Misalignment correction and lossless compression of pixelated images

Abstract

Pixelated images are used to carry high data over optical wireless channels. The pixelated images may experience rotational misalignment error due to the lack of alignment between the transmitter and the receiver. Such images also need to be compressed for future offline processing. This thesis proposes a new technique for correction of rotational error for pixelated images. Unlike the existing misalignment correction algorithms which insert separate border or corner pixels, the proposed scheme estimates the border line and the corner points of the pixelated images. The desired imagesare then found by cropping or reforming the image pixels. In addition to the misalignment correction algorithm, this thesis proposes two image compression methods for pixelated image. The first one is termed as edge-based transformation and entropy coding (ETEC) and the second one as hierarchical prediction based transformation and entropy coding (PTEC). In the first stage of the proposed ETEC method, the intensity difference of neighboring pixels is calculated in the horizontal or vertical direction depending on the presence of a horizontal or vertical edge. In the second stage of the ETEC method, the intensity differences are used to form two matrixes – one containing the absolute intensity difference and the other having the polarity of differences. Next, Huffman or Arithmetic entropy coding is applied on the generated matrixes. In the first stage of the proposed PTEC method, hierarchical based prediction is done to predict the intensity difference of pixels. These differential intensity values are then mapped to two matrixes. In the second stage of the PTEC method, entropy coding is applied on the generated matrixes. Both ETEC and PTEC are compared to existing lossless compression techniques: Joint Photographic Experts Group Lossless (JPEG-LS), Set Partitioning in Hierarchical Trees (SPIHT) and Differential Pulse Code Modulator (DPCM). Simulation results show that the proposed ETECscheme can provide better compression compared to JPEG-LS and SPIHT algorithms for pixelated images, and superior compression than SPIHT for non-pixelated images. However, ETEC has higher computation time than SPIHT method. On the other hand, PTEC method has better and approximately near computation time compared to ETEC/DPCM and SPIHT, respectively. In addition, PTEC has better compression ratio compared to SPIHT and DPCM for both pixelated and standard images, and have better compression ratio to JPEG-LS only for pixelated images. So, for the case of pixelated images, ETEC is preferable to PTEC when compression ratio is an issue but PTEC is better in terms of computation time. However, for the case of standard non-pixelated images, PTEC is a better choice than ETEC.