Using image analysis for quantitative assessment of needle bladder rust disease of Norway spruce

Abstract

High elevation spruce forests of the European Alps are frequently infected by the needle rust Chrysomyxa rhododendri, a pathogen causing remarkable defoliation, reduced tree growth and limited rejuvenation. Exact quantification of the disease severity on different spatial scales is crucial for monitoring, management and resistance breeding activities. Based on the distinct yellow discolouration of attacked needles, it was investigated whether image analysis of digital photographs can be used to quantify disease severity and to improve phenotyping compared to conventional assessment in terms of time, effort and application range. The developed protocol for preprocessing and analysis of digital RGB images enabled identification of disease symptoms and healthy needle areas on images obtained in ground surveys (total number of analysed images n = 62) and by the use of a semiprofessional quadcopter (n = 13). Obtained disease severities correlated linearly with results obtained by manual counting of healthy and diseased needles for all approaches, including images of individual branches with natural background (R 2 = 0.87) and with black background (R 2 = 0.95), juvenile plants (R 2 = 0.94), and top views and side views of entire tree crowns of adult trees (R 2 = 0.98 and 0.88, respectively). Results underline that a well‐defined signal related to needle bladder rust symptoms of Norway spruce can be extracted from images recorded by standard digital cameras and using drones. The presented protocol enables precise and time‐efficient quantification of disease symptoms caused by C. rhododendri and provides several advantages compared to conventional assessment by manual counting or visual estimations.

From visual estimates to fully automated sensor-based measurements of plant disease severity: status and challenges for improving accuracy

“…Image capture using mobile platforms (UAVs, ground robots etc) is being studied in the field, although disease detection is the primary focus (Johnson et al 2003;Garcia-Ruiz et al 2013;de Castro et al 2015). Measurement of severity with VIS spectrum image analysis using mobile platforms is less common (Lelong et al 2008;Sugiura et al 2016;Duarte-Carvajalino et al 2018;Franceschini et al 2019;Ganthaler et al 2018;Liu et al 2018), but is an area of research need. An automated VIS image analysis system on a UAV for measuring severity had moderate precision compared to visual rating (R 2 = 0.73), but was deemed acceptable for rating potato resistance to late blight (Sugiura et al 2016).…”

RUST: A Robust, User-Friendly Script Tool for Rapid Measurement of Rust Disease on Cereal Leaves

“…When the calibration and training steps were performed, the reliability and accuracy obtained by RUST was very high and comparable to that obtained by previous work. For instance, Ganthaler et al [ 44 ] reported coefficients of determination between 0.87 (natural background) and 0.95 (black background) when they compared the evaluation of the distinct yellow discoloration of rust attacked needles of spruce forest by image analysis and conventional methods. Similarly, Bock et al [ 45 ] reported bias correction factors between 0.93 and 0.99 in their comparison of the number of citrus canker lesions on grapefruit leaves estimated with Assess or by visual ratings, indicating that image analysis was more reliable when repeated compared to visual raters.…”

UAV-Based Forest Health Monitoring: A Systematic Review

“…Soil sampling was conducted to determine possible origins of nutritional deficiencies [109] and to measure the field capacity [147]. Ground-based photographs were taken to calculate canopy cover [148], for the documentation of weather conditions [137], and for an improved categorization of pest infestation [44,118,142,146], disease [122], and fire [149] severity classes. Smigaj et al [136] collected data from intratrunk water flow, canopy temperature, soil moisture, and incident and reflected light using an array of sensors.…”

“…In some cases, spectral data were enriched with structural information derived from point clouds, DSMs or CHMs based on either LiDAR [116][117][118] or image data [96,102,119,120], further improving classification results. In a few papers, the raw drone images were directly analyzed without further processing [44,108,[121][122][123][124][125][126]. To create photogrammetric products, the researchers predominantly implemented commercial SfM software such as Agisoft Metashape (Agisoft LLC, St. Petersburg, Russia) and Pix4D (Pix4D S.A., Lausanne, Switzerland).…”