We report the experimental demonstration of a photorefractive dynamic holographic memory that has a simple feedback architecture stabilized through a novel spatial sampling technique. By using dynamic copying to compensate for photorefractive decay, this memory provides nondestructive readout, selective erasure and rewriting , and real-time adaptation of… (More)
An optical network is described that is capable of recognizing at standard video rates the identity of faces for which it has been trained. The faces are presented under a wide variety of conditions to the system and the classification performance is measured. The system is trained by gradually adaptingphotorefrac tive holograms.
The stochastic signal model, hidden Markov model (HMM), is a probabilistic function of the Markov chain. In this letter, we propose a general nth-order HMM based dynamic texture description and classification method. Specifically, the pixel intensity sequence along time of a dynamic texture is modeled with a HMM that encodes the appearance information of… (More)
An anti-Hebbian local learning algorithm for two-layer optical neural networks is introduced. With this learning rule, the weight update for a certain connection depends only on the input and output of that connection and a global, scalar error signal. Therefore the backpropagation of error signals through the network, as required by the commonly used back… (More)
Aurora-A kinase functions mainly in centrosome maturation, separation and spindle formation. It has also been found to be amplified or overexpressed in a range of solid tumors, which is linked with tumor progression and poor prognosis. Importantly, Aurora-A inhibitors are being studied in a number of ongoing clinical trials. However, whether and how… (More)
Dynamic texture segmentation is to partition a video (image sequence) comprising of different dynamic texture into disjointed regions with uniformity and consistency property. This paper introduces a Markov random field based dynamic texture segmentation algorithm using the inter-scale context. The method models the label field and the observed field with… (More)
The opening positioning for railway oil tank wagon using machine vision technology is an effective way to load petroleum automatically. To overcome the adverse effects produced by illumination changes, various types of railway tank wagon and various positions of camera, a shape matching positioning method is proposed in this paper. Due to the shape of the… (More)
Dynamic texture is an extension of texture to the spatio-temporal domain. The motion property is one of the most important characteristics in dynamic texture analysis. This paper proposes a dynamic texture classification method based on the block-based motion estimation method and the statistical feature matrix. The method makes use of the block-based… (More)