Identification of mature Nutmeg using Colour Space segmentation Algorithm

Abstract

Automation in Agriculture is essential to achieve a better quality harvest and alleviating the dependency on human workers. As the southern part of India is rich in spices cultivation, the work emphases on one of the spice which has got more medicinal value and commercial viability. The spice considered in this is nutmeg which is mostly cultivated in the mountain ranges of Kerala and Kanyakumari District. The identification of mature nutmeg in a large group is a bit time consuming task. The current fruit-picking methods takes a long time in fixing the clamp on the right fruit from the cluster of fruits, these leads to the cervical spondylitis problem. Recent developments in image processing and the extensive usage of autonomous platforms have provided the opportunity for fast and automatic harvesting machines. This paper proposes an image segmentation algorithm to identify the mature nutmeg. The datasets used are KAU Kochukudy, IISR Keralashree, Punnathanam and local clone. The matured fruit is identified by it’s boundaries using the boundary edge detection algorithm. The color detection method and colour space method were considered. Boundary Edge Detection algorithm is focused on identifying the edges or boundaries present in an image. Color Detection Method relies on identifying objects based on their color characteristics.Colour Space method involves converting the image from RGB color space (Red, Green, Blue) to a different color space, which may be more effective for certain types of analysis. Colour space methods outperforms image segmentation algorithm in terms of identification of matured nutmeg images with 96% accuracy. The average elapsed time was 1.150 secs. Also, the processing time to identify the right matured fruit requires much lesser time than other methods.