The final size on disk will depend on the cluster / sector size of the underlining filesystem used by the storage device (actually, the partition) the file is in. It is rounded up to the upper nearest value multiple of the sector size of your storage (for NTFS it is 4KB by default, but can be higher / smaller).
Therefore, a single byte extra could mean a file taking up multiple kilbytes of your storage. In fact, that's the very cause for the problem of "slack space" on storage media, that is, the file wasted by unused space on the last sector of each file, a problem that is very noticieable when there are lots of really small files on a partition with a big sector size.
As to your question, there is no definite answer, and I believe it is a non-issue, unless you really need image files with such a precision in file size where a 4KB difference is significative.
possible duplicate of Batch "resize" JPEG images to a fixed file size
– Excellll – 2015-06-18T14:45:31.510@ethanjyx: This is an old post that just resurfaced because a possible duplicate question was suggested that links to more answers. Two questions. 1st, can you post a comment on the current status of the problem (did you find a solution & if so, what was it? are you still looking for an answer or has this been overtaken by events?)? 2nd, once questions are posted, they take on a life of their own as they help others with a similar problem. Can you clarify whether your goal was a precise file size or a maximum file size? Compression can be hard to predict so most options target a max size. – fixer1234 – 2015-06-19T12:51:53.093
Why do you need image files with such a precision in file size where a 4KB difference is so relevant? What are you trying to accomplish? – That Brazilian Guy – 2013-12-14T00:51:34.783