Deep Learning Fruit Segmentation - Mangoes
Deep learning a fruit detection model using a hand labelled dataset of 21 images of size 2400 x 1600. Tested on a video stream of mango crop data (recorded at roughly 15Hz).
This project is part of the Precision Agriculture team at Kumar Lab (GRASP, University of Pennsylvania).
Please ignore the constant heat region in the lower right bottom corner.
For more details refer to the publication:
S. W. Chen et al., "Counting Apples and Oranges With Deep Learning: A Data-Driven Approach," in IEEE Robotics and Automation Letters, vol. 2, no. 2, pp. 781-788, April 2017.
URL: http://ieeexplore.ieee.org/stamp/stam...
Precision Agriculture at Penn:
http://label.ag/wiki/index.php/Penn_A...