35results about How to "Guaranteed subjective quality" patented technology

The invention discloses a variable-bit-rate encoder based on AMR (adaptive multi-rate)-NB (narrow band) voice signals. The variable-bit-rate encoder comprises a preprocessing unit, a voice frame quality judging unit, an encoding mode selecting unit, a bit rate determining unit and a code-excited linear predictive encoding unit, wherein the preprocessing unit is used for numeralization of voice signals to form voice frames; the voice frame quality judging unit is used for judging quality grades of the current voice frames to provide respective encoding modes and target bit rates for the voice frames; the encoding mode selecting unit is used for selecting the voice frame encoding modes according to the quality grades; the bit rate determining unit is used for determining the target bit rates of the voice frames according to the encoding modes; the code-excited linear predictive encoding unit is used for encoding the voice frames according to the target bit rates of the voice frames to form encoded voice frames. The invention further discloses a variable-bit-rate decoder used correspondingly to the encoder, and discloses a variable-bit-rate encoding method and a variable-bit-rate decoding method. The variable-bit-rate encoder is lower in bit rate than AMR, capable of achieving variable bit rates according to voice frame content and capable of selecting required encoding bit rate modes according to voice frame content importance judgment by setting channel voice quality.

The invention provides a monitoring video background image modeling method, which adopts a block boundary detection method, image residual gradient calculation and self-adaptive training set length adjustment method to ensure excellent subjective quality. The background image modeling method of the present invention is block-based, stores YUV three components at the same time, and the generated background image ensures high-quality consistency of chromaticity and brightness. By using the modeled background image as a global reference image for video encoding, the bit rate can be significantly saved, and a background image for global encoding reference can be established while maintaining subjective quality.

The invention discloses a visual perceptual coding method based on a multi-domain JND (Just Noticeable Difference) model and relates to video information processing. The method comprises the following steps: respectively calculating a space-domain basic JND threshold value, a luminance masking modulation factor, a contrast masking modulation factor and a time domain masking modulation factor of each transformation coefficient in a DCT (Discrete Cosine Transform) block by utilizing a time-space-domain multi-domain JND model so as to obtain the time-space-domain multi-domain JND threshold value of each transformation coefficient; introducing a block perception-based distortion probability evaluative criteria in the transform coding process, and searching a correction factor of each coefficient relative to the JND threshold value through an adaptive searching algorithm so as to obtain a transformation coefficient suppression value; and finally, subtracting the original transformation coefficient from the most appropriate suppression value obtained through corresponding calculation, and taking the coefficient as a novel coefficient to be put at an entropy coding stage. According to the coded suppression strategy of the multi-domain JND model and the block perception-based distortion probability, the coding rate can be effectively reduced on the premise of guaranteeing certain subjective quality, and the compression ratio of the current coding standard is further improved.

The invention discloses a method and a device for coding a video based on region of interest (ROI) and just-noticeable difference (JND). The method comprises the following steps that whether the current macro block for frame image processing belongs to the ROI or not is judged; if the current macro block for frame image processing does not belong to an ROI, a transform domain JND model of the current macro block is built, and based on the transform domain JND model, forecasted residual error self-adjustment is conducted on the current macro block. The ROI and non-ROI in a frame image are separately processed, namely, conventional coding processing is conducted on macro blocks belonging to the ROI, and coding processing is conducted on macro blocks belonging to the non-ROI in the coding mode that forecasted residual error self-adjustment is carried out through transform domain JND; therefore, subjective coding quality is guaranteed, and coding rate is reduced.