In many regions of the world, wheat is vulnerable to severe yield and quality losses from the fungus disease of Fusarium head blight (FHB). The development of resistant cultivars is one means of ameliorating the devastating effects of this disease, but the breeding process requires the evaluation of hundreds of lines each year for reaction to the disease. These field evaluations are laborious, expensive, time-consuming, and are prone to rater error. A phenotyping cart that can quickly capture images of the spikes of wheat lines and their level of FHB infection would greatly benefit wheat breeding programs. In this study, mask region convolutional neural network (Mask-RCNN) allowed for reliable identification of the symptom location and the disease severity of wheat spikes. Within a wheat line planted in the field, color images of individual wheat spikes and their corresponding diseased areas were labeled and segmented into sub-images. Images with annotated spikes and sub-images of individual spikes with labeled diseased areas were used as ground truth data to train Mask-RCNN models for automatic image segmentation of wheat spikes and FHB diseased areas, respectively. The feature pyramid network (FPN) based on ResNet-101 network was used as the backbone of Mask-RCNN for constructing the feature pyramid and extracting features. After generating mask images of wheat spikes from full-size images, Mask-RCNN was performed to predict diseased areas on each individual spike. This protocol enabled the rapid recognition of wheat spikes and diseased areas with the detection rates of 77.76% and 98.81%, respectively. The prediction accuracy of 77.19% was achieved by calculating the ratio of the wheat FHB severity value of prediction over ground truth. This study demonstrates the feasibility of rapidly determining levels of FHB in wheat spikes, which will greatly facilitate the breeding of resistant cultivars.

https://www.mdpi.com/2072-4292/13/1/26/pdf

Automatic evaluation of wheat resistance to fusarium head blight using dual mask-rcnn deep learning frameworks in computer vision

With the rapid development of genetic analysis techniques and crop population size, phenotyping has become the bottleneck restricting crop breeding. Breaking through this bottleneck will require phenomics, defined as the accurate, high-throughput acquisition and analysis of multi-dimensional phenotypes during crop growth at organism-wide levels, ranging from cells to organs, individual plants, plots, and fields. Here we offer an overview of crop phenomics research from technological and platform viewpoints at various scales, including microscopic, ground-based, and aerial phenotyping and phenotypic data analysis. We describe recent applications of high-throughput phenotyping platforms for abiotic/biotic stress and yield assessment. Finally, we discuss current challenges and offer perspectives on future phenomics research.

High-throughput phenotyping: Breaking through the bottleneck in future crop breeding

Automatic detection and severity analysis of grape black measles disease based on deep learning and fuzzy logic

An approach for characterization of infected area in tomato leaf disease based on deep learning and object detection technique

Assessment of the levels of damage caused by Fusarium head blight in wheat using an improved YoloV5 method

. Wheat crop productivity is susceptible to decrease from Fusarium head blight (FHB). To select a highly resistant cultivar, different wheat lines would be evaluated in a crop cycle. Shortages of field laborers for manual analysis of FHB severity is inspiring the design of more efficient assessment tools. A reliable phenocart with a high-throughput camera that works in different wheat crops to automatically determine the resistance of wheat lines to FHB would bring economic benefits to cereal production. In this study, Mask region convolutional neural network (Mask-RCNN) obtained acceptable results for detecting wheat FHB disease, yielding the overall rates of precision (P), recall (R), and F1-score of 72.10%, 76.16%, and 74.04%, respectively. The proposed method enabled the rapid recognitions of diseased wheat spikes with the detection rate of 98.81%. This study demonstrates the feasibility to rapidly determine the optimal wheat line from the field, which is of great significance for ensuring food safety and sustainable supply.

Jiajing Zhang

Papers

Automatic evaluation of wheat resistance to fusarium head blight using dual mask-rcnn deep learning frameworks in computer vision

Evaluation of mask RCNN for learning to detect fusarium head blight in wheat images