AN IOT BASED AGRICULTURAL FULLY AUTOMATED SYSTEM USING AI

Abstract

In the agriculture industry, artificial intelligence and sensor technology have become essential in recent years. Regardless of any insights into agricultural processes, agricultural research diligence today is more thorough, precise, data-driven, and aggressive than ever before. There has been shift in almost every industry since the introduction of Internet of Things (loT)-based technology. These include smart houses, smart grids, smart cities, "smart agriculture or precision agriculture," and smart health. The application of machine learning through loT data analytics will provide the agriculture sector with new benefits in order to satisfy the growing global need for food. As a result, crop field productivity will increase in both quantity and quality. With a few notable exceptions, these ground breaking findings are upending traditional farming methods and creating exciting new opportunities. Conventional farming methods make it impossible to assess plant development on a regular basis. Issues pertaining to nutrition, water supply, and early plant disease detection are also not properly monitored. Additionally, the old process uses a lot of manpower. It's also difficult to keep the soil's humidity level in the current one. For instance, an excessive amount of cultivating land makes it difficult to maintain ideal levels of nutrition, moisture, and disease prevention, among other things, which lowers crop yield. These issues with agricultural field automation are resolved by this approach. It makes use of artificial intelligence and the Internet of Things to decrease human labor and boost crop yields through the use of sensors and communication modules. When compared to current approaches, this technology offers a progressive rate of improved crop yield since it allows us to detect disease at an earlier stage, measure plant growth, and maintain the amount of nutrients.

Specification

1. DESCRIPTION
Field of the Invention
The present invention is related to the field of Artificial Intelligence
Detailed Description of the invention
In· the agriculture industry, artificial intelligence and sensor technology have become essential in
recent years. Regardless of any insights into agricultural processes, agricultural research
diligence today is more thorough, precise, data-driven, and aggressive than ever before. There
has been shift in almost every industry since the introduction of Internet of Things (IoT)-based
technology. These include smart houses, smart grids, smart cities, "smart agriculture or precision
agriculture," and smart health. The application of machine learning through loT data analytics
will provide the agriculture sector with new benefits in order to satisfy the growing global need
for food. As a result, crop field productivity will increase in both quantity and quality. With a
few notable exceptions, these groundbreaking findings are upending traditional farming methods
and creating exciting new opportunities. Conventional farming methods make it impossible to
assess plant development on a regular basis. Issues pertaining to nutrition, water supply, and
early plant disease detection are also not properly monitored. Additionally, the old process uses a
lot of manpower. It's also difficult to keep the soil's humidity level in the current one. For
instance, an excessive amount of cultivating land makes it difficult to maintain ideal levels of
nutrition, moisture, and disease prevention, among other things, which lowers crop yield. These
issues with agricultural field automation are resolved by this approach. It makes use of artificial
intelligence and the Internet of Things to decrease human labor and boost crop yields through the
use of sensors and communication modules. When compared to current approaches, this
technology offers a progressive rate of improved crop yield since it allows us to detect disease at
an earlier stage, measure plant growth, and maintain the amount of nutrients.


WORKING OF ARTIFICIAL INTELLIGENCE
The definition of AI is "automation based on associations." Two fundamental shifts in computing
that go beyond traditional educational technologies occur when computers automate reasoning
based on associations in data (or associations inferred from expert knowledge): (I) from
capturing data to detecting patterns in data and (2) from providing access to instructional resources to automating decisions about instruction and other educational processes. The range
of tasks that can be assigned to a computer system has ~xpanded with the ability to identify
patterns and automate judgments. The process of creating an AI system could result in unfair
decision-making and biased pattern recognition. As a result, the use of AI systems in education
must be regulated. This study outlines opportuniti~s for utilizing AI to enhance education,
acknowledges potential difficulties, and formulates suggestions to direct future policy
development.


DEEP LEARNING
Deep learning is a part of machine learning. Contrary to traditional machine learning algorithms,
many of which have a finite ability to learn regardless of the quantity of data they get, deep
learning systems can perform better with access to more data, which is the machine equivalent of
more experience. Once they have accumulated enough experience through deep learning,
machines can be trained to perform specific tasks like driving a car, identifying weeds in a field
of crops, diagnosing diseases, looking for weaknesses in machinery, and other duties. Deep
learning networks learn by spotting intricate patterns in the material they analyse. In order to
describe the data, the networks can achieve various levels of abstraction by building
computational models that are composed of several processing layers.
For example, a convolutional neural network, a sort of deep learning model, can be trained using
numerous (hundreds of thousands or millions) of images, such as ones of cats. This particular
neural network frequently gathers data from the pixels in the images it collects. With sets like
claws, ears, and eyes indicating the presence of a cat in a photo, it has the capacity to classify
sets of pixels that are typical of cat characteristics. There are significant differences between
deep learning and conventional machine learning. In order to recognize the traits of a cat in this
scenario, a domain expert would have to put a lot of effort into creating a standard machine
learning system. With deep learning, all that is necessary to teach the system the characteristics
of a cat is to feed it a very large number of images of cats. For many tasks, such as computer
vision, speech recognition (also known as natural language processing), machine translation, and
robotics, deep learning systems outperform typical machine learning systems by a large margin.
This does not imply that creating regular machine learning systems is easier than creating deep
~ learning systems.


ARCHITECTURE OF DEEP LEARNING
Deep learning is a subset of machine learning, which is a subset of artificial intelligence.
Generally speaking, "artificial intelligence" refers to techniques that enable computers to mimic
human behavior. Machine learning, a group of algorithms that have been taught on data, enables
all of this. Deep learning is just one type of machine learning that models the structure of the
human brain.

Using a predetermined logical structure, deep learning algorithms continuously examine data in
an effort to come to conclusions that are akin to those made by people. Neural networks, a multilayered
structure of algorithms, are used by deep learning to do this.

The architecture of the neural network was inspired by the structure of the human brain. Similar
to how human brains automatically recognize patterns and group different types of information,
neural networks may be trained to do the same. Using the distinct neural network layers as a
form of filter that goes from coarse to fine increases the likelihood of recognizing and generating
an appropriate outcome. The human brain acts in a similar manner. Every time we receive new knowledge, the brain tries to draw parallels with familiar objects. The same concept is also used
by deep neural networks. Neural netwo.rks can be used for a range of tasks, such as grouping,
classification, and regression.


DESCRIPTION OF DRAWINGS
FIG 1: ARCHITECTURE OF DEEP LEARING
FIG 2: ARCIDTECTURE OF NEURAL NETWORK
FIG 3: ARCHITECTURE OF AN lOT BASED AGRICULTURAL FULLY AUTOMATED
SYSTEM USING AI





CLAIMS
We Claim:

a. An loT Based Agricultural Fully Automation System Using AI, uses high speed
network connections to communicate between the smart farm and the farmer. Here, a
person does not require carrying any physical gadgets.
b. An loT Based Agricultural Fully Automation System Using AI uses deep learning and
machine learning methods, also it includes Digital Camera, Sensors, and communicating
module with high speed network.
c. An loT Based Agricultural Fully Automation System Using AI uses live camera device,
which share the images offarm field. This device requires embedded devices like Digital
camera, Sensors, location tracker etc.

Documents