This repository contains my solution to the coding assignment for Skylark Drones. The assignment was to find 'L' shaped white Ground Control Points (GCPs) from the given images.
The method I have used to solve this problem is using Scale-Invariant Feature Transform (SIFT) method along with Blunt Force Matching.
-> First the Image is converted to binary so as to remove major noise and extra features.
-> Then it is passed through the SIFT function.
-> The resulting image is matched using KNN with Blunt Force Method with GCP at 4 different angles kept in the template folder in data.
-> The resulting matched points are returned. Please note that the array obtained is first checked for similar points or points in same area.
A point to be noted is that the results can be improved by changing the feature extraction and matching methods, but the main idea is the same.
The following is a correctly placed marker by the method mentioned above.
The following is an image containing a GCP without a marker because of obstruction from tree.
The following is an image without a GCP with a marker.
The following is an image with multiple GCP but the result shows only one. This can be improved by a little more processing and changing methods. (I am sure I will be able to improve this but could not because lack of time.)
