Study On Lossy Compression For Halftoning Image

Halftoning image is widely used for printing and publishing. In order to save storage capacity and transmission time, and to achieve rapid transmission and immediate processing, it is necessary to compress, which can be realized by reducing the amount of data, and the quality of decoding image is good.The conventional binary image coding techniques does not consider the property of halftoning image. Vector Quantization is on the basis of block encoding, so it can be used for ordered-dither image, which has periodic clumps texture. An initial codebook algorithm that can distribute uniformly in training set and combine with lossless compress method is proposed. To overcome disadvantages of existing initial codebook algorithms, the initial codebook with uniform distribution is produced, and iteratively produce final codebook using Linde-Buzo-Gray(LBG). The experimental results show that new method can get better visual effect.The detail of halftoning image is located in medium and high frequency band, which will be discarded if directly using the wavelet transformation. According to texture of the halftone image, A new wavelet packet halftone image compression is proposed, in which the scan scheme of the wavelet packet coefficients and the selection of threshold function are improved. The experimental results show that new method can get better effect.