Scribd
Upload
50 views18 pages

NIELIT Evaluation

The document outlines a Work Based Learning (WBL) program focused on empowering SC/ST/EWS/Women graduate engineers through training in Artificial Intelligence and Machine Learning. Key tasks include a research paper on an AI-based fruit classification and grading system, data pre-processing for meter reading, and corporate training presentations. The project aims to automate fruit processing and enhance data acquisition techniques, with future plans for scalability and practical applications in agriculture.

Uploaded by

Storm Shadow
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
50 views18 pages

NIELIT Evaluation

The document outlines a Work Based Learning (WBL) program focused on empowering SC/ST/EWS/Women graduate engineers through training in Artificial Intelligence and Machine Learning. Key tasks include a research paper on an AI-based fruit classification and grading system, data pre-processing for meter reading, and corporate training presentations. The project aims to automate fruit processing and enhance data acquisition techniques, with future plans for scalability and practical applications in agriculture.

Uploaded by

Storm Shadow
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 18

ARTIFICIAL

INTELLIGENCE
MACHINE LEARNING
DEEP LEARNING
Work Based Learning (WBL) programme to strengthen
and empower SC/ST/EWS/WOMEN graduate engineers
through MeitY institutions.
Mentor Name: Shri Bhaskar Banerjee & Shri. Dr. Anirban Jyoti Hati
Intern Name: Sayantan Sikdar
AICTE Internship Student Registration ID:
STU62e00645a14501658848837
Work Area: Artificial Intelligence
and Machine Learning
Date of Joining: 4th September 2023
•Task I: Training of Trainers and
Invigilation of NIELIT Exams
•Task II: Research Paper on AI based
Model- Fruit Classification and Grading
•Task III: Data Pre-processing on dataset
provided by WBSEDCL
Work Undertaken •Task IV: Dataset Fetching under
Anirban Sir
•Task V: PPT on Corporate Training on
Artificial Intelligence(AI) and Machine
Learning(ML)
Task I
• Training of Trainers Programme on Cyber
Security & Cyber Hygiene for the CVC
Officials at NIELIT Salt Lake Kolkata
Campus: I delivered a comprehensive
PowerPoint presentation (PPT) on
Information Gathering, covering
techniques, tools, and methodologies to
collect and analyze data effectively. The
session provided insights into enhancing
information retrieval strategies for various
domains, emphasizing the importance of
accurate and ethical data acquisition.
• Invigilation in NIELIT recruitment
examinations for the post of STA and JA on
24th September 2023
Task II: Research Paper on Artificial Intelligence based
Yield Phenotyping Framework: Fruit Classification and
Grading using Deep Neural Network
Problem Statement
“AI based yield phenotyping framework to establish a real-time, non-destructive
system that automates fruit segmentation, classification and grading system
using Deep Neural Network”
 How can a robust Machine Learning model be developed for fruit classification and
grading, addressing challenges related to the homogeneity in color, shape, and size of
fruit species?
 The goal is to automate the fruit processing system, eliminating manual sorting through
the implementation of state-of-the-art technologies like Computer Vision and Deep
Learning.
 Also making a physical prototype for the Machine Learning Model
Task II: Research Paper on Artificial Intelligence based
Yield Phenotyping Framework: Fruit Classification and
Grading using Deep Neural Network – Dataset Fetching
Dataset I: Mango Varieties Dataset II: FruitNet
Classification & Grading

Sample images of Dataset I: (a) Anwar Ratol, (b)


Chaunsa (Black), (c) Chaunsa (Summer Bahisht), (d)
Chaunsa (White), (e) Dosehri, (f) Fajri, (g) Langra,
(h) Sindhri Sample images of Dataset II: (a) Good Apple, (b)
Good Banana, (c) Good Guava, (d) Good Lime, (e)
Good Orange, (f) Good Pomegranate, (g) Bad
Source: www.kaggle.com Apple, (h) Bad Banana, (i) Bad Guava, (j) Bad
Lime, (k) Bad Orange, (l) Bad Pomegranate
Task II: Research Paper on Artificial Intelligence based Yield
Phenotyping Framework: Fruit Classification and Grading
using Deep Neural Network – Data Pre-processing

 Background Image Removal


 Image Resizing
 Noise Reduction
 Brightness Reduction
Task II: Research Paper on Artificial Intelligence based Yield
Phenotyping Framework: Fruit Classification and Grading
using Deep Neural Network – Workflow and Architectures

Architecture of AlexNet

Flowchart of the proposed


Architecture of ResNet50 v2
implementation methodology
Task II: Research Paper on Artificial Intelligence based Yield
Phenotyping Framework: Fruit Classification and Grading
using Deep Neural Network – Model Execution for Dataset II
AlexNet ResNet

Epoch 50 Epoch 50

Epoch 75 Epoch 75

Epoch 100 Epoch 100

Epoch 150 Epoch 150

Epoch 200 Epoch 250

Epoch 250
Task II: Research Paper on Artificial Intelligence based Yield
Phenotyping Framework: Fruit Classification and Grading
using Deep Neural Network – Physical Model in Practical

Structure of Conveyer : (a) Base of the conveyer (b)


Lateral Wooden board (c) Roller (d) Base of Motor (e)
Motor

Working Principle of Fruit Classification and


Grading System

Conveyer belt: (a) Sand Paper as Conveyer belt


Task II: Research Paper on Artificial Intelligence based
Yield Phenotyping Framework: Fruit Classification and
Grading using Deep Neural Network – Miscellaneous
 Writing the research paper:
• Abstract section (w/ Pratik Sarkar)
• Flowchart of the proposed implementation methodology
• ResNet architecture description and flowchart and architecture creation
 Cutting and drilling of the plywood boards for the physical model.
 Fitting of rods for smooth rotation of the conveyor belt for the fruits to pass on.
Future Scope: Research Paper on Artificial Intelligence
based Yield Phenotyping Framework: Fruit Classification
and Grading using Deep Neural Network
 Accepted at the 2024 IEEE International Conference on Contemporary
Computing and Communications(InC4), offering an opportunity for refinement.
 Extend the model to include more fruit varieties for enhanced versatility.
 Optimize scalability to adapt to diverse environmental conditions.
 Explore integration with edge computing and IoT for advanced capabilities.
 Collaborate with stakeholders for practical implementations in precision
agriculture.
 Contribute to sustainable and efficient fruit yield management.
Task III: Data Pre-processing on WBSEDCL Dataset
Problem Statement
“Meter display segmentation and reading the digits using OCR”

Packages required:
1. OS import os
2. OpenCV v3.4.3 import cv2
3. NumPy v1.16 import numpy as np
4. PIL from PIL import Image
Original Meter Image

After background removal using


OCR
Task III: Data Pre-processing on WBSEDCL Dataset
Following step has been used to extract the meter display:
1.Read the image
2.Convert the image from RGB to HSV color space
3.Create a mask of the image where the color which falls in a particular range
4.Find the largest contour in the mask image as the display will be the largest contour
5.Create bounding box over the largest contour
6.Extract the bounding box portion from the original image
7.Use adaptive thresholding on the extracted meter display to get a binary image with less
amount of noise
Further Processing required:
8.Skew Correction of the extracted image
9.Read the digits using OCR to get the meter reading
Task IV: Dataset
Fetching
 Dataset fetching assigned
by Dr. Anirban Jyoti Hati
Sir. Fetching of remote
sensing data or satellite
images of mapping,
spraying, irrigation,
monitoring, growth,
photosynthetic activity of
multiple Indian crops
Task V: Corporate
Training on Artificial
Intelligence(AI) and
Machine Learning

• Making of PowerPoint
Presentation:
• Data Collection
• Data Cleaning
• Data Exploration
• Feature Selection
• Data Transformation
Acknowledgement
I extend my heartfelt gratitude to NIELIT Kolkata for providing exceptional
infrastructure and valuable resources, essential for the successful completion
of my projects. Special thanks to our esteemed Head of the Institute, Shri V.
Krishnamurthy, for his insightful guidance and leadership. Additionally, I am
deeply grateful to Shri Bhaskar Banerjee and Shri Anirban Jyoti Hati, my
esteemed teachers, for their invaluable guidance and mentorship throughout
my projects, and to my dedicated teammates for their collaborative efforts,
without which our projects wouldn't have been possible.
Thank You

You might also like

pFad - Phonifier reborn

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.

Note: This service is not intended for secure transactions such as banking, social media, email, or purchasing. Use at your own risk. We assume no liability whatsoever for broken pages.


Alternative Proxies:

Alternative Proxy

pFad Proxy

pFad v3 Proxy

pFad v4 Proxy