1. A system and method for real-time detection and classification of wheat crop diseases, comprising an image acquisition module, an edge computing unit, and a deep learning–based disease classification model.

2. The system as claimed in claim 1, wherein the image acquisition module captures images of wheat crop leaves using cameras or imaging sensors deployed in an agricultural field.

3. The system as claimed in claim 1, wherein the captured images are processed locally by an edge computing device to perform real-time disease detection without requiring cloud-based processing.

4. The system as claimed in claim 1, wherein the deep learning model analyzes visual features of wheat leaves to classify the crop condition into healthy or diseased categories.

5. The system as claimed in claim 1, wherein the deep learning model is optimized to operate efficiently on resource-constrained edge devices with reduced computational overhead.

6. The system as claimed in claim 1, further comprising a communication interface configured to transmit disease detection results to a user through a mobile application or monitoring interface.

7. The system as claimed in claim 1, wherein the system supports continuous crop monitoring and provides real-time alerts for early detection of wheat crop diseases.

Description:The present invention relates generally to the field of precision agriculture, artificial intelligence, and edge computing systems. More particularly, the invention pertains to an edge-based intelligent system and method for the real-time detection and classification of wheat crop diseases using a deep learning model deployed on edge devices. The proposed system integrates image acquisition, edge-based processing, and deep learning–based disease classification to identify common wheat leaf diseases directly in the field environment. By processing crop images locally at the edge device, the system reduces dependency on cloud infrastructure and enables faster diagnosis with minimal network latency. The invention further focuses on improving the efficiency of crop health monitoring by providing an automated and scalable solution for early disease detection. Such a system assists farmers and agricultural stakeholders in taking timely preventive actions, thereby reducing crop losses and improving agricultural productivity. The invention can be implemented in smart farming environments and integrated with mobile or IoT-enabled agricultural monitoring platforms.
BACKGROUND OF THE INVENTION
Agriculture plays a vital role in ensuring global food security and sustaining the economic stability of many countries. Among various cereal crops, wheat is one of the most widely cultivated and consumed staple crops worldwide, providing a major source of carbohydrates and nutrients for a large portion of the global population. However, wheat production is significantly affected by a variety of plant diseases, including leaf rust, powdery mildew, stripe rust, and other fungal or bacterial infections. These diseases can severely reduce crop yield and quality if they are not detected and controlled at an early stage. Therefore, timely and accurate identification of wheat crop diseases is critical for maintaining agricultural productivity and minimizing economic losses for farmers.
Traditionally, the detection and diagnosis of crop diseases rely heavily on manual field inspection by farmers or agricultural experts. In this process, farmers visually examine plant leaves and stems to identify symptoms of disease. However, manual inspection methods have several limitations. The process is often time-consuming, labor-intensive, and subjective, as it depends on the experience and expertise of the individual performing the inspection. In many rural areas, farmers may not have immediate access to agricultural specialists who can accurately diagnose crop diseases. As a result, diseases may remain unnoticed until they reach an advanced stage, leading to substantial crop damage and yield reduction.In recent years, the advancement of computer vision and artificial intelligence technologies, particularly deep learning, has created new opportunities for automated plant disease detection. Deep learning models, such as convolutional neural networks (CNNs), have demonstrated strong performance in analyzing plant leaf images and identifying disease patterns. These models can learn complex visual features from large datasets of crop images and provide highly accurate classification results. Consequently, deep learning-based plant disease detection systems have gained considerable attention in the field of smart agriculture and precision farming.
Despite these advancements, many existing crop disease detection systems rely on cloud-based architectures, where images captured in the field are transmitted to remote servers for processing and analysis. Although cloud computing provides high computational power and storage capacity, it also introduces several challenges. The transmission of large image data to cloud servers requires stable internet connectivity, which may not always be available in rural agricultural areas. Additionally, cloud-based processing can result in network latency, higher bandwidth consumption, and potential privacy concerns related to data transmission. These limitations may hinder the practical deployment of real-time crop monitoring systems in remote farming environments.To address these issues, edge computing technologies have emerged as an effective solution for performing data processing closer to the source of data generation. In an edge computing environment, computational tasks such as image analysis and disease classification can be executed directly on edge devices, including embedded systems, smart cameras, mobile devices, or edge-enabled IoT platforms deployed in the field. By processing data locally, edge computing significantly reduces communication delays and minimizes dependence on continuous internet connectivity. This approach enables real-time decision-making and rapid disease detection, which is essential for effective crop management.
Several research efforts have attempted to combine deep learning with edge computing for agricultural monitoring applications. These approaches aim to deploy trained deep learning models on edge hardware platforms capable of processing crop images captured from cameras or mobile devices. However, existing systems often face challenges related to model efficiency, computational resource limitations, and optimization for edge environments. Many deep learning models are computationally intensive and require large memory and processing resources, which may exceed the capabilities of low-power edge devices used in agricultural settings.Furthermore, current crop monitoring solutions may lack a fully integrated framework that combines image acquisition, optimized deep learning inference, and real-time disease classification in a single edge-based system. Without proper optimization techniques, the deployment of deep learning models on edge platforms may result in slow inference speeds and reduced detection accuracy. Therefore, there is a need for an improved system that can efficiently perform real-time disease detection and classification directly at the edge while maintaining high accuracy and low computational overhead.
Accordingly, there exists a need for an edge-based intelligent system and method capable of detecting and classifying wheat crop diseases in real time using an optimized deep learning model. Such a system would allow farmers to monitor crop health continuously, receive rapid diagnostic feedback, and take timely corrective actions to prevent disease spread. The development of an efficient edge-enabled disease detection system can significantly enhance the adoption of smart farming technologies, reduce crop losses, and contribute to sustainable agricultural production.

OBJECTIVE OF THE INVENTION
The primary objective of the present invention is to develop an edge-based intelligent system and method for the real-time detection and classification of wheat crop diseases using a deep learning model. The invention aims to provide an automated, accurate, and efficient solution for monitoring crop health directly in the agricultural field environment. By integrating edge computing with deep learning techniques, the system is designed to identify disease symptoms at an early stage, enabling farmers and agricultural stakeholders to take timely preventive actions to minimize crop losses and improve overall agricultural productivity.
Another important objective of the invention is to eliminate the dependence on cloud-based processing systems that require continuous internet connectivity for data transmission and analysis. In many rural and remote farming areas, reliable internet access may not be available, which limits the effectiveness of conventional cloud-based crop monitoring solutions. The proposed invention addresses this limitation by implementing the disease detection and classification process directly on edge computing devices, thereby enabling local data processing with minimal latency and reduced network requirements.A further objective of the invention is to develop a deep learning–based image analysis mechanism capable of accurately identifying multiple types of wheat crop diseases from leaf images captured in real-time. The system utilizes image acquisition modules, such as cameras or mobile devices, to capture visual information from wheat plants in the field. The captured images are then processed using an optimized deep learning model that extracts relevant features and classifies the disease type based on visual symptoms present on the leaf surface.
Another objective of the invention is to improve the speed and efficiency of disease detection through optimized deep learning inference on edge devices. Since many deep learning models require high computational resources, the invention focuses on designing an optimized model architecture that can operate efficiently on resource-constrained edge platforms. This ensures that the system can deliver rapid classification results while maintaining high accuracy and reliability.The invention also aims to provide a scalable and practical solution that can be deployed in smart agriculture and precision farming environments. The system can be integrated with Internet of Things (IoT) devices, mobile applications, or agricultural monitoring platforms to enable continuous crop surveillance and automated disease detection. By combining edge computing with intelligent image analysis, the system supports real-time monitoring without requiring centralized data processing infrastructure.
Another objective of the invention is to assist farmers, agricultural researchers, and farm management systems in making informed decisions regarding crop disease management. By providing timely disease detection and classification results, the system helps farmers apply appropriate treatment measures such as targeted pesticide application or crop management strategies. This contributes to reducing unnecessary pesticide use, lowering production costs, and improving crop yield quality.Finally, the invention aims to enhance the overall efficiency, reliability, and accessibility of crop disease monitoring technologies by offering a cost-effective and easy-to-deploy solution that can operate directly in agricultural environments. The integration of edge computing and deep learning in the proposed system supports sustainable agricultural practices and contributes to the advancement of intelligent farming technologies.
SUMMARY OF THE INVENTION
The present invention discloses an edge-based intelligent system and method for the real-time detection and classification of wheat crop diseases using an optimized deep learning model. The invention provides an automated agricultural monitoring solution designed to assist farmers and agricultural stakeholders in identifying plant diseases at an early stage directly in field environments. By integrating edge computing technology with advanced deep learning techniques, the proposed system enables efficient, real-time crop disease diagnosis without requiring continuous internet connectivity or centralized cloud processing.
In one aspect, the invention provides a hardware–software integrated system that includes an image acquisition module, an edge computing unit, a deep learning–based disease detection model, and a communication interface for delivering diagnostic results to the user. The image acquisition module captures visual images of wheat plant leaves using cameras, mobile devices, or other imaging sensors deployed within agricultural fields. These captured images contain visible symptoms of potential plant diseases such as discoloration, lesions, spots, or abnormal patterns present on leaf surfaces.Once the images are captured, they are transmitted directly to the edge computing device, which may include embedded computing platforms such as edge processors, IoT-enabled microcontrollers, or low-power computing modules capable of performing on-device data processing. Unlike traditional cloud-based crop monitoring systems, the proposed invention performs image analysis locally at the edge device, thereby minimizing network dependency and reducing communication latency.
The core component of the invention is the optimized deep learning model responsible for detecting and classifying wheat crop diseases. The model is designed to analyze visual features extracted from the captured images and identify patterns associated with specific plant diseases. The deep learning model may include convolutional neural network architectures or similar image-processing frameworks that are capable of learning complex disease characteristics from training datasets. Through training and optimization, the model becomes capable of distinguishing between healthy leaves and leaves affected by different disease types.To ensure efficient operation on resource-constrained edge devices, the invention incorporates model optimization techniques that reduce computational overhead while maintaining high classification accuracy. These optimization techniques may include model compression, parameter reduction, feature selection, and efficient inference mechanisms suitable for edge environments. As a result, the system can perform rapid disease detection and classification in real time without requiring high-performance computing infrastructure.
The invention further provides a real-time decision support mechanism that communicates the classification results to the user. After processing the captured images, the edge device generates a disease classification output indicating whether the wheat plant is healthy or affected by a particular disease. This output can be displayed through a user interface such as a mobile application, agricultural monitoring dashboard, or wireless notification system. In certain embodiments, the system may also provide recommendations or alerts that assist farmers in applying appropriate crop protection measures.Another aspect of the invention relates to the continuous monitoring capability of the system. The edge-based architecture allows the system to repeatedly capture and analyze crop images over time, enabling continuous surveillance of crop health conditions across agricultural fields. This capability supports early disease detection and prevents the spread of infections to surrounding plants, thereby improving crop management efficiency and reducing potential yield losses.
The invention is particularly suitable for deployment in smart agriculture and precision farming environments, where automated monitoring systems are used to improve farm productivity and resource management. The edge-based design makes the system adaptable to various field conditions, including remote agricultural areas where internet connectivity may be limited or unreliable. By processing data locally, the system ensures consistent operation regardless of network availability.
In addition, the invention contributes to sustainable agricultural practices by enabling targeted disease management. Early identification of crop diseases allows farmers to apply pesticides or treatment measures only when necessary, thereby reducing excessive chemical usage and minimizing environmental impact. The system also helps farmers improve crop yield quality by preventing disease outbreaks at an early stage.Overall, the present invention provides a reliable, scalable, and efficient technological solution for automated wheat crop disease detection and classification. By combining edge computing with optimized deep learning techniques, the system addresses the limitations of conventional crop monitoring methods and offers a practical approach for real-time agricultural disease management. The invention supports the advancement of intelligent farming technologies and contributes to improving productivity and sustainability in modern agricultural systems.
BRIEF DESCRIPTION OF THE DRAWINGS
The accompanying drawings illustrate exemplary embodiments of the present invention and are intended to provide a clearer understanding of the structural and functional aspects of the proposed system. The drawings are presented for illustrative purposes and should not be interpreted as limiting the scope of the invention. Various modifications and variations may be implemented without departing from the spirit and scope of the invention.

Figure 1: System Architecture of the Edge-Based Intelligent Wheat Crop Disease Detection System
Figure 1 illustrates the overall system architecture of the proposed edge-based intelligent system for real-time detection and classification of wheat crop diseases. The architecture diagram presents the key components and their interconnections within the system. The system primarily consists of an image acquisition module, an edge computing unit, a deep learning–based disease detection module, and a user communication interface.The image acquisition module includes cameras or imaging sensors that capture high-resolution images of wheat plant leaves directly from the agricultural field environment. These images represent the visual data required for detecting disease symptoms such as leaf spots, discoloration, fungal growth, or abnormal patterns on the crop surface.The captured images are transmitted to the edge computing unit, which performs local data processing. The edge computing unit may consist of an embedded processing device, a microcontroller-based edge processor, or an IoT-enabled computing platform capable of executing deep learning inference tasks. This component eliminates the need for remote cloud processing and ensures faster data analysis.Within the edge computing device, the optimized deep learning model processes the captured images to extract relevant features and classify the disease type. The classification results are then transmitted to a user interface module, which may include a mobile application, monitoring dashboard, or wireless notification system that informs the farmer or agricultural operator about the detected disease condition.

Figure 2: Operational Workflow of the Edge-Based Wheat Crop Disease Detection Method
Figure 2 illustrates the operational workflow of the proposed system for real-time disease detection and classification. The workflow diagram describes the sequential stages involved in processing crop images and generating diagnostic results.The process begins with the image capture stage, where visual images of wheat crop leaves are collected using field-deployed cameras or mobile devices. These images are then transferred to the edge processing unit for further analysis.In the next stage, the system performs image preprocessing, which may include resizing, normalization, noise removal, and enhancement of relevant visual features. This step ensures that the captured images are suitable for accurate analysis by the deep learning model.Following preprocessing, the system executes the deep learning–based disease classification stage, where the optimized model analyzes the leaf images to identify disease patterns. The model compares extracted features with trained disease characteristics to classify the crop condition.Finally, the result generation and notification stage provides the detected disease information to the user through a digital interface. The system may display the classification results, generate alerts, or provide decision-support information to assist farmers in taking appropriate crop management actions.
DETAILED DESCRIPTION OF THE INVENTION
The present invention relates to an edge-based intelligent system and method for the real-time detection and classification of wheat crop diseases using an optimized deep learning model. The invention provides a comprehensive technological framework designed to monitor crop health conditions directly within agricultural environments and to assist farmers in identifying plant diseases at an early stage. The system integrates image acquisition, edge computing, and deep learning–based analysis to provide accurate and timely disease diagnosis without relying on centralized cloud processing systems.
In one embodiment of the invention, the system includes an image acquisition module responsible for capturing visual images of wheat plant leaves from the field environment. The image acquisition module may include cameras, mobile imaging devices, or sensor-based imaging units deployed within agricultural fields. These devices capture images of wheat crop leaves that may exhibit visible symptoms of diseases such as leaf rust, powdery mildew, stripe rust, or other fungal or bacterial infections. The captured images represent the primary input data used by the system for disease detection and classification.Once the images are captured, they are transmitted to an edge computing device that performs local processing and analysis. The edge computing unit may include embedded computing platforms such as microcontrollers, edge processors, or Internet of Things (IoT)–enabled devices capable of executing machine learning models. By processing the data locally at the edge device, the system significantly reduces the need for transmitting large volumes of image data to remote cloud servers. This approach minimizes network latency, reduces bandwidth consumption, and ensures continuous system operation even in areas with limited or unstable internet connectivity.The edge computing device includes a data preprocessing module that prepares the captured images for analysis by the deep learning model. In this stage, the system performs several preprocessing operations such as image resizing, normalization, noise reduction, and feature enhancement. These preprocessing steps ensure that the images are formatted appropriately and that relevant visual characteristics associated with crop diseases are preserved for accurate classification.
After preprocessing, the system executes the deep learning–based disease detection and classification module. The deep learning model is trained using a dataset containing images of healthy and diseased wheat leaves. Through the training process, the model learns to recognize complex visual features associated with different disease conditions. In certain embodiments, the model may utilize convolutional neural network architectures or other image analysis frameworks capable of extracting hierarchical features from plant leaf images.
To enable efficient deployment on edge computing devices, the invention incorporates model optimization techniques that reduce computational complexity and memory requirements. These techniques may include parameter optimization, model compression, feature selection, and lightweight inference architectures that are suitable for resource-constrained edge hardware platforms. The optimized model ensures that the system can perform disease classification in real time while maintaining a high level of detection accuracy.During operation, the deep learning model analyzes the captured leaf images and classifies them into predefined categories such as healthy wheat leaf or specific wheat disease types. The classification results are generated within the edge device and subsequently communicated to the user through a notification or display interface. The user interface may include a mobile application, a web-based monitoring dashboard, or an agricultural management system capable of receiving and displaying disease detection results.
In some embodiments, the system may also generate alerts or recommendations that assist farmers in implementing appropriate crop protection measures. For example, the system may notify the user when a disease is detected and may provide suggestions regarding possible treatment strategies or preventive actions. This feature supports more efficient farm management and helps reduce the spread of diseases across agricultural fields.The invention further supports continuous monitoring of crop health conditions by periodically capturing images from the agricultural environment and performing automated disease detection. This continuous monitoring capability allows the system to identify early-stage disease symptoms before they become severe. Early detection enables farmers to apply targeted treatments and avoid widespread crop damage.
Additionally, the system architecture is designed to be scalable and adaptable to different agricultural environments. The edge-based architecture allows multiple edge devices to be deployed across large farming areas, enabling distributed monitoring of crop health conditions. The system can also be integrated with existing smart farming technologies such as IoT-based agricultural sensors, automated irrigation systems, or farm management platforms.Overall, the present invention provides a practical and efficient technological solution for automated wheat crop disease detection and classification. By combining edge computing with optimized deep learning techniques, the system offers real-time analysis, improved accessibility for farmers, and reduced reliance on centralized computing infrastructure. The invention contributes to the advancement of intelligent agricultural technologies and supports sustainable farming practices by enabling early disease detection and efficient crop management.
, Claims:1. A system and method for real-time detection and classification of wheat crop diseases, comprising an image acquisition module, an edge computing unit, and a deep learning–based disease classification model.
2. The system as claimed in claim 1, wherein the image acquisition module captures images of wheat crop leaves using cameras or imaging sensors deployed in an agricultural field.
3. The system as claimed in claim 1, wherein the captured images are processed locally by an edge computing device to perform real-time disease detection without requiring cloud-based processing.
4. The system as claimed in claim 1, wherein the deep learning model analyzes visual features of wheat leaves to classify the crop condition into healthy or diseased categories.
5. The system as claimed in claim 1, wherein the deep learning model is optimized to operate efficiently on resource-constrained edge devices with reduced computational overhead.
6. The system as claimed in claim 1, further comprising a communication interface configured to transmit disease detection results to a user through a mobile application or monitoring interface.
7. The system as claimed in claim 1, wherein the system supports continuous crop monitoring and provides real-time alerts for early detection of wheat crop diseases.