| 摘要 |
PURPOSE:To obtain a picture of excellent quality inexpensively with less number of bits by halving a step width of gradation excluding the most significant bit (MSB) in a part having a small input signal change and quantizing the result, adding the stored MSB to the input signal to apply the demodulation at a half gradation thereby doubling the resolution at a part with a smaller signal change. CONSTITUTION:When an equivalent detection circuit 3 in a quantization section 1 detects same adjacent quantized value with less analog input change, a latch circuit 6 is set and a switching signal HS is outputted. Then a variable gradation AD converter 5 halves the step width of the gradation and applies AD conversion to an upper half or a lower half of the dynamic range with an accuracy of a half the gradation step width according to the content of the MSB latch 7 representing the state of the MSB just before in the accuracy being a half of the gradation step width. When a part having a larger analog input change comes, a difference detection circuit 4 is operated to reset the latch circuit 6 and the MSB latch 7 via an OR gate 10 to apply the conventional AD conversion. In receiving adjacent signals of the same equivalence, a variable gradation DA converter 13 in reference to the quantization in the demodulation section 2 adds the MSB information just before stored in the MSB latch 7 to the input signal and the result is subject to DA conversion with an accuracy of half gradation.
|
