Optical character recognition of machine-printed documents is an effective means for extracting textural material. While the level of effectiveness for handwritten documents is much poorer, progress is being made in more constrained applications such as personal checks and postal addresses. In these applications a series of steps is performed for recognition beginning with removal of skew and slant. Slant is a characteristic unique to the writer and varies from writer to writer in which characters are tilted some amount from vertical. The second attribute is the skew that arises from the inability of the writer to write on a horizontal line. Several methods have been proposed and discussed for average slant estimation and correction in the earlier papers. However, analysis of many handwritten documents reveals that slant is a local property and slant varies even within a word. The use of an average slant for the entire word often results in overestimation or underestimation of the local slant. This paper describes three methods for local slant estimation, namely the simple iterative method, high-speed iterative method, and the 8-directional chain code method. The experimental results show that the proposed methods can estimate and correct local slant more effectively than the average slant correction.

A shape matching framework designed for an industrial application is presented. The task of the proposed system is to identify and sort plastic teeth, used by dentists, based on their 2D shape only. A sorting machine puts each tooth on a predefined location under a camera which is equipped with a telecentric lens. From the resulting image the contour of the object is extracted and compared with a database of reference teeth. In order to cope with the problem that a tooth may rest on several (typically 4 - 15) stable positions when placed under the camera, all its proper contours are stored as valid tooth representations. In total the tooth database used for tests contained 171 teeth represented by 1257 contours of 1000 points each. Under the constraint that one contour out of 1257 has to be identified in less than one second, we describe the algorithmic approach which has successfully led to the implementation of the system. A fast pre-selection of shapes and the repeated calculation of point transforms to match them with the reference contours makes up the underlying principle of the proposed system. In addition to the decisions actually made during the design of the system we describe several possible enhancements which can further improve the speed and generality of our matching approach.

In this paper the authors present two methods to extract handwritten components from a personal bank check. The first technique utilizes a blank check image designated as the reference image and a filled-in check image designated as the sample image. These checks differ only at specific regions of the check reserved for insertion of handwritten or machine printed components. A technique called morphological subtraction is used to extract the user-entered components. However, because the background patterns of reference image and sample images will not be identical, an affine transform is used to ensure better alignment between the two images. To eliminate the need for a blank check of every kind, a second method is proposed that uses grayscale morphology to extract the handwritten information. Because some background features are similar to the user-entered information, some residual background information is likely to be present in the processed image; however these can be removed with post processing.

An overview of the oyster industry in the U. S. with emphasis in Virginia shows oyster grading occurs at harvest, wholesale and processing markets. Currently whole oysters, also called shellstock, are graded manually by screening and sorting based on diameter or weight. The majority of oysters harvested for the processing industry are divided into three to four main grades: small, medium, large, and selects. We have developed a shape analysis method for an automatic oyster grading system. The system first detects and removes poor quality oysters such as banana shape, broken shell, and irregular shapes. Good quality oysters move further into grades of small, medium and large. The contours of the oysters are extracted for shape analysis. Banana shape and broken shell have a specific shape flaw (or difference) compared to the ones with good quality. Global shape properties such as compactness, roughness, and elongation are suitable and useful to measure the shape flaw. Image projection area or length of the major axis measured as global properties for sizing. Incorporating a machine vision system for grading, sorting and counting oysters supports reduced operating costs. The savings produced from reducing labor, increasing accuracy in size, grade and count and providing real time accurate data for accounting and billing would contribute to the profit of the oysters industry.

Fish migration is being monitored year round to provide valuable information for the study of behavioral responses of fish to environmental variations. However, currently all monitoring is done by human observers. An automatic fish recognition and migration monitoring system is more efficient and can provide more accurate data. Such a system includes automatic fish image acquisition, contour extraction, fish categorization, and data storage. Shape is a very important characteristic and shape analysis and shape matching are studied for fish recognition. Previous work focused on finding critical landmark points on fish shape using curvature function analysis. Fish recognition based on landmark points has shown satisfying results. However, the main difficulty of this approach is that landmark points sometimes cannot be located very accurately. Whole shape matching is used for fish recognition in this paper. Several shape descriptors, such as Fourier descriptors, polygon approximation and line segments, are tested. A power cepstrum technique has been developed in order to improve the categorization speed using contours represented in tangent space with normalized length. Design and integration including image acquisition, contour extraction and fish categorization are discussed in this paper. Fish categorization results based on shape analysis and shape matching are also included.

As the cost/performance Ratio of vision systems improves with time, new classes of applications become feasible. One such area, automotive applications, is currently being investigated. Applications include occupant detection, collision avoidance and lane tracking. Interest in occupant detection has been spurred by federal automotive safety rules in response to injuries and fatalities caused by deployment of occupant-side air bags. In principle, a vision system could control airbag deployment to prevent this type of mishap. Employing vision technology here, however, presents a variety of challenges, which include controlling costs, inability to control illumination, developing and training a reliable classification system and loss of performance due to production variations due to manufacturing tolerances and customer options. This paper describes the measures that have been developed to evaluate the sensitivity of an occupant detection system to these types of variations. Two procedures are described for evaluating how sensitive the classifier is to camera variations. The first procedure is based on classification accuracy while the second evaluates feature differences.

In this paper we show a new automated measuring system based on a fast optical co-ordinate acquisition device that performs the calculation of all the parameters characterizing the strain of an industrial piece. These results are obtained with high accuracy and in real time. An application of this improved system is illustrated and the whole measuring procedure is described in the following steps: grid marking on the metal sheet, stamping, co-ordinate acquisition, calculation of the strain parameters and their visualizaiton.

This paper describes a novel optical system capable of measuring 5-axis (five degrees of freedom) object surface movement without the need for special surface preparation or stringent alignment. The compact optical system is based on electronic speckle photography (ESP) and is designed to be insensitive to out-of-plane movement. Experiments were conducted to measure the 5-axis motion simulated by a 6-axis motion system. The results show that the optical system accurately resolves the motion on every axis successfully, with the expected insensitivity to out-of-plane displacements. A possible application of the technique in strain measurement is also addressed in the paper.

A thermal technique and system for measuring the depth of a crack in metal is described. The transient heat signature of a moving infrared laser source is analyzed using three intensity-based algorithms and the limitations of this approach are examined. Experimental results demonstrate capability to measure EDM-notches 0.06 inches deep and shallower, and the variability of the system agrees with the theoretical. The advantages and disadvantages of this technique over other non-destructive testing (NDT) methods are also outlined.

In the field of industrial vision, extracting the 3D shape information of highly reflective metallic objects is still a delicate task. This paper presents a new application of "Shape from Polarization" method to specular metallic surfaces. Studying the state of polarization of the reflected light is very useful to get information on the normals of the surface. This article demonstrates how to extend the commonly used method for dielectric to metallic surfaces. Applications on shape defects detection are also discussed, and the efficiency of the system to discriminate defects on metallic objects made by stamping and polishing is presented.

A research undertaken to help blind people to navigate autonomously or with minimum assistance is termed as "Blind Navigation". In this research, an aid that could help blind people in their navigation is proposed. Distance serves as an important clue during our navigation. A stereovision navigation aid implemented with two digital video cameras that are spaced apart and fixed on a headgear to obtain the distance information is presented. In this paper, a neural network methodology is used to obtain the required parameters of the camera which is known as camera calibration. These parameters are not known but obtained by adjusting the weights in the network. The inputs to the network consist of the matching features in the stereo pair images. A back propagation network with 16-input neurons, 3 hidden neurons and 1 output neuron, which gives depth, is created. The distance information is incorporated into the final processed image as four gray levels such as white, light gray, dark gray and black. Preliminary results have shown that the percentage errors fall below 10%. It is envisaged that the distance provided by neural network shall enable blind individuals to go near and pick up an object of interest.

A large number of 3D cameras suffer from so-called holes in the data, i.e. the measurement lattice is affected by invalid measurements and the range image has undefined values. Conventional image filters used for removing the holes perform not well in presence of holes with large varying hole sizes. The novel hole-filling method presented in this paper operates on reliability attributed range images featuring unwanted holes with wide varying sizes. The method operates according to a multi resolution scheme where the image resolution is decreased at the same time as the range reliability is successively increased until sufficient confidence is reached. It builds on three main components. First, the described process performs a weighted local neighbourhood filter where the contribution of each pixel stands for its reliability. Second, the filtering combines filters with different kernel sizes and implements therefore the multi resolution schema. Third, the processing requires a complete travel from high resolution down to the resolution of satisfactory confidence and back again to the highest resolution. The algorithm for the described method was implemented in a efficient way and was widely applied in the hole-filling of range images from a depth from focus process where reliability is obtainable non-linearly from the local sharpness measurement. The method is valid in a very general way for all range imagers providing reliability information. It seems therefore well suited to depth cameras like time-of-flight, stereo and other similar rangers.

A new technique is proposed for calibrating a 3D modeling system with variable zoom based on multi-view stereo image analysis. The 3D modeling system uses a stereo camera with variable zoom setting and a turntable for rotating an object. Given an object whose complete 3D model (mesh and texture-map) needs to be generated, the object is placed on the turntable and stereo images of the object are captured from multiple views by rotating the turntable. Partial 3D models generated from different views are integrated to obtain a complete 3D model of the object. Changing the zoom to accommodate objects of different sizes and at different distances from the stereo camera changes several internal camera parameters such as focal length and image center. Also, the parameters of the rotation axis of the turntable changes. We present camera calibration techniques for estimating the camera parameters and the rotation axis for different zoom settings. The Perspective Projection Matrices (PPM) of the cameras are calibrated at a selected set of zoom settings. The PPM is decomposed into intrinsic parameters, orientation angles, and translation vectors. Camera parameters at an arbitrary intermediate zoom setting are estimated from the nearest calibrated zoom positions through interpolation. A performance evaluation of this technique is presented with experimental results. We also present a refinement technique for stereo rectification that improves partial shape recovery. And the rotation axis of multi-view at different zoom setting is estimated without further calibration. Complete 3D models obtained with our techniques are presented.

Structured light techniques have been used in a lot of applications. As a two-dimensional optical measurement method, structured light sensors are faster than one-dimensional point triangulation sensor while easier to calibrate and move than full-field three-dimensional sensors. The accuracy of structured light sensors mainly depends on the accuracy of both calibration and beam center extraction. In some applications with complicated surface shapes, the extracted center may not be the actual “true” center, which results in image bias. This paper presents a method to compensate the image bias and improve the measurement accuracy of structured light sensors. The basic concepts of image bias correction are given and some initial results are provided.

We investigate lateral and axial chromatic confocal microscopy using supercontinuum white light, and its application to surface profile measurement. In the systems that we describe here, the lateral or the axial scanning is effectively realized by focusing different wavelengths of the supercontinuum to either different lateral or axial positions through purposely introduced chromatic dispersion and aberration respectively. As a result, the imaging speed can be greatly improved. We use this system to demonstrate the surface profile measurement of a microcircuit chip, with a sensitivity of 8.5 nm and a depth measurement range of about 7 microns.

One key problem of fringe projection techniques for 3D shape measurements is the limited phase unambiguity range when only one grating period is used. Dual-frequency patterns in which involves two grating periods can easily extend the unambiguity range. A method to fabricate accurate dual-frequency patterns is presented. The advantage of using digital dual-frequency patterns for projected fringe profilometry are (1) high geometrical accuracy (< 0.5μm); (2) high contrast ratio; (3) very low high order harmonic distortions; and (4) extended unambiguity range.

We propose a novel structured light method, namely trapezoidal phase-shifting method, for 3-D shape measurement. This method uses three patterns coded with phase-shifted, trapezoidal-shaped gray levels. The 3-D information of the object is extracted by direct calculation of an intensity ratio. Theoretical analysis showed that this new method was significantly less sensitive to the defocusing effect of the captured images when compared to the traditional intensity-ratio based methods. This important advantage makes large-depth 3-D shape measurement possible. If compared to the sinusoidal phase-shifting method, the resolution is similar, but the processing speed is at least 4.5 times faster. The feasibility of this method was demonstrated in a previously developed real-time 3-D shape measurement system. The reconstructed 3-D results showed similar quality as those obtained by the sinusoidal phase-shifting method. However, since the processing speed was much faster, we were able to not only acquire the images in real time, but also reconstruct the 3-D shapes in real time (40 fps at a resolution of 532 x 500 pixels). This real-time capability allows us to measure dynamically changing objects, such as human faces. The potential applications of this new method include industrial inspection, reverse engineering, robotic vision, computer graphics, medical diagnosis, etc.

Small features such as bevels and edge profiles are a common and often critical feature on many manufactured parts. In the past, the only ways to measure such features was to make a wax mold of the feature and slice it for measurement on an optical comparator, or to do a slow mechanical tracing with a stylus or CMM type measurement system. This paper describes the application of machine vision tools, using controlled lighting to highlight shape information such as curvature, combined with 2D vision processing to extract the 3D shape based upon surface modeling.

A long-established distance sensing technique is based on measurement of the modulation phase shift of an intensity-modulated laser beam that is reflected from a remote target. This technique is capable of distance measuring precision in the order of several microns in conjunction with a co-operative target and modulation frequencies up to 1GHz, but generally suffers from severe performance degradation on natural surfaces due to the signal-to-noise ratio limitations of available fast optical detectors. In a novel variant of the intensity-modulated phase shift technique, the measuring beam reflected back off the target is modulated a second time at a slightly different frequency to achieve modulation frequency translation in the optical domain prior to optical detection. The resulting component of optical modulation at the difference frequency preserves the phase shift carried on the high frequency measuring beam, permitting the use of a sensitive, low-bandwidth optical detector to measure the critical phase shift. Following a review of the measuring principle, its practical implementation and current stage of development, examples are provided of the measurement performance achievable in various applications.

We proposed a boundary extraction method by utilizing the scale space in order to extract boundaries flexibly from the cross-sectional SEM images. The Gaussian scale space is constructed by convolving the original image with Gaussians of increasing standard deviation, that is, multiple Gaussian filtering. Setting the appropriate range of standard deviations for a target CD measurement part gives a high degree of flexibility to the edge detection method. Though the locations of the edges at the coarse scales may be shifted form their true locations, the true edge location is estimated by tracing the edges detected at coarse scales back to finer scales. By utilizing the edge-location-scale map, the blurring of image acquisition system is also estimated. By applying the proposed method to a model image, the edge location and the blurring estimation accuracies were evaluated. The accuracy in estimated edge locations was less than 0.5 pixels and the blurring very close to the true value was obtained. Next, the proposed method was applied to a cross-sectional SEM image of the DRAM that had a very low contrast boundary to show its validity.

In this paper, we propose a method to find out the intrinsic parameters of the camera using the matrix rank constrain (MRC) of the relation matrix for absolute conic W. At the end of this paper, experimental results are presented and are compared with the other methods, which show the good performance of this proposed method.