Sep 16th 2023

Try Plant-E

We’re introducing a neural network called Plant-E which efficiently learns visual concepts from a large dataset of plant images.This allows the system to recognize the type and name of the plant  The system is based on a deep-learning algorithm that learns the differences between plants by learning what each plant looks like under different light and in different environments. 
Our first version of Plant-E called Plant-E (basic) has over 3,000 Plant images from the internet This version has 159 species of plants from all over the world included.  
Strengths:  Plant-E’s strengths are most apparent with the dataset we trained it on. We taught the system on data with different backgrounds and noise and blurry images. This allows the system to understand user inputs without having to have a white background or super clear image.  Limitations:  Plant-E is limited by its data. There are over 320,000 plants in the world and we won’t be able to add them all to our system. We will try to add as many plants as we can to future versions but we likely won’t be able to completely add everything.Scores:  We have implemented scores to plant-e  instead of other plant identification systems where you just get an answer Plant-E goes way more in depth. Plant-E will give you an answer and a score of how confident the system is in that answer. 
Plant-E will be free for our testing round but we plan to launch a paid version in the near future   Open Source Version:  As we are a group that has built off of other images and data we are making sure to give back. The model will have a fully open source version launching SEP 25th 2023 When launched our open source  version will include the models training data the model weights in a tensorflow.js binary format and finally a full version of the model in tensorflow format Conclusion:  We look forward to Plant-E becoming a valuable tool in improving the world of nature. There’s still a lot of work to do, and we look forward to improving this model through the collective efforts of the community building on top of, exploring, and contributing to the model.