Adaptive Image Steganalysis for LSB Embedding Technique using Enhanced Canny Operator

Objectives: Adaptive image steg analysis retrieves concealed content from the adaptable regions of cover image. To identify adaptive regions, Enhanced canny operator is used and which identifies the false edges accurately. Method/Analysis: Adaptive image steganography is the method of hiding the content, based on the adaptable regions of the colour image. The edges in the cover image are used for hiding the secret information by considering two LSB (Least Significant Bit) bits. In the existing method, canny edge detectors were used to extract the features of the image but it fails to identify the false edges and smoothes the boundaries with noise. Findings: In the proposed method, Adaptive regions are identified using enhanced canny operator which identifies the false edges accurately and thus reduces the overhead in payload location identification and content retrieval. This enhanced canny operator outperforms the other edge detectors for the retrieval of content which are embedded using LSB embedding method during steganography. The performance of the operator is measured using Positive Predictive Value (Precision).The precision is calculated after identifying the adaptive region with its payload location and hidden content using ensemble classifier. Applications/Improvements: The performance of the method can be improved by using different classifier combinations as ensemble classifier for multi class classification.