An improved scheme for newspaper block segmentation and classification is described. The newspaper image is first segmented into blocks using three passes of a run-length smoothing algorithm. Blocks may have any shape and need not be non-overlapped rectangles. The height H between the top-line and base-line of lower case letters, and the number of pixels that have values differing from their four neighboring pixels, are measured for simple and reliable block classification. Blocks of different types are compressed based on their own characteristics. Unlike conventional methods, halftone image blocks are treated differently from black and white graphic blocks for better compression. A lossless compression scheme for halftoned images is proposed. Reconstruction of gray-tones from halftone images employing information of both smooth and edgy areas is presented.

Neural networks can be powerful tools for nonlinear signal processing and systems modeling. The authors present a class of discrete-time, neural-network-based, nonlinear models suitable for such applications in a systems identification framework. The parameters for these models include a local interconnection neighborhood size, a time constant characterizing the neurons, a weight matrix, and input output connection matrices. The prediction error identification method trains the net to solve two problems: the prediction of a chaotic time series generated by the logistic function, and the demodulation of FSK (frequency-shift keying) signals. The proposed approach allows a speedup of 1 to 2 orders of magnitude while preserving important properties of its continuous-time analog, and its speed permits global minima to be determined by simulated annealing. The model allows multichannel applications for control or prediction

A novel technique for setting up the Benes switch that has essentially optimal time complexity when implemented on a neural network is presented. This approach consists of defining distances between input or output ports and expressing the Banyan network's ability to realize a given permutation in terms of a distance matrix. This description allows a neural network implementation of the setup algorithm. It is shown that the neural network converges to a solution that will allow the switch to self-route the information according to the desired permutation. The time complexity of this operation is closer to optimal and the hardware simpler to implement than with previously presented algorithms.>

The problem of determining a shaping filter for nonstationary colored noise is considered. The shaping filter transforms white noise into a possibly nonstationary random process (having no white noise component) with a specified covariance function. A set of conditions to be satisfied by the covariance function leads to the determination of a shaping filter. The shaping filter coefficients are simply related to the solution of a matrix Riccati equation. In order to formulate the Riccati equation, basic results concerning the mean-square differentiability of a random process are developed. If the Riccati equation can not be defined, an autonomous (zero-input) shaping filter may be easily determined.