Dinesh Nair

Bayesian recognition of targets by parts in second generation forward looking infrared images

Abstract This paper presents a system for the recognition of targets in second generation forward looking infrared images (FLIR). The recognition of targets is based on a methodology for recognition of two-dimensional objects using object parts. The methodology is based on a hierarchical, modular structure for object recognition. In the most general form, the lowest level consists of classifiers that are trained to recognize the class of the input object, while at the next level, classifiers are trained to recognize specific objects. At each level, the objects are recognized by their parts, and thus each classifier is made up of modules, each of which is an expert on a specific part of the object. Each modular expert is trained to recognize one part under different viewing angles and transformations. A Bayesian realization of the proposed methodology is presented in this paper, in which the expert modules represent the probability density functions of each part, modeled as a mixture of densities to incorporate different views (aspects) of each part. Recognition relies on the sequential presentation of the parts to the system, without using any relational information between the parts. A new method to decompose a target into its parts and results obtained for target recognition in second generation FLIR images are also presented here.

Control strategies and image processing

This paper presents three real-world applications that highlight the benefits of using image processing and machine vision techniques for control applications. These examples also illustrate how classical and modern control techniques can be used to solve image processing and machine vision problems. For many applications the use an appropriate combination of image processing and control theory offers advantages over alternative approaches.

Induced well-distributed sets in Riemannian spaces

The concept of Riemannian geometries is used to construct induced homogeneous point sets on manifolds that are based on well-distributed point sets in unit cubes of an appropriately chosen Euclidean space. These well-distributed point sets in unit cubes are based on standard low-discrepancy sequences. The approach is algorithmic, that is, the methods developed in this article have been implemented and tested. Applications in image processing, graph theory and measurement-based exploration are presented.

Pattern matching based on a generalized Fourier transform

In a two-dimensional pattern matching problem, a known template image has be located in another image, irrespective of the templates position, orientation and size in the image. One way to accomplish invariance to the changes in the template is by forming a set of feature vectors that encompass all the variations in the template. Matching is then performed by finding the best similarity between the feature vector extracted from the image to the feature vectors in the template set. In this paper we introduce a new concept of a generalized Fourier transform. The generalized Fourier transform offers a relatively robust and extremely fast solution to the described matching problem. The application of the generalized Fourier transform to scale invariant pattern matching is shown here.

Automatic inspection of textured surfaces by support vector machines

Automatic inspection of manufactured products with natural looking textures is a challenging task. Products such as tiles, textile, leather, and lumber project image textures that cannot be modeled as periodic or otherwise regular; therefore, a stochastic modeling of local intensity distribution is required. An inspection system to replace human inspectors should be flexible in detecting flaws such as scratches, cracks, and stains occurring in various shapes and sizes that have never been seen before. A computer vision algorithm is proposed in this paper that extracts local statistical features from grey-level texture images decomposed with wavelet frames into subbands of various orientations and scales. The local features extracted are second order statistics derived from grey-level co-occurrence matrices. Subsequently, a support vector machine (SVM) classifier is trained to learn a general description of normal texture from defect-free samples. This algorithm is implemented in LabVIEW and is capable of processing natural texture images in real-time.

The SIVA Image Processing Demos

Publisher Summary This chapter explains the popular courseware for image processing education known as SIVA—the signal, image, and video audiovisualization—gallery. The image and video processing section of the SIVA gallery consists of a suite of special purpose LabVIEW-based programs known as Virtual Instruments (VIs). Equipped with informative visualization and a user-friendly interface, these VIs were carefully designed to facilitate a gentle introduction to the fascinating concepts in image and video processing. To make things easier, the demos are accompanied by a comprehensive set of help files that describe the various controls, and that highlight some illustrative examples and instructive parameter settings. A demo can be activated by clicking the rightward pointing arrow in the top menu bar. Help for the demo can be activated by clicking the ‘?’ button and moving the cursor over the icon that is located immediately to the right of the ‘?’ button. In addition, when the cursor is placed over any other button/control, the help window automatically updates to describe the function of that button/control. The user will find this visual, hands-on, interactive introduction to image processing to be a fun, enjoyable, and illuminating experience.

A new algorithm for real-time multi-stage image thresholding

Image binarization under non-uniform lighting conditions is required in many industrial machine vision applications. Many local adaptive thresholding algorithms have been proposed in the literature for this purpose. However, existing local adaptive thresholding algorithms are either not robust enough or too expensive for real-time implementation due to very high computation costs. This paper presents a new algorithm for local adaptive thresholding based on a multi-stage framework. In the first stage, a mean filtering algorithm, with kernel-size independent computation cost, is proposed for background modeling to eliminate the non-uniform lighting effect. In the second stage, a background-corrected image is generated based on the background color. In the final stage, a global thresholding algorithm is applied to the background-corrected image. The kernel-size independent computation algorithm reduces the order of computation cost of background modeling from NML2 to ML+NL+6NM for an N x M image with an L x L kernel, which enables the real-time processing of objects of arbitrary size. Experiments show that the proposed algorithm performs better than other local thresholding algorithms, such as the Niblack algorithm, in terms of both speed and segmentation results for many machine vision applications under non-uniform lighting conditions.

Color characterization for image indexing and machine vision

Color representation and comparison based on the histogram has proved to be very efficient for image indexing in content-based image retrieval and machine vision applications. However, the issues of color constancy and accurate color similarity measures remain unsolved. This paper presents a new algorithm for intensity- insensitive color characterization for image retrieval and machine vision applications. The color characterization algorithm divides the HSI (hue, saturation and intensity) color space into a given number of bins in such a way that the color characterization represents all the colors in the hue/saturation plane as well as black, white and gray colors. The color distribution in these bins of the HSI space is represented in the form of a one-dimensional vector called Color Spectrum Vector (CSV). The color information that is stored in the CSV is insensitive to changes in the luminance. A weighted version of CSV called WCSV is introduced to take the similarity of the neighboring bins into account. A Fuzzy Color Spectrum Vector (FCSV) color representation vector that takes into account the human uncertainty in color classification process is also introduced here. The accuracy and speed of the algorithm is demonstrated in this paper through a series of experiments on image indexing and machine vision applications.