摘要
利用单个信息源进行苹果分类,常常出现分类错误,即使分类正确,也不能给出综合标准下的分类结果。本文利用苹果大小及颜色两个信息源,首先提取大小及颜色的特征,利用各自的特征对苹果等级进行初步决策,然后应用DSmT进行决策级融合,给出了基于决策级融合的苹果等级分类算法。实验分析表明,本文算法可实现大小及颜色两个指标的苹果等级分类,分类精度高,鲁棒性好。
Using single information source for apple grading always leads to false classifying. However, it can not give the grading results with an integrated standard. In this paper, the size and color information source are utilized. Firstly, the size and color features are extracted, and the apples are graded according to each feature. Then Dezert- Smarandache Theory （DSmT） fusion rule is used to achieve the decision fusion. Thus, the apple grading algorithm based on decision fusion is proposed. The experimental results show that this algorithm can carry out the apple grading with size and color features, and has good performance on veracity and robustness.
出处
《电子测量技术》
2008年第1期10-12,24,共4页
Electronic Measurement Technology
基金
国家自然科学基金资助(60372085)
2004~2005西北工业大学本科毕业设计重点扶持项目资助
