Automatic Counting of Wheat Spikes from Wheat Growth Images

Alharbi, Najmah, Zhou, Ji and Wang, Wenjia (2018) Automatic Counting of Wheat Spikes from Wheat Growth Images. In: 7th International Conference on Pattern Recognition Applications and Methods. SCITEPRESS – Science and Technology Publications, pp. 346-355. (In Press)

[img] PDF (Accepted manuscript) - Submitted Version
Restricted to Repository staff only until 31 December 2099.

Download (4MB)

    Abstract

    This study aims to develop an automated screening system that can estimate the number of wheat spikes (i.e. ears) from a given wheat plant image acquired after the flowering stage. The platform can be used to assist the dynamic estimation of wheat yield potential as well as grain yield based on wheat images captured by the CropQuant platform. Our proposed system framework comprises three main stages. Firstly, it transforms the wheat plant raw image data using colour index of vegetation extraction (CIVE) and then segments wheat ear regions from the image to reduce the influence of the background signals. Secondly, it detects wheat ears using Gabor filter banks and K-means clustering algorithm. Finally, it estimates the number of wheat spikes within extracted wheat spike region through a regression method. The framework is tested with a real-world dataset of wheat growth images equally distributed from flowering to ripening stages. The estimations of the wheat ears were benchmarked against the ground truth produced in this study by human manual counting. Our automatic counting system achieved an average accuracy of 90.7% with a standard deviation of 0.055, at a much faster speed than human experts and hence the system has a potential to be improved for agricultural applications on wheat growth studies in the future.

    Item Type: Book Section
    Uncontrolled Keywords: wheat spikes counting,gabor filter,k-means,segmentation,regression analysis
    Faculty \ School: Faculty of Science > School of Computing Sciences
    University of East Anglia > Faculty of Science > Research Groups > Computational Biology (subgroups are shown below) > Machine learning in computational biology
    ?? RGMLS ??
    Depositing User: Pure Connector
    Date Deposited: 12 Jan 2018 12:30
    Last Modified: 15 Jan 2018 00:37
    URI: https://ueaeprints.uea.ac.uk/id/eprint/65922
    DOI:

    Actions (login required)

    View Item