{"id":344,"date":"2022-06-30T20:29:46","date_gmt":"2022-06-30T20:29:46","guid":{"rendered":"https:\/\/cse.unl.edu\/~rquinones\/?page_id=344"},"modified":"2026-01-30T11:00:02","modified_gmt":"2026-01-30T17:00:02","slug":"my-dissertation","status":"publish","type":"page","link":"https:\/\/cs.siue.edu\/~rquinon\/my-dissertation\/","title":{"rendered":"My Dissertation"},"content":{"rendered":"\n

As the global population is projected to exceed 9 billion by 2050 and crop production is growing at just 1.3% annually amid climate change, developing crops that can withstand stress environments is crucial for food security. Testing crop resilience requires fast, efficient, and accurate data collection, which can be achieved through high-throughput imaging systems and advanced segmentation techniques.<\/p>\n\n\n\n

First, I defend the novel integration of co-segmentation (2008) and plant phenotyping by performing a performance study against state-of-the-art segmentation methods. Link to work<\/a>. I also provide an open-sourced multi-dimensional dataset, CosegPP<\/a>.<\/p>\n\n\n\n

Second, I advance co-segmentation by leveraging the latest state-of-the-art algorithms to create a framework that increases accuracy in plant images by 3% to 45%. Link to work<\/a>. My novel framework can handle high-dimensional datasets, is end-to-end trainable, and is unsupervised. Our code<\/a> is open-sourced, and our novel CosegPP+ dataset<\/a> is open-sourced.<\/p>\n\n\n\n

Third, I introduced concurrent imaging and unidimensional phenotypes that are capable of capturing dynamic morphological and physiological responses that are not observable with conventional single-image or cumulative metrics. Link to work<\/a>. I also provide two open-sourced datasets, SIMID and SIPID<\/a>.<\/p>\n\n\n\n

\n
\"\"<\/figure>\n\n\n\n
\"\"<\/figure>\n\n\n\n
\"\"<\/figure>\n
These are images of an automated, high-throughput plant phenotyping platform (LemnaTec Scanalyzer 3D) at the University of Nebraska-Lincoln. For more information on the system (and the source of the images) visit: https:\/\/ard.unl.edu\/phenotyping\/nebraska-innovation-campus-greenhouse <\/a><\/figcaption><\/figure>\n\n\n\n

Citations<\/strong><\/h3>\n\n\n\n

If you use any of my dissertation work, please cite the appropriate citations: <\/strong><\/em><\/p>\n\n\n\n