Plants are necessary for life to exist on this planet. There are several plant species to pick from, and the number of plant species is increasing every year. Programmed species identification has a variety of advantages over traditional species identification. The majority of plant-programmed identification methods today concentrate on leaf morphology, venation, and surface traits, which have shown to be useful in recognising some plant species. Due to their year-round availability, especially in tropical climates, leaves are widely utilised in plant species identification. In terms of shape, texture, breathability, and colour, a single leaf can serve a number of purposes. A variety of methodologies, such as classic moral assessment methods or machine learning, can be employed to extract these functions. For non-specialists who have little or no comprehension of common natural notions, this, on the other hand, demands expert knowledge and becomes a time-consuming and challenging task. Nonetheless, advancements in the fields of machine learning and computer vision may be able to assist in making this labour more doable. This study provided an effective approach for recognising plant leaf species based on SVM and SMO. A random data set was used to select the leaves. The performance is assessed using accuracy, error rate, and other measures. Nonetheless, advances in the domains of machine learning and computer vision can help to make this work more manageable. Based on SVM and SMO, this research proposed an efficient method for identifying plant leaf species. The leaves were chosen at random from a random data set. The accuracy, error rate, and other metrics are used to evaluate the performance.