BackgroundField phenotyping by remote sensing has received increased interest in recent years with the possibility of achieving high-throughput analysis of crop fields. Along with the various technological developments, the application of machine learning methods for image analysis has enhanced the potential for quantitative assessment of a multitude of crop traits. For wheat breeding purposes, assessing the production of wheat spikes, as the grain-bearing organ, is a useful proxy measure of grain production. Thus, being able to detect and characterize spikes from images of wheat fields is an essential component in a wheat breeding pipeline for the selection of high yielding varieties. ResultsWe have applied a deep learning approach to accurately detect, count and analyze wheat spikes for yield estimation. We have tested the approach on a set of images of wheat field trial comprising 10 varieties subjected to three fertilizer treatments. The images have been captured over one season, using high definition RGB cameras mounted on a land-based imaging platform, and viewing the wheat plots from an oblique angle. A subset of in-field images has been accurately labeled by manually annotating all the spike regions. This annotated dataset, called SPIKE, is then used to train four region-based Convolutional Neural Networks (R-CNN) which take, as input, images of wheat plots, and accurately detect and count spike regions in each plot. The CNNs also output the spike density and a classification probability for each plot. Using the same R-CNN architecture, four different models were generated based on four different datasets of training and testing images captured at various growth stages. Despite the challenging field imaging conditions, e.g., variable illumination conditions, high spike occlusion, and complex background, the four R-CNN models achieve an average detection accuracy ranging from 88 to across different sets of test images. The most robust R-CNN model, which achieved the highest accuracy, is then selected to study the variation in spike production over 10 wheat varieties and three treatments. The SPIKE dataset and the trained CNN are the main contributions of this paper.ConclusionWith the availability of good training datasets such us the SPIKE dataset proposed in this article, deep learning techniques can achieve high accuracy in detecting and counting spikes from complex wheat field images. The proposed robust R-CNN model, which has been trained on spike images captured during different growth stages, is optimized for application to a wider variety of field scenarios. It accurately quantifies the differences in yield produced by the 10 varieties we have studied, and their respective responses to fertilizer treatment. We have also observed that the other R-CNN models exhibit more specialized performances. The data set and the R-CNN model, which we make publicly available, have the potential to greatly benefit plant breeders by facilitating the high throughput selection of high yielding varieties.Electronic su...

Abstract:This study evaluates an aerial and ground imaging platform for assessment of canopy development in a wheat field. The dependence of two canopy traits, height and vigour, on fertilizer treatment was observed in a field trial comprised of ten varieties of spring wheat. A custom-built mobile ground platform (MGP) and an unmanned aerial vehicle (UAV) were deployed at the experimental site for standard red, green and blue (RGB) image collection on five occasions. Meanwhile, reference field measurements of canopy height and vigour were manually recorded during the growing season. Canopy level estimates of height and vigour for each variety and treatment were computed by image analysis. The agreement between estimates from each platform and reference measurements was statistically analysed. Estimates of canopy height derived from MGP imagery were more accurate (RMSE = 3.95 cm, R 2 = 0.94) than estimates derived from UAV imagery (RMSE = 6.64 cm, R 2 = 0.85). In contrast, vigour was better estimated using the UAV imagery (RMSE = 0.057, R 2 = 0.57), compared to MGP imagery (RMSE = 0.063, R 2 = 0.42), albeit with a significant fixed and proportional bias. The ability of the platforms to capture differential development of traits as a function of fertilizer treatment was also investigated. Both imaging methodologies observed a higher median canopy height of treated plots compared with untreated plots throughout the season, and a greater median vigour of treated plots compared with untreated plots exhibited in the early growth stages. While the UAV imaging provides a high-throughput method for canopy-level trait determination, the MGP imaging captures subtle canopy structures, potentially useful for fine-grained analyses of plants.

The morphology of plant root anatomical features is a key factor in effective water and nutrient uptake. Existing techniques for phenotyping root anatomical traits are often based on manual or semi-automatic segmentation and annotation of microscopic images of root cross sections. In this article, we propose a fully automated tool, hereinafter referred to as RootAnalyzer, for efficiently extracting and analyzing anatomical traits from root-cross section images. Using a range of image processing techniques such as local thresholding and nearest neighbor identification, RootAnalyzer segments the plant root from the image’s background, classifies and characterizes the cortex, stele, endodermis and epidermis, and subsequently produces statistics about the morphological properties of the root cells and tissues. We use RootAnalyzer to analyze 15 images of wheat plants and one maize plant image and evaluate its performance against manually-obtained ground truth data. The comparison shows that RootAnalyzer can fully characterize most root tissue regions with over 90% accuracy.

In this paper we report on an automated procedure to capture and characterize the detailed structure of a crop canopy by means of stereo imaging. We focus attention specifically on the detailed characteristic of canopy height distribution—canopy shoot area as a function of height—which can provide an elaborate picture of canopy growth and health under a given set of conditions. We apply the method to a wheat field trial involving ten Australian wheat varieties that were subjected to two different fertilizer treatments. A novel camera self-calibration approach is proposed which allows the determination of quantitative plant canopy height data (as well as other valuable phenotypic information) by stereo matching. Utilizing the canopy height distribution to provide a measure of canopy height, the results compare favourably with manual measurements of canopy height (resulting in an R2 value of 0.92), and are indeed shown to be more consistent. By comparing canopy height distributions of different varieties and different treatments, the methodology shows that different varieties subjected to the same treatment, and the same variety subjected to different treatments can respond in much more distinctive and quantifiable ways within their respective canopies than can be captured by a simple trait measure such as overall canopy height.

BackgroundOne of the main challenges associated with image-based field phenotyping is the variability of illumination. During a single day’s imaging session, or between different sessions on different days, the sun moves in and out of cloud cover and has varying intensity. How is one to know from consecutive images alone if a plant has become darker over time, or if the weather conditions have simply changed from clear to overcast? This is a significant problem to address as colour is an important phenotypic trait that can be measured automatically from images.ResultsIn this work we use an industry standard colour checker to balance the colour in images within and across every day of a field trial conducted over four months in 2016. By ensuring that the colour checker is present in every image we are afforded a ‘ground truth’ to correct for varying illumination conditions across images. We employ a least squares approach to fit a quadratic model for correcting RGB values of an image in such a way that the observed values of the colour checker tiles align with their true values after the transformation.ConclusionsThe proposed method is successful in reducing the error between observed and reference colour chart values in all images. Furthermore, the standard deviation of mean canopy colour across multiple days is reduced significantly after colour correction is applied. Finally, we use a number of examples to demonstrate the usefulness of accurate colour measurements in recording phenotypic traits and analysing variation among varieties and treatments.Electronic supplementary materialThe online version of this article (10.1186/s13007-018-0308-5) contains supplementary material, which is available to authorized users.