Volume 20 No 12 (2022)
 Download PDF
Adaptive thresholdingStein’s unbiased probability Estimate + CNN Classification Based Plant Leaf Diseases Detection
E.Banu , Dr. A. Geetha
In the last two decades, lack of agricultural yield has caused various diseases caused by fungi, bacteria and viruses. These illnesses, which directly interfere with photosynthesis in plants, are among the fastest increasing ones. Early signs are difficult to discern and can't be seen with the naked eye, and they vary greatly with variety and cultivation. Only once acervuli are built to assign host-pathogen connections are microscopic studies carried out. In the end, this causes a lack of time and subpar disease management. Therefore, early and accurate detection of the plant illness is crucial for prompt disease management and lower initial costs.The proposed research will lead to image processing based early detection of plant diseases, which will be more efficient and reduce the subjectivity arising from human experts in plant disease diagnosis.Analyzing plant leaf photos of various standards involves a lot of image processing. Images of agricultural plant leaves are frequently analysed using the method of image segmentation under the plant image enhancement technique, which is one of the image enhancement, reconstruction, and compression pre-processing approaches. new adaptive thresholding methods for classifying plant leaf diseases from images Stein's impartial probability This study suggests estimate preprocessing methods using convolutional neural networks. The plant leaf image dataset was enhanced and its noise was reduced using four data pre-processing approaches. Convolutional neural networks were then utilised to classify illnesses.Pre-processing performance parameters such as MSE. PSNR, co-relation and time consumption for noise removal considered to evaluation of proposed and existing methods. proposed Adaptive thresholding Stein’s unbiased probability Estimate with CNNproduce better leave diseases diagnose with the efficiency of 90.22%. Performance parameters compared with other existing methods median+CNN, Wiener+CNN,Gaussian+CNNand proposed ATSUPE+CNNtechnique.
convolutional neural network, pre-processing techniques, plant leaves diseases detection, Adaptive thresholding.
Copyright © Neuroquantology

Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Articles published in the Neuroquantology are available under Creative Commons Attribution Non-Commercial No Derivatives Licence (CC BY-NC-ND 4.0). Authors retain copyright in their work and grant IJECSE right of first publication under CC BY-NC-ND 4.0. Users have the right to read, download, copy, distribute, print, search, or link to the full texts of articles in this journal, and to use them for any other lawful purpose.