摘要
从红富士苹果的3个面上样本中获得最佳样本,再运用主成分分析算法从最佳样本的16个特征中提取出面积、周长等7个特征的最优特征子空间是;采用支持向量机分类器对红富士苹果进行分类,识别率达到95.3%。经大量实验证明该方法是可行的。
The optimum sample is obtained by samples form the three surfaces of Fuji apple ,and then the PCA algorithm was applied to the feature extraction of the optimum sample. The algorithm extracted seven features that were composed of the optimal feature space from the 16 morphological features, such as area and perimeter. Finally, the four species of the Fuji apples were recognized by the support vector machine classifier, and the correct identification ratio was 95.3%. Experimental results on the Fuji apple database show the method in this paper is feasible.