Lossless medical image compression

摘要: Lossless coding is widely applied in medical image compression because of its feasibility. This thesis offers two major contributions for lossless compression; (i) the relationship between minimum-mean-squared-error (MMSE) and minimum-entropy-of-error (MEE) prediction has been revealed, (ii) novel methods improving rates operation using Shape-Vector Quantization (VQ) have presented. These new schemes a simpler implementation, more computational efficiency, lower memory requirement than other have. The proposed are capable providing significant improvement over traditional predictive coders adaptive coders. One goal any pursuit to minimize MEE. Realizing this valuable terms performance minimizing MMSE. Most techniques, however, centered on MMSE MEE limitation linear backbone Shape-VQ-based introduced thesis. concepts presented detail analyzed mathematically It shown that one conditions reaching minimum entropy local stationarity, which makes feasible. explains reason behind effectiveness where rather Predictive techniques well accepted coding. main advantages technique simplicity encoder