Environmental pollution is a serious issue of this generation. The predominant reason is the degradation of plant resources. However, there is a significant amount of increase in plantation because of awareness. Though we plant millions of trees every year, many saplings die an untimely death because of a lack of maintenance and care. These saplings require attention to uncommon vicinity conditions like temperature, moisture, humidity, and proper health monitoring. Sustainable enhancement of plants requires monitoring the young plant's health by looking into external conditions that impact the environment. Looking at this perturbing situation, we have designed and developed a sustainable enhancement system for saplings. The research involves developing an autonomous, robust, and non-contact system for the saplings to monitor their growth, vicinity, and health conditions using deep learning models like Convolutional Neural Network and MQTT IoT protocol. The designed prototype consists of a drone for geo-tagging each sapling, collecting sensor, and image data for health monitoring and classification. The developed system reduces human efforts by automating the system and minimizes the number of the untimely death of planted saplings.

The dependencies are listed under requirements.txt and are all purely python based. To install them simply run

pip install -r requirements.txt

The datset link is provided here Bay Leaf Dataset .

Run the Cnn_final.ipynb jupyter notebook

The research is potential enough to curb serious problems like environmental pollution through sustainable development. The Central and the State government takes a large number of initiatives to improve such situations. If there is a successful development of the prototype to a complete product with high accuracy, it will lead the stepping stone towards sustainable research. The current system demonstrates the control of the drones through the RX/TX remote controller. An enhancement to the system is possible by controlling the drone from the remote station through the internet. In most of the research, the UAVs are controlled from the ground station. The implementation of the autonomous skyrocketing with the help of path planning will enhance the current research. In the current system, the server is running in the local machine. The raspberry pi present on the drone and the local machine connects to the same network. The port forwarding method will help to implement the prototype in the real scenario and to access raspberry pi over the internet. The hosting of the web application is possible in any hosting platform on the public server. The people can get access to the plant data, including the geolocation of all plantation areas, in the city through an android application. They can voluntarily go to the nearest plantation areas and take the necessary steps. The current model acts as a prototype for the innovation, with an accuracy of 88%. It requires further development to reach higher accuracy with a better model. The continuous research in the field of Sustainable development will lead our world to a better place.