I have a big set of images where I want to measure the size of an object in each image, and if possible obtain the overall shape by extracting the object boundary. I tried to convert the RGB image into binary and also tried to play around with greyscales and other RGB combinations to extract the object but the background (a grid) interferes with the boundary pixels of the object. How can I ignore the background or obtain what I want? Attached I upload two images in which one has the object I want (a pebble) in grey and other in pink. Thank you in advanced.