BibTeX:

@article{IJIRSTV3I3069,
     title={Detection of Double JPEG Compression on Color Image using Neural Network Classifier},
     author={Geethu Varghese and Dr. Arun Kumar M N},
     journal={International Journal for Innovative Research in Science & Technology},
     volume={3},
     number={3},
     pages={175--181},
     year={},
     url={http://www.ijirst.org/articles/IJIRSTV3I3069.pdf},
     publisher={IJIRST (International Journal for Innovative Research in Science & Technology)},
}

Identification of double JPEG compression with same quantization lattice is a testing issue. Recognition of double JPEG compression assumes an essential part in computerized picture legal sciences. Some effective approaches have been proposed to identify double JPEG compression when primary and secondary compression have diverse quantization matrix. Nonetheless, distinguishing double JPEG compression with same quantization matrix is still a testing issue. Here, a powerful error based statistical feature extraction scheme is introduced to tackle this issue. Initial, a given JPEG file is decompressed to frame a reconstructed image. An error image is obtained by processing the contrasts between the inverse discrete cosine transform coefficients and pixel values in the reconstructed image. Two classes of blocks in the error image, in particular, rounding error block and truncation error block, are analyzed. Then a group of features are proposed to differentiate between single and double JPEG images. At last artificial neural network classifier is used to identify whether a given JPEG image is double compressed or not. The process is implemented in a MATLAB 2014.

Computerized Picture, Legal Sciences, Rounding Error, Truncation Error