摘要
为使树叶识别达到更好的效果,对分形维度作为纹理特征的方法进行了研究,提出把分形维度应用于叶片识别领域,将其作为叶片纹理特征的提取方法,来表示叶片图像的纹理复杂程度及其自相似性。使用了26种叶片(390幅图像)的图像库进行实验,提取了包括分形维度在内的9项特征进行训练测试,达到了91.54%的正确率。实验结果表明,提出的方法优于传统的叶片识别方法,证实了该方法的有效性。
In order to get a better result of leaves recognition,the method of using fractal dimension as texture feature is researched.Fractal dimension which shows texture complication and self similarity is used in the leaves recognition as the part of texture feature.The experiment tested on 390 images which belong to 26 kinds of leaves and then extracted 9 feature including fractal dimension.The classification accuracy is 91.54% which is better than traditional method of leaves recognition.The experimental results have demonstrated the effectiveness of the proposed method.
出处
《计算机工程与设计》
CSCD
北大核心
2012年第2期670-673,共4页
Computer Engineering and Design
关键词
分形
分形维度
纹理特征
特征提取
叶片识别
fractal
fractal dimension
texture feature
feature abstraction
leaves recognition
