Determination of wheat kernels damaged by Fusarium head blight using monochromatic images of effective wavelengths from hyperspectral imaging coupled with an architecture self-search deep network
نویسندگان
چکیده
Wheat kernels damaged by Fusarium head blight (FHB) lose moisture, protein and starch carry dangerous toxins. Classification of degree damage can be helpful to customise the use wheat kernels, reduce financial losses ensure grain safety. In this study, hyperspectral imaging (HSI) deep learning network were explored determine sound, mildly, moderately severely kernels. Effective wavelengths (EWs) selected from reflectance spectroscopy HSI images ReliefF, uninformative variable elimination, random frog shuffled leaping algorithm, monochromatic different combinations EWs adopted develop classification models combined with an architecture self-search (ASSDN). ASSDN composed 941, 876 732 nm achieved best determination average accuracy 100% 98.31% in training prediction sets, respectively, outperforming other or methods. And area under curve 0.9985 indicated its excellent robustness. Using sporadic wavelengths, computation operation complexity evidently reduced, a simple custom-built instrument easily designed for practical recognition FHB-damaged Meanwhile, generate optimise high-performance itself, which is user–friendly largely expands application potential network.
منابع مشابه
Fusarium Head Blight of Wheat
Nebraska–Lincoln cooperating with the Counties and the United States Department of Agriculture. University of Nebraska–Lincoln Extension educational programs abide with the nondiscrimination policies of the University of Nebraska–Lincoln and the United States Department of Agriculture. © 2008, The Board of Regents of the University of Nebraska on behalf of the University of Nebraska–Lincoln Ext...
متن کاملFusarium Damaged Kernels Detection Using Transfer Learning on Deep Neural Network Architecture
The present work shows the application of transfer learning for a pre-trained deep neural network (DNN), using a small image dataset (≈ 12,000) on a single workstation with enabled NVIDIA GPU card that takes up to 1 hour to complete the training task and archive an overall average accuracy of 94.7%. The DNN presents a 20% score of misclassification for an external test dataset. The accuracy of ...
متن کاملDetection of Sprouted and Midge-damaged Wheat Kernels Using Near-infrared Hyperspectral Imaging
Sprout damage which results in poor bread making quality due to enzymatic activity of alpha-amylase is one of the important grading factors of wheat in Canada. The potential of near-infrared (NIR) hyperspectral imaging was investigated to detect sprouting of wheat kernels. Artificially sprouted, midge-damaged and healthy wheat kernels were scanned using NIR hyperspectral imaging system in the r...
متن کاملFusarium Head Blight Symptoms and Mycotoxin Levels in Single Kernels of Infected Wheat Spikes
متن کامل
Phylogeny and genetic diversity of Fusarium graminearum species complex associated with Fusarium head blight of wheat in Moghan plain (Iran)
Thirty-seven isolates of Fusarium graminearum species complexobtained from wheat heads with Fusarium head blight symptoms were selected and used for phylogenetic studies. They were collected from different localities of Moghan plain (Ardebil province, Iran). Partial sequences of translation elongation factor 1-alpha (TEF), putative reductase (RED) and UTP-ammonia ligase (URA) genes were amplifi...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Food Control
سال: 2022
ISSN: ['0956-7135', '1873-7129']
DOI: https://doi.org/10.1016/j.foodcont.2022.108819