Sinhgad Technical Education Society’s
SINHGAD INSTITUTE OF
TECHNOLOGY,LONAVALA
Department OF Engineering Sciences
PROJECT BASED LEARNING
WORK BOOK
ACADEMIC YEAR: 2020/2021 Semester: II
Division: C Batch: 2020/21 Group:CG-2
Project Title:
ICEBERG-SHIP CLASSIFICATION OF SATELLITE
RADAR IMAGES
Department OF Engineering Sciences SINHGAD INSTITUTE OF
TECHNOLOGY
Kusgaon (Bk), Off Mumbai-
Pune Expressway Lonavala –
410401
SINHGAD INSTITUTE OF TECHNOLOGY
Kusgaon (Bk), Off Mumbai-Pune Expressway, Lonavala – 410 401.
Department of Engineering Sciences
Certificate
This is to certify that, following students,
1. PRAGATI GUPTA Roll No: C-42
2. BHAUSAHEB SHINDE Roll No: C-50
3. RUPAL TARJULE Roll No: C-54
4. KARAN UPARE Roll No: C-55
5. KARTIK WATILE Roll No: C-56
6. RISHVI KUMARI Roll No: C-58
has completed all the Term Work & Practical Work in the subject Project Based Learning
satisfactorily in the department of Engineering Sciences as prescribed by Savitribai Phule
Pune University, in the academic year 20 -20 2 1 .
Faculty-in-charge Head of Department Principal
Date: / / .
Rules & Regulations:
1. Handle the workbook very carefully.
2. All students must enter the correct information in the work book.
3. All entries in the PBL work book must be verified by the concerned
Supervisor/Mentor.
4. Activities planned should be completed as per the instructions and schedule given
by Supervisor/Mentor.
5. Assessment of TW for Project Based Learning (PBL) is out of 25 Marks which
are based on attendance, regularity of completion of activities on given time and
students involvement.
6. Assessment of PR for PBL is out of 50 Marks which are based on idea inception,
outcomes of PBL, problem solving skills, solution provided, final product,
documentation, demonstration, contest participation, and awareness.
7. Students need to submit final report of 5 to 10 pages in the prescribed format
given at the end of this workbook.
Course Objectives:
1. To emphasizes learning activities that are long-term, interdisciplinary and student-
centric.
2. To inculcate independent learning by problem solving with social context.
3. To engages students in rich and authentic learning experiences.
4. To provide every student the opportunity to get involved either individually or as a
group so as to develop team skills and learn professionalism.
Course Outcomes:
CO1: Project based learning will increase their capacity and learning through shared
cognition.
CO2: Students able to draw on lessons from several disciplines and apply them in
practical way.
CO3: Learning by doing approach in PBL will promote long-term retention of
material and replicable skill, as well as improve teachers' and students' attitudes
towards learning.
Group Structure:
Working in supervisor/mentor monitored groups; the students plan, manage, and
complete a task/project/activity which addresses the stated problem.
1. There should be team/group of 5 -6 students
2. A supervisor/mentor teacher assigned to individual groups
Selection of Project/Problem:
The problem-based project oriented model for learning is recommended. The model
begins with the identifying of a problem, often growing out of a question or
“wondering”. This formulated problem then stands as the starting point for learning.
Students design and analyze the problem within an articulated interdisciplinary or
subject frame.
A problem can be theoretical, practical, social, technical, symbolic, cultural, and/or
scientific and grows out of students’ wondering within different disciplines and
professional environments. A chosen problem has to be exemplary. The problem may
involve an interdisciplinary approach in both the analysis and solving phases.
By exemplarity, a problem needs to refer back to a particular practical, scientific,
social and/or technical domain. The problem should stand as one specific example or
manifestation of more general learning outcomes related to knowledge and/or modes
of inquiry.
There are no commonly shared criteria for what constitutes an acceptable project.
Projects vary greatly in the depth of the questions explored, the clarity of the learning
goals, the content, and structure of the activity.
1. A few hands-on activities that may or may not be multidisciplinary.
2. Use of technology in meaningful ways to help them investigate, collaborate,
analyze, synthesize, and present their learning.
3. Activities may include- Solving real life problem, investigation, /study and
Writing reports of in depth study, field work.
Group Information:
Division: C Batch: 2020-2021 Group: CG-2
Roll No. PRN No. Name of Student Mobile No.
42 72150216E PRAGATI GUPTA 8329511867
50 721502162J BHAUSAHEB SHINDE 9145465854
54 72150282C RUPAL TARJULE 9075461176
55 72150291B KARAN UPARE 9503485837
56 72150298K KARTIK WATILE 9689053798
58 72150230L RISHVI KUMARI 7004653981
Name of Faculty/Mentor: PROF. GHODICHOR
E-mail: _
Mobile No.: _
Initial Survey for Finalization of Title (Literature Survey):
WE GET THIS IDEA THROUGH NEWSPAPER ARTICLE.
ONE OF OUR TEAMMATE WAS READING NEWSPAPER AND SAW AN DEATH
ARTICLE RELATED TO ICEBERG.
SO WE DECIDED TO FINILISE OUR TOPIC AS ICEBERG SHIP CLASSIFICATION
THROUGH SATELLITE RADAR IMAGES.
Required H/W & S/W:
WE HAD DEEP LEARNING ON ICE NERG SHIP CLASSSIFICATION.
References: (Website/Books/Papers):
Statoil/c-core radar images of iceberg and ship dataset.
https://www.kaggle.com/c/statoil-iceberg-classifier-challenge
Ankita Rana and Vadivel Sangili. Implementation of improved ship-iceberg
Jun 2017.
Deep learning approach in automatic iceberg – ship detection with sar remote
sensing data. University Of Bristol DEC 2018.
Figure/Circuit Diagram/Block Diagram/Flow Chart:
ARCHITECTURE:-
Abstract:
Drifting Icebergs present threats to navigation and activities in areas such as
offshore. In remote areas with particularly harsh weather, aerial reconaissances
methods are not fea- sible so use of machine learning techniques is needed.
However the major challenges that still remain is lack of sufficient training data
and integrating additional features. Here in this Bachelors Project Report we go
through some of the challenges faced and detect Iceberg in radar satellite
images. Different Machine Learning Techniques were used like Support vector
machine to achieve accuracy of 82 percent and transfer learning to achieve
accuracy of 91 percent in ResNet model.
Area & Scope:
Drifting Icerbergs present threats to navigation and activities in areas such as
offshore. Companies use aerial reconnaissance and shore-based support to
monitor environmental conditions and assess risks from icebergs.In remote
areas with particularly harsh weather, these methods are not feasible, and the
only viable monitoring option is via satellite.Considering this need to bring in
the advancements of Machine Learning and Deep Learning truly in
real-life problems.
Final Title of Project:
ICEBERG-SHIP CLASSIFICATION OF SATELLITE RADAR
IMAGES
Signature of PBL Coordinator/HOD,FE
WeeklyPlanningSheet
Week Signatureof Signatureof
ActivityPlanned ActivitiesCompleted
No. Students Faculty/Mentor
With the help our mentor we Completed
successfully formed our group
We did several research like Completed
net surfing , newspaper
reading etc.,for finalising our
topic for project
2
We discussed our ideas and Completed
motive of our project with our
mentor
Activity 4: We all group Completed
members collected
information on different parts
of our project and sub-topics
like
4 1.Simple Convolution Neural
Network
2.Convolution Neural
Network with Data
Augmentation
3.Convolution Neural
Network with incidence angle
4.Transfer Learning
SignatureofPBLCoordinator/HOD,FE
WeeklyPlanningSheet
5.Resent model
6.VGG16 model
7.Feature Extraction and
Different Classifiers
SignatureofPBLCoordinator/HOD,FE
WeeklyPlanningSheet
Week Signatureof Signatureof
ActivityPlanned ActivitiesCompleted
No. Students Faculty/Mentor
We collected all the data & Completed
information and prepared our
report
After that we prepared our Completed
PPT
We wound up project by Completed
doing all the finalization
which were required
We presented our project Completed
SignatureofPBLCoordinator/HOD,FE
