Abstract:
A method for texture compressing images having a plurality of color components (R, G, B) includes the step of defining color representatives for use in encoding by defining groups of colors for each said color component (R,G,B), and selecting for each said group a representative color for the group, the median color being chosen as the representative color of the group. Each said group is preferably composed of 3 to 15 colors and the method includes the step of computing, for each group, an error between each member of the group and said representative color of the group. Typically, the error is computed as the sum of the absolute differences (SAD) between each member of the group and said representative color of the group. In order to select each said group and then extract therefrom said representative color, a criterium is used selected from the group consisting of:
selecting the group that minimizes said error by assuming each group comprised of the lower colors sorted in ascending order, wherein the same applies for the groups comprised of the higher colors, accruing the error as computed separately for two groups in all possible combinations and finding the minimum of the composite error.
Abstract:
A digital AM demodulator, particularly for demodulating a signal (x) originating from a tuner, comprising:
means (13) for generating a first carrier (ω a t) which is not correlated with the input signal (x) to be demodulated; first means (3) for multiplying the first carrier (ω a t) by the input signal (x) to be demodulated; filtering means (1, 2), arranged upstream and downstream of said multiplier means (3) and adapted to eliminate unwanted spectral repeats; the particularity of which is the fact that it comprises means (15) for detecting the phase shift between the pulse (ω i ) of the input signal (x) to be demodulated and a local carrier (ω o t) and means (18) for correlating the first carrier with the input signal, the first carrier and the local carrier being mutually correlated, the local carrier not being correlated with the input signal to be demodulated.
Abstract:
A digital demodulator for angle-modulated (PM, FM) signals comprises:
an input module (1, 2) for generating, starting from the modulated signal (s fm ), a first component (cosφ(n)) and a second component (sinφ(n)) in quadrature with respect to one another, said components in quadrature being representative of the real part and of the imaginary part of the modulating signal (s'(t); s"(t)) represented as a complex variable; and a processing module (4) able to generate an output signal, calculating the arctangent function of the ratio between said imaginary part and said real part, the output signal identifying the demodulated signal (s'(t); s"(t)).
The processing module (4) acts on the components in quadrature either directly, in the case of phase demodulation, or indirectly, in the case of frequency demodulation. In the latter case, the above-mentioned real component and imaginary component are calculated, respectively, as the sum of the products of each of the components in quadrature with a corresponding delayed version, and as the difference of the cross products of each of the two components in quadrature with a version delayed in time of the other component in quadrature. The demodulator may be implemented either as a dedicated hardware unit, or else resorting to a computer program product for DSP or microprocessor.