Agriculture is essential for human survival and remains a major economic driver in many countries around the world. Most of the living things around the world feed on vegetation produced by agriculture. Therefore, researchers have to work on developing agriculture using technology. We need to know the class of plant before making decisions about development and improvement. Previous machine vision systems for selective weeding have struggled to identify weeds reliably and accurately. Traditional classification workflows are sluggish and error-prone; classification expertise is held by a small number of expert taxonomists; and, to make matters worse, classification expertise is held by a small number of expert taxonomists. In recent years, the number of taxonomists has gradually decreased. Automated organism identification has thus become more than a wish, but a necessity for better understanding, using, and preserving biodiversity. This paper gives an overview of recent attempts to classify species using computer vision and machine learning techniques. It concentrates on identifying plant species using leaf images. With a dataset containing 4,275 photos of 12 species at various growth stages, we present approaches for plant seedling classification. We compare the results of two commonly used image classification algorithms: The Convolutional Neural Network (CNN) and transfer learning. Our proposed model achieved 0.9754,0.9742,0.9766,0.9754 In terms of, Accuracy, Sensitivity, Specificity, F-score, respectively. Both standard machine learning approaches and those using Convolution Neural Networks compare the results.