A High Capacity Steganographic Algorithm in Color Images

In this paper, we propose a steganographic method which is conducted on the quantized discrete cosine transform (DCT) coefficients of YUV. Three coefficients at the same frequency in YUV components are selected respectively to compose a triplet as the message carrier, then decide the embedding capacity in each triplet according to the property of the triplet. The modulus 2 or modulus 4 arithmetic operations are applied to the valid triplet to embed I bit or 2bits secret messages. When modification is required for data embedding, the random number is produced to decide the flipping direction and the improved shortest route modification scheme (SRM) is applied to reduce distortion. To compare with other existing methods, the proposed method has high embedding capacity while approximately preserving the histogram of quantized DCT coefficients, and the spatial blockiness of stego image has no obvious changes, so the method can resist against histogram-based and blockiness-based steganalysis.