Hevea leaf features extraction and recognition algorithm for hevea clones classification using image / Mohamad Faizal Ab Jabal, Suhardi Hamid, Salehuddin Shuib

Planting a plant is one of a method to control a current globe temperature. Plant features recognition has been performed by several researchers previously. Wu et al. [1] has performed the leaf recognition by using Probabilistic Neural Network (PNN) in order to classify the plants. As a result Wu et al. [1] was successful developed an efficient algorithm for the plant classification. 32 kinds of plants have been classified by using the algorithm. The basic leaf features considered by the algorithm had been defined by Wu et al. [1] involved diameter of the leaf, physiological length, physiological width, leaf area and leaf perimeter. Moreover, from the basic leaf features, Wu et al. [1] had defined several digital morphological features which are involved smooth factor, aspect ratio, form factor, rectangularity, narrow factor, vein features and perimeter ratio of diameter, physiological length and width. Final result produced by the algorithm is 92.312% of average accuracy and the classification for the leaf was based on the leaf-shape information.