Segmentation Algorithm for Green Apples Recognition Based on K-means Algorithm
For the green apples that have similar green color with leaves, an apple recognition method based on K-means algorithm is proposed. The image is divided into 8*8 pixels blocks and the algorithm takes the block as Segmentation unit. Color difference R-G was selected as color feature and mean value, standard deviation and regional entropy of gray scale images are selected as texture features. The feature vectors which include color feature and texture features are extracted. Gap statistic was applied to calculate the best number of clusters. The algorithm is applied to 200 images that taken in different illumination conditions. The experiments results show that the apple fruits can be recognized successfully in front light conditions and back light conditions. The recognition rate reached 81％.
Apple Machine vision Image recognition K-means algorithm Gap Statistic
Si Yongsheng Liu Gang Gao Rui
Key Laboratory of MOE on Modern Precision Agriculture System Integration Research,China Agricultural Key Laboratory of MOE on Modern Precision Agriculture System Integration Research,China Agricultural
国际会议
北京
英文
1-7
2009-10-14（万方平台首次上网日期，不代表论文的发表时间）