In this article, you will learn how to use GEWEL to find a place where a photo is created in a suburban area. In this example, we will use two weaknesses: [[GEWEL-4. Used Language and Inscriptions]] and [[GEWEL-2. Notable Large Objects]], and apply a number of techniques to them ![[uog-source-1.png]] >[!abstract] >- [[UoG-1. Geolocating in a Suburban Area#^ca4b33|Getting input data]] >- [[UoG-1. Geolocating in a Suburban Area#^260d5b|GEWEL-4. Used Language and Inscriptions]] > - [[UoG-1. Geolocating in a Suburban Area#^6dc352|GETL-4.4. Get Countries with Similar License Plates]] > - [[UoG-1. Geolocating in a Suburban Area#^1a4e11|GETL-4.15. Search for the Country the License Plate Belongs to]] > - [[UoG-1. Geolocating in a Suburban Area#^dd6a80|GETL-4.2. Search for the License Plate Algorithm]] > - [[UoG-1. Geolocating in a Suburban Area#^719ca6|GETL-4.16. Draw a Silhouette of the Inscription]] > - [[UoG-1. Geolocating in a Suburban Area#^ddde7e|GETL-4.17. Compare the Silhouette with Possible Variants]] > - [[UoG-1. Geolocating in a Suburban Area#^7191f3|GETL-4.5. Use the License Plate Algorithm]] >- [[UoG-1. Geolocating in a Suburban Area#^1abdee|GEWEL-2. Notable Large Objects]] > - [[UoG-1. Geolocating in a Suburban Area#^bdd06d|GETL-2.8. Determine Places with Huge Buildings or Structures]] > - [[UoG-1. Geolocating in a Suburban Area#^c4c79d|GETL-2.5. Search for the Same Huge Buildings or Structures]] > - [[UoG-1. Geolocating in a Suburban Area#^6751fd|GETL-2.1. Determine Places with Similar Topography Features]] > - [[UoG-1. Geolocating in a Suburban Area#^377877|GETL-2.4. Search for the Same View of the Landform]] >- [[UoG-1. Geolocating in a Suburban Area#^594263|Conclusion]] ## Getting input data ^ca4b33 Once a photograph is obtained, the objects that will be used as inputs need to be selected and collected ![[uog-coord-1.png]] ## Applying weakness: Used Language and Inscriptions ^260d5b Understanding the weakness [[GEWEL-4. Used Language and Inscriptions]] we pay attention to the license plate number ![[uog-coord-2.png]] ### GETL-4.4. Get Countries with Similar License Plates ^6dc352 Making the hypothesis that the license plate was issued in the country we are looking for, let's use the [[GETL-4.4. Get Countries with Similar License Plates]] technique and see how different license plates look like to create a list of hypothetical countries. Notice that European license plates are not similar and don't fit, so we move on to Asian license plates ![[uog-coord-3.png]] Looking at license plates in Asia, we find similarities with the format of license plates in China ![[uog-coord-4.png]] ### GETL-4.15. Search for the Country the License Plate Belongs to ^1a4e11 Let's apply the [[GETL-4.15. Search for the Country the License Plate Belongs to]] technique and find an example of a Chinese license plate. The format is the same, so the presumed search area is narrowed down to China ![[uog-coord-5.jpg]] ### GETL-4.2. Search for the License Plate Algorithm ^dd6a80 License plates often contain information about the region where they were registered. Let's use the [[GETL-4.2. Search for the License Plate Algorithm]] technique and find [article](https://jenxi.com/china-vehicle-registration-plates/), which describes the algorithm of license plate formation in China. We learn that the first character indicates municipality, province, or autonomous region, and the second letter gives us the district, county, or city where the vehicle was registered ![[uog-coord-6.png]] ![[uog-coord-7.png]] ### GETL-4.16. Draw a Silhouette of the Inscription ^719ca6 Since the quality of the photo allows us to accurately identify only the second character, let's use the [[GETL-4.16. Draw a Silhouette of the Inscription]] technique and draw a silhouette of the first character ![[uog-coord-8.png]] ![[uog-coord-9.png]] ### GETL-4.17. Compare the Silhouette with Possible Variants ^ddde7e Once we have a character silhouette, we use the [[GETL-4.17. Compare the Silhouette with Possible Variants]] technique and search through all the possible variants of the characters that are listed in the article we found. Find the most similar character for the province of Fujian — 闽 ![[uog-coord-10.png]] Let's take a Fujian provincial license plate and overlay the resulting silhouette on it, which matched up ![[uog-coord-12.png]] ### GETL-4.5. Use the License Plate Algorithm ^7191f3 This allows us to use the [[GETL-4.5. Use the License Plate Algorithm]] technique and significantly narrow the search area to Fujian Province and Fuzhou City, since the character is followed by the letter A ![[uog-coord-13.png]] ## Applying weakness: Notable Large Objects ^1abdee The weakness [[GEWEL-2. Notable Large Objects]] draws our attention to large objects in the background ![[uog-coord-14.png]] Zooming in, we can see skyscrapers and a bridge that has distinctive supports. Let's focus on finding the bridge ![[uog-coord-[15]-[19].png]] ### GETL-2.8. Determine Places with Huge Buildings or Structures ^bdd06d Open the [Baidu maps](https://map.baidu.com/) and using the [[GETL-2.8. Determine Places with Huge Buildings or Structures]] technique, go through the map and select all the bridges over the river between urban and suburban areas ![[uog-coord-16.png]] ### GETL-2.5. Search for the Same Huge Buildings or Structures ^c4c79d Following the [[GETL-2.5. Search for the Same Huge Buildings or Structures]] technique, examine each bridge, paying attention to the supports ![[uog-coord-17.png]] Find a bridge matching the appearance of the supports, compare with our image ![[uog-coord-18.png]] ![[uog-coord-[15]-[19].png]] Obtaining the location of the bridge allows us to narrow our search to the Mawei district ![[uog-coord-20.png]] ![[uog-coord-21.png]] Let's pay attention to the terrain presented in the picture. Note that the location of the picture is on a hillside ![[uog-coord-22.png]] ### GETL-2.1. Determine Places with Similar Topography Features ^6751fd Let's use the [[GETL-2.1. Determine Places with Similar Topography Features]] technique and find roads on a topographic map that are located on a hill in the area ![[uog-coord-23.png]] ### GETL-2.4. Search for the Same View of the Landform ^377877 Using the technique [[GETL-2.4. Search for the Same View of the Landform]] we can find the location and set the exact coordinates: `25.99565829616495, 119.44233660815982` ![[uog-[coord-24]-[date-time-2].png]] ## Conclusion ^594263 So, having decomposed and structured the search process, we solved the problem in a relatively short time. Here's a list of the techniques that were used for that case: - [[GETL-4.4. Get Countries with Similar License Plates]] - [[GETL-4.15. Search for the Country the License Plate Belongs to]] - [[GETL-4.2. Search for the License Plate Algorithm]] - [[GETL-4.16. Draw a Silhouette of the Inscription]] - [[GETL-4.17. Compare the Silhouette with Possible Variants]] - [[GETL-4.5. Use the License Plate Algorithm]] - [[GETL-2.8. Determine Places with Huge Buildings or Structures]] - [[GETL-2.5. Search for the Same Huge Buildings or Structures]] - [[GETL-2.1. Determine Places with Similar Topography Features]] - [[GETL-2.4. Search for the Same View of the Landform]]