Machine Vision-Based Assessment of Fall Color Changes and its Relationship with Leaf Nitrogen Concentration

TL;DR

The paper addresses non-destructive, tree-level estimation of leaf nitrogen status in apple orchards by linking fall canopy color changes to nitrogen content using a machine-vision system. It collects RGB-D data from a ground vehicle, segments canopies into Green, Yellow, and Trunk regions with a gradient-boosted classifier to compute a yellowness index. The authors report that the yellowness index can be estimated with $R^2 = 0.72$ and that its relationship with leaf nitrogen is generally weak but can show moderate associations in specific weeks, with timing influenced by temperature. The approach offers a scalable, low-cost alternative for precision nitrogen management, with potential gains from fusing yellowness with additional canopy metrics.

Abstract

Apple(\textit{Malus domestica} Borkh.) trees are deciduous, shedding leaves each year. This process is preceded by a gradual change in leaf color from green to yellow as chlorophyll is degraded prior to abscission. The initiation and rate of this color change are affected by many factors including leaf nitrogen (N) concentration. We predict that leaf color during this transition may be indicative of the nitrogen status of apple trees. This study assesses a machine vision-based system for quantifying the change in leaf color and its correlation with leaf nitrogen content. An image dataset was collected in color and 3D over five weeks in the fall of 2021 and 2023 at a commercial orchard using a ground vehicle-based stereovision sensor. Trees in the foreground were segmented from the point cloud using color and depth thresholding methods. Then, to estimate the proportion of yellow leaves per canopy, the color information of the segmented canopy area was quantified using a custom-defined metric, \textit{yellowness index} (a normalized ratio of yellow to green foliage in the tree) that varied from -1 to +1 (-1 being completely green and +1 being completely yellow). Both K-means-based methods and gradient boosting methods were used to estimate the \textit{yellowness index}. The gradient boosting based method proposed in this study was better than the K-means-based method (both in terms of computational time and accuracy), achieving an $R^2$ of 0.72 in estimating the \textit{yellowness index}. The metric was able to capture the gradual color transition from green to yellow over the study duration. Trees with lower leaf nitrogen showed the color transition to yellow earlier than the trees with higher nitrogen. Keywords: Fruit Tree Nitrogen Management, Machine Vision, Point Cloud Segmentation, Precision Nitrogen Management

Figures (15)

• Figure 1: A representative high density apple orchard plantation in Prosser, Washington. The trees have been trained in a tall spindle architecture and spaced closely ( 4 foot) forming a highly dense planar “wall”.
• Figure 2: Overall data collection and processing steps used in the study; Each light-colored box represents a distinct process and has been described further in the subsequent sections.
• Figure 3: Study site and sensing setup; (a) An aerial view of the test plot. The red outline shows the location of the plot used in this study., and b) Ground vehicle with a camera mounted on top (zoomed portion shows camera and axes orientations)
• Figure 4: Images of a sample apple tree acquired over the data collection period in 2023. The foliage can be seen gradually changing from green to yellow.
• Figure 5: Distribution of a) Hue angles and b) a* and c) b* values for one of the sample trees during the five-week study period of 2023. The color chart shows the color associated with different values for (a) hue angles, (b) thompson_how_2017 a*, and b*. The hue angle decreased in a) and a* and b* increased in b) and c) as the weeks progressed during the study all signifying the shift from green to yellow.