| dc.description.abstract | Rice crop disease detection and its diagnosis methods are vitally important for
the agriculture field to be sustainable. For that many researchers finding solutions to
minimize or avoid the rice plant disease to take the best yield for formers. Because
this disease led to a more than 38% yearly drop in paddy production. Due to a lack of
awareness and digital knowledge in fast identifying and best remedy for rice crop
diseases. In that, automated and artificial intelligence (AI) based rice crop disease
detection and prevention method is a key research solution needed for the current
agriculture field. The internet of things (IoT), has plenty of opportunities and
contributing a vital role in wireless networks, especially in the last 15 years. Using
IoT in the agriculture industry is growing up rapidly as it receives complex contextual
information about water irrigation, crop disease detection, fertilizer utilization, and
soil rate. Various crop disease detection methods need more accuracy and
dimensionality corrections. Disease detection is indispensable for agriculture to be
maintainable. Meantime automated rice plant disease detection systems also face
various problems to detect diseases in the current situation. Regular machine learning based image-wise disease detection methods are following preprocessing input values,
necessary feature extraction, image segmentation, and disease classifications steps.
The proposed research work provides solutions for the above-mentioned
problems and requirements with a novel approach which is the combination of the
laurent series with intelligent multidimensional object optimization (LIMO
classification framework) based on generative adversarial network (GAN) and swarm
intelligence optimal classification through cognitive attribute selection (SIOC-CAS)
to recognize various types of crop diseases in an agricultural field. Through this
proposed research work IoT nodes are senses the values of a field, and crop, and the
gathered information is shared with the processing unit with base station
communication. Multi-objective and cognitive learning-based routing (MOCLEAR)
protocol supports choosing optimal paths for data transmission betterment. After
MOCLEAR protocol communication, receives crop data such as image then data
reduction and dimensional sets creation to improve the eminence of the input values
for segmentation and classification. Then, image segmentation using GAN combined
with cognitive residual convolution network (CRCNet) is modified to segment values
from input images. After receiving segment input images perform under feature
iv
extraction and image classification with significant attributes. Generally, a corpus of
attributes is extracted to better characterize a pattern recognition problem. Attribute
selection plays a vital role in image classification to the complexity of handling many
features and increases the possibility of prediction rate. From a broader perspective,
this stage greatly influences the recognition process, in terms of both predictive
performance and the design of a computationally simple classifier. | en_US |
