Impact of Training Data Distribution on Plant Diseases Recognition Based on Deep Learning
【Objective】The study was carried out to improve the accuracy of the deep learning model through adjusting the distribution of training dataset of lab-condition and field-condition images, to reduce the dependence of plant diseases recognition models on field-condition data.【Method】The plant diseases...
| Main Authors: | , , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Guangdong Academy of Agricultural Sciences
2022-06-01
|
| Series: | Guangdong nongye kexue |
| Subjects: | |
| Online Access: | http://gdnykx.cnjournals.org/gdnykx/ch/reader/view_abstract.aspx?file_no=202206013 |
| _version_ | 1797802590853922816 |
|---|---|
| author | Hongle WANG Xinglin WANG Wenbo LI Quanzhou YE Yonghai LIN Hui XIE Lie DENG |
| author_facet | Hongle WANG Xinglin WANG Wenbo LI Quanzhou YE Yonghai LIN Hui XIE Lie DENG |
| author_sort | Hongle WANG |
| collection | DOAJ |
| description | 【Objective】The study was carried out to improve the accuracy of the deep learning model through adjusting the distribution of training dataset of lab-condition and field-condition images, to reduce the dependence of plant diseases recognition models on field-condition data.【Method】The plant diseases recognition model was optimized through adjusting the distribution of images of lab-conditions and field-conditions in training datasets. Deep learning models of plant diseases trained by three artificial neural networks of ResNeSt-50, VGG-16 and ResNet-50 were tested and compared.【Result】In a training dataset composed of a certain number of plant disease images, it had an impact on the model accuracy through adjusting the distribution of images of different conditions. When the proportion of images of the fieldconditions reached 30%, the accuracy of the model was improved by more than 18%. Through adding field-conditions images at a number ratio of 30% into a training dataset composed of 100% lab-condition images, the accuracy of the model was improved by more than 17%. Through adding lab-conditions images into a training dataset composed of 100% field-condition images, the accuracy of the model was improved with the increasing number of images, and the improved ranges were between 2% and 4%.【Conclusion】This method is suitable for the rapid establishment of high-accuracy plant diseases recognition models in the complex agricultural environment. It could reduce the dependence of plant recognition models on field-condition images, shorten the field data collection cycle at the beginning of model establishment and reduce the cost of field-condition images collection. It promotes a more effective application of artificial intelligence in unmanned farms and smart agriculture. |
| first_indexed | 2024-03-13T05:07:57Z |
| format | Article |
| id | doaj.art-6ea1234b3f244173a30b71cd03037f52 |
| institution | Directory Open Access Journal |
| issn | 1004-874X |
| language | English |
| last_indexed | 2024-03-13T05:07:57Z |
| publishDate | 2022-06-01 |
| publisher | Guangdong Academy of Agricultural Sciences |
| record_format | Article |
| series | Guangdong nongye kexue |
| spelling | doaj.art-6ea1234b3f244173a30b71cd03037f522023-06-16T06:41:39ZengGuangdong Academy of Agricultural SciencesGuangdong nongye kexue1004-874X2022-06-0149610010710.16768/j.issn.1004-874X.2022.06.013202206013Impact of Training Data Distribution on Plant Diseases Recognition Based on Deep LearningHongle WANG0Xinglin WANG1Wenbo LI2Quanzhou YE3Yonghai LIN4Hui XIE5Lie DENG6School of Environment and Energy, South China University of Technology, Guangzhou 510006, ChinaShenzhen Fengnong Holding Co., ltd, Shenzhen 518055, ChinaShenzhen Yuzhong Union Science and Technology Co., Ltd, Shenzhen 518126, ChinaShenzhen Wugu Network Technology Co., Ltd, Shenzhen 518055, ChinaShenzhen Wugu Network Technology Co., Ltd, Shenzhen 518055, ChinaShenzhen Wugu Network Technology Co., Ltd, Shenzhen 518055, ChinaShenzhen Fengnong Holding Co., ltd, Shenzhen 518055, China【Objective】The study was carried out to improve the accuracy of the deep learning model through adjusting the distribution of training dataset of lab-condition and field-condition images, to reduce the dependence of plant diseases recognition models on field-condition data.【Method】The plant diseases recognition model was optimized through adjusting the distribution of images of lab-conditions and field-conditions in training datasets. Deep learning models of plant diseases trained by three artificial neural networks of ResNeSt-50, VGG-16 and ResNet-50 were tested and compared.【Result】In a training dataset composed of a certain number of plant disease images, it had an impact on the model accuracy through adjusting the distribution of images of different conditions. When the proportion of images of the fieldconditions reached 30%, the accuracy of the model was improved by more than 18%. Through adding field-conditions images at a number ratio of 30% into a training dataset composed of 100% lab-condition images, the accuracy of the model was improved by more than 17%. Through adding lab-conditions images into a training dataset composed of 100% field-condition images, the accuracy of the model was improved with the increasing number of images, and the improved ranges were between 2% and 4%.【Conclusion】This method is suitable for the rapid establishment of high-accuracy plant diseases recognition models in the complex agricultural environment. It could reduce the dependence of plant recognition models on field-condition images, shorten the field data collection cycle at the beginning of model establishment and reduce the cost of field-condition images collection. It promotes a more effective application of artificial intelligence in unmanned farms and smart agriculture.http://gdnykx.cnjournals.org/gdnykx/ch/reader/view_abstract.aspx?file_no=202206013plant diseasedeep learningconvolutional neural networkdata distributionlab-conditionfield-condition |
| spellingShingle | Hongle WANG Xinglin WANG Wenbo LI Quanzhou YE Yonghai LIN Hui XIE Lie DENG Impact of Training Data Distribution on Plant Diseases Recognition Based on Deep Learning Guangdong nongye kexue plant disease deep learning convolutional neural network data distribution lab-condition field-condition |
| title | Impact of Training Data Distribution on Plant Diseases Recognition Based on Deep Learning |
| title_full | Impact of Training Data Distribution on Plant Diseases Recognition Based on Deep Learning |
| title_fullStr | Impact of Training Data Distribution on Plant Diseases Recognition Based on Deep Learning |
| title_full_unstemmed | Impact of Training Data Distribution on Plant Diseases Recognition Based on Deep Learning |
| title_short | Impact of Training Data Distribution on Plant Diseases Recognition Based on Deep Learning |
| title_sort | impact of training data distribution on plant diseases recognition based on deep learning |
| topic | plant disease deep learning convolutional neural network data distribution lab-condition field-condition |
| url | http://gdnykx.cnjournals.org/gdnykx/ch/reader/view_abstract.aspx?file_no=202206013 |
| work_keys_str_mv | AT honglewang impactoftrainingdatadistributiononplantdiseasesrecognitionbasedondeeplearning AT xinglinwang impactoftrainingdatadistributiononplantdiseasesrecognitionbasedondeeplearning AT wenboli impactoftrainingdatadistributiononplantdiseasesrecognitionbasedondeeplearning AT quanzhouye impactoftrainingdatadistributiononplantdiseasesrecognitionbasedondeeplearning AT yonghailin impactoftrainingdatadistributiononplantdiseasesrecognitionbasedondeeplearning AT huixie impactoftrainingdatadistributiononplantdiseasesrecognitionbasedondeeplearning AT liedeng impactoftrainingdatadistributiononplantdiseasesrecognitionbasedondeeplearning |