|Reese, Daniel - UNIV OF MD, COLLEGE PARK|
|Lo, Martin - UNIV OF MD, COLLEGE PARK|
|Narayanan, Priya - UNIV OF MD, BALTIMORE CO|
Submitted to: Proceedings of SPIE
Publication Type: Proceedings
Publication Acceptance Date: September 8, 2007
Publication Date: October 12, 2007
Citation: Reese, D.Y., Lefcourt, A.M., Kim, M.S., Lo, M., Narayanan, P. 2007. Whole surface image reconstruction for machine vision inspection of fruit. Proceedings of SPIE Conference, Boston, MA 09/12/2007. Vol. 6761. Technical Abstract: Abstract: Automated imaging systems offer the potential to inspect the quality and safety of fruits and vegetables consumed by the public. Current automated inspection systems allow fruit such as apples to be sorted for quality issues including color and size by looking at a portion of the surface of each fruit. However, to inspect for defects and contamination, the whole surface of each fruit must be imaged. The goal of this project is to develop an economical method for whole surface imaging of apples using mirrors and a single camera to reduce costs. Challenges include mapping the concave stem and calyx regions. To allow the entire surface of an apple to be imaged, apples are suspended or rolled over the mirrors using two parallel piano wires. A camera above the mirrors captures 60 images per sec (640 by 480 pixels). Images can be used to create line-scan data at any selected offset by looking at the appropriate row in sets of sequential images. Single or multiple flat, concave, or convex mirrors are mounted under the apple at selected inclinations and positions to maximize imaging of the surface. Preliminary data suggest that the use of two flat mirrors provides inadequate coverage of a fruit but using three mirrors allows the entire surface to be mapped. Systems performance is analyzed in terms of localized pixel resolution and distortion in mapped images. The preliminary results suggest that a single camera with three mirrors can be a cost-effective method for whole surface imaging.