How to Align two images?
Hi!
Hope you're all doing well! I am working on a project two fuse two images (RGB + IR). I am using FLIR dataset. RGB image resolution is 1800x1600, whereas IR images are 640x512. Additionally, RGB images have slightly wider angle and has more info, so when the images go through image fusion, it produces shadowy output. I am looking to crop the images, so the details match infrared images. I came across a medium article where the author used feature matching to identify and crop parts. It works great for daylight scenes, however, it produces very wrong results when night scenes images are used. I notice the feature matching is not correct either. I have tried other methods (SURF, SIFT as well, but they don't seem to work well either, unless I am missing something? I am new to CV).
1) Do you know what could be causing this issue to occur? 2) Is there any other straight forward method I can use to crop RGB images? Because the position of rgb camera is fixed, so maybe cropping by certain factor or identifying height, width points can be used to crop the images?
Cropped Image Daylight (1) - https://imgur.com/0r6gQkX
Cropped Image Nightscene (2) - https://imgur.com/PTlFCue
Best Regards
Results Compared (night scene matching is not accurate)
Daylight Scene matches and cropped properly.
Original Image Night Scene C:\fakepath\apicture.jpg
Original image of day scene
can you put your images here, please, noton an external site ?
(imgur is also not visible / blocked from several countries)
added pictures. Thanks!
the night scene RGB picture is simply of very poor quality. it's blurry and noisy, and the scene is mostly dark. discard that pair of images. it's not worth spending time on that.
maybe... post the original pictures of the night scene. it seems as if you just posted the output of the failed execution.
Alright! I updated the original night scene image as an attachment. It wasn't displaying the image with image function in the forum. Thanks!
Anyone got any suggestions?
For what it's worth, I was able to find fixed coordinates that makes two images align, but it seems they change throughout the dataset, so I don't think I can batch process them.