on vertical angular displacement of the pixels relative to a horizon line to account for spherical distortion; calculating 26 a resolution coefficient for the pixels based on resolution of the first layer image and the second layer image; assigning 28 a binary outcome corresponding to whether the pixels differ between the first layer image and second layer image; and calculating 30 area of visual difference by multiplying the binary outcome by the projection coefficient weightings for each the pixels and multiplying by the resolution coefficient to normalise the area of visual difference to be expressed as steradians.

[0053] The method 10 further comprises repeating for further viewpoints in the selection of viewpoints.

[0054] It will be appreciated by those persons skilled in the art that further aspects of the method 10 will be apparent from the below description of a system 31, shown in Figure 2, for assessing human visual impact of a three-dimensional (3D) digitally modelled object on an environment.

[0055] The system 31 shown in Figure 2 comprises a computing device 33 comprising a memory 32 and a processor 34. The computing device 33 is configured to implement a 3D modelling module 36 and an image processing module 38 to perform the assessment.

[0056] Persons skilled in the art will appreciate that at least part of the method 10 could be embodied in software (e.g., program code) that is implemented by the processor 34 that is configured to control the computing device 33 for performing the human visual impact assessment. Also, the software could be supplied in a number of ways, for example on a tangible computer readable medium, such as a disc, or in the memory 32 of the computing device 33. The software could also be supplied by any possible data transfer method too, such as downloading.

[0057] The 3D modelling module 36 is configured to receive a 3D model of an environment and receive a 3D model of an object to be located in the environment. In an embodiment, the 3D modelling module 36is further configured to generate the 3D model of an environment. The 3D modelling module 36 may do so using survey data or LIDAR data of the environment. Photogrammetry or other alternative inputs may also be used to develop this 3D digital environment, with the level of precision of these inputs having direct impact on the prevision of the outputs.

[0058] The 3D modelling module 36 may also be configured to generate the 3D model of the object. As mentioned, modern architectural and engineering processes utilise 3D digital software to produce digital images, and this software can be utilised to generate the 3D models of the object and the environment.

[0059] Also as mentioned, in an example, the object to be located in the environment may be a proposed infrastructure project, such as turbines of a wind farm. A 3D model of the turbines may therefore be received from an engineering firm designing the turbines for the wind farm project or the 3D modelling module 36 may be configured to generate the 3D model of turbines from data obtained from the engineering firm.

[0060] Establishing an accurate 3D model of the existing environment is critical for the assessing of human visual impact of the 3D object on the environment for at least three reasons. Firstly, to establish the baseline for comparison, particularly if existing infrastructure elements are also to be isolated for comparative analysis. Secondly, to determine the extent and nature of any occluding foreground elements to the proposed digital infrastructure. Third, the ground terrain at the digital camera location will determine the height of the digital camera, which has a material impact on the relative angle of the proposed infrastructure to the viewpoint.

[0061] The 3D modelling module 36 is configured to integrate the 3D model of the object into the 3D model of the environment to create an integrated model. Integrating the 3D models of both the proposed infrastructure objects and surrounding environment is then undertaken. For some infrastructure, this may be a solely additive activity, such as for windfarms on greenfield sites, however for more complex infrastructure, this may include removal of existing infrastructure or integration with existing elements.

[0062] The 3D modelling module 36 is further configured to receive a selection of viewpoints of the environment for assessing human visual impact of the object on the environment. Once the 3D Digital integrated model has been developed, virtual digital cameras will be placed in the scene based on the viewpoint selection. The cameras will not require lens or sensor size as 360-degree, spherical renders will be produced; however, the height of the cameras will be based on user selection.

[0063] Spherical renders are critical for this process, as taking select viewpoints with select fields of view limits the effectiveness for locations where proposed infrastructure is sufficiently close or where the extent of development, such as a transmission line along a road, does not take into account the total range of impact.

[0064] It is considered that best practice is to use the average human eye-height for the height of the cameras, however this may differ based on local population or specific stakeholder requirements.

[0065] Given the existing scene and proposed infrastructure have now been developed, layered render images can now be produced. The image processing module 38 is configured to render the images and calculate an area of visual difference from the images to perform the assessment.

[0066] For a given viewpoint, image processing module 38 is configured to render a first layer image of the integrated model of the environment using a spherical render with equirectangular projection, render a second layer image of the 3D model of the environment using a spherical render with equirectangular projection, and compute a third image comprising pixels that differ between the first layer image and second layer image. The third image is a "difference map" between the two images, produced either by rendering off the occluded proposed geometry with alpha layers, or by rendering both and performing a different calculation to produce a further image that highlights the difference between geometry.

[0067] This image can also be decomposed further so that different elements are rendered onto different layers of the image to enable independent calculation. This mayinclude separating out the infrastructure of a new substation from proposed transmission lines to evaluate each independently and together.

[0068] The image processing module 38 is further configured to determine the quantitative extent of the visual impact of the object on the environment. For a given viewpoint, the image processing module 38 is configured to assign a binary outcome corresponding to whether the pixels differ between the first layer image and second layer image. That is, to computationally calculate the area of visual difference, the number of pixels that differ are calculated. As each pixel is weighted differently based on equirectangular project, each pixel coordinate is assigned a binary outcome, either 0 for no difference, or 1 for difference.

[0069] The image processing module 38 is further configured to calculate a resolution coefficient for the pixels based on resolution of the first layer image and the second layer image. Due to the above step, the resolution of the image is critical to the precision of the assessment.

[0070] The image processing module 38 is further configured to calculate a projection coefficient weighting for each of the pixels that differ based on vertical angular displacement of the pixels relative to a horizon line to account for spherical distortion. To understand the area in terms of field of view, each pixel must be weighted based upon the field-of-view that is disrupted. For instance, with equirectangular projection, the area described by 1 degree near the poles is different to 1 degree at the horizon line. To account for this, the coefficient is calculated based on the cosine of the area at the mid-point of the pixel, as if it is a trapezoid. The projection coefficient thus compensates for distortion effects used in equirectangular projection.

[0071] The image processing module 38 is then configured to calculate an area of visual difference by multiplying the binary outcome by the projection coefficient weightings for each the pixels and multiplying by the resolution coefficient to normalise the area of visual difference to be expressed as steradians or radians-squared. That is, once the pixel with image have been added together with their weightings, the total is multiplied by a fixed metric to convert to radians-squared or steradians. This is based on the level of resolution of the images provided.

[0072] The processor 34 further implements a metric calculation module 40, which is configured to calculate one or more metrics to provide an assessment of the human visual impact of the object on the environment from the area of visual difference. These metrics comprise: spherical visual impact, maximum field of view impact, minimum field of view impact, average field of view impact, and constrained fields of view impact.

[0073] Different metrics are calculated based on the visual impact data and coefficients for a variety of reasons. The spherical visual impact corresponds to the total amount of impact from a viewpoint. The field of view impact for a fixed field of view could include maximum impact, minimum impact or average impact. The constrained fields of view impact correspond to limiting the field of view impact across the vertical and horizontal axes based on a human field of interest.

[0074] Each of these metrics produce different results for the assessment of human visual impact across a variety of scenarios, particularly when adjacent to large scale infrastructure objects.

[0075] To determine the maximum impact for a fixed field of view, the metric can be calculated using brute force approaches, or optimisation algorithms to reduce computational time.

[0076] The viewpoint selection for visual impact assessments is often driven by the sensitivity of locations relative to the local stakeholder environment. This may include community spaces, residential or commercial developments, or through engagement with Traditional Owners. These metrics enable quantitative descriptions from each of the viewpoints.

[0077] The selected viewpoints can be used in an intensity map image generated by the metric calculation module 40. The intensity map image is indicative of the assessment of the human visual impact of the object on the environment across a grid of viewpoints of the selection of viewpoints. The size of the grid determines the precision of the map, with values interpolated between different points.

[0078] The colour of the intensity map indicates extent of the human visual impact of the object on the environment. In the embodiment, the intensity map is a heatmap with redindicating a high level of human visual impact and black indicating a very low level of human visual impact. The intensity map is also called a Visual Amenity Intensity Map, and it enables a quick comparison of existing and proposed infrastructure on an environment, including on brownfield sites.

[0079] The metrics also enable comparisons of existing and proposed infrastructure objects, particularly for sites where there is already visual disruption. This may include adding more turbines to an existing windfarm, or a rail overpass in a residential environment. To calculate comparative metrics, layers of each development need to be isolated so that the change in each metric can be evaluated.

[0080] As the process can be developed using 3D digital plans and survey data, different layouts and configurations can be evaluated to understand the visual impact across the design process. This can be applied at the earliest stages of development, from massing and concept design, right through to detailed design and development approval.

[0081] In an alternative embodiment, the processor 34 implements a photo-composite module configured to receive spherical photographs of an environment from a plurality of viewpoints. The photo-composite module is also configured to receive a 3D model of an object to be located in the environment and to composite the 3D model of the object on the spherical photographs of the environment to create an integrated model. The integrated model is used in the above-described manner by the image processing module 38 and the metric calculation module 40 to perform and output the visual assessment.

[0082] The alternative pathway for development is using photomontages as a foundation for the existing environment. This approach is usually undertaken later in the process as it is more time and labour intensive, however it provides a significantly more precise view of foreground elements, such as trees, landscaping, geographical features and other elements, particularly in the near-field environment.

[0083] This process is typically undertaken by stitching together photos taken with a50mm lens, however for our purposes, an equirectangular projected 360-sphere is the best practice method. To capture this, a mechatronic head is advised to cycle the cameraincrementally around a centre-point in the middle of the camera lens to minimise distortion and aberration effects.

[0084] Once the image has been stitched together, manual compositing is required whereby the rendered image of the proposed infrastructure is processed to be occluded by foreground elements previously described.

[0085] The outcome of the pathway is the same as the 3D Digital Pathway; however, the existing environment render is based on the stitched photography.

[0086] An example of assessing the human visual impact of a three-dimensional (3D) digitally modelled object on an environment is shown in Figures 4 and 5. In this example, the object is a proposed electrical infrastructure project.

[0087] Figure 4 shows intensity map images of the 3D digitally modelled objects of the proposed electrical infrastructure project on an environment. These images are heatmap images where colour indicates extent of visual impact on the environment; i.e. red is high impact and black is little or no impact.

[0088] The three images of Figure 4 show the assessment of human visual impact based on different metrics of unconstrained view, maximum constrained view and average constrained view. The left image of Figure 4 shows the intensity map image indicative of the assessment being based on an unconstrained view and shows the impact area of visual difference of the object on the environment across a grid of viewpoints expressed as steradians. As mentioned, the values of the intensity map between the grid of viewpoints are interpolated.

[0089] The middle image of Figure 4 shows the intensity map image indicative of the assessment being based on a maximum constrained view and also shows the impact area of visual difference of the object on the environment across the grid of viewpoints. The right image of Figure 4 shows the intensity map image indicative of the assessment being based on an average constrained view, which is clearly less impactful than the maximum or unconstrained view.

[0090] Figure 5 shows an intensity map image with three viewpoints on the grid selected: VPl, VP2 and VP3. The corresponding viewpoint images of these viewpoints are shown, showing the 3D digitally modelled objects on the environment.

[0091] In the example shown in Figure 5, the objects are wind turbines. The viewpoints of the wind farm project have varying levels of visual impact. VPl has the highest visual impact and is shown in orange on the intensity map image. It is also the closest to the turbines. VP3 is the furthest from the turbines, has the least visual impact and is shown in dark blue on the intensity map image.

[0092] Those skilled in the art will also appreciate that the invention described herein is susceptible to variations and modifications other than those specifically described. It is to be understood that the invention comprises all such variations and modifications.

[0093] Where any or all of the terms "comprise", "comprises", "comprised" or "comprising" are used in this specification (comprising the claims) they are to be interpreted as specifying the presence of the stated features, integers, steps or components, but not precluding the presence of one or more other features, integers, steps or components.

Claims

The claims defining the invention are as follows:

1. A computer-implemented method of assessing human visual impact of a three- dimensional (3D) digitally modelled object on an environment, comprising: receiving a 3D model of an environment; receiving a 3D model of an object to be located in the environment; integrating the 3D model of the object into the 3D model of the environment to create an integrated model; receiving a selection of viewpoints of the environment for assessing human visual impact of the object on the environment; for a given viewpoint: rendering a first layer image of the integrated model from the given viewpoint of the environment using a spherical render with equirectangular projection; rendering a second layer image of the 3D model of the environment from the given viewpoint using a spherical render with equirectangular projection; computing a third image comprising pixels that differ between the first layer image and second layer image; calculating a projection coefficient weighting for each of the pixels that differ based on vertical angular displacement of the pixels relative to a horizon line to account for spherical distortion; calculating a resolution coefficient for the pixels based on resolution of the first layer image and the second layer image; assigning a binary outcome corresponding to whether the pixels differ between the first layer image and second layer image; and calculating area of visual difference by multiplying the binary outcome by the projection coefficient weightings for each the pixels and multiplying by the resolution coefficient to normalise the area of visual difference to be expressed as steradians.

2. A computer-implemented method of claim 1, wherein the third image comprises rows of pixels and the projection coefficient is based on which row each of the pixels that differ are located to account for equirectangular projection.

3. A computer-implemented method of claim 2, further comprising, for each of the rows of pixels, calculating a number of pixels that differ between the first layer image and second layer image, multiplying the number of pixels by the projection coefficient, and multiplying by the resolution coefficient to derive a value.

4. A computer-implemented method of claim 3, further comprising adding the value for each of the rows of pixels to derive the area of visual difference.

5. A computer-implemented method of any one of claims 1 to 4, wherein the projection coefficient is based on the cosine of the vertical displacement of each of the pixels that differ to account for equirectangular projection.

6. A computer-implemented method of any one of claims 1 to 5, further comprising providing an assessment of the human visual impact of the object on the environment from the area of visual difference based on one or more of: spherical visual impact, maximum field of view impact, minimum field of view impact, average field of view impact, and constrained fields of view impact.

7. A computer-implemented method of claim 6, wherein the spherical visual impact comprises a total amount of visual impact from a viewpoint.

8. A computer-implemented method of claim 6, wherein the constrained fields of view comprise limiting field of view impact across a vertical axis based on human field of interest.

9. A computer-implemented method of any one of claims 1 to 8, further comprising repeating for further viewpoints in the selection of viewpoints.

10. A computer-implemented method of claim 9, when dependent on claims 6 to 8, further comprising generating an intensity map image indicative of the assessment of the human visual impact of the object on the environment across a grid of viewpoints of the selection of viewpoints, wherein colour of the intensity map indicates extent of the human visual impact of the object on the environment.

11. A computer-implemented method of claim 10, further comprising interpolating values of the intensity map between the grid of viewpoints.

12. A computer-implemented method of claim 9, further comprising: calculating a volume of visual difference by integrating the area of visual difference at sampled viewpoints over a geographic or planimetric area to be expressed as steradians metres-squared to produce a cumulative assessment of the human visual impact of the object on the environment.

13. A computer-implemented method of any one of claims 1 to 12, wherein computing the third image comprises determining a boundary extent of the object on the environment.

14. A computer-implemented method of claim 13, further comprising determining a range of motion of components of the object and updating the boundary extent of the object based on the range of motion.

15. A computer-implemented method of any one of claims 1 to 14, further comprising generating the 3D model of the environment from data comprising one or more of: survey data, LIDAR scans of the environment, photogrammetry data and 3D mesh data.

16. A computer-implemented method of any one of claims 1 to 15, further comprising receiving one or more images of the environment to establish an equirectangular projected spherical image of the environment.

17. A computer-implemented method of claim 17, further comprising stitching the images to establish the equirectangular projected spherical image.

18. A computer-implemented method of any one of claims 1 to 17, wherein the viewpoints of the environment correspond to a modelled digital camera at an average human eye-height.

19. A computer-implemented method of assessing human visual impact of a three- dimensional (3D) digitally modelled object on an environment, comprising: receiving spherical photographs of an environment from a plurality of viewpoints; receiving a 3D model of an object to be located in the environment; compositing the 3D model of the object on the spherical photographs of the environment to create an integrated model;receiving a selection of viewpoints of the environment for assessing human visual impact of the object on the environment; for a given viewpoint: rendering a first layer image of the integrated model from the given viewpoint of the environment using a spherical render with equirectangular projection; rendering a second layer image of the 3D model of the environment from the given viewpoint using a spherical render with equirectangular projection; computing a third image comprising pixels that differ between the first layer image and second layer image; calculating a projection coefficient weighting for each of the pixels that differ based on vertical angular displacement of the pixels relative to a horizon line to account for spherical distortion; calculating a resolution coefficient for the pixels based on resolution of the first layer image and the second layer image; assigning a binary outcome corresponding to whether the pixels differ between the first layer image and second layer image; and calculating area of visual difference by multiplying the binary outcome by the projection coefficient weightings for each the pixels and multiplying by the resolution coefficient to normalise the area of visual difference to be expressed as steradians.

20. A system for assessing human visual impact of a three-dimensional (3D) digitally modelled object on an environment, the system comprising: a computing device comprising a memory and a processor, the computing device configured to implement a 3D modelling module and an image processing module, wherein the 3D modelling module is configured to: receive a 3D model of an environment; receive a 3D model of an object to be located in the environment; integrate the 3D model of the object into the 3D model of the environment to create an integrated model; and receive a selection of viewpoints of the environment for assessing human visual impact of the object on the environment, wherein the image processing module is configured, for a given viewpoint, to:render a first layer image of the integrated model from the given viewpoint of the environment using a spherical render with equirectangular projection; render a second layer image of the 3D model of the environment from the given viewpoint using a spherical render with equirectangular projection; compute a third image comprising pixels that differ between the first layer image and second layer image; calculate a projection coefficient weighting for each of the pixels that differ based on vertical angular displacement of the pixels relative to a horizon line to account for spherical distortion; calculate a resolution coefficient for the pixels based on resolution of the first layer image and the second layer image; assign a binary outcome corresponding to whether the pixels differ between the first layer image and second layer image; and calculate area of visual difference by multiplying the binary outcome by the projection coefficient weightings for each the pixels and multiplying by the resolution coefficient to normalise the area of visual difference to be expressed as steradians.

Images