FAKE IMAGE DETECTION USING MACHINE LEARNING

Computer Science Engineering

Image editing is now so widespread because to the availability of image processing tools like Adobe Photoshop or GIMP. Detecting such phone photos is unavoidable if image-based cybercrime is to be exposed. Because of its ubiquity, a photograph captured with a digital camera or smartphone is frequently saved in the JPEG format. The JPEG method works with 8x8 pixel picture grids that are compressed individually. Unmodified photos have a comparable amount of inaccuracy. Due to a comparable quantity of faults over the whole picture, each block should deteriorate at about the same pace during the resaving procedure. Error Level Analysis detects that the compression ratio of this false picture differs from that of the genuine image. Our paper's goal is to create a picture forensics programme that can detect any type of photo modification. The vertical and horizontal histograms of the error level analysis image were then used to determine the site of the alteration. The suggested method was able to recognize the changed picture while also displaying the specific position of the adjustments, according to the results.

This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.