Nader Mohsenian

Single-pass constant- and variable-bit-rate MPEG-2 video compression

Most real-time MPEG-2 encoders are designed to perform in a constant-bit-rate (CBR) mode, in which buffer constraints are imposed to circumvent large deviations from a desired rate at any instant in time. Although such streams are generally good-quality sequences, certain types of operations or environments call for a more efficient real-time CBR encoder. The first part of the paper describes how a better-quality CBR video stream can be produced by estimating the relative complexity of a picture in comparison with the average complexity of the partially encoded stream and using it to adjust the compression parameters in a single-pass mode of operation. Our CBR encoder is particularly attractive for digital broadcast and editing environments, in which representations of higher-fidelity video objects in both display and freeze modes are constantly pursued. The second part of the paper describes the real-time generation of video streams with a variable-bit-rate (VBR) encoder. This mode of operation is highly desirable for home entertainment and recreational events. We propose a robust single-pass VBR video encoder algorithm which is capable of learning and adapting itself to the complexity of image segments and thereafter creating streams which have constant visual picture quality. The new VBR scheme displays a better performance than the CBR encoder, particularly when special effects such as scene transitions, fades, or luminance changes are to be compressed. Both CBR and VBR encoders are fully compliant with the MPEG-2 standard and are easily implementable with IBM encoder architecture. Compression results for the new single-pass encoding algorithms and comparisons with previous CBR schemes are provided. The result suggests the suitability of our VBR approach for record/playback in storage media such as digital video disc (DVD) players, disk-based camcorders, and digital videocassette recorders (DVCRs). It further reflects the importance of our single-pass CBR scheme for providers of broadcast services, for which it allows more video programs to be allocated to a selected communication link, and for in-studio applications, for which it greatly facilitates visual analysis of captured streams.

Interframe coding of magnetic resonance images

Presents a new interframe coding method for medical images, in particular magnetic resonance (MR) images. Until now, attempts in using interframe redundancies for coding MR images have been unsuccessful. The authors believe that the main reason for this is twofold: unsuitable interframe estimation models and the thermal noise inherent in magnetic resonance imaging (MRI). The interframe model used here is a continuous affine mapping based on (and optimized by) deforming triangles. The inherent noise of MRI is dealt with by using a median filter within the estimation loop. The residue frames are quantized with a zero-tree wavelet coder, which includes arithmetic entropy coding. This particular method of quantization allows for progressive transmission, which aside from avoiding buffer control problems is very attractive in medical imaging applications.

Predictive vector quantization using a neural network approach

A new predictive vector quantization (PVQ) technique capable of exploring the nonlinear dependencies in addition to the linear dependencies that exist between adjacent blocks (vectors) of pixels is introduced. The two components of the PVQ scheme, the vector predictor and the vector quantizer, are implemented by two different classes of neural networks. A multilayer perceptron is used for the predictive cornponent and Kohonen self-organizing feature maps are used to design the codebook for the vector quantizer. The multilayer perceptron uses the nonlinearity condition associated with its processing units to perform a nonlinear vector prediction. The second component of the PVQ scheme vector quantizes the residual vector that is formed by subtracting the output of the perceptron from the original input vector. The joint-optimization task of designing the two components of the PVQ scheme is also achieved. Simulation results are presented for still images with high visual quality.

Edge-based subband VQ techniques for images and video

A key issue in subband coding is the efficient compression of the less informative but perceptually important upper frequency bands of the decomposed image. A new approach capable of effectively encoding the upper-bands is described. An intra-band vector quantization (VQ) technique is employed for compression of the base-band while the upper frequency bands are encoded by a hierarchical inter-band VQ method. The proposed inter-band vector quantization scheme exploits the redundancies that exist between pels of significant perceptual importance across the upper frequency bands of the same resolution. Such pels are observed to be at or around the edge-locations, displaying similar discontinuity behavior across the subbands. Therefore, an edge-detector is applied on the reconstructed base-band to extract the edge locations, and as a result no overhead information is transmitted to identify the position of these pels in the upper-bands. Furthermore, a residual subband, being the difference between the original base-band of each layer and its encoded version, is incorporated in the proposed inter-band VQ model. Thus, the proposed interband VQ scheme is simply the vector quantization of pixels across the upper-bands together with their corresponding residual band. Compression results are presented for both digital images and video sequences which demonstrate high subjective visual qualities. >

Interslice coding of magnetic resonance images using deformable triangular patches

We present a new inter-frame coding for medical images, in particular magnetic resonance (MR) images. Until now, attempts in using inter-frame redundancies for coding MR images have been unsuccessful. We contend that the main reason for this is twofold: bad inter-frame estimation models and ignoring the thermal noise inherent in MRI. Our inter-frame model is a continuous affine mapping based on (and optimized by) deforming triangles. The inherent noise of MRI is dealt with by using a median filter within the estimation loop. Simulations demonstrate the viability of this algorithm.<<ETX>>

Predictive vector quantization using a neural network

Predictive vector quantization (PVQ) of images using two novel coding approaches is considered. The first scheme, namely, address-PVQ, exploits the inter-vector (block) dependencies by predicting the VQ address of the current block from the addresses of the previously encoded blocks. A three-layer perceptron was used as an address-predictor with the position of the residual address being encoded. The second scheme is a vector extension of a differential pulse code modulation (DPCM) system. It exploits the inter-vector dependencies by predicting the current block of pixels. The predictive phase utilizes a three-layer perceptron while the residual blocks are vector quantized using the Kohonen self-organizing feature maps (KSOFM) clustering algorithm. The joint-optimization problem for design of the two components of PVQ was also considered. Coding results are presented for monochrome images. The joint optimization procedure improved the peak signal-to-noise ratio result by more than 1 dB.<<ETX>>

An object-based approach to color subsampling

We propose an object-based approach to image subsampling in the chroma domain which outperforms the 4:2:0/4:2:2 chroma formats of MPEG-2 and other techniques proposed in the literature. This scheme is motivated by the fact that sharp color edges and detailed areas in the color domain represent significant visual information, and therefore, require better color definition. The object-based color subsampling approach is based on segmenting the C/sub B/ C/sub R/ components of an image into high and low-activity areas. The relatively large and less informative low activity regions are decimated to one pixel while high-activity objects are preserved entirely. A fuzzy filter is incorporated in our segmentation scheme to circumvent the effect of noise which could lead to isolated pixels. The object-based approach demonstrates superior results when measured perceptually. This scheme is particularly attractive as a preprocessing step to object-based coding schemes such as MPEG-4 where segmentation can be already available for content-based manipulation and generation of hybrid video data.

Scalar-vector quantization of medical images

A new coding scheme based on the scalar-vector quantizer (SVQ) is developed for compression of medical images. The SVQ is a fixed rate encoder and its rate-distortion performance is close to that of optimal entropy-constrained scalar quantizers (ECSQs) for memoryless sources. The use of a fixed-rate quantizer is expected to eliminate some of the complexity of using variable-length scalar quantizers. When transmission of images over noisy channels is considered, our coding scheme does not suffer from error propagation that is typical of coding schemes using variable-length codes. For a set of magnetic resonance (MR) images, coding results obtained from SVQ and ECSQ at low bit rates are indistinguishable. Furthermore, our encoded images are perceptually indistinguishable from the original when displayed on a monitor. This makes our SVQ-based coder an attractive compression scheme for picture archiving and communication systems (PACS). PACS are currently under study for use in an all-digital radiology environment in hospitals, where reliable transmission, storage, and high fidelity reconstruction of images are desired.

An interframe dynamic FSVQ codec for video sequence coding

An interframe vector quantization scheme with finite memory called dynamic finite-state vector quantization (DFSVQ) is presented. The encoding of a current input vector consists of a subcodebook, representing the best matchable code vectors, selected from a larger codebook, i.e., a supercodebook. The choice for the entries in the supercodebook is based on the information obtained from the previously encoded blocks where directional conditional block probability matrices are used in the selection of the entries. An adaptive DFSVQ scheme is also proposed in which, when encoding an input vector, first the supercodebook is searched for a matching code vector to satisfy a prespecified waveform distortion. If such a code vector is not found then the whole supercodebook is checked for a better match and a signaling flag along with the corresponding address of the code vector is transmitted to the receiver. The performances of the DFSVQ encoder and its adaptive version are evaluated for several video sequences.<<ETX>>

Color image compression using neural network prediction of color components

In this paper we present a new scheme for color image compression. The proposed scheme exploits the correlation between the basic color components (red, green, and blue: RGB) by predicting two color components given one color component. Specifically, this scheme employs neural network predictors to predict the red and blue color components using the encoded (reconstructed) green color component. The prediction error is further quantized using vector quantization. The performance of the proposed scheme is evaluated and compared with that of the JPEG.