Maize(Zea mays L.), also known as corn, is the third most cultivated crop in the world. Northern corn leaf blight(NCLB) is a globally devastating maize foliar disease caused by Setosphaeria turcica(Luttrell) Leonard a...Maize(Zea mays L.), also known as corn, is the third most cultivated crop in the world. Northern corn leaf blight(NCLB) is a globally devastating maize foliar disease caused by Setosphaeria turcica(Luttrell) Leonard and Suggs. Early intelligent diagnosis and warning is an effective and economical strategy to control this disease. Today, deep learning is beginning to play an essential role in agriculture. Notably, deep convolutional neural networks(DCNN) are amongst the most successful machine learning techniques in plant disease detection and diagnosis. Our study aims to identify NCLB in the maize-producing area in Jilin Province based on several DCNN models. We established a database of 985 leaf images of healthy and infected maize and applied data augmentation techniques including image segmentation, image resizing, image cropping, and image transformation, to expand to 30 655 images. Several proven convolutional neural networks, such as AlexNet, Google Net, VGG16, and VGG19, were then used to identify diseases. Based on the best performance of the DCNN pre-trained model Google Net, some of the recent loss functions developed for deep facial recognition tasks such as Arc Face, Cos Face, and A-Softmax were applied to detect NCLB. We found that a pre-trained Google Net architecture with the Softmax loss function can achieve an excellent accuracy of 99.94% on NCLB diagnosis. The analysis was implemented in Python with two deep learning frameworks, Pytorch and Keras. The techniques, training, validation, and test results are presented in this paper. Overall, our study explores intelligent identification technology for NCLB and effectively diagnoses NCLB from images of maize.展开更多
The northem corn leaf blight is one of the most serious diseases in maize production. The research progress on etiology, generation law and control strategy, antigen identification, resistance mechanism and inheritanc...The northem corn leaf blight is one of the most serious diseases in maize production. The research progress on etiology, generation law and control strategy, antigen identification, resistance mechanism and inheritance of this disease was discussed. And the research work which should be enhanced in China was pointed out, such as mechanism of resistance inheritance, developing function marker, gene mining, screening resistance resource and dominant physiological race in different areas.展开更多
Maize is cultivated extensively throughout the world and has the highest production among cereals.However,Northern corn leaf blight(NCLB)disease caused by Exherohilum turcicum,is the most devastating limiting factor o...Maize is cultivated extensively throughout the world and has the highest production among cereals.However,Northern corn leaf blight(NCLB)disease caused by Exherohilum turcicum,is the most devastating limiting factor of maize production.The disease causes immense losses to corn yield if it develops prior or during the tasseling and silking stages of crop development.It has a worldwide distribution and its development is favoured by cool to moderate temperatures with high relative humidity.The prevalence of the disease has increased in recent years and new races of the pathogen have been reported worldwide.The fungus E.turcicum is highly variable in nature.Though different management strategies have proved effective to reduce economic losses from NCLB,the development of varieties with resistance to E.turcicum is the most efficient and inexpensive way for disease management.Qualitative resistance for NCLB governed by Ht genes is a race-specific resistance which leads to a higher level of resistance.However,some Ht genes can easily become ineffective under the high pressure of virulent strains of the pathogen.Hence,it is imperative to understand and examine the consistency of the genomic locations of quantitative trait loci for resistance to NCLB in diverse maize populations.The breeding approaches for pyramiding resistant genes against E.turcicum in maize can impart NCLB resistance under high disease pressure environments.Furthermore,the genome editing approaches like CRISPR-cas9 and RNAi can also prove vital for developing NCLB resistant maize cultivars.As such this review delivers emphasis on the importance and current status of the disease,racial spectrum of the pathogen,genetic nature and breeding approaches for resistance and management strategies of the disease in a sustainable manner.展开更多
基金financially supported by the Key Planning Projects on Science and Technology of Jilin Province,China(20180201012NY)the Inter-Governmental International Cooperation Special Project of National Key R&D Program of China(2019YFE0114200)the National Key R&D Program of China(2017YFD0201802)。
文摘Maize(Zea mays L.), also known as corn, is the third most cultivated crop in the world. Northern corn leaf blight(NCLB) is a globally devastating maize foliar disease caused by Setosphaeria turcica(Luttrell) Leonard and Suggs. Early intelligent diagnosis and warning is an effective and economical strategy to control this disease. Today, deep learning is beginning to play an essential role in agriculture. Notably, deep convolutional neural networks(DCNN) are amongst the most successful machine learning techniques in plant disease detection and diagnosis. Our study aims to identify NCLB in the maize-producing area in Jilin Province based on several DCNN models. We established a database of 985 leaf images of healthy and infected maize and applied data augmentation techniques including image segmentation, image resizing, image cropping, and image transformation, to expand to 30 655 images. Several proven convolutional neural networks, such as AlexNet, Google Net, VGG16, and VGG19, were then used to identify diseases. Based on the best performance of the DCNN pre-trained model Google Net, some of the recent loss functions developed for deep facial recognition tasks such as Arc Face, Cos Face, and A-Softmax were applied to detect NCLB. We found that a pre-trained Google Net architecture with the Softmax loss function can achieve an excellent accuracy of 99.94% on NCLB diagnosis. The analysis was implemented in Python with two deep learning frameworks, Pytorch and Keras. The techniques, training, validation, and test results are presented in this paper. Overall, our study explores intelligent identification technology for NCLB and effectively diagnoses NCLB from images of maize.
文摘The northem corn leaf blight is one of the most serious diseases in maize production. The research progress on etiology, generation law and control strategy, antigen identification, resistance mechanism and inheritance of this disease was discussed. And the research work which should be enhanced in China was pointed out, such as mechanism of resistance inheritance, developing function marker, gene mining, screening resistance resource and dominant physiological race in different areas.
文摘Maize is cultivated extensively throughout the world and has the highest production among cereals.However,Northern corn leaf blight(NCLB)disease caused by Exherohilum turcicum,is the most devastating limiting factor of maize production.The disease causes immense losses to corn yield if it develops prior or during the tasseling and silking stages of crop development.It has a worldwide distribution and its development is favoured by cool to moderate temperatures with high relative humidity.The prevalence of the disease has increased in recent years and new races of the pathogen have been reported worldwide.The fungus E.turcicum is highly variable in nature.Though different management strategies have proved effective to reduce economic losses from NCLB,the development of varieties with resistance to E.turcicum is the most efficient and inexpensive way for disease management.Qualitative resistance for NCLB governed by Ht genes is a race-specific resistance which leads to a higher level of resistance.However,some Ht genes can easily become ineffective under the high pressure of virulent strains of the pathogen.Hence,it is imperative to understand and examine the consistency of the genomic locations of quantitative trait loci for resistance to NCLB in diverse maize populations.The breeding approaches for pyramiding resistant genes against E.turcicum in maize can impart NCLB resistance under high disease pressure environments.Furthermore,the genome editing approaches like CRISPR-cas9 and RNAi can also prove vital for developing NCLB resistant maize cultivars.As such this review delivers emphasis on the importance and current status of the disease,racial spectrum of the pathogen,genetic nature and breeding approaches for resistance and management strategies of the disease in a sustainable manner.
