Volume 10 Issue 2 - April 2018

  • 1. Plant leaf diseases detection and classification using machine learning

    Authors : Vijeta Shrivastava, Pushpanjali, Samreen Fatima, Indrajit Das

    Pages : 233-239

    DOI : http://dx.doi.org/10.21172/1.102.37

    Keywords : Alternaria AlternataAnthracnoseBacterial BlightCercospora Leaf SpotAnisotropic diffusion filter K-means clustering Support Vector Machine

    Abstract :

    Indian economy highly depends on agricultural productivity. Plant diseases detection has a major role in improving the production rate, as plant diseases causes serious effect on plant productivity and quality. Manually detection of infected plants requires immense knowledge about plant diseases, enormous time and huge amount of work. Hence, this can be done using image processing plant disease detection technique and machine learning. Generally plant disease symptoms are seen on the stem, fruits and leaves. So, we considered using plant leaf for detection of disease symptoms. This disease detection technique involves image acquisition, image filtering, segmentation, feature extraction and classification. This paper is focused towards the design of an optimal and more accurate way for the detection of plant diseases from leaf image and if it confirms the presence of disease then it is focused on evaluating its type among Alternaria Alternata, Anthracnose, Bacterial Blight and Cercospora Leaf Spot. We have experimentally performed this process and found that this process gives us almost accurate result as the minimum accuracy was 95.774 percent and maximum was 99.874 percent. It detects the disease by which plant is affected considering the affected region and disease is recognized precisely despite of having low affected region.

    Citing this Journal Article :

    Vijeta Shrivastava, Pushpanjali, Samreen Fatima, Indrajit Das, "Plant leaf diseases detection and classification using machine learning", https://www.ijltet.org/journal_details.php?id=930&j_id=4497, Volume 10 Issue 2 - April 2018, 233-239, #ijltetorg