AI / ML Project

Crop Disease Detection System

An image classification system that identifies plant leaf diseases from a photograph and suggests documented treatment steps.

Intermediate8–10 weeksCSE · IT · MCAAI & Machine LearningComputer Vision

Project overview

A farmer or field officer photographs an affected leaf, and the system identifies the crop and the most likely disease from a trained set of classes, returning a confidence value and standard treatment guidance sourced from agricultural extension material.

The project uses transfer learning on a public plant disease dataset, so it is achievable without heavy compute, and the demo — point a camera at a leaf image, get a diagnosis — is immediately clear to any reviewer.

Problem statement

Crop disease diagnosis in rural areas depends on the availability of an extension officer, and delays of even a few days can spread infection across a field. Visual symptoms are diagnosable, but the expertise is not evenly distributed.

Proposed solution

Fine-tune a pretrained convolutional network on labelled leaf images across several crops and disease classes, and serve it behind a mobile-friendly web interface. Pair each prediction with treatment guidance from published agricultural sources, and be explicit about which crops and diseases the model was actually trained on.

Key features

  • Image upload or direct camera capture on mobile
  • Transfer learning on a pretrained CNN backbone
  • Multi-crop, multi-disease classification with confidence scores
  • Top-three predictions rather than a single forced answer
  • Treatment and prevention guidance per detected disease
  • Offline-tolerant interface that queues uploads when connectivity is poor
  • History of past diagnoses with images

Technology stack

Vision

PythonTensorFlow / KerasOpenCVMobileNetV2 or ResNet50

Training

Data augmentationGoogle ColabTransfer learning

Backend

FlaskSQLite

Front end

ReactProgressive Web App shell

System architecture

The flow below is what goes into your architecture diagram and the corresponding chapter of the report.

  1. Capture layer takes an image from upload or the device camera
  2. Preprocessing resizes, normalises and optionally segments the leaf region
  3. The fine-tuned CNN produces class probabilities
  4. Post-processing applies a confidence threshold and returns the top three classes
  5. Advisory layer maps the predicted class to stored treatment guidance
  6. Result view shows image, prediction, confidence and guidance

Expected output

  • Predicted disease with confidence and two alternatives
  • Treatment and prevention notes for the detected class
  • Training and validation accuracy curves for the report
  • Confusion matrix across all disease classes

Suitable branches, difficulty and duration

Suitable branches

Difficulty level

Intermediate. Comfortable for a team where at least one member has worked in this stack before.

Development duration

8–10 weeks including documentation, tested and ready to demonstrate. Milestones are planned around your review dates.

Documentation, review, demo and viva support

Documentation support

Abstract, problem statement, literature survey guidance, system architecture, ER and UML diagrams, module descriptions, test cases and screenshots — prepared in your department format.

Project review support

Delivery is split across your review calendar so review one has a design and plan, review two has working modules, and the final review has the integrated, tested system.

Demo support

A rehearsed demo path, seeded sample data, a reset script, and a recorded walkthrough kept as a backup in case something fails in the review room.

Viva preparation

A written question bank for this domain, a code walkthrough session, and practice at explaining each module in your own words — because the examiner will ask you.

Customisation options

Colleges reject duplicate topics, so most projects are adjusted before we start. These are the usual directions for this one.

  • Restrict to crops grown locally in Tamil Nadu for a focused, higher-accuracy model
  • Add severity estimation from the affected leaf area percentage
  • Add a Tamil language interface for field use
  • Deploy the model to a mobile app with on-device inference using TensorFlow Lite
Talk to CodeLab69

Enquire about the Crop Disease Detection System project

Send your branch, college and first review date. You will get confirmation of scope, what the customisation would look like, and a delivery plan mapped to your reviews.