Thepade's Hartley Error Vector Rotation For Codebook Generation In Vector Quantization

Vector quantization is simple and efficient technique of lossy image compression. Codebook generation has important role in Vector Quantization. The paper presents an effective clustering algorithm in Vector Quantization for codebook generation. Traditional Kekre's Error Vector Rotation (KEVR) resulting in slow cluster orientation. The Kekre's Error Vector Rotation using Walsh Sequence (KEVRW) gives rapid change in cluster orientation. The proposed codebook generation technique Thepade's Hartley Error Vector Rotation (THtEVR) is based on KEVR algorithm. Here Hartley Transform matrix is used to prepare error vector. The proposed THtEVR is tested for various codebook sizes on different training images. The results obtained indicated that THtEVR gives less MSE and less distortion as compared to KEVR and KEVRW giving better image compression.