Jednipat Moonrinta

3D modeling for agricultural field inspection robot - Pathum Thani, Thailand : Asian Institute of Technology, 2010 - 33 p. : ill. - Thesis ; no. CS-10-04 . - Asian Institute of Technology. Thesis ; no. CS-10-04 .

A thesis submitted in partial fulfillment of the requirements for the degree of Master of Science in Computer Science, School of Engineering and Technology

Thesis (M.Eng.) - Asian Institute of Technology, 2010

Through automated agricultural inspection, farmers can potentially achieve better produc-tivity and accurately predict yields and crop quality. The cheapest device for collecting richinformation about a crop is the video camera. This thesis focuses on the design, implemen-tation, and evaluation of algorithms for extracting information about a pineapple crop froma monocular webcam fixed to a mobile inspection platform. The process begins with fruitdetection and tracking in video sequence. I propose a 3D modeling algorithm that workswith the tracking history to generate a 3D point cloud for each fruit using structure-from-motion techniques in computer vision. The 3D modeling algorithm consists of two mainprocedures: 3D point cloud estimation and 3D fruit shape reconstruction. Once the 3D pointcloud for a fruit is obtained, the fruit shape reconstruction method estimates the shape andsize of the fruit using a robust ellipsoid estimation technique. A series of experiments showsthat the method produces a reasonably accurate 3D field map despite only partial and oc-cluded views of the fruit. The prototype is thus a promising step towards the realization ofautomated mobile in-field agricultural inspection systems.


Three-dimensional imaging
Robot vision
Camcorders