Hi Jaeyoung,
For the first question, you can use SIFT (Scale Invariant Feature Transform) to extract features from the images. Using the extracted features, you can then compare both the images having different scale. To extract SIFT features in MATLAB, you can use “detectSIFTFeatures” function: https://www.mathworks.com/help/vision/ref/detectsiftfeatures.html?s_tid=doc_ta
For the second question, you can follow the process mentioned below:
- Detect four marker points using a feature detection algorithm like SURF. You can use “detectSURFFeatures” MATLAB function to extract the SURF features. Refer to: https://www.mathworks.com/help/vision/ref/detectsurffeatures.html?s_tid=doc_ta
- Establish a coordinate system by defining the positions of four reference markers using “fitgeotform2d” MATLAB function. Refer to: https://www.mathworks.com/help/images/ref/fitgeotform2d.html
- Estimate the geometric transformation between the new frames and the referenced frame. You can use “estgeotform2d” MATLAB function for the same. Refer to: https://www.mathworks.com/help/vision/ref/estgeotform2d.html
- Once you have the estimated transformation, you can apply it to the position of the desired point in each frame to obtain its position in the reference frame.
I hope this helps.
