Design and development of mango sorter with machine vision system
Poonpat Poonnoy and Ampawan Tansakul
Proceedings of the APEC symposium on postharvest handling systems Bangkok, Thailand September 1-3, 2003. p. 167
2003
บทคัดย่อ
Design and development of mango sorter with machine vision system
Mango is one of most favorite fruits for Thai exporting. The qualityof the exported mango follows Thai Standard for Mango which the mango is classified into 3 categories due to its size, shape, maturity and defect. Therefore, the sorting process is necessary for quality control purpose. This sorting task could be done by man but some errors may be found due to fatigue and experience of individual. This research applied machine vision system with number of selected colour pixel to classify size and defect. The co-ordinates that represent the position of stem, end, front and back sides of the mango supplying detailed data to classify its shape and maturity. Every step of sorting process has been controlled by computer software which was well designed and created on visual basic 6 program. The experiments were carried out with Nam Dok Mai mangoes. It was found that the software could be applied to estimate the size and shape. In comparison between results from machine vision system and those from human classifying capability, it appeared 94.9 percent coincidence for size and 80.3 percent coincidence for shape. For separating process, immature mango was resulted by using W1/L ratio, it was able to classify with 64.51 percent coincidence. For sorting ripe mango by specifying the level of yellow area at 10 percent of total area, it allowed 93.4 percent coincidence. For external defect inspection, the results from single-sided inspection and those from double-sided inspection were in good agreement within 94.1 percent.