Publications by authors named "Kaaviya Velumani"

Applying deep learning to images of cropping systems provides new knowledge and insights in research and commercial applications. Semantic segmentation or pixel-wise classification, of RGB images acquired at the ground level, into vegetation and background is a critical step in the estimation of several canopy traits. Current state of the art methodologies based on convolutional neural networks (CNNs) are trained on datasets acquired under controlled or indoor environments.

View Article and Find Full Text PDF
Article Synopsis
  • SegVeg is a new approach for segmenting high-resolution RGB images into three vegetation classes: background, green, and senescent, by using a U-net model combined with SVM for enhanced accuracy.
  • The method shows good performance, accurately estimating the green fraction (0.94) but struggles slightly with the senescent (0.70) and background (0.73) fractions, especially in varying light regions.
  • The SegVeg model and the annotated dataset are publicly available for researchers to utilize, aiming to improve the assessment of vegetation traits.
View Article and Find Full Text PDF

The Global Wheat Head Detection (GWHD) dataset was created in 2020 and has assembled 193,634 labelled wheat heads from 4700 RGB images acquired from various acquisition platforms and 7 countries/institutions. With an associated competition hosted in Kaggle, GWHD_2020 has successfully attracted attention from both the computer vision and agricultural science communities. From this first experience, a few avenues for improvements have been identified regarding data size, head diversity, and label reliability.

View Article and Find Full Text PDF