基于AdaBoost集成学习的烟丝组分识别

目的：提高烟丝的识别效率。方法：利用F-score特征选择方法和AdaBoost集成学习方法对烟丝组分进行识别，提取烟丝的纹理、颜色、形状特征作为模型的输入，通过F-score特征选择方法降低特征维度，以支持向量机(Support Vector Machine，SVM)作为基分类器，再利用AdaBoost集成学习方法，得到烟丝的分类模型。结果：该方法能够有效区分不同组分烟丝，每种烟丝的识别准确率都在95%以上。结论：AdaBoost集成学习方法比传统方法更快捷、方便，也更安全、有效。

Identification of cut tobacco components based on AdaBoost ensemble learning

Objective:In order to improve the identification efficiency of cut tobacco.Methods:F-score feature selection method and AdaBoost ensemble learning method were used to recognize cut tobacco components. The texture, color and shape features of cut tobacco were extracted as the input of the model. The feature dimension is reduced by F-score feature selection method, and the support vector machine (SVM) was used as the base classifier, then AdaBoost ensemble learning method was used to get the classification model of cut tobacco.Results:This method could effectively distinguish different components of cut tobacco, and the recognition accuracy of each kind of cut tobacco was more than 95%.Conclusion:AdaBoost ensemble learning method is faster and more convenient than traditional methods, and also safer and more effective.