Data Analysis And Prediction On Cloud Computing For Enhancing Productivity In Agriculture

• Published in: IOP conference series. Materials Science and Engineering 2019-10, Vol.590 (1), p.12014
• Main Authors: Ashokkumar, K., Divya Chowdary, D., Divya Sree, Ch
• Format: Article
• Language: English
• Subjects: Agriculture; Bilateral filtered image; Classifiers; Cloud computing; Data analysis; Extracted Regions; Gaussian image; GLCM; Gray image; Homogeneity; Image classifications; Image processing; K-means clustering; Leadership; Leaves; Neural Network; Neural networks; Pesticides; Pests; Preprocessing; Pricing; Productivity; Regions detected image; Segmented image; Shading
• ISSN: 1757-8981; 1757-899X
• DOI: 10.1088/1757-899X/590/1/012014

Introducing a concept to increase productivity and predict the crops from the diseases. To find the leaf characteristics using image processing to detect the diseases and pests that are present in leaves. In agriculture the main important aspect is to take proper steps to increase the production, which means to know the state of leaves condition in the field. In this undertaking the programmed system recognizes the qualities of the leaves. The programmed leaf attributes recognition is basic one in observing extensive fields of yields, and to naturally distinguish indications of leaf attributes when they show up on takes off. The basic leadership framework uses the picture portrayal and managed classifier of Neural Network. Picture preparing procedures for this sort of choice examination includes preprocessing, highlight extraction and characterization organize. In this procedure, the picture is taken and that can be resized if required the locale of intrigue choice can be performed. Here, shading and surface highlights are extricated from a contribution for organize preparing and grouping. Shading highlights like mean, standard deviation of HSV shading space and surface highlights like vitality, differentiation, homogeneity and connection. The framework will be utilized to order the test pictures naturally to choose leaf attributes. For this approach, programmed classifier Neural Network (NN) is utilized for grouping in light of learning with some preparation tests of that same classification. This system utilizes digression sigmoid capacity as portion work. At long last, the mimicked result demonstrates that utilized system classifier gives least blunder amid preparing and better exactness in order. This helps to read the classifier (Characteristics difference of the input image).By using the Threshold values we can detect the pests/diseases and also helps to give the required data about the diseases, details of pesticides and other like required quantity of the pesticides and displays the present market prices for the selected crop.