background image
ISO/IEC 10918-1 : 1993(E)
F.2.1.3.1
Decoding model for DC coefficients
The decoded difference, DIFF, is added to PRED, the DC value from the most recently decoded 8
8 block from the
same component. Thus ZZ(0)
=
PRED
+
DIFF.
At the beginning of the scan and at the beginning of each restart interval, the prediction for the DC coefficient is
initialized to zero.
F.2.1.3.2
Decoding model for AC coefficients
The AC coefficients are decoded in the order in which they occur in ZZ. When the EOB is decoded, all remaining
coefficients in ZZ are initialized to zero.
F.2.1.4
Dequantization of the quantized DCT coefficients
The dequantization of the quantized DCT coefficients as described in Annex A, is accomplished by multiplying each
quantized coefficient value by the quantization table value for that coefficient. The decoder shall be able to use up to four
quantization table destinations.
F.2.1.5
Inverse DCT (IDCT)
The mathematical definition of the IDCT is given in A.3.3.
After computation of the IDCT, the signed output samples are level-shifted, as described in Annex A, converting the
output to an unsigned representation. For 8-bit precision the level shift is performed by adding 128. For 12-bit precision
the level shift is performed by adding 2 048. If necessary, the output samples shall be clamped to stay within the range
appropriate for the precision (0 to 255 for 8-bit precision and 0 to 4 095 for 12-bit precision).
F.2.2
Baseline Huffman Decoding procedures
The baseline decoding procedure is for 8-bit sample precision. The decoder shall be capable of using up to two DC and
two AC Huffman tables within one scan.
F.2.2.1
Huffman decoding of DC coefficients
The decoding procedure for the DC difference, DIFF, is:
T
=
DECODE
DIFF
=
RECEIVE(T)
DIFF
=
EXTEND(DIFF,T)
where DECODE is a procedure which returns the 8-bit value associated with the next Huffman code in the compressed
image data (see F.2.2.3) and RECEIVE(T) is a procedure which places the next T bits of the serial bit string into the low
order bits of DIFF, MSB first. If T is zero, DIFF is set to zero. EXTEND is a procedure which converts the partially
decoded DIFF value of precision T to the full precision difference. EXTEND is shown in Figure F.12.
104
CCITT Rec. T.81 (1992 E)
[1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72] [73] [74] [75] [76] [77] [78] [79] [80] [81] [82] [83] [84] [85] [86] [87] [88] [89] [90] [91] [92] [93] [94] [95] [96] [97] [98] [99] [100] [101] [102] [103] [104] [105] [106] [107] [108] [109] [110] [111] [112] [113] [114] [115] [116] [117] [118] [119] [120] [121] [122] [123] [124] [125] [126] [127] [128] [129] [130] [131] [132] [133] [134] [135] [136] [137] [138] [139] [140] [141] [142] [143] [144] [145] [146] [147] [148] [149] [150] [151] [152] [153] [154] [155] [156] [157] [158] [159] [160] [161] [162] [163] [164] [165] [166] [167] [168] [169] [170] [171] [172] [173] [174] [175] [176] [177] [178] [179] [180] [181] [182] [183] [184] [185] [186]