| 摘要 |
<p>PURPOSE:To improve the fidelity and to prevent bit error, by grouping picture elements constituting a picture into a prescribed number and representing the group with a predetermined pattern code and a gradation code of a specific picture element in the group. CONSTITUTION:Picture elements of a picture to be transmitted are taken as one group at each 4-bit. As to the first picture element of each group, the contrast is represented in multiple gradations and the pattern of the change in the contrast of each group is represented as a pattern code in succeeding 3 bits. The pattern code is introduced from the gradation in 4-bit at each picture element in the group of a DFF. That is, a signal B from a DFF23 is multiplied by -2 at a negative doubler 25, summed at an adder 28 with a signal A from an FF24 and with a signal C from an FF22 at an adder 29, compared with a threshold value (k) at a discriminator 34 and one element of the code is determined.</p> |
