ABSTRACT
For any country, the backbone of its economy depends on the percentage of people involved in agriculture. However, many major agricultural regions are termed underdeveloped because of many factors, like lack or no use of modernized technologies. Seed classification is still done with the farmers’ basic knowledge, which is proven to have no mechanical validations and hence is considered inefficient. The present research presents a mechanism for assessing the quality of rice seed that can help provide better crop production. Machine learning (ML), a subset of artificial intelligence (AI), is used for learning the data that is used for making predictions, recognizing patterns, making simulations in the real world, and classifying the input data. Rice is the major source of food for a total of 80% of the population of the world. Rice is grown in different varieties; hence, detecting faulty seeds and distinguishing among the varieties is another important farming aspect. This research elaborates on a method that can be used efficiently for extracting the features of rice seeds by their identification and classification using digital imaging. The process involved is filtering, segmentation, and edge detection as pre-processing techniques.
