Computer Algorithms for Automatic Segmentation of Abnormal Surfaces of Agricultural Products

Conduct research to develop efficient computer algorithms for automatic segmentation of computer images of fruits and poultry obtained by ISL multispectral and hyperspectral imaging systems and to test the algorithms on-line at pilot-scale processing systems.
<p>
<ol> <li>Thoroughly survey current available algorithms that perform automatic thresholding on monochrome image surfaces.
<li>Writing computer software in C++ language under Windows environment for automatic thresholding of image surfaces of apple and poultry meat.
<li>Implement the algorithms on the prototype detection systems.
<li>Test the algorithms for robustness and a real-time operation.</ol>