Adapting a Stereo Camera System to Determine Coconut Ages

A coconut is one of the most well-known fruit of Thailand as indicated by the Office of Agricultural Economics, Ministry of Agriculture and Cooperatives and the national production volume of coconuts is significantly high as compared to other agricultural products. Particularly, in 2016 the value of the exportation of coconuts in Thailand is around 13,400 million baht [1]. At present, information technology plays a vital role as a driven force in various business environment. The technology is hence brought into various agricultural areas in order to efficiently improve the production of agricultural products. Examples of an IT adoption in the agricultural fields include strategy planning, plantation and farming, logistics, storage, and marketing. Coconuts were generally classified by using Saturation (S) channel in HSV color system for inconsistence light environment [2], [3]. To improve the quality of the classification process, Density parameter (D), unaffected by the light, will be used in this study. This is because existing literature suggests that density parameter can be a potential feature for classification, even if general IF-THEN rules is applied as a classifier. In addition, color data will still be used as supportive data to enhance the quality of the classification results. Based on the knowledge of local farmers, this research provides customers an assistance on coconut selection by using an adapted stereo camera system to identify three different types of coconuts, i.e., young, mature, and old coconuts.