C2_Batch (4)[1].pptx skdljwekdjlaksdjwiodhoiwd

lab

1. ARTIFICIAL INTELLIGENCE AND IOT BASED DETECTION OF PESTICIDE IN APPLE AND POTATO Batch No:C2 Internal Guide Swetha Amireddy -20251A04H5 Ms.Ch.Anusha Kuruva Vaishnavi -20251A04D3 Pantham Apurva -20251A04G7 Vulli Varshini -20251A04F0 1

2. CONTENTS Introduction Hardware Requirements Software Requirements Literature Survey Existing Systems Plan of work Algorithms Applications References 2

3. Introduction Objective of the project Develop a system that can accurately detect the presence and quantity of pesticide residues in fruits and vegetables. Develop an early warning system that alerts farmers and other stakeholders when pesticide residue levels exceed safe limits. Promote sustainable agricultural practices by encouraging the use of alternative pest control methods that reduce the need for chemical pesticides. 3

4. Hardware Requirements Chemiresistive Gas Sensors pH Sensors Moisture Sensors Liquid Crystal Display Arduino Uno Microcontroller 4

5. Software Requirements Database Management System IoT Platform AI and Machine Learning Frameworks 5

6. Literature Survey 1. Explored various techniques to find pesticides in fruits and vegetables. The techniques like Color Identification Technology Sensor Integration Method 2. Determination of Fruits and vegetables by calculating NDVI (Normally Difference Vegetation Index). The values of NDVI which indicates the inorganic substances . 6

7. E xisting Systems 1. Laboratory-Based Testing: • Gas Chromatography-Mass Spectrometry (GC-MS) • High-Performance Liquid Chromatography (HPLC) 2. Portable Testing Kits: • Test Strips • Handheld Spectrometers 3. Electronic Nose and Electronic Tongue 7

8. Plan of work System will be developed with a detailed architecture, including sensor specifications and data flow diagrams. Hardware and Software Setup. Data Collection and Sensors implementation. Deploy sensors in the selected agricultural fields, storage facilities, and transportation vehicles. AI Model Development. Testing and Validation. 8

9. Algorithms 1. Convolutional Neural Networks (CNNs): Input images of produce can be processed through a CNN to identify areas with pesticide residue. 2. Regression Models: Regression algorithms, such as Linear Regression or Random Forest Regression, can be used to predict pesticide residue levels based on environmental conditions, pesticide application data, and historical residue measurements. 3. Deep Learning Models: Recurrent Neural Networks (RNNs), like LSTM and GRU, can be applied to time-series data and sequential data, such as sensor readings, for anomaly detection or pattern recognition related to pesticide residues. 9

10. Applications Real-time Monitoring and Early Detection Precision Agriculture Alerts and Notifications Research and Development 10

11. References [1]. Bhavini J.Samajpati , Sheshang D. Degadwala “Hybrid Approach for Apple Fruit Diseases Detection and Classification” IEEE International Conference on Communication and Signal Processing, pp. 978- 5090-0396, 2016. [2]. IOT-Based Prediction of Pesticides and Fruits Diseases Detection Using SVM and Neural Networks D.Nagajyothi , P. Hema Sree, M.Rajani Devi, G.Madhavi , Sk Hasane Ahammad , International Journal of Mechanical Engineering, January 2022. [3]. Shalini Gnanavel et all “Quality Detection of Fresh Fruits and Vegetables to Improve Horticulture and Argo-industries” IEEE Xplore2017. 11

12. 12