Image processing of aerial imagery for energy infrastructure analysis using pre-processing image selection

1. A computer-implemented method for selecting aerial images for image processing to identify Energy Infrastructure (EI) features, the method to be carried out by at least one processor executing computer instructions, the method comprising:

receiving a first plurality of aerial images spanning a portion of global terrain, the first plurality of aerial images associated with a first time of image capture;

receiving a second plurality of aerial images spanning the portion of global terrain, the second plurality of aerial images associated with a second time of image capture;

identifying one or more differences in terrain content between the first time of image capture and the second time of image capture according to a comparison of image content in at least one aerial image from the first plurality of aerial images and image content in at least one aerial image from the second plurality of aerial images;

identifying one or more geographical locations at which the respective one or more differences in terrain content have been identified;

selecting a set of aerial images from the first plurality of aerial images, the second plurality of aerial images, or a third plurality of aerial images based on the one or more geographical locations having a difference in terrain content;

applying an EI feature recognition model to the set of aerial images to identify at least one EI feature.

2. The computer-implemented method of claim 1 , further comprising determining a level of relevance of the difference in terrain content identified at at-least one of the one or more geographical locations, wherein the level of relevance is representative of one of;

a magnitude or scale of the difference in terrain content;

a determined likelihood that the difference in terrain content is related to EI feature development.

3. The computer-implemented method of claim 2 wherein selecting the set of aerial images from the first plurality of aerial images, the second plurality of aerial images, or a third plurality of aerial images is further based on one or more of:

the determined level of relevance;

a ranking of a plurality of determined levels of relevance;

a comparison of one or more levels of relevance with a predetermined threshold.

4. The computer-implemented method of claim 2 wherein determining the level of relevance of the difference observed at at-least one of the one or more geographical locations further comprises:

determining a value for each of one or more difference attributes associated with an identified difference in terrain content, and

determining one or more weighting factors associated with each of the respective one or more difference attributes.

5. The computer-implemented method of claim 4 , wherein the level of relevance of the identified difference in terrain content is based on a sum over the one or more difference attributes, of the product of the value of each difference attribute and an associated weighting factor.

6. The computer-implemented method of claim 1 , wherein the third plurality of aerial images has an image resolution that is different to the image resolution of either the first or second pluralities of aerial images.

7. The computer-implemented method of claim 6 , wherein the third plurality of aerial images has an image resolution that is higher than the image resolution of both the first plurality of aerial images and the second plurality of aerial images.

8. The computer-implemented method of claim 1 further comprising obtaining the third plurality of aerial images.

9. The computer-implemented method of claim 1 , wherein identifying the one or more differences in terrain content includes:

adjusting a scale or a resolution of a first aerial image from the first plurality of aerial images to align with a scale or resolution of a second aerial image from the second plurality of aerial images,

aligning a first set of pixel positions in the first aerial images with a second set of pixel positions in the second aerial image such that the first and second sets of pixel positions correspond to a same geographical area and,

performing a comparison between image content of the first aerial image associated with the first set of pixel positions and image content of the second aerial image associated with the second set of pixel positions.

10. The computer-implemented method of claim 1 , wherein the difference in pixel values between at least one aerial image from the first plurality of aerial images and at least one aerial image from the second plurality of aerial images is representative of the development of an EI feature.

11. The computer-implemented method of any preceding claim further comprising providing information relating to the identified at least one EI feature to a user of an online platform.