חיפוש מתקדם
Bulanon, D.M., Agricultural and Biological Engineering Department, University of Florida, Gainesville, FL 32611, United States
Burks, T.F., Agricultural and Biological Engineering Department, University of Florida, Gainesville, FL 32611, United States
Alchanatis, V., ARO-The Volcani Center, Bet Dagan, Israel
One of the major challenges in developing a machine vision system for robotic fruit harvesting is fruit visibility. Fruit trees such as oranges have a dense canopy, which can often result in partial or complete occlusion. This paper discusses improved visibility of oranges using multiple viewing angles. Fruit visibility was defined as the ratio of the number of fruits recognized in the image to the total number of fruits inside the region of interest, which was a section of tree canopy enclosed by a bounding box. Multiple images of the region of interest from different viewing angles were acquired and analyzed. Results from both manual and automatic recognition approaches showed 90% and 87% fruit visibility respectively. These levels are a significant improvement over earlier reports in literature which ranged from 65 to 80%.
פותח על ידי קלירמאש פתרונות בע"מ -
הספר "אוצר וולקני"
אודות
תנאי שימוש
Study on fruit visibility for robotic harvesting
8 BOOK
Bulanon, D.M., Agricultural and Biological Engineering Department, University of Florida, Gainesville, FL 32611, United States
Burks, T.F., Agricultural and Biological Engineering Department, University of Florida, Gainesville, FL 32611, United States
Alchanatis, V., ARO-The Volcani Center, Bet Dagan, Israel
Study on fruit visibility for robotic harvesting
One of the major challenges in developing a machine vision system for robotic fruit harvesting is fruit visibility. Fruit trees such as oranges have a dense canopy, which can often result in partial or complete occlusion. This paper discusses improved visibility of oranges using multiple viewing angles. Fruit visibility was defined as the ratio of the number of fruits recognized in the image to the total number of fruits inside the region of interest, which was a section of tree canopy enclosed by a bounding box. Multiple images of the region of interest from different viewing angles were acquired and analyzed. Results from both manual and automatic recognition approaches showed 90% and 87% fruit visibility respectively. These levels are a significant improvement over earlier reports in literature which ranged from 65 to 80%.
Scientific Publication
You may also be interested in