A Review of Design and Development of an Intelligent Groundnut Threshing Machine Using Machine Vision Technique

Abstract

Groundnut (Arachis hypogaea) is a vital leguminous crop and holds great importance in Nigeria, where it plays a key role in food security and supports rural livelihoods. Traditional threshing practices, including hand shelling, stick beating, and animal trampling, are labor-intensive, slow, and error-prone, thereby reducing productivity and market quality. Existing mechanical threshers, though an improvement, still face drawbacks such as seed breakage, minimal automation, and susceptibility of sensors to environmental conditions. This review focuses on the design and development of an intelligent groundnut threshing machine that applies machine vision techniques to enhance efficiency, minimize seed damage, and advance precision agriculture. By incorporating image processing, fuzzy logic control, and machine learning models, the system is designed to automatically identify groundnut varieties and dynamically adjust threshing parameters according to pod characteristics like size, shape, and texture. The review also examines recent progress in threshing technologies, noting persistent challenges in concave construction, motor regulation, and sensor reliability, while underlining the potential of artificial intelligence in agricultural mechanization. The integration of Raspberry Pi, high-resolution imaging, and supervised learning for feature extraction and classification makes the intelligent thresher a promising innovation. Ultimately, this approach promotes sustainable farming practices, increases productivity, and supports the overarching objectives of precision agriculture and food security in developing nations.

Keywords: Machine learning, Fuzzy logic control, Groundnut, intelligent, extraction, classification