Abstract. Many modern computer vision systems are built by chaining together standard vision procedures, often in graphical programming environments such as Khoros, CVIPtools or IUE. Typically, these procedures are selected and sequenced by an ad-hoc combination of programmer’s intuition and trial-and-error. This paper presents a theoretically sound method for constructing object recognition strategies by casting object recognition as a Markov Decision Problem (MDP). The result is a system called ADORE (Adaptive Object Recognition) that automatically learns object recognition control policies from training data. Experimental results are presented in which ADORE is trained to recognize five types of houses in aerial images, and where its performance can be (and is) compared to optimal. 1

This paper presents the Cameron Project, which aims to provide a high level, algorithmic language and optimizing compiler for the development of image processing applications on Reconfigurable Computing Systems (RCSs). SA-C, a single assignment variant of the C programming language, is designed to exploit both coarse-grain and fine-grain parallelism in image processing applications. Khoros, a software development environment commonly used for image processing, has been modified to support SA-C program development. SA-C supports image processing with true multidimensional arrays, and with sophisticated array access and windowing mechanisms. Reduction operators such as medians and histograms are also provided. The optimizing compiler targets RCSs, which are fine-grained parallel processors made up of Field Programmable Gate Arrays (FPGAs), memories and interconnection hardware. They can be used as inexpensive co-processors with conventional workstations or PCs. This paper discusses compiler optimizations to generate optimal FPGA code using dataflow analysis techniques applied to data dependence graphs. Initial results are presented.

High information redundancy and correlation in face images result in ineciencies when such images are used directly for recognition. In this paper, discrete cosine transforms are used to reduce image information redundancy because only a subset of the transform coecients are necessary to preserve the most important facial features such as hair outline, eyes and mouth. We demonstrate experimentally that when DCT coecients are fed into a backpropagation neural network for classi cation, a high recognition rate can be achieved by using a very small proportion of transform coecients. This makes DCT-based face recognition much faster than other approaches. Key words: Face recognition, neural networks, feature extraction, discrete cosine transform. 1 Introduction High information redundancy present in face images results in ineciencies when these images are used directly for recognition, identi cation and classi cation. Typically one builds a computational model to transform pixel i...

This paper presents Gamera, a new toolkit for the creation of domain-specific structured document recognition applications by domain experts with limited programming experience. The goal of the Gamera system is to leverage the user’s knowledge of the target documents to create custom applications rather than attempting to meet the needs of diverse users with a monolithic application. The system allows a knowledgeable user to combine image processing and recognition tools in an intuitive, interactive, graphical scripting environment based on Python. The use of Python in Gamera creates a simple yet powerful and flexible programming environment for novice programmers. Additionally, the resulting applications are suitable for a large-scale digitization project because they can be run in a batch-processing mode and easily integrated into a digitization framework. Finally, the Python module system has been extended to allow the easy creation of plugins using Python or C++. 1

In this paper, efficient biometric security techniques for iris recognition system with high performance and high confidence are described. The system is based on an empirical analysis of the iris image and it is split in several steps using local image properties. The system steps are capturing iris patterns; determine the location of the iris boundaries; converting the iris boundary to the stretched polar coordinate system; extracting the iris code based on texture analysis using wavelet transforms; and classification of the iris code. The proposed system use the wavelet transforms for texture analysis, and it depends heavily on knowledge of the general structure of a human iris.

Computer vision is a broad-based field of computer science that requires students to understand and integrate knowledge from numerous disciplines. Computer science [CS] majors, however, do not necessarily have an interdisciplinary background. In the rush to integrate, we can forget, or fail to plan for the fact that our students may not possess a broad undergraduate education. To explore the appropriateness of our education materials, this paper begins with a discussion of what we can expect CS majors to know and how we can use that knowledge to make a computer vision course a more enriching experience. The paper then provides a review of a number of the currently available computer vision textbooks. These texts differ significantly in their coverage, scope, approach, and audience. This comparative review shows that, while there are an increasing number of good textbooks available, there is still a need for new educational materials. In particular, the field would benefit from both an undergraduate computer vision text aimed at computer scientists and from a text with a stronger focus on color computer vision and its applications.

This paper presents Sassy, a single-assignment variant of the C programming language developed in concert with Khoral Inc. and designed to exploit both coarse-grain and ne-grain parallelism in image processing applications. Sassy programs are written in the Khoros software development environment, and can be manipulated inside Cantata (the Khoros GUI). The Sassy language supports image processing with true multidimensional arrays, sophisticated array access and windowing mechanisms, and built-in reduction operators (e.g. histogram). At the same time, Sassy restricts C so as to enable compiler optimizations for parallel execution environments, with the goal of reducing data traffic, code size and execution time. In particular, the Sassy language and its optimizing compiler target reconfigurable systems, which are fine-grain parallel processors. Recongurable systems consist of field-programmable gate arrays (FPGAs), memories and interconnection hardware, and can be used as inexpensive co-processors with conventional workstations or PCs. The compiler optimizations needed to generate highly optimal host, FPGA, and communication code, are discussed. The massive parallelism and high throughput of reconfigurable systems makes them well-suited to image processing tasks, but they have not previously been used in this context because they are typically programmed in hardware description languages such as VHDL. Sassy was developed as part of the Cameron project, with the goal of elevating the programming level for reconfigurable systems from hardware circuits to programming language.

Techniques based on thresholding of wavelet coefficients are gaining popularity for alenoising data.

This paper presents a new toolkit for the creation of customized structured document recognition applications by domain experts. This open-source system, called Gamera, allows a user, with particular knowledge of the documents to be recognized, to combine image processing and recognition tools in an easy-to-use, interactive, graphical scripting environment. Gamera is one of the key technology components in a proposed international project for the digitization of diverse types of humanities documents. Categories and Subject Descriptors I.4 [Computing Methodologies]: Image processing and computer vision

The aim of this study is to detect masses in mammograms on the basis of textural features. Suspicious regions are identified following the bilateral image subtraction of left and right breast image pairs. The study uses the nipple as a common rotational point thereby facilitating an alignment with the highest correlation prior to subtraction. Within this study, 144 breast images from the MIAS database are considered. Five cooccurrence matrices are constructed at four different distances for each suspicious region. Twelve texture features defined by Haralick et. al. [5], angular second moment, correlation, contrast, entropy, inverse difference moment, sum average, sum entropy, sum variance, difference entropy, difference variance and two information measures of correlation. Two further features defined by Chan et. al [2], inertia and difference average, are also computed giving a total of fourteen texture measures. Following classification of six principal components calculated for the ...