Page Banner

United States Department of Agriculture

Agricultural Research Service

Title: 3D Surface Reconstruction and Analysis in Automated Apple Stem-End/Calyx Identification

item Zhu, Bin
item Lu, Jiang
item Luo, Yaguang - Sunny
item Tao, Yang
item Cheng, Xuemei

Submitted to: International Journal of Pattern Recognition and Artificial Intelligence (IJRAI)
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 3/25/2008
Publication Date: 11/20/2009
Citation: Zhu, B., Lu, J., Luo, Y., Tao, Y., Cheng, X. 2009. 3D Surface Reconstruction and Analysis in Automated Apple Stem-End/Calyx Identification. International Journal of Pattern Recognition and Artificial Intelligence (IJRAI). 52(5):1775-1784.

Interpretive Summary: The apple industry has a high demand for automated apple defect sorting, because this technology significantly improves process efficiency and reduces labor cost. However, the difficulty in accurately differentiating naturally occurring apple stem and calyx ends from true defects using traditional two-dimensional, near-infrared imaging presents a major technical challenge for the apple industry. In this research, we employed a novel three dimensional data analysis strategy to distinguish the true defects from the stem and calyx ends which significantly improved detection accuracy. The implementation of this new process will improve machine vision technology so that it can be used successfully, enabling apple growers and shippers to accurately and efficiently sort out defective apples and provide high quality products to consumers.

Technical Abstract: Machine vision methods are widely used in apple defect detection and quality grading applications. Currently, 2D near-infrared (NIR) imaging of apples is often used to detect apple defects because the image intensity of the defect is different from the normal apple tissue. However, a drawback of this method is that the apple stem-end/calyx also exhibits similar image intensity to the apple defect. Since an apple stem-end/calyx often appears in the NIR image, the false alarm rate is high with the 2D NIR imaging method. In this paper, a novel two-step 3D data analysis strategy is developed so that the apple stem-end/calyx can be differentiated from apple defect according to their different 3D shape information. In the first step, a 2D NIR imaging method is extended to a 3D reconstruction according to the Shape-From-Shading (SFS) approach. Given successfully recovered 3D information, in the second step, a quadratic facet model is introduced to do the 3D concave shape fitting such that the identification of apple stem-end and calyx can be achieved based on their different 3D structure. Significant improvement in terms of the detection rate can be obtained based on 3D shape fitting comparing to the traditional 2D intensity fitting approach. The reconstructed 3D apple surface maps as well as the identified stem-end/calyx are shown in the results.

Last Modified: 10/19/2017
Footer Content Back to Top of Page