An Automated paddy variety identification system

Abstract

An automated system has been developed to analyze and identify the variety of paddy largely grown in Sri Lanka by using digital image processing and statistical methods. In this design, the process is divided into three stages namely: Filtering, Feature Extraction and analyzing. In the first stage, the acquired image is separated from noise and background area. Feature extraction stage includes techniques to gather useful data in the form of numerical values. In the final stage, the extracted data is compared with the data which is already collected to identify paddy varieties. In this work, about six varieties were taken into consideration and were analyzed. This research work achieved a satisfactory level of success. Future expansion of this work using neural network is also suggested.