Compound Image Segmentation and Compression for Secure Transmission

Increase in the use of images in the continuous expansion of multimedia applications has increased the demand for efficient techniques that can store and transmit visual information. This demand has made image compression a vital factor. This work proposes compression techniques for compound images. For this purpose, the primary goal was set as "to develop compression schemes that can compress color compound images quickly, which can reduce the compression rate, while maintaining a good level of visual quality in a secure fashion". Here two types of segmentation techniques (i) layer-based segmentation and (ii) block-based segmentation are analyzed and two hybrid models are proposed. The first hybrid model combines histogram-threshold block-based technique with wavelets and the second model combines layer and block-based techniques for efficient compression. Security to image content is provided in three ways, namely, Block Transformation Coding (BTC), XML transformation and conversion for text block and Encryption of image blocks. Experiments are conducted with the traditional compressors which prove that the proposed models give better results.