Tico_tech Posted January 28, 2017 Report Share Posted January 28, 2017 I’m trying to automate the process for aligning two images vertically. The images are from different sensors mounted on the same PCA. As shown below in the "aligned_images.png", images overlayed are vertically aligned: Currently the alignment acceptance criteria is determined by the operator performing the alignment. My intent is to remove the subjective criteria and have software decide if alignment is correct. Things attempted: 1. I compared the distance from bottom to first edge; shown in Right_image.png and Left_image.png. The lower the difference the better the alignment (in theory). Is this a good approach? Is there a better process for doing this? Thanks Quote Link to comment
JKSH Posted January 29, 2017 Report Share Posted January 29, 2017 On 1/28/2017 at 0:52 PM, Tico_tech said: 1. I compared the distance from bottom to first edge; shown in Right_image.png and Left_image.png. The lower the difference the better the alignment (in theory). Is this a good approach? Is there a better process for doing this? That's a simple but valid approach, assuming that both sensors are very close together and face the same direction at all times. The technical term for alignment is "image registration" or "image fusion". The Advanced Signal Processing toolkit contains an example for doing image fusion -- do you have it installed? 1 Quote Link to comment
Recommended Posts
Join the conversation
You can post now and register later. If you have an account, sign in now to post with your account.