Community Service Project
External Viva-Voce Examination
“Smart Pest Control System”
Internal Guide:
Team Members:
Mr. K. Bhushnam 1) Rapolu Sravani-21BQ1A05J5
Assistant Professor 2) Syed Suhana Khatun – 21BQ1A05M7
3) Tadiparthi Jhansi Yasaswini–21BQ1A05M9
4) Padavala Tarun – 21BQ1A05P1
Department of CSE (NBA Accredited),
Vasireddy Venkatadri Institute of Technology (Autonomous), Nambur, Pedakakani Mandal,
Guntur Dist., Andhra Pradesh
INDEX
• Abstract
• Introduction
• Survey Problems
• Field work
• Problem Statement
• Technologies Used
• Implementation and Result
• Conclusion
• References
ABSTRACT
Globally, pest infestations are a major challenge affecting agricultural crop production, leading to
significant yield losses and increased financial burdens for farmers. In developing nations, the lack of
knowledge about pest identification and effective control measures further exacerbates the problem,
resulting in excessive reliance on harmful chemical pesticides. Addressing this issue requires innovative
solutions that provide real-time pest detection and eco-friendly management strategies to ensure
sustainable farming practices.
In recent years, the agricultural sector has increasingly adopted advanced technologies to enhance pest
monitoring and control. Among these technologies, deep learning-based approaches have demonstrated
remarkable success in identifying pest species with high accuracy. This project focuses on the application
of Convolutional Neural Networks (CNNs) for automated pest detection. A web-based platform will be
developed, allowing farmers to upload images of pest-affected crops for analysis. The system will quickly
classify pests and provide targeted recommendations, including biological and eco-friendly control
measures.
By using deep learning-based system for pest detection and real-time recommendations, this system helps
farmers quickly and accurately identify pests and take effective action. It reduces the need for excessive
pesticides, protects crops, and supports sustainable farming. With advanced pest management powered by
deep learning, agriculture becomes more efficient, eco-friendly, and technology-driven.
Introduction
A community service project is an initiative undertaken by individuals or groups to address a specific need or issue within
a community. These projects are typically aimed at improving the well-being, quality of life, or circumstances of
community members.
Community service projects can take various forms and may involve activities such as:
Volunteer Work: Engaging in hands-on activities such as cleaning up parks, serving meals at homeless shelters, tutoring
students, or providing companionship to elderly residents.
Fundraising: Organizing events or campaigns to raise funds for local charities, schools, or other community organizations
that support important causes.
Awareness Campaigns: Creating and promoting initiatives to raise awareness about social, environmental, or health issues
within the community, such as mental health awareness campaigns or recycling drives.
Skill-building Workshops: Offering workshops or training sessions to teach valuable skills to community members, such
as financial literacy, job readiness, or computer skills.
Environmental Initiatives: Undertaking projects to protect and improve the environment, such as planting trees, cleaning
up litter, or promoting recycling and conservation efforts.
Community Development Projects: Working on projects that aim to enhance the infrastructure or facilities within the
community, such as building playgrounds, renovating community centers, or improving access to clean water.
Support Services: Providing support services to vulnerable populations within the community, such as organizing food
drives, offering free healthcare clinics, or assisting with disaster relief efforts.
Objective:
• Early Disease Detection: Detect plant diseases at early stages before they cause significant damage to crops. Early
detection allows for timely intervention and management, preventing widespread outbreaks and crop losses.
• Accurate Disease Identification: Utilize advanced technologies such as deep learning algorithms, convolutional neural
networks (CNNs), and image processing techniques to accurately identify and classify different types of plant diseases.
Achieving high accuracy ensures reliable disease diagnosis.
• Automated Diagnosis: Develop a system that can automatically analyze images of plant samples (e.g., leaves, stems) to
detect disease symptoms and patterns indicative of specific diseases. Automation reduces the need for manual inspection
and speeds up the diagnosis process.
• Remedy Recommendation: Provide actionable insights and recommendations for disease management and treatment
based on the detected plant diseases. This may include suggesting specific pesticides, fungicides, fertilizers, or cultural
practices to control or mitigate the spread of diseases.
• User-Friendly Interface: Design an intuitive and user-friendly interface for farmers, agronomists, or plant health
professionals to interact with the system. The interface should allow users to upload images, view disease diagnosis
results, and access recommended remedies and management strategies.
Community service projects can focus upon the following areas :
1. Plant diseases
2. Pest Detect & Control
3. Nutrition
4. Plantation
5. Horticulture
6. Herbal plants
7. Botanical survey
8. Zoological survey
9. Food habits etc.
Survey Problems
Day / Week No Activity Done Learning outcome
We went to the Peddakakani village center and observed the people , We study how people are involved in the work and
1 they are doing every day chores and ask them about their village adjusted to help others
We visited to other people's house and asked them how they do We learned how farmers are making the best use of natural
agriculture, what they use to farm the land resources to cultivate their land, yet they struggle to
2
identify and control pest infestations that threaten their
crops.
As we conducted our assessment, we identified a significant issue: Recognizing healthcare disparities led to efforts to
3 Limited access to healthcare facilities and medical services within the enhance medical services and infrastructure, ensuring
community. equitable access for all community members.
We saw people are meeting at location near the farm to learn how they We learned the we’ve to gathering to discuss ways to
4 employ water usage for their crops from pond’s tackle our difficulties
We encounter people near the end of the hamlet who have the capacity We can simply learn that in village, people support each
5 to work in factories and run their own business other even when they have limited work
Engaged in cleaning the surroundings of a local pond where a lot of Due to lack of proper knowledge about keeping the
plastic bottles and polythene bags were accumulated. community clean, people neglected the need for proper
6 maintenance of the pond.
As we surveyed in the village, we came to know that there are limited They are only RTC buses
public transportation facilities and drainage is not cleaned regularly. service available to guntur from the village and most of
7 There are only few RTC bus services available to Guntur from this the people really on private vehicles to reach the nearest
village and most of the people rely on private vehicles to reach the towns
nearest towns.
Conducted a health awareness workshop at the village focusing on Empowered individuals with knowledge to improve
8
preventive measures to overcome seasonal diseases. personal and community health outcomes.
Visited a community park in the village and planted some trees along As trees are very essential for environmental
the boundary of the park with a motive of increasing the greenery. sustainability, we requested the local people to take care
9
of those plants by watering them daily and protect them.
With all the surveys that we have done, we came to know that major If the drainage problem is solved, growth of mosquitoes
problem in the village is improper drainage cleaning, Identification of will rapidly decrease and consequently diseases such as
diseases of plants, Difficulty in identifying and control pest dengue and malaria will reduce gradually.
10
infestations, which significantly impact crop health and yield .
Field Work
Day 1:
Respondents Name : Sk.Nazeema
Age : 28
Education : Intermediate
Major Source of Income : Electrician
Children : 2 Daughters
Family Details : 1 male and 3 female
Family Income : approx. 3 lakhs
Notable Issues in the village : Stagnant Drainage water
Photo with the Respondent :
Respondents Name : Vara Lakshmi
Age : 30
Education : N/A
Major Source of Income : Masonary
Children : 1 Daughter
Family Details : 1 male and 2 female
Family Income : approx. 2 lakhs
Notable Issues in the village : Transportation
Photo with the Respondent :
Day 2:
Respondents Name : Mahalakshmi
Age : 60
Education : Third Standard
Major Source of Income : Fruit Business
Children : 1 Son
Family Details : 2 male and 3 female
Family Income : approx. 1 lakh
Notable Issues in the village : Lack of Health Care
Photo with the Respondent :
Day 3:
Respondents Name : Nagendram
Age : 50
Education : N/A
Major Source of Income : Farmer
Children : 1 Daughter, 1 son
Family Details : 2 male and 2 female
Family Income : approx. 3 lakhs
Notable Issues in the village : Irrigation Problem
Photo with the Respondent :
Respondents Name : K. Siva Kumar
Age : 32
Education : N/A
Major Source of Income : Pest Control Worker
Children : 2 Daughters
Family Details : 1 male and 1 female
Family Income : approx. 3 lakhs
Notable Issues in the village : Stagnant Drainage water
Photo with the Respondent :
Respondents Name : Suresh
Age : 45
Education : 10
Major Source of Income : Shop Keeper
Children : 2 Daughters
Family Details : 1 male and 3 female
Family Income : approx. 3 lakhs
Notable Issues in the village : Transportation
Photo with the Respondent :
Respondents Name : Sk.Reena
Age : 30
Education : 10
Major Source of Income : House Wife
Children : 1 Son
Family Details : 2 male and 1 female
Family Income : approx. 1 lakhs
Notable Issues in the village : Water Stagnation, Dogs in streets
Photo with the Respondent :
Respondents Name : U.Jagadesh
Age : 20
Education : 10
Major Source of Income : Mechanic
Children : 2 Sons
Family Details : 2 male and 1 female
Family Income : approx. 1 lakhs
Notable Issues in the village : Lack of Health Care
Photo with the Respondent :
Problem Statement
Based on the above Field Survey we want to discuss and want to find a solution regarding the domain
Agriculture.
Technologies used
Python
Used for data preprocessing, machine learning model training, and backend development due to its rich
ecosystem of libraries for AI and data analysis
HTML5, CSS3, JavaScript
Used for front-end development to create an interactive user interface.
Deep learning Model
• Convolutional Neural Network (CNN) is used for pest image classification by extracting features from
images.
• PyTorch / TensorFlow → Used for training and deploying the CNN model for pest identification.
Web Development Frameworks
Flask: A lightweight backend framework that handles image processing, AI model execution, and result
serving.
Implementation
• Data Collection: Gathered a dataset of images containing various pests and healthy crops. The
dataset includes different types of pests affecting multiple crops to ensure diversity. The images are
labeled to indicate the pest type for accurate classification.
• Data Preprocessing: Resized all images to a consistent dimension for uniform processing.
Normalized pixel values to enhance model learning and reduce computation complexity. - *Applied
data augmentation techniques (rotation, flipping, contrast adjustments) to improve model
generalization.
• Model Selection:Chose Convolutional Neural Networks (CNNs) for pest classification due to their
strong feature extraction capabilities.
• Model Training: Split the dataset into training, validation, and testing sets. Train the model on the
training data and validate its performance using the validation set. Fine-tune hyperparameters as
necessary to improve performance and prevent overfitting.
Model Evaluation: Assessed the model using accuracy, precision, recall, and F1-score. Achieved high
accuracy in pest classification while minimizing false predictions. Identified misclassified images and
adjusted model parameters accordingly.
Deployment: Deploy it for real-world use. This involve integrating it into a web, setting up an API to
handle user requests for pest classification. Developed a frontend interface where users can upload crop
images and receive pest identification results along with pest control suggestions.
Monitoring and Maintenance: Continuously monitor the performance of the model in the production
environment and update it as needed. As new pest species emerge or additional data becomes available
there is need to retrain the model or fine-tune its parameters to ensure accurate detection.
Compiled a dataset of images containing healthy crops and pest-affected crops, covering 10
pest species across 5 different crop types. The dataset consists of 10,000 images, ensuring a well-
balanced and diverse representation. To maintain consistency, all images were resized and normalized
for better model performance. A Convolutional Neural Network (CNN) model was selected for pest
classification. The dataset was divided into training , validation , and testing sets. The model was
trained on 8,000 images over 10 epochs, using a batch size of 16 and a learning rate of 0.001. After
training, the model achieved an accuracy of 92.5%, demonstrating reliable pest detection capabilities.
Results
Plant Disease Detection Page Result Page
Conclusion
The Smart Pest Control System is a step forward in modern agriculture, using deep learning to make
pest detection more accurate and efficient. By leveraging technology, this system helps farmers identify
pests early and take the right measures to protect their crops.
Effective Use of Technology: This project demonstrates the feasibility of using deep learning and
computer vision for accurately identifying pests and recommending suitable pesticides. By integrating
AI-based detection with a user-friendly web interface, farmers can efficiently monitor pest infestations
and take timely action.
Challenges and Solutions: During development, we faced challenges like collecting data, training an
accurate model, and making real-time predictions. However, by improving our CNN model and fine-
tuning the system, we created a reliable and efficient pest detection solution.
Impact on Agriculture: The implementation of this system has the potential to revolutionize pest
management by enabling early detection, reducing excessive pesticide use, and ultimately improving
crop yield and quality. It also provides accurate recommendations, helping to protect the environment
and support sustainable farming.
Continuous Improvement: While this project marks a significant milestone, it also lays the foundation
for future enhancements. Further improvements may include integrating IoT devices for real-time
monitoring, expanding the pest database, and incorporating drone-based surveillance to enhance
accuracy and efficiency.
In conclusion, the Smart Pest Control System combines technology and agriculture to address a key
farming issue. By using AI-driven pest detection, it helps farmers identify pests early, reduce pesticide
use, and improve crop health. This not only boosts food security but also promotes eco-friendly farming
and supports sustainable agriculture, demonstrating how technology can transform modern farming.
References
1. Chaudhary, A., Jha, K., & Tripathi, R. (2020). Deep learning-based pest detection and classification in agricultural crops. Computers and Electronics in
Agriculture, 175, 105527.
2. Mittal, S., Chouhan, R., Agarwal, S., & Dey, A. (2021).* A deep learning-based approach for pest detection in crops using UAV images. Expert
Systems with Applications, 182, 115121.
3. Dhaka, V., Meena, S. V., Rani, G., Sinwar, D., Kavita, & Singh, A. (2021). A deep learning approach for pest detection using convolutional neural
networks. Artificial Intelligence in Agriculture, 5, 24-30.
4. Chen, J., Zhang, D., Zhao, Y., Chen, L., & Li, Y. (2021).* Automatic pest detection and recognition using deep learning in agricultural fields.
Computers and Electronics in Agriculture, 184, 106081.
5. Fuentes, A., Yoon, S., Kim, S. C., & Park, D. S. (2017).* High-performance deep learning-based method for real-time insect pest recognition on field
crops. Sensors, 17(12), 2932.
6. Larios, N., Deng, H., Zhang, H., Sarpola, M., Yuen, J., Paasch, R., & Dietterich, T. (2008). Automated insect identification through concatenated
histograms of local appearance features: A case study of bees. Pattern Recognition, 42(5), 814-824.
7. Ramesh, D., & Vydeki, D. (2018). Recognition and classification of paddy leaf diseases using deep learning algorithms. International Journal of
Computer Applications, 180(20), 20-25
8. Jadhav, P., & Patil, J. (2022). Deep learning approach for pest identification and classification in agriculture. Applied Artificial Intelligence, 36(1), 204-
219.
9. Ramesh, D., & Vydeki, D. (2018). Recognition and classification of paddy leaf diseases using deep learning algorithms. International Journal of
Computer Applications, 180(20), 20-25.
10. Kumar, A., Singh, A., & Pandey, A. (2021).* CNN-based automatic pest detection system for precision agriculture. Multimedia Tools and Applications,
80(5), 7319-7335.
