Abstract: The present invention improves image quality by detecting and enhancing edges in an image. Images often include blurry or fuzzy edges that can obscure an image. An edge is a portion of an image separating two regions of substantially constant image intensity. An image can be examined on a pixel-by-pixel basis to find a candidate edge. When a candidate edge is found, a determination can made as to whether the candidate edge is a true edge. A true edge can be enhanced by amplifying the image intensity differences between pixels on the true edge and adjacent pixels not on the true edge. The present invention also provides a novel image processing filter for eliminating well-known noise from an image. The image processing filter can further improve edge enhancement by eliminating such noise prior to the edge detection and enhancement.

Abstract: A method and system can automatically select the optimal size of a visual object that can be displayed on an display device. For example, the present invention can determine the best or optimum size to display a visual object received from a computer network to a computer screen such that image degradations due to compression or transmission defects are less noticeable or perceivable. Typically, the best or optimum size to display a visual object can comprise a size that is also the maximum in which the visual object can be displayed without any degradations or defects.

Abstract: The present invention provides video signal compression that efficiently groups pictures in a video stream into variably-sized groups of pictures (GOPs), thereby providing lower achievable output signal bit rates and higher output signal quality. The video signal compression maximizes the output signal quality by appropriately allocating bits among individual pictures and GOPs in the output signal. The video signal compression of the present invention also applies compression methods that reduce noise in the output signal, by utilizing a macroblock-based tunable conditional replenishment technique. The conditional replenishment technique exploits the similarities among images in the variably-sized GOPs to further minimize output bit rate and maximize the output signal quality. An analysis-by-synthesis method is also provided to select a best asynchronous sampling method among various generated candidate output streams.

Abstract: The present invention improves image quality by detecting and enhancing edges in an image. Images often include blurry or fuzzy edges that can obscure an image. An edge is a portion of an image separating two regions of substantially constant image intensity. An image can be examined on a pixel-by-pixel basis to find a candidate edge. When a candidate edge is found, a determination can made as to whether the candidate edge is a true edge. A true edge can be enhanced by amplifying the image intensity differences between pixels on the true edge and adjacent pixels not on the true edge. The present invention also provides a novel image processing filter for eliminating well-known noise from an image. The image processing filter can further improve edge enhancement by eliminating such noise prior to the edge detection and enhancement.

Abstract: The device and method according to the present invention permute the order of signals representing information intended for transmission over a communication medium. The signal ordering is permuted prior to or at the time of transmission over the communication medium such that when the information or data is transmitted in the permuted order, it is more tolerant to data corrupting defects, such as a faulty communications line, or events, such as an electromagnetic pulse, that exist or may occur during the communication process.

Abstract: The device and method according to the present invention permute the order of signals representing information intended for storage in a memory. The signal ordering is permuted prior to or at the time of storage in the memory such that when the information or data is stored in the permuted order, it is more tolerant to errors or defects that existed or may occur in the memory.

Abstract: The apparatus and method presented use a localized model-based approach to compress video image signals for applications such as video teleconferencing. Preferably, the apparatus and method use three-dimensional vector quantization (3D VQ), recent localized history of the image signals, and a pyramidal hierarchial encoding technique to compress the image signals.

Abstract: Compression of signals is achieved through a simple decision of whether or not to encode certain frequency bands; not how well to encode all of the frequency bands. Based on the input signal and on a preselected perceptual model, a "just noticeable difference" (jnd) noise spectrum is computed. This spectrum is applied to a selector where it is used in the decision to select a chosen number of frequency bands of the input signal. The bands selected are the bands with the greatest energy relative to the jnd energy in the same band. Each of the selected bands is encoded and transmitted to the receiver. Both analog and digital realizations are presented.

Abstract: A system for and method of enhancing image/video signals to be decoded is disclosed. The system for and method of post-filtering uses, preferably, a temporal filter and a spatial filter, both of which are adaptive but neither is required to be adaptive. However, the system for and method of post-filtering may also be used with only an adaptive spatial post-filter. In this case, the performance is upper-bounded by the performance of systems using both the adaptive temporal and adaptive spatial post-filter.

Abstract: A method of tailored error protection coding is disclosed. An illustrative coding method according to the present invention concerns encoding a first signal and a second signal to provide error protection for the signals, wherein the first signal requires greater error protection than the second signal. According to the method, the first signal is encoded with a first error protection code. The encoded first signal is then combined with the second signal. The combination of the encoded first signal and the second signal is further encoded with a second error protection code to generate a combined error protected signal, wherein the first signal is coded with both the first and second error protection codes. A second embodiment concerns the use of separate coders which apply different error protection codes to two signals. The two encoded signals are then combined into a single signal.

Abstract: A geometric vector quantizer coding technique is illustrated in the context of a full motion video coder based on a three-dimensional sub-band framework. The original image data is decomposed into different spatial-temporal frequency bands and based on the data in each band, different quantization strategies are applied to the bands. The lowest spatial-temporal frequency band is coded using a traditional three-dimensional switched predictor and optimum scaler quantizer. The non-dominant sub-bands are coded using the vector quantization approach to efficiently encode the images while appropriately exploiting the sparse, highly structured nature of the data to design the codebooks. Variable resolution is achieved using these techniques and no training is needed to establish or maintain the codebook. A fast codebook search technique is provided for identifying the optimal codebook vector for representing a block of input data. Examples of two and three level geometric vector quantizers are also provided.

Abstract: A geometric vector quantizer coding technique is illustrated in the context of a full motion video coder based on a three-dimensional sub-band framework. The original image data is decomposed into different spatial-temporal frequency bands and based on the data in each, different quantization strategies are applied to the bands. The lowest spatial-temporal frequency band is coded using a traditional three-dimensional switched predictor and optimum scaler quantizer. The non-dominant sub-bands are coded using the vector quantization approach to efficiently encode the images while appropriately exploiting the sparse, highly structured nature of the data to design the codebooks. Variable resolution is achieved using these techniques and no training is needed to establish or maintain the codebook.

Abstract: A method for resisting the effects of channel noise in the digital transmission of information by means of the technique known as vector quantization, in which the codebook for binary index code assignment is generated by picking a vector quantized codeword with high probability and low perceptually-related distance from a required group of nearest neighbors, assigning that codeword and those neighbors binary index codes differing only in one bit, repeating the steps just outlined for assigned binary index codes to residual codewords until the lasts assignments must be made arbitrarily. Applications include transmission of speech by coded LPC parameters and transmission of image intensity or chrominance blocks for visual images. Such systems also make possible and are attractive for memory-efficient storage of such signals.

Abstract: A digital speech signal communication system includes a decoder that generates a predictively coded speech signal responsive to received digital codes and predicitve parameters corresponding thereto. A signal representative of the quality of said predictively coded speech signal is formed and the predictively coded speech signal is modified responsive to the quality representative signal and the predictive parameters, The quality representative signal may be generated by combining a signal corresponding to the step of each predictive code with the predictive parameters for the received code to form a signal representing an estimate of the noise in the predictive code.

Abstract: A digital communication system includes a predictive decoder that is operative to convert received digital codes into a predictively decoded signal, e.g., speech signal, and to generate a set of predictive parameter signals and a signal representative of the communication system bit rate. A set of filter control signals is generated responsive to the communication bit rate signal. The predictively decoded signal is passed through a filter which modifies the output signal responsive to the filter control signals and the decoder predictive parameter signals to improve the output signal quality. The filter control signals selectively alter the predictive parameter signals to optimize the predictively decoded signal modification for the current transmission bit rate.

Abstract: Subband signals are formed each corresponding to a portion of an input signal frequency spectrum. The subband signals are partitioned into subband time segments. The subband time segment signals are permuted. The permuted signals are combined to form a scrambled signal.

Abstract: A differential pulse code modulation system is made adaptive to quantization noise in a way which permits variable-rate operation as either a conventional R-bit embedded code or as a combination of a conventional R-bit code and a further explicit noise code with instantaneous quantization. Additional improvement in signal-to-noise ratio is attainable by using a noninstantaneous variable-bit allocation procedure for equal-length sampling blocks with an averaged fixed bit code.

Abstract: In a transmission system, reconstruction of a portion of the signal lost in transmission is based on signal encoding at the transmitter, and decoding at the receiver; at the transmitter, sequential portions of the signal are intermixed to form encoded portions, and autocorrelation factors (waveform feature signals) of the original portions are added to each encoded portion to form a packet; at the receiver, any missing packet is reconstructed from the adjacent received packet.

Abstract: An arrangement reduces transmission errors in a communication system that includes a transmitter operative to convert an input signal into blocks of coded signals and a receiver operative to convert the coded blocks into a replica of the input signal. In the transmitter, a signal representative of first features of the input signal and a signal corresponding to second features of the input signal are generated. In the receiver, corrupted received coded signals are detected responsive to the first feature signal and the received blocks of coded signals. Each corrupted received coded signal is modified jointly responsive to the received blocks of coded signals and the second feature signal to reduce errors in the input signal replica.