Abstract | ||
---|---|---|
In this paper, an address block segmentation approach based on fractal dimension FD is proposed. After computing the fractal dimension of each image pixel by the 2D variation procedure, a clustering technique based on K-means is used to label pixels into semantic objects. The evaluation of the efficiency is carried out from a total of 200 postal envelope images with no fixed position for the address block, postmark and stamp. A ground-truth strategy is used to achieve an objective comparison. Experiments showed significant and promising results. By using the 2D variation procedure for three ranges of neighbor window sizes (r = {3, 5}, r = {3, 5, 7}, and r = {3, 5, 7, 9}), the proposed approach reached a success rate over than 90% on average. |
Year | DOI | Venue |
---|---|---|
2004 | 10.1007/978-3-540-30463-0_57 | Lecture Notes in Computer Science |
Keywords | Field | DocType |
k means,ground truth,fractal dimension | k-means clustering,Fractal dimension,Pattern recognition,Segmentation,Computer science,Fractal,Algorithm,Ground truth,Pixel,Artificial intelligence,Digital mammogram,Cluster analysis | Conference |
Volume | ISSN | Citations |
3287 | 0302-9743 | 2 |
PageRank | References | Authors |
0.48 | 4 | 3 |
Name | Order | Citations | PageRank |
---|---|---|---|
Luiz Felipe Eiterer | 1 | 11 | 1.16 |
Jacques Facon | 2 | 67 | 15.67 |
David Menoti | 3 | 13 | 1.44 |