1 of 10

PRESENTATION ON PROJECT OF CROP DISEASES DETECTION ��� PRESENTED BY:-� DIKSHA GAUTAM,� SHANVI AND� RIDHAM CHAUDHARY

2 of 10

  • INTRODUCTION
  • PROBLEM STATEMENT
  • PROPOSED SOLUTION
  • TECHNOLOGY STACK
  • WORKING
  • RESULT
  • CONCLUSION AND FUTURE SCOPE
  • REAL WORLD APPLICATIONS

3 of 10

INTRODUCTION :-

AGRICULTURE IS THE BACKBONE OF OUR COUNTRY, BUT DISEASES IN CROPS OFTEN LEAD TO SIGNIFICANT LOSSES. OUR PROJECT, 'CROP DISEASES DETECTION', AIMS TO BRING THE POWER OF ARTIFICIAL INTELLIGENCE DIRECTLY INTO THE HANDS OF FARMERS, HELPING THEM MAKE INFORMED DECISIONS AT THE RIGHT TIME.

​

4 of 10

PROBLEM STATEMENT :-�

  • LACK OF EARLY DETECTION

​

  • LIMITED ACCESS TO EXPERTISE

​

​

  • INEFFICIENCY OF MANUAL INSPECTION

​

​

  • ECONOMIC IMPACT

5 of 10

�PROPOSED SOLUTION:-

  • AUTOMATED IMAGE CLASSIFICATION: USING A DEEP LEARNING MODEL (LIKE CNN) TO ANALYZE LEAF IMAGES AND INSTANTLY IDENTIFY SPECIFIC DISEASES.
  • REAL-TIME DETECTION: PROVIDING QUICK DIAGNOSTICS, ALLOWING FARMERS TO TAKE ACTION BEFORE THE DISEASE SPREADS ACROSS THE FIELD.
  • USER-FRIENDLY INTERFACE A SIMPLE PLATFORM WHERE USERS CAN UPLOAD A PHOTO AND RECEIVE AN IMMEDIATE REPORT AND SUGGESTED REMEDIES.
  • SCALABLE SOLUTION: A COST-EFFECTIVE TOOL THAT CAN BE ACCESSED VIA MOBILE OR WEB, BRIDGING THE GAP BETWEEN EXPERT KNOWLEDGE AND LOCAL FARMER.

OUR SYSTEM FOLLOWS A STRUCTURED PIPELINE WHERE THE INPUT IMAGE UNDERGOES FEATURE EXTRACTION TO IDENTIFY DISEASE PATTERNS, FOLLOWED BY A CLASSIFICATION LAYER THAT ACCURATELY PREDICTS THE CROP CONDITION.

6 of 10

TECHNOLGY STACK:-

​Programming Language: Python

​Why: Ease of use and powerful libraries for AI/ML.

​Machine Learning Framework: TensorFlow / Keras (CNN)

​Why: For accurate image classification and pattern recognition.

​Backend API: FastAPI

​Why: To handle high-speed requests between the frontend and the model.

​Database: SQLite / PostgreSQL

​Why: For storing user data, disease records, and treatment logs.

​Frontend/Interface: HTML, CSS, JavaScript (or Streamlit)

​Why: To provide a clean, simple interface for the end-user (farmers).

6

7 of 10

WORKING METHODOLOGY:-

    • Image Upload (Frontend): The user (farmer) uploads a picture of the crop leaf via our web interface.
    • API Request (Backend): The frontend sends the image to our Fast API backend via an HTTP request.
    • Model Prediction (AI Engine): Our pre-trained CNN model processes the image, extracts features, and classifies the disease.
    • Result & Remedy (Output): The system returns the detected disease name along with the recommended treatment/remedy to the user.

FRONTEND

​

USER

UPLOAD

PROCESS

​

​

BACKEND

​

RESULT

​

7

FAST API

SERVER

AI MODEL

8 of 10

CONCLUSION:-

OUR PROJECT, 'CROP DISEASES DETECTION', PROVIDES A ROBUST AND EFFICIENT AI-BASED SOLUTION TO ADDRESS THE CRITICAL ISSUE OF CROP HEALTH. BY INTEGRATING DEEP LEARNING WITH A SIMPLE USER INTERFACE, WE EMPOWER FARMERS TO PROTECT THEIR YIELDS AND MINIMIZE ECONOMIC LOSSES, PROVING THAT TECHNOLOGY CAN BE A VITAL PARTNER IN MODERN AGRICULTURE.

FUTURE SCOPE :-

MULTILINGUAL SUPPORT: ADDING SUPPORT FOR REGIONAL LANGUAGES TO ENSURE ACCESSIBILITY FOR FARMERS ACROSS INDIA.

EXPANDED DATABASE: INCLUDING MORE CROP VARIETIES AND A BROADER RANGE OF DISEASES TO IMPROVE DIAGNOSTIC ACCURACY.

8

9 of 10

REAL WORLD APPLICATIONS :-

  • FIELD DEPLOYMENT FOR FARMERS: EMPOWERS FARMERS TO INSTANTLY IDENTIFY CROP DISEASES DIRECTLY IN THE FIELD USING THEIR SMARTPHONES, ENABLING TIMELY INTERVENTION BEFORE THE SPREAD.
  • AGRICULTURAL ADVISORY SERVICES: ASSISTS AGRICULTURAL EXPERTS AND LOCAL CENTERS IN PROVIDING QUICK, ACCURATE DIAGNOSIS AND EXPERT RECOMMENDATIONS TO FARMERS.
  • EDUCATIONAL TOOL: SERVES AS A DIGITAL LEARNING RESOURCE FOR AGRICULTURAL STUDENTS TO VISUALLY IDENTIFY AND STUDY VARIOUS CROP DISEASES.
  • SMART FARMING INTEGRATION: CAN BE SCALED AND INTEGRATED WITH DRONE TECHNOLOGY OR IOT FIELD SENSORS FOR AUTOMATED LARGE-SCALE CROP MONITORING.

9

10 of 10

Thank you

​

​