| 摘要 |
A method for encoding a sequence of data values comprises: generating, from the input data values, respective complementary most-significant data portions and least-significant data portions, such that the most-significant data portion of a value represents a plurality of most significant bits (MSB) of that value, and the least-significant data portion represents the remaining least significant bits (LSB) of that value; generating one or more data sets indicative of positions, relative to an array of values, of most-significant data portions of predetermined magnitudes; encoding the data sets to an output data stream using binary encoding; and including data defining the less-significant portions in the output data stream. The data set may be a significance map, indicating positions of non-zero most-significant portions, the last of which may be indicated by a data flag. Also disclosed is an image data encoding method comprising frequency-transforming input data, generating a frequency-transformed coefficient array by a matrix multiplication process, according to a maximum dynamic range and using transform matrices having a data precision; maximum dynamic range and transform matrix data precision are selected according to image data bit depth. |
