14-12-2012, 06:31 PM
Project report BLOCK TRUNCATION IMAGE CODER IMPLEMENTATION ON TMS320C6713.
1TRUNCATION IMAGE CODER.doc (Size: 75.5 KB / Downloads: 23)
ABSTRACT
Block truncation coding (BTC) is a successful image compression technique due to its simple and fast computational burden. The bit rate is fixed to 2.0 bits/pixel, whose performance is moderate in terms of compression ratio compared to other compression schemes such as discrete cosine transform (DCT), vector quantization (VQ), wavelet transform coding (WTC), etc. Two kinds of overheads are required for BTC coding: bit plane and quantization values, respectively. A new technique is presented to reduce the bit plane overhead. Conventional bit plane overhead is 1.0 bits/pixel; we decrease it to0.734 bits/pixel while maintaining the same decoded quality as absolute moment BTC (AMBTC) does for the "Lena" image. Compared to other published bit plane coding strategies, the proposed method outperforms all of the existing methods.
Block Truncation Coding (BTC) is a lossy image compression. In the simplest possible terms: BTC is a block-adaptive binary encoder scheme based on moment preserving quantization. The standard block truncation coding (BTC) technique is a simple block-based image compression, which preserves the block mean and the block standard deviation.
The basic BTC algorithm is a lossy fixed length compression method that uses a Q level quantiser to quantize a local region of the image. The quantiser levels are chosen such that a number of the moments of a local region in the image are preserved in the quantized output. In its simplest form, the objective of BTC is to preserve the sample mean and sample standard deviation of a grayscale image. Additional constraints can be added to preserve higher order moments. For this reason BTC is a block adaptive moment preserving quantiser.