Automatic Detection of Diseased Tomato Plants Using Thermal and Stereo Visible Light Images

• DOI: 10.1371/journal.pone.0123262
• Publication Date: 2015-04-11T03:11:56Z
• Authors: Shan-e-Ahmed Raza; Gillian Prince; John P. Clarkson; Nasir M. Rajpoot
• Keywords: Oidium neolycopersici; Light; multi resolution stereo matching algorithm; Erysiphales; tomato; Solanum lycopersicum; graph cut based stereo matching algorithm; Image Processing, Computer-Assisted; 2. Zero hunger; 0303 health sciences; predictive value; autoanalysis; block based stereo matching algorithm; classification; controlled study; depth estimation; fungus; fungus transmission; graph cut based stereo matching algorithm; humidity; image analysis; image processing; image registration; imaging; imaging and display; machine learning; multi resolution semi global matching algorithm; multi resolution stereo matching algorithm; natural transmission; non local cost aggregation algorithm; nonhuman; Oidium neolycopersici; powdery mildew; predictive value; semi global matching algorithm; sensitivity and specificity; stereo visible light image; temperature dependence; thermal aging; thermal visible light image; tomato; validation process; Erysiphales; Fungi; Lycopersicon esculentum; Oidium neolycopersici; fungus; Q; R; Temperature; imaging; imaging and display; fungus transmission; machine learning; classification; TA; natural transmission; Medicine; multi resolution semi global matching algorithm; Algorithms; Research Article; stereo visible light image; non local cost aggregation algorithm; Science; 03 medical and health sciences; Ascomycota; image analysis; depth estimation; controlled study; Lycopersicon esculentum; temperature dependence; thermal visible light image; autoanalysis; Plant Diseases; block based stereo matching algorithm; nonhuman; QK; humidity; Fungi; 15. Life on land; image processing; Plant Leaves; image registration; semi global matching algorithm; sensitivity and specificity; validation process; powdery mildew; thermal aging

Accurate and timely detection of plant diseases can help mitigate the worldwide losses experienced by the horticulture and agriculture industries each year. Thermal imaging provides a fast and non-destructive way of scanning plants for diseased regions and has been used by various researchers to study the effect of disease on the thermal profile of a plant. However, thermal image of a plant affected by disease has been known to be affected by environmental conditions which include leaf angles and depth of the canopy areas accessible to the thermal imaging camera. In this paper, we combine thermal and visible light image data with depth information and develop a machine learning system to remotely detect plants infected with the tomato powdery mildew fungus Oidium neolycopersici. We extract a novel feature set from the image data using local and global statistics and show that by combining these with the depth information, we can considerably improve the accuracy of detection of the diseased plants. In addition, we show that our novel feature set is capable of identifying plants which were not originally inoculated with the fungus at the start of the experiment but which subsequently developed disease through natural transmission.