Greenhouse Plant Growth Supervision with the LED Lights using Machine Learning

What is it about?

Smart planting is a new approach for making use of recent technology in the agriculture sector. Meanwhile, it’s a challenging research topic. This paper proposes an Automated greenhouse to provide the plants with a favorable atmosphere to make them grow faster and healthier. This has been accomplished by the contribution of both Image Processing techniques and Deep Learning. The system is in action when the real-time cameras start taking frames from the greenhouse then an Arduino reads the DHT11 and soil moisture sensors’ readings. After this is done, these frames are passed to the pre-processing stage to detect the desired green range of the plant and the desired color range of the fruit/vegetable. This is done using masking techniques with Hue-Saturation-Value(HSV). By the usage of Histogram of oriented gradient(HOG), the features of fruits/vegetables will be extracted such as it’s shape and color. Finally, the classifier detects if there is any fruit/vegetable appearing in the image. This helped the system to detect the correct stage of the plant; whether it’s the seeding, flowering or harvesting stage. The classifier used in this process is One-Class Support Vector Machine (OC-SVM). Our experiments were conducted on tomato plants.

Featured Image