Approach to plant species identification in natural grasslands based on deep learning

The classification and identification of plant species plays an important role in the investigation of natural grassland vegetation. The traditional classification of plant species is time-consuming and laborious and requires high professional knowledge. With the rapid development of computer vision and deep learning, plant image classification and recognition based on deep learning algorithms has become a popular research topic. Based on the image data set of 293 natural grassland plants, the TF-slim module of the TensorFlow deep learning framework was used to construct the image recognition model of natural grassland plants by fine-tuning the training parameters of the Inception V3 model. The training results show that the model has a recognition accuracy of 89.41% in Top1 and 97.71% in Top5. Compared with the recognition effects of applications such as Xingse, Huabanlv, The Flower Recognition, and Paizhaoshihua, the results show that the plant recognition model trained in this study can recognize more species of natural grassland plants, and the recognition accuracy is higher.