From the Publisher:
This is the first comprehensive treatment of feed-forward neural networks from the perspective of statistical pattern recognition. After introducing the basic concepts, the book examines techniques for modelling probability density functions and the properties and merits of the multi-layer perceptron and radial basis function network models. Also covered are various forms of error functions, principal algorithms for error function minimalization, learning and generalization in neural networks, and Bayesian techniques and their applications. Designed as a text, with over 100 exercises, this fully up-to-date work will benefit anyone involved in the fields of neural computation and pattern recognition.

/pdf/neural-networks-for-pattern-recognition-34puigiau1.pdf

Neural networks for pattern recognition

The core component of most modern trackers is a discriminative classifier, tasked with distinguishing between the target and the surrounding environment. To cope with natural image changes, this classifier is typically trained with translated and scaled sample patches. Such sets of samples are riddled with redundancies—any overlapping pixels are constrained to be the same. Based on this simple observation, we propose an analytic model for datasets of thousands of translated patches. By showing that the resulting data matrix is circulant, we can diagonalize it with the discrete Fourier transform, reducing both storage and computation by several orders of magnitude. Interestingly, for linear regression our formulation is equivalent to a correlation filter, used by some of the fastest competitive trackers. For kernel regression, however, we derive a new kernelized correlation filter (KCF), that unlike other kernel algorithms has the exact same complexity as its linear counterpart. Building on it, we also propose a fast multi-channel extension of linear correlation filters, via a linear kernel, which we call dual correlation filter (DCF). Both KCF and DCF outperform top-ranking trackers such as Struck or TLD on a 50 videos benchmark, despite running at hundreds of frames-per-second, and being implemented in a few lines of code (Algorithm 1). To encourage further developments, our tracking framework was made open-source.

High-Speed Tracking with Kernelized Correlation Filters

We present a neural network-based upright frontal face detection system. A retinally connected neural network examines small windows of an image and decides whether each window contains a face. The system arbitrates between multiple networks to improve performance over a single network. We present a straightforward procedure for aligning positive face examples for training. To collect negative examples, we use a bootstrap algorithm, which adds false detections into the training set as training progresses. This eliminates the difficult task of manually selecting nonface training examples, which must be chosen to span the entire space of nonface images. Simple heuristics, such as using the fact that faces rarely overlap in images, can further improve the accuracy. Comparisons with several other state-of-the-art face detection systems are presented, showing that our system has comparable performance in terms of detection and false-positive rates.

Neural network-based face detection

A major new professional reference work on fingerprint security systems and technology from leading international researchers in the field Handbook provides authoritative and comprehensive coverage of all major topics, concepts, and methods for fingerprint security systems This unique reference work is an absolutely essential resource for all biometric security professionals, researchers, and systems administrators

/pdf/handbook-of-fingerprint-recognition-16mh887j24.pdf

Handbook of Fingerprint Recognition

http://romisatriawahono.net/lecture/rm/survey/machine%20learning/Chandrashekar%20-%20Feature%20Selection%20Methods%20-%202014.pdf

A survey on feature selection methods

Several recent advances are described that use neural-network methods to produce the higher-order decision surface required for difficult pattern recognition discrimination problems. Work at Carnegie Mellon University is emphasized and includes new hyperspherical Ho-Kashyap neural nets and new piecewise quadratic neural nets. Also addressed are Fourier neural-net interconnections to handle multiple objects and achieve morphological, image processing, and enhancement functions. >

http://ieeexplore.ieee.org/iel2/632/5901/00227039.pdf

Higher-order decision surfaces in neural nets

Various papers on hybrid image and signal processing are presented. The topics addressed include: optical matrix-vector laboratory data for finite element problems, solving ill-posed algebra problems using the bimodal optical computer, robust texture extractors for real-time pyramidal architectures, modified algorithm for scanning tomographic acoustic microscopy, identifying metal surfaces in color images, interpolative vector quantization, and moving image signal processing by Markovian random walk approach. Also discussed are: parallel optical pyramidal image processing, optical laboratory realization of distortion invariant filters, hybrid methods to compute image moments, hybrid techniques for postal address location, automatic visual inspection system for IC bonding wires, image processing on hypercube multiprocessors, interactive three-dimensional display and analysis, flexible mask subtraction for digital topography, least squares filtering via systolic array and light scattering for testing roughness of formed surfaces.

Hybrid image and signal processing III; Proceedings of the Meeting, Orlando, FL, Apr. 23, 24, 1992

Three general optical approaches to multiple degree of freedom object pattern recognition (where no stable object rest position exists) are advanced. These techniques include: feature extraction, correlation, and artificial intelligence. The details of the various processors are advanced together with initial results.

/pdf/multiple-degree-of-freedom-optical-pattern-recognition-1z9zhcaqcv.pdf

Multiple degree of freedom optical pattern recognition

A new algorithm based on scale-space median and gabor filtering is used to find boundaries in electron microscopy images under noise and low contrast. Boundary information from different scales are fused to find triple junctions and dihedral angles that are of use in material science. The proposed algorithm has been tested on various electron microscopy images and has been shown to be robust to changing imaging conditions, noise and contrast.

Scale-space median and Gabor filtering and fuzzy unification for boundary detection electron microscopy images.

Three aspects of image quality and their effects in an optical matched spatial filter correlator are described. These concern operation on: digitized imagery, data with low modulation and low space bandwidth product, and synthetic reference imagery. To address these practical problems, we employ: spatial filtering, edge enhancement, use of different apertures, photoreduced imagery, and digital preprocessed data. The first exuerimental data on optical matched spatial filter correlations with synthetic reference imagery and the first comparative data on digitally and optically processed multisensor image correlations are included. From these experiments, we find an optical weighted matched spatial filter correlator to be adequate for most multisensor data and that advanced digital preprocessing operators are necessary when presently available synthetic reference imagery is used.

David P. Casasent

Papers

Higher-order decision surfaces in neural nets

Hybrid image and signal processing III; Proceedings of the Meeting, Orlando, FL, Apr. 23, 24, 1992

Multiple degree of freedom optical pattern recognition

Scale-space median and Gabor filtering and fuzzy unification for boundary detection electron microscopy images.

Image Quality Effects In Optical Correlators