Color-based surface reflectance separation for scene illumination estimation and rendering

Given the importance of color processing in computer vision and computer graphics, estimating and rendering illumination spectral reflectance of image scenes is important to advance the capability of a large class of applications such as scene reconstruction, rendering, surface segmentation, object recognition, and reflectance estimation. Consequently, this dissertation proposes effective methods for reflection components separation and rendering in single scene images. Based on the dichromatic reflectance model, a novel decomposition technique, named the Mean-Shift Decomposition (MSD) method, is introduced to separate the specular from diffuse reflectance components. This technique provides a direct access to surface shape information through diffuse shading pixel isolation. More importantly, this process does not require any local color segmentation process, which differs from the traditional methods that operate by aggregating color information along each image plane. ^ Exploiting the merits of the MSD method, a scene illumination rendering technique is designed to estimate the relative contributing specular reflectance attributes of a scene image. The image feature subset targeted provides a direct access to the surface illumination information, while a newly introduced efficient rendering method reshapes the dynamic range distribution of the specular reflectance components over each image color channel. This image enhancement technique renders the scene illumination reflection effectively without altering the scene’s surface diffuse attributes contributing to realistic rendering effects. ^ As an ancillary contribution, an effective color constancy algorithm based on the dichromatic reflectance model was also developed. This algorithm selects image highlights in order to extract the prominent surface reflectance that reproduces the exact illumination chromaticity. This evaluation is presented using a novel voting scheme technique based on histogram analysis. ^ In each of the three main contributions, empirical evaluations were performed on synthetic and real-world image scenes taken from three different color image datasets. The experimental results show over 90% accuracy in illumination estimation contributing to near real world illumination rendering effects. ^

Clash of identities in South Florida: An examination of the complexity of racial and national identity and their impact on ethnic solidarity between black Americans and Jamaicans

Immigrants from the West Indies and other nations challenge the simple United States dichotomy of blacks versus whites. Many apparently black Caribbean immigrants proclaim that they did not know they were “black” until they arrived in the U.S. They seek to maintain their national identity and resist identity and solidarity with Black Americans. In response, many Black Americans respond that the immigrants are simply being naive, that U.S. society demands simple racial identity. Regardless of one's self-identity and personal history, in the U.S., if you look black, you are black, was their thinking. ^ This study examines the contemporary struggle of identity and solidarity among and between Black Americans and Jamaicans living in South Florida (Broward and Miami-Dade counties). Even though the primary focus of this study is to examine the relationship between Black Americans and Jamaicans, other West Indian nationals will be addressed more generally. The primary research problem of this study is to determine why the existence of common ancestry and physical traits are insufficient for an assumption of ethnic solidarity between Black Americans and Jamaicans. ^ In examining this problem, I felt that depth rather than breadth would provide insight into the current state of polarization between Black Americans and Jamaicans. To this end, a qualitative study was designed. A non-random snowball sample consisting of forty-seven informants was selected for this study. Realizing that such a technique presents problems with generalizations beyond the sample, this approach was, nonetheless, the most suitable for the current research problem. One of the initial challenges of this research was the use of the label “black” in discussing Caribbean immigrants. Unlike America, where distinctions based on skin color were at the bedrock of America's formation, this was not the case in the Caribbean. In the Caribbean skin color was an important marker as an indicator of class, rather than of race. Therefore, I refrained from using the label, “black Jamaicans,” but rather used Jamaicans throughout. ^