Affiliation:
1. Sambalpur University, India
2. Guru Ghasidas Vishwavidyalaya, Bilaspur, India
3. Veer Surendra Sai University of Technology, India
Abstract
India is the second-largest food producer globally, trailing only in China. However, significant agricultural losses occur because of the lack of skilled laborers. Harvested commodities often go into waste. Additionally, the imprecise nature of crop identification, classification, and quality inspection, which is influenced by human subjectivity, poses challenges. To address these issues and reduce labor costs, the agricultural sector must embrace automation. Developing an automated system capable of distinguishing between various crops based on their texture, shape, and color is feasible by employing appropriate image-processing techniques and machine-learning methods. This study focuses on advancing the state-of-the-art research in this field. It briefly explores recent research publications' methodologies, comparing them using diverse techniques, such as k-nearest neighbors (KNN), artificial neural networks (ANN), random forest (RF), naive bayes (NB), backpropagation neural networks (BPNN), support vector machines (SVM), and convolutional neural networks (CNN).
Reference75 articles.
1. Machine learning approach for the classification of corn seed using hybrid features
2. Identification of haploid and diploid maize seeds using convolutional neural networks and a transfer learning approach
3. Agriculture sector performance: An analysis through the role of agriculture sector share in GDP. Journal of Agricultural Economics;M.Anwer;Extension and Rural Development,2015
4. An image processing technique for coffee black beans identification
5. Wheat class identification using computer vision system and artificial neural networks.;A.Arefi;International Agrophysics,2011