Detection of Fruits in Natural Background (Part 1)

Image Analysis for Detection of Fruits Edge

Abstract

Detection of fruits in a natural background is difficult when the colors of the fruits and back grounds, such as the leaves and stems, are similarly greenish. In this research, we tried to analyze the shape information as an alternative and used the papaya fruits (Carica papaya L.) in the orchard as the sample image to be investigated. The shape information could be evaluated from the boundary of the fruits. However, the straightforward application of the boundary detection technique can be sensitive to the edges of every object in the input image. Therefore, filtering method has to be performed to remove the extraneous edges. By using the edge and color information and the Maximum and Minimum filters simultaneously, the papaya boundaries could be detected.