Conservation of Water in Urban Gardening Using Machine Learning

The rate of urbanization in India largely surpasses the pace of providing basic amenities like water and electricity, thereby causing scarcity of resources. One such water intensive activity is urban gardening. Earlier studies examine scarcity of water as a major challenge in urban gardening, and use of technologies to improve efficiencies. This paper presents managing the use of water as per plant need in urban gardening through the use of Internet of Things (IoT) and Machine Learning (ML) models. The approach aligns with the Goal 11 - Sustainable cities and communities of the United Nations Sustainable Development Goals. The Transfer Learning model on Tensorflow Lite was used to classify plants types. A modest dataset consisting of 124 72x72 pixels images of 16 indoor plant species was used to train the model to classify the plants into three watering frequency categories viz. frequent, regular and infrequent. This model was deployed in a smart system to classify plants and efficiently schedule watering. Thereby, significantly minimizing overwatering and water wastage.