Plant identification from photos is a complex task because of the variability in the image data and small distinctiveness between different plant species. However, there is a limited amount of information that can be retrieved from the photo metadata and user knowledge. Therefore, the aim of the thesis is to demonstrate how to support the plant idenfitication with the plant photo context, such as time, location, user demands, and other aspects in the identification process. By analyzing image metadata, this thesis intends to improve the performance of the plant identification service and enrich the neural network model.