Mark R. Bolin

A perceptually based adaptive sampling algorithm

A perceptually based approach for selecting image samples has been developed. An existing image processing vision model has been extended to handle color and has been simplified to run efficiently. The resulting new image quality model was inserted into an image synthesis program by first modifying the rendering algorithm so that it computed a wavelet representation. In addition to allowing image quality to be determined as the image was generated, the wavelet representation made it possible to use statistical information about the spatial frequency distribution of natural images to estimate values where samples were yet to be taken. Tests on the image synthesis algorithm showed that it correctly handled achromatic and chromatic spatial detail and that it was able predict and compensate for masking effects. The program was also shown to produce images of equivalent visual quality while using different rendering techniques. CR

A frequency based ray tracer

A ray tracer has been developed that synthesizes images directly into the frequency domain. This makes it possible to use a simple vision model to control where rays are cast into a scene and to decide how rays should be spawned once an object is intersected. In this manner the most visible artifacts can be removed first and noise can be channeled into those areas of an image where it is least noticeable. The resulting image is produced in a format that is consistent with many image compression and transmission schemes. CR

Visual difference metric for realistic image synthesis

An accurate and efficient model of human perception has been developed to control the placement of sample in a realistic image synthesis algorithm. Previous sampling techniques have sought to spread the error equally across the image plane. However, this approach neglects the fact that the renderings are intended to be displayed for a human observer. The human visual system has a varying sensitivity to error that is based upon the viewing context. This means that equivalent optical discrepancies can be very obvious in one situation and imperceptible in another. It is ultimately the perceptibility of this error that governs image quality and should be used as the basis of a sampling algorithm. This paper focuses on a simplified version of the Lubin Visual Discrimination Metric (VDM) that was developed for insertion into an image synthesis algorithm. The sampling VDM makes use of a Haar wavelet basis for the cortical transform and a less severe spatial pooling operation. The model was extended for color including the effects of chromatic aberration. Comparisons are made between the execution time and visual difference map for the original Lubin and simplified visual difference metrics. Results for the realistic image synthesis algorithm are also presented.

An Error Metric for Monte Carlo Ray Tracing

A method is presented for characterizing the error in Monte Carlo realistic image synthesis calculations. An error metric has been developed that can be used to control the variance in the final picture by choosing both the number of rays to be cast into the image plane and the number of rays to be spawned at each bounce in the environment. The method provides specific guidance in how to apply Russian Roulette and splitting at each level of the ray tree. An initial implementation of the method has been done to test the theory and to illustrate its mechanics.

Automatic target segmentation in color dental images

Automatic target segmentation is critical to computerized dental imaging systems, which are designed to reduce human effort and error. We have developed an automatic algorithm that is capable of outlining an intra-oral reference bar and the tooth of interest. In particular, the algorithm first locates the reference bar using unique color and shape cues. The located reference bar provides an estimate for the tooth of interest in terms of both its scale and location. Next, the estimate is used to initialize a trained active shape model (ASM) consisting of the bar and the tooth. Finally, a search process is performed to find a match between the ASM and the local image structures. Experimental results have shown that our fully automatic algorithm provides accurate segmentation of both the reference bar and the tooth of interest, and it is insensitive to lighting, tooth color, and tooth-shape variations.

Inducing node specification in active shape models for accurate lung-field segmentation

We have developed an active shape model (ASM)-based segmentation scheme that uses the original Cootes et al. formulation for the underlying mechanics of the ASM but improves the model by fixating selected nodes at specific structural boundaries called transitional landmarks. Transitional landmarks identify the change from one boundary type (such as lung-field/heart) to another (lung-field/diaphragm). This results in a multi-segmented lung-field boundary where each segment correlates to a specific boundary type (lung-field/heart, lung-field/aorta, lung-field/rib-cage, etc.). The node-specified ASM is built using a fixed set of equally spaced feature nodes for each boundary segment. This allows the nodes to learn local appearance models for a specific boundary type, rather than generalizing over multiple boundary types, which results in a marked improvement in boundary accuracy. In contrast, existing lung-field segmentation algorithms based only on ASM simply space the nodes equally along the entire boundary without specification. We have performed extensive experiments using multiple datasets (public and private) and compared the performance of the proposed scheme with other contour-based methods. Overall, the improved accuracy is 3-5 &percent; over the standard ASM and, more importantly, it corresponds to increased alignment with salient anatomical structures. Furthermore, the automatically generated lung-field masks lead to the same fROC for lung-nodule detection as hand-drawn lung-field masks. The accurate landmarks can be easily used for detecting other structures in the lung field. Based on the related landmarks (mediastinum-heart transition, heart-diaphragm transition), we have extended the work to heart segmentation.