Estimation of Physical Characteristics of Peach Leaves Using K-means Clustering in the L*a*b* Color Space.

Evaluating the health of a peach tree based on the characteristics of its leaves is a common practice in botany and agriculture. The length, width, area, and perimeter of peach leaves represent their basic physical characteristics, which can be used further to determine the healthy status of a peach tree from the affected status based on leaf symptoms of diseases or nutrient deficiencies. In this study, we compare the segmentation outcomes across twenty-five different color spaces, and L*a*b* color space yields the best result using the proposed metric function scores. Furthermore, seven segmentation algorithms in the L*a*b* color space are compared using the same metric function and K-means algorithm is identified as the most effective one among assessed algorithms. Based on these results, we employ the proposed procedure using K-means clustering in the L*a*b* color space to estimate physical characteristics of peach leaves.