Abstract:
Everyday an enormous amount of information is stored, processed, and
transmitted. Because much of this information is graphical or pictorial in nature, the
storage and communications requirements are immense. Though in recent time’s
bandwidth capacities got much higher and cost of mass storage space got lower but
still a lot of problems are faced during transmitting and storing images. Image
compression plays vital role in terms of saving storage space and reduction of
transmission time. Wavelet transform is considered as landmark in the field of image
compression due to the feature that it represents a signal in terms of functions those
are localized in both frequency and time domain, as not in case of other
Transformation techniques. Various techniques have been explored by different
authors to employ wavelet transform for image compression e.g. EZW, SPIHT etc.
The idea of any scheme is to remove the correlation present in the data. Tensor
product orthogonal wavelet bases are unable to adapt towards directional geometric
features.
The purpose of this work is to develop an algorithm that exploits spatial
correlation between pixel values and then compresses the image using wavelet
transform. Images are connected regions of similar texture and intensity levels that
combine to form objects. Typically, magnitude of pixels relate very closely to each
other, thereby having less difference between them. Main idea is what already
suggested in chain codes but contrary to it encoding is with respect to direction.
Applying such a technique reduces entropy of geometrical features.
