US 11,880,981 B2
Method and system for leaf age estimation based on morphological features extracted from segmented leaves
Prakruti Vinodchandra Bhatt, Thane West (IN); Sanat Sarangi, Thane West (IN); Srinivasu Pappula, Hyderabad (IN); and Avil Saunshi, Bangalore (IN)
Assigned to TATA CONSULTANCY SERVICES LIMITED, Mumbai (IN)
Filed by Tata Consultancy Services Limited, Mumbai (IN)
Filed on Sep. 2, 2021, as Appl. No. 17/446,725.
Claims priority of application No. 202021046224 (IN), filed on Oct. 23, 2020.
Prior Publication US 2022/0130051 A1, Apr. 28, 2022
Int. Cl. G06T 7/13 (2017.01); G06T 7/187 (2017.01); G06V 10/56 (2022.01); G06V 10/26 (2022.01); G06T 5/00 (2006.01); G06F 18/2413 (2023.01)
CPC G06T 7/13 (2017.01) [G06F 18/24147 (2023.01); G06T 5/002 (2013.01); G06T 5/009 (2013.01); G06T 7/187 (2017.01); G06V 10/267 (2022.01); G06V 10/56 (2022.01); G06T 2207/30188 (2013.01)] 18 Claims
OG exemplary drawing
 
1. A processor implemented method comprising the steps of:
receiving, via one or more hardware processors, an image including one or more leaves that need to be monitored;
segmenting, via the one or more hardware processors, the received image to identify veins in the one or more leaves, wherein identifying veins comprises:
performing edge detection on the received image to obtain edges representative of veins and contours of the one or more leaves;
determining straight lines from the obtained edges; and
identifying the determined straight lines as veins if a difference between average pixel intensity on either side of the straight lines is lesser than an empirically determined first threshold;
obtaining, via the one or more hardware processors, a de-veined image by replacing each pixel on the identified veins with an average pixel intensity of neighboring pixels;
selecting, via the one or more hardware processors, a seed point for each of the one or more leaves using distance transform on a binary image of the de-veined image; and
performing region growing, via the one or more hardware processors, using an k-neighborhood method, to obtain contours of one or more full leaves, wherein the region growing starts from the seed point selected for each of the one or more leaves.