Abstract:Recent extensions to the standard difference-of-Gaussians (DoG) edge detection operator have rendered it less susceptible to noise and increased its aesthetic appeal. Despite these advances, the technical subtleties and stylistic potential of the DoG operator are often overlooked. This paper offers a detailed review of the DoG operator and its extensions, highlighting useful relationships to other image processing techniques. It also presents many new results spanning a variety of styles, including pencil-shad… Show more
“…Parametric image filters such as the Canny edge detector [1] and the difference-of-Gaussians filter [2][3][4] are widely used in image abstraction for generating line drawing images. However, the quality of output may vary significantly when adjusting the associated control parameters, leading to either excessive or insufficient details.…”
Line drawing is a style of image abstraction where the perceptual content of the image is conveyed using distinct straight or curved lines. However, extracting semantically salient lines is not trivial and mastered only by skilled artists. While many parametric filters have successfully extracted accurate and coherent lines, their results are sensitive to parameter choice and easily lead to either an excessive or insufficient number of lines. In this work, we present an interactive system to generate concise line abstractions of arbitrary images via a few user specified strokes. Specifically, the user simply has to provide a few intuitive strokes on the input images, including tracing roughly along edges and scribbling on the region of interest, through a sketching interface. The system then automatically extracts lines that are long, coherent and share similar textural structures to form a corresponding highly detailed line drawing. We have tested our system with a wide variety of images. Our experimental results show that our system outperforms state-of-the-art techniques in terms of quality and efficiency.
“…Parametric image filters such as the Canny edge detector [1] and the difference-of-Gaussians filter [2][3][4] are widely used in image abstraction for generating line drawing images. However, the quality of output may vary significantly when adjusting the associated control parameters, leading to either excessive or insufficient details.…”
Line drawing is a style of image abstraction where the perceptual content of the image is conveyed using distinct straight or curved lines. However, extracting semantically salient lines is not trivial and mastered only by skilled artists. While many parametric filters have successfully extracted accurate and coherent lines, their results are sensitive to parameter choice and easily lead to either an excessive or insufficient number of lines. In this work, we present an interactive system to generate concise line abstractions of arbitrary images via a few user specified strokes. Specifically, the user simply has to provide a few intuitive strokes on the input images, including tracing roughly along edges and scribbling on the region of interest, through a sketching interface. The system then automatically extracts lines that are long, coherent and share similar textural structures to form a corresponding highly detailed line drawing. We have tested our system with a wide variety of images. Our experimental results show that our system outperforms state-of-the-art techniques in terms of quality and efficiency.
“…Where ( ) v v > , to get more precise edges, the fractional differential mask is twice employed to enhance edge and texture with order 1 v and 2 v respectively, and the subtractive result of the two enhanced image is the edge of the image, and the process can be expressed as the following formula.…”
“…Edge detection is not only the basic problem of image processing, but also an important part of computer recognition and understanding. It plays an important role in human and computer vision [1] [2].…”
Section: Introductionmentioning
confidence: 99%
“…The difference of Gaussians (DoG) operator with strong robustness for noise is mentioned by Lowe has been shown to yield aesthetically pleasing edge lines without postprocessing [3]. The XDoG(eXtended Difference of Gaussians) operator presented by Winnem¨oller Obtains a range of subtle artistic effects, such as ghosting, speed-lines, negative edges, indication, and abstraction [1] [4]. SUSAN (Small Univalue Segment Assimilating Nucleus) operator based on local gray level difference with strong robustness for noise is an acronym standing for smallest univalue segment assimilating nucleus.…”
Abstract. In order to extract detailed edge information, a multi-scale fractional differential edge detection algorithm is proposed in this paper. Firstly, the G-L factional differential is applied to enhance image with two different fraction differential orders (one is small and the other is big), then, the edges can be gotten by subtraction the two enhanced images. Experiments and results showed that the proposed method can not only efficiently detect the edges information of simple objects, but also can detect the edges of complex objects.
“…The first paper "XDoG: An eXtended Difference-of-Gaussians Compendium including Advanced Image Stylization" [6] responds to the growing volume of the NPR literature adopting edge enhancing filtering operators to emphasize and stylize edges to create a variety of artistic effects in images. For the first time, the paper draws together the many works in this area to present both a comprehensive overview of these techniques, and a new formulation of edge enhancing filtering-the extended Difference of Gaussian (XDoG).…”
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.