Synergetic Innovation in Gherkin Cultivation Based on Machine Learning Algorithms

Gherkin cultivation, while a significant contributor to the agricultural sector, remained underutilized in Sri Lanka despite high international demand, particularly for pickles. Gherkin farmers faced many challenges from pests, diseases, and unpredictable harvests. The main reasons identified were farmers' lack of knowledge, different weather conditions, and inability to identify the danger. The proposed system harnesses cutting-edge machine learning algorithms, computer vision, and object detection to address key challenges in Gherkin production. The solution enabled pest identification and control, disease diagnosis, and accurate harvest estimation addressing the utilization of gherkin cultivation in Sri Lanka despite high international demand, especially for pickles. By leveraging YOLOv8, Random Forest, and CNN architectures, the system returned reliable predictions and recommendations for the farmers. The unique approach of this research lies in its capability to deliver high-accuracy predictions, practical recommendations provide quick solutions for pest and disease problems that arise during cultivation and give an overview of the yield at the start of cultivation to the farmers. The innovations presented not only address the progression of Gherkin cultivation but also establish a foundation for future smart agriculture advancements.