AI-Powered Drone Innovation for Crop Yield Estimation in the Era of Climate Change
Description: This innovation utilizes an AI-powered drone system to estimate paddy yields in agricultural fields based on aerial imagery data. The system employs computer vision and machine learning techniques to analyze field images captured by the drone, enabling accurate yield predictions to address rice shortages and price fluctuations.
In the face of dwindling rice stocks and soaring prices, exacerbated by climate change and crop failures, this research aims to develop an innovative AI-powered drone solution for paddy yield estimation. The proposed system leverages the rapidly advancing drone technology, integrating artificial intelligence capabilities to tackle agricultural challenges. The core objective is to create a drone system capable of estimating paddy yields in rice fields based on aerial imagery data acquired by the drone. While previous research has explored the application of deep learning, machine learning, and optical flow methods in drone-based agriculture, this study focuses specifically on leveraging AI to predict crop yields accurately. The lack of AI-driven drone technologies tailored for agricultural yield estimation is a significant gap, especially in light of Indonesia's rice shortage and price hikes. By utilizing aerial field images captured by the drone, the system will implement computer vision and machine learning techniques to analyze the color patterns of paddy crops, employing the camera's field of view (FOV) method. The FOV method will convert the image data into actual field area measurements. Subsequently, grayscaling techniques will be applied to separate the yellow and green areas within the field, representing different crop maturity stages. By combining the field area information with the color-based segmentation, the system can estimate the crop yield, which will be converted into weight measurements. This AI-powered drone solution holds the potential to address agricultural challenges by providing accurate and timely yield predictions, enabling better planning and resource allocation to mitigate rice shortages and stabilize prices.
Organisation: Universitas Nahdlatul Ulama Sidoarjo
Innovator(s): Iyung Fachrur Rozi, Muhammad Badruz Zaman, Habib Maulana Syah, M Faizal Zhafran Farros, Norma Zaneta Lia Karisma
Category: Agriculture and Aquaculture
Country: Indonesia
