Identification And Rehabilitation of Plant Disease by Image Processing and Electro culture Treatment

Abstract

India is a one of the largest cultivated countries in the world about 70% of the population depends on the agriculture. And our nation is exporting certain corps to many countries. Plant disease is major significant reduction in both quality and quantity of agriculture output. Cause of quality reduction by disease infection on plants that makes unhealthy to humans. The farmers are unaware lagging to find about the consequence of virus severity and difficult to identify what kind of virus affected on their corps. In this research interprets a clear ideological solution for the current state of problem using unique methodologies for the detection and examining of plant disease by regular monitoring in real-time cultivation. the CNN modalities and their modifying based deep learning of the image processing are using to detection and diagnosis with image classifications of plant pathogens with the precise separation of pathogens relativity between humans and plants. Finding in the early stages of the disease in plants are the major time consumption complexity challenges in the huge farming agriculture land. This proposed method also deploys to detect fungal range of infection area in the plants because fungal disease can easily spread in air to neighbour plants to stop this activity need to initiate the spreading cluster management this helps to indicate infection levels. An enforcement of convolutional neural network (CNN) approach for fungal disease severity classification analysis through scanning image and comparing day by day with their growth. An automated CNN model is designed for range of infection on plants classified into four severity levels(L) as mild(L1), moderate(L2), severe(L3), and critical(L4) including symptoms of various plant pathogens with a moderate accuracy. Fig 1 Utilizing a sufficient large number of sample images of various affected plants capturing from cultivating land for to collect diagnosing dataset. These predictions are indicating best economical way to process to detect disease. And also, Final stage for rehabilitation for disease affected plants to terminating the disease cells by injecting flow of electrons, electromagnetic waves and to conduct small amount of electricity through plant inside on affected area of tissues in roots, stems, leaves. The shock does not seem to harm the plants. It gives rehabilitate to the plant and also this method may act as a vaccine for plants to defend from various disease it would be the less expensive way of cell incitement of plants production. The results have been erratic and the electrical conditions leading to definite benefits on a large scale could not be confidently predicted from early studies. Finally, this terminology stands for terminating plant disease without chemical which is harmful to both humans and plants finally cultivating healthy agriculture. Fig 2

BLOCK DIAGRAM FOR CLASSIFICATION OF INFECTION RANGE:

To calculate and diagnosis the infection range of the plant tissue by primary and secondary diagnosis unit. Target integration unit validate the both units for accurate output. Classification filter banks classifies into 3 categories of infections (Mild, Moderate, Severe). An enforcement of convolutional neural network (CNN) approach for disease severity classification analysis although AI and DL based Framework. Convolutional Neural Network (CNN) and Long Short-Term Memory (LSTM) architecture to identify virus infected plants Image classifications to distinguish between disease and non-disease areas in the leaf, the variance of the Gray-level of each region is computed and used as feature to detect region of disease which are distinguished by low variance compare to non-disease regions.

PROPOSED ELECTRO CULTURE TREATMET

The construction of above electrified field on cultivated land is to conduct electro culture treatment structure. Natural phenomenon effects on plant growth response of environmental factors needs such as moisture, light, temperature, and similar functions are well documented on existing theory. some Literature relating almost all aspects of these environmental conditions is also quite extensive. [2] Very rare impactable researchers said is known of the physiological influence on plant growth of the artificial electric field environment which prevails at all times everywhere. This experiment shows that sufficiently high electric fields have a definite effect on plant growth and the growth response. that observed for conventional electrostatic and electrokinetic field growth. [3] While the plant and leaf polarities for all three modes of electrification were found to be essentially consistent. Fig 2