A new method for locating data hiding in image steganography

In the steganographic picture, it remains a challenging problem to determine the best place for inserting the hidden message with a minimum distortion of the host medium. However, there is a long way to go to select the right embedding position with less distortion. To accomplish this goal, we suggest a new high performance image steganography method in which extreme machine learning algorithms (ELM) are updated to build a supervised mathematical model. The ELM is initially checked in regression mode on a portion of the picture or host medium. This helped us to determine the best place to embed the message with the best values in the expected assessment measurements. For practicing on a new metric, contrast, homogeneity and other texture characteristics are used. In addition, the established ELM is used to tackle over fitting during workout. The findings are analyzed using the correlation, the structural similarity measure, fusion matrices and mean square error in the efficiency of the proposed steganography approach. In terms of imperceptibility, the adjusted ELM has been found to transcend current approaches. Excellent characteristics of the findings indicate that the proposed steganographic method is highly capable of retaining the visual image detail. In accordance with the current state of the art approaches, an increase of 28 per cent of imperceptibility is achieved.