US 12,147,504 B2
Systems and methods for classifying mosquitoes based on extracted masks of anatomical components from images
Sriram Chellappan, Tampa, FL (US); Mona Minakshi, DeKalb, IL (US); Pratool Bharti, DeKalb, IL (US); and Ryan M Carney, Temple Terrace, FL (US)
Assigned to University of South Florida, Tampa, FL (US)
Filed by University of South Florida, Tampa, FL (US)
Filed on Aug. 31, 2021, as Appl. No. 17/462,809.
Prior Publication US 2023/0077353 A1, Mar. 16, 2023
Int. Cl. G06K 9/00 (2022.01); G06F 18/21 (2023.01); G06F 18/211 (2023.01); G06F 18/213 (2023.01); G06F 18/24 (2023.01); G06F 18/25 (2023.01); G06N 3/02 (2006.01); G06N 3/045 (2023.01); G06N 3/0464 (2023.01); G06N 3/09 (2023.01); G06N 20/20 (2019.01); G06V 10/32 (2022.01); G06V 10/422 (2022.01); G06V 10/82 (2022.01); G06V 40/10 (2022.01)
CPC G06F 18/24 (2023.01) [G06F 18/211 (2023.01); G06F 18/213 (2023.01); G06F 18/217 (2023.01); G06F 18/253 (2023.01); G06F 18/254 (2023.01); G06N 3/02 (2013.01); G06N 3/045 (2023.01); G06N 3/0464 (2023.01); G06N 3/09 (2023.01); G06N 20/20 (2019.01); G06V 10/32 (2022.01); G06V 10/422 (2022.01); G06V 10/82 (2022.01); G06V 40/10 (2022.01)] 16 Claims
OG exemplary drawing
 
1. A system for identifying a genus and species of an insect, the system comprising:
an imaging device configured to generate images of the insect;
a computer processor connected to a memory storing computer implemented commands in software, the memory receiving the images, wherein the software implements the following computerized method with respective images:
inputting the respective images to a first convolutional neural network, wherein at least a first feature map is obtained from a first layer of the first convolutional neural network and a second feature map is obtained from a second layer of the first convolutional neural network, wherein each of the first feature map and the second feature map is based on a subset of anatomical pixels at a corresponding image location, said subset of anatomical pixels corresponding to a body part of the insect, and wherein the first convolutional neural network is trained with a plurality of training images that each depict a single insect body part;
assigning a respective weight to each of the first feature map and the second feature map based on a class associated with the respective body part of the insect;
calculating an outer product of the first feature map and the second feature map;
forming an integrated feature map from the outer product of the first feature map and the second feature map; and
identifying the genus and the species of the insect based, at least in part, on the integrated feature map.