科研之友
微信
新浪微博
Facebook
分享链接
摘要
Multiple fractal features estimated by box-counting method are applied to fabric defects detection with a view to getting over the difficulty encountered in tasks of textures differentiating using only one single fractal feature. Extraction of the optimum multiple fractal features is based on the following fact that has been observed for the first time: when box-counting method is adopted to estimate the fractal dimension of an image or a data series, the involved measuring box size series have a considerable influence on the estimated fractal dimension and in turn the ultimate value of the separability between normal samples and defective ones. Testing results indicate that, using the proposed algorithm, actual false alarm probability and missing rate can be controlled simultaneously below 10% under an appropriate range of threshold, proving with confidence the effectiveness of the fractal features extracted in this study.
- 出版日期2008
- 单位东华大学
