Show simple item record

dc.contributor.authorRamcharan, A.
dc.contributor.authorMcCloskey, P.
dc.contributor.authorBaranowski, K.
dc.contributor.authorMbilinyi, N.
dc.contributor.authorMrisho, L.
dc.contributor.authorNdalahwa, M.
dc.contributor.authorLegg, J.P.
dc.contributor.authorHughes, D.P.
dc.date.accessioned2019-12-04T11:39:08Z
dc.date.available2019-12-04T11:39:08Z
dc.date.issued2019-03-20
dc.identifier.citationRamcharan, A., McCloskey, P., Baranowski, K., Mbilinyi, N., Mrisho, L., Ndalahwa, M., ... & Hughes, D.P. (2019). A mobile-based deep learning model for cassava disease diagnosis. Frontiers in Plant Science, 10, 272.
dc.identifier.issn1664-462X
dc.identifier.urihttps://hdl.handle.net/20.500.12478/6542
dc.descriptionOpen Access Journal; Published online: 20 March 2019
dc.description.abstractConvolutional neural network (CNN) models have the potential to improve plant disease phenotyping where the standard approach is visual diagnostics requiring specialized training. In scenarios where a CNN is deployed on mobile devices, models are presented with new challenges due to lighting and orientation. It is essential for model assessment to be conducted in real world conditions if such models are to be reliably integrated with computer vision products for plant disease phenotyping. We train a CNN object detection model to identify foliar symptoms of diseases in cassava (Manihot esculenta Crantz). We then deploy the model in a mobile app and test its performance on mobile images and video of 720 diseased leaflets in an agricultural field in Tanzania. Within each disease category we test two levels of severity of symptoms-mild and pronounced, to assess the model performance for early detection of symptoms. In both severities we see a decrease in performance for real world images and video as measured with the F-1 score. The F-1 score dropped by 32% for pronounced symptoms in real world images (the closest data to the training data) due to a decrease in model recall. If the potential of mobile CNN models are to be realized our data suggest it is crucial to consider tuning recall in order to achieve the desired performance in real world settings. In addition, the varied performance related to different input data (image or video) is an important consideration for design in real world applications.
dc.description.sponsorshipHuck Institutes at Penn State University
dc.description.sponsorshipCGIAR
dc.format.extent1-8
dc.language.isoen
dc.rightsCC-BY-4.0
dc.subjectCassava
dc.subjectDiseases
dc.subjectPlant Diseases
dc.subjectDiagnosis
dc.subjectPlant Condition
dc.subjectTanzania
dc.titleA mobile-based deep learning model for cassava disease diagnosis
dc.typeJournal Article
dc.description.versionPeer Review
cg.contributor.crpRoots, Tubers and Bananas
cg.contributor.affiliationPenn State University
cg.contributor.affiliationInternational Institute of Tropical Agriculture
cg.coverage.regionAfrica
cg.coverage.regionEast Africa
cg.coverage.countryTanzania
cg.creator.identifierJames Legg: 0000-0003-4140-3757
cg.researchthemePLANT PRODUCTION & HEALTH
cg.isijournalISI Journal
cg.authorship.typesCGIAR and advanced research institute
cg.iitasubjectCassava
cg.iitasubjectDisease Control
cg.iitasubjectPlant Diseases
cg.journalFrontiers in Plant Science
cg.howpublishedFormally Published
cg.accessibilitystatusOpen Access
local.dspaceid109925
cg.targetaudienceScientists
cg.identifier.doihttps://dx.doi.org/10.3389/fpls.2019.00272


Files in this item

Thumbnail

This item appears in the following Collection(s)

Show simple item record