Font Size: a A A

Research And Implementation On Line Drawing Generation Method For Ancient Chinese Frescoes

Posted on:2017-09-19Degree:MasterType:Thesis
Country:ChinaCandidate:D XuFull Text:PDF
GTID:2348330515967337Subject:Computer Science and Technology
Abstract/Summary:PDF Full Text Request
Edge outlining is one of the basic pictorial skills in traditional Chinese paintings,which forms a unique artistic work named “line drawing”.High-quality line drawing requires accurate positioning of lines and line width which retains the original style of lines.For generating high-quality line drawings from frescoes using image processing technology,the paper conducts in-depth research and practice.Line drawing is obtained by a two-step process: generating lines' skeletons,which locates lines,and rendering strokes with width,which reflects the style of the fresco.Interactive image segmentation is used to extract the region of interest and the skeletons of lines is extracted by an edge detection operator,then the paper use two different methods to render strokes with width.The first method combines edge detection and thresholding to render strokes with width.The paper use an image enhancement based on a high-boost filtering image,and then whether the pixel is located at stokes can be determined by its grey level and position.When the range of strokes' grey level changes little,the first method shows good results.In the second method,a non-photorealistic rendering algorithm is used to render stylish lines.Firstly,pixels are divided into two categories according to the edge tangent flow,secondly the category of the pixel and edge tangent flow are used for the generation of line drawing.When there is only little change of the width of stokes,a good line drawing can be obtained.The experimental results indicate that the proposed method retains the original stroke's style and it shows good results in extraction of creative materials in ancient Chinese frescoes.
Keywords/Search Tags:Fresco, Line drawing, Edge detection, High-boost filtering, Edge tangent flow
PDF Full Text Request
Related items