RGB color space based machine vision system for Classification of guava fruits

In the present work, a general approach is developed to estimate the ripeness level of Guava fruit without touching it. This workdeals with the study of using an artificial computer based vision system as a non-destructive instrument to analyze fruit ripenessstage. The cultivar chosen for this study is Guava fruit. This paper presents a simple method that uses a combination of digital web camera, computer, and self developed graphics software to measure and analyze the surface color of fruits. The method has also the advantages of being versatile and affordable. The images of the fruits can be displayed on computer screen or printed on paper for qualitative analysis of color and structure. Quantitative information such as RGB color distribution, Chromaticity coordinates according to CIE1931 standard and averages (in terms of L_, a_ and b_ values) can also be determined readily.