FINAL PROJECT

Linear algebra

December 30, 1899

GROUP MEMBERS
HOORIA MINHAJ,EMAAN ASHRAF,AREEBA SAEED, LAIBA KABIR NOUSHEEN B
 APPLICATION OF LINEAR TRANSFORMATION
Linear transformations are widely used in various fields. They manipulate images and objects in
computer graphics, reduce dimensionality in data science, and filter signals in signal processing.
They model control systems, describe quantum states and electromagnetic fields in physics, and
optimize economic models. In robotics, they help with kinematics and path planning, and they
simplify solve linear differential equations. These applications underscore their importance
across many domains.
Here, we will discuss use of linear transformation in image transformation, cryptography,
coding, IT and data analysis.

1. Image Transformation

The geometric transformation of the image coordinate system in order to align partially
overlapping images taken at different times or with different devices.
In digital image transformation/image processing system, we will develop a system that whose
input would be an image and output would be an image too. And the system would perform
some processing on the input image and gives its output as a processed image. It is shown
below:

Now function applied inside this digital system that process an image and convert it into output
can be called as transformation function.
As it shows transformation or relation, that how an image 1 is converted to image 2.
Consider this equation; G ( x,y ) = T { f ( x,y ) }
In this equation,
f( x,y ) = Input image on which transformation function has to be applied.
G( x,y ) = The output image or processed image.
T is the transformation function.
This relation between input image and the processed output image can also be represented as.
s = T (r)
where r is actually the pixel value of f ( x,y ) at any point. And s is the pixel value of g ( x,y ) at
any point.
 Matrix Representation of Image Transformation:
A simple two-dimensional matrix has m rows and n columns as shown below, and so is the
image in the computer, where the image has pixels of m rows and n columns with each pixel
having an attribute of color in RGB format. For simplicity, an image as a two-dimensional matrix.
This paper will use house images and heart images to show matrix usage as an example.
By applying addition operation matrices, we can merge two images.

We get an image in the dark if we subtract the heart image from the house image.

By applying multiplication operation, we can add filters to the image such as Gray-scale
Conversion and Color/Channel Transformation.

Now, we are going to discuss some of the very basic transformation functions.
 Resize:
This transformation often referred to as expansion, dilation, or contraction, is widely used when
we want to make an object appear larger or smaller, in other words, to change its scale. When
we zoom in or out on a photograph to enlarge or reduce it, we are applying a scaling
transformation. Example of a scaling linear transformation.
T( x,y)= (ax,by)

(20 02)
 Rotation:
This transformation rotates the object in the plane around a base point. A real-world scenario
would be when you have a photo that’s upside down on its side, and you want to set it straight,
that’s when you apply a rotation transformation. In robotics rotation matrices are fundamental
for representing and controlling the orientation of robotic components.
Equation of rotation transformation.
T(x,y) =(x.cos θ – y.sinθ, x.sinθ + y.cos θ)
Matrix of Transformation

(cosθ−sinθ
sin θ cosθ) Rotation of an
object in a plane.
 Translation:
Translation allows us to move objects within the plane, either horizontally, vertically, or a
combination of both. When an object undergoes a translation, it maintains its size and shape,
changing solely in its position. This concept is exemplified in the figure.
Matrix equation representing the new position of the object.

[ ][ ][ ] [ ][ ][ ]
' '
Ax Ax + ∆ A x Bx Bx + ∆ Bx
' = ' =
Ay Ay ∆ Ay By By ∆ By
[ ][ ][ ] [ ][ ][ ]
' '
Cx Cx + ∆ Cx D x = Dx + ∆ Dx
' =
Cy Cy ∆ Cy Dy ' Dy ∆ Dy
 Mirror
The reflection transformation applied to an object produces a mirrored object concerning one or
both axes.
Reflection across the x-axis:
Equation and matrix of mirror transformation across x-axis.
Visual Example of reflection in x-axis.

T(x,y)= (x,-y)

[ 0−1 ] [ y ] = [−xy ]
10 x

Reflection across the y-axis:

Equation and matrix of mirror transformation across y-axis.

T(x,y)= (x,-y)

[−10 10][ xy ] = [−xy ]

Visual example of mirror transformation in y-axis.
 Shearing:
The shear transformation distorts the object in one or both of its coordinates. It is widely used in
image distortion filters in graphic design programs.
Equation and Matrix of horizontally shearing;
T(x,y)= (x+ay, y)

[ 101∞]
2. Cryptography

Cryptography is the science of securing information by transforming it into an unreadable

format, ensuring that only authorized parties can access and understand it. Here are the key
concepts and components explained in simple terms:
 Daily Life Examples:
i. Online Shopping: When you enter your credit card information on an e-commerce site, it's
encrypted to prevent hackers from stealing it.
ii. Banking: Online banking transactions are encrypted to protect your financial information
from being intercepted.
iii. Secure Websites: Websites use HTTPS (which relies on SSL/TLS, a form of cryptography) to
secure data transferred between your browser and the website.
 Types of linear transformations and their use in cryptography in simpler terms
 Rotation:
 Shifting Letters:
Imagine you rotate each letter by a certain number of places in the alphabet. For example, 'A'
becomes 'D' if rotated by 3 places. This is a simple form used in some ciphers.
Example: Imagine you and a friend want to keep your messages secret. You decide to shift each
letter in your message by 3 places in the alphabet. So, "HELLO" becomes "KHOOR" (H becomes
K, E becomes H, and so on). You send "KHOOR" to your friend, who knows to shift the letters
back by 3 to read the original message "HELLO". This way, anyone who intercepts the message
sees only the scrambled letters.
 AES Encryption:
In the AES encryption method, data is put into a grid. Rows of this grid are rotated to mix up the
data, making it harder for anyone to figure out the original message.
Example: Imagine you use an online banking app to check your balance. AES encryption
protects your data by turning your information into a complex code when you send it over the
internet. Only the bank's server, which has the key, can decode it back to your readable
information. This keeps your financial details safe from hackers during transmission.
 Reflection/Magnification:
 Elliptic Curve Math:
When working with elliptic curves (a type of math used for encryption), points on the curve can
be flipped over an axis. This helps in creating secure keys for encryption.
Example: Imagine you send a digital signature to verify an online document. Elliptic Curve
Cryptography (ECC) helps create this signature using complex math based on elliptic curves.
Only the intended recipient, using a corresponding key, can verify that the signature is valid.
This ensures that your document is authentic and hasn't been tampered with, keeping your
online transactions secure.
 Data Mixing:
In some encryption methods, small parts of the data are reflected or magnified to create
complex patterns. This makes it tough for an attacker to reverse-engineer the data.
Example: Imagine you're mixing ingredients in a smoothie. Just like blending fruits and
vegetables hides the individual ingredients, cryptography mixes data to hide the original
information. For example, in AES encryption, data mixing ensures that every piece of the
original message is spread throughout the encrypted message. This way, even if someone sees
the encrypted data, they can't figure out the original message, keeping your information secure.
 Translation:
 Adding Numbers:
In some ciphers, each number or letter is shifted by adding a constant value. Think of it like
moving every point on a number line by a fixed amount.
Adding numbers in cryptography involves shifting each letter or number by a fixed amount to
scramble the original message. For instance, in an affine cipher, each letter is shifted by adding a
specific number (like adding 3 to 'A' makes it 'D'). This transformation makes it harder for
unauthorized people to read the message without knowing the exact shift. In everyday
scenarios, this method is like shifting each piece of a puzzle by a certain number of places to
create a new pattern that only those with the right instructions can decipher.
 Random Sequence Generators:
For creating pseudo-random numbers (used to secure data), simple shifts and adds are used to
produce sequences that look random but can be reproduced by someone who knows the
starting point.
Random sequence generators are crucial in cryptography for creating unpredictable numbers or
patterns used in encryption keys and secure communications. In daily life, think of them like a
dice roll or a lottery draw where each outcome is random and unpredictable. Cryptographic
algorithms use sophisticated methods to generate these sequences, ensuring that no one can
predict the next number or sequence without knowing the specific algorithm and initial
conditions. This randomness adds a layer of security by making it extremely difficult for
attackers to guess or replicate the encrypted data or keys, protecting sensitive information in
digital transactions and communications.
 Inversion:
 Mixing Data in AES:
In AES, parts of the data are mixed by processes that include flipping numbers around in a
specific way. This ensures that every output part depends on multiple input parts, making it
more secure.
In AES (Advanced Encryption Standard), data mixing is crucial for securing digital information.
Imagine you're packing items into a suitcase for a trip. AES works similarly by organizing your
data into blocks and then mixing them up using mathematical operations. This mixing ensures
that every bit of your information interacts with every other bit, creating a complex pattern that
is difficult for unauthorized individuals to decipher. It's like jumbling up puzzle pieces so that
even if someone sees them, they can't put them back together without the right instructions
(the decryption key). This method keeps sensitive information, like passwords or financial
details, safe from prying eyes during transmission over the internet.
 Key Math:
When creating keys for encryption methods like RSA and ECC, inverting numbers in a special
mathematical way is necessary. This inversion helps in verifying signatures and decrypting
messages securely.
In cryptography, key mathematics is crucial for securing and decrypting data. For example, in
RSA encryption, two keys are generated—a public key for encrypting data and a private key for
decrypting it. Think of the public key like a mailbox where anyone can drop a letter, but only the
person with the private key (the mailbox key) can open and read those letters. This ensures that
sensitive information, such as passwords or financial transactions, remains confidential and
protected from unauthorized access. In everyday life, similar principles apply when using secure
websites that encrypt your information before transmitting it, ensuring privacy and security
online.
 3. Coding

Coding refers to the process of creating instructions for computers using programming
languages. It involves writing, testing, and maintaining the source code of computer programs,
which are then executed by computers to perform specific tasks or functions. Coding is essential
in software development, enabling the creation of applications, websites, and various digital
tools that we use in everyday life.
 Usage of linear transformations in coding
It is diverse and widespread, with applications across various fields. Here are some common
ways in which linear transformations are utilized in coding:
 Computer Graphics: In computer graphics, linear transformations are used to manipulate
objects, images, and animations. Operations like rotation, scaling, translation, and shearing
are essential for creating visual effects in video games, simulations, and graphic design
applications.
 Image Processing: Linear transformations play a crucial role in image processing tasks such
as resizing, filtering, and geometric transformations. Techniques like rotation, reflection, and
scaling are applied to enhance and analyze images in applications like photo editing
software and medical imaging systems.
 Machine Learning: Linear transformations are integral in machine learning algorithms for
tasks such as dimensionality reduction, feature extraction, and data normalization.
Techniques like principal component analysis (PCA) and linear discriminant analysis (LDA)
rely on linear transformations to preprocess and analyze data effectively.
 Signal Processing: In signal processing applications, linear transformations are used to
analyze and manipulate signals efficiently. Operations like Fourier Transform, Discrete Cosine
Transform, and Wavelet Transform are applied for filtering, compression, and feature
extraction in audio, video, and communication systems.
 Data Visualization: Linear transformations are employed in data visualization techniques to
project high-dimensional data into lower-dimensional spaces for visualization and analysis.
Methods like multidimensional scaling and dimensionality reduction algorithms use linear
transformations to visualize complex datasets in a more interpretable form.
 Robotics and Control Systems: Linear transformations are essential in robotics and control
systems for tasks like coordinate transformations, path planning, and robot kinematics. By
applying linear transformations, robots can navigate their environment, manipulate objects,
and perform tasks accurately.
 Data Compression: Linear transformations are utilized in data compression techniques to
reduce redundancy and extract essential features from data. Methods like Singular Value
Decomposition (SVD) and discrete cosine transform (DCT) use linear transformations to
compress data efficiently in applications like image and video compression.
 Mathematical Modeling: Linear transformations are used in mathematical modeling to
represent and solve systems of equations, analyze data patterns, and simulate real-world
phenomena. Linear algebraic techniques involving matrices and transformations are applied
in various mathematical and scientific computations.

 Benefits of Utilizing Linear Transformations in Coding

 Efficiency: Linear transformations enable efficient manipulation and transformation of data
by utilizing matrix operations, which are computationally optimized for high performance.
 Versatility: Linear transformations offer a wide range of operations such as rotation,
reflection, translation, and inversion, providing programmers with versatile tools to handle
various coding tasks effectively.
 Simplicity: Linear transformations simplify complex coding tasks by representing
transformations through matrices, making it easier to understand and implement different
operations.
 Modularity: Linear transformations allow for modular coding practices where
transformations can be combined, decomposed, and reused in different parts of the
codebase, promoting code reusability and maintainability.
 Geometric Intuition: Linear transformations provide a geometric intuition for understanding
data transformations, especially in fields like computer graphics, image processing, and data
visualization, enhancing the programmer's visualization skills.
 Optimization: Linear transformations can be optimized and parallelized for efficient
computation, making them suitable for handling large datasets and complex algorithms in
coding applications.
 Scalability: Linear transformations can scale to handle varying data sizes and dimensions,
making them suitable for applications that require dynamic data processing and
transformation.
 Interdisciplinary Applications: Linear transformations find applications in diverse fields such
as machine learning, robotics, signal processing, and data analysis, allowing programmers to
leverage their skills across different domains.
By leveraging the benefits of linear transformations in coding, programmers can enhance their
coding efficiency, improve data manipulation capabilities, and develop innovative solutions in
various domains effectively.
 Techniques
Rotation, reflection, translation, and inversion are fundamental linear transformations that play
a significant role in various coding applications. Let's explore each of these techniques in the
context of coding:
 Rotation:
 Description: Rotation is a linear transformation that involves changing the orientation of an
object or data points around a fixed point (usually the origin) by a certain angle.
 Application: In coding, rotation is commonly used in computer graphics, image processing,
and geometric transformations to rotate objects, images, or coordinate systems.
 Implementation: Rotation can be implemented using transformation matrices or
trigonometric functions to rotate points or vectors in 2D or 3D space.
 Reflection:
 Description: Reflection is a linear transformation that involves flipping an object or data
points across a line (mirror line).
 Application: In coding, reflection is used in computer graphics, image processing, and
symmetry operations to create mirror images or reflect objects across a specified axis.
 Implementation: Reflection can be implemented by applying transformation matrices that
involve negating specific coordinates based on the reflection axis.
 Translation:
 Description: Translation is a linear transformation that involves shifting an object or data
points by a certain distance in a specified direction.
 Application: In coding, translation is used in computer graphics, image processing, and
geometric transformations to move objects or images to new positions.
 Implementation: Translation can be implemented by adding or subtracting constant values
to the coordinates of points or vectors to shift them in the desired direction.
 Inversion:
 Description: Inversion is a linear transformation that involves finding the inverse of a matrix
or function that undoes the original transformation.
 Application: In coding, inversion is used in solving systems of linear equations, data
normalization, and error correction.
 Implementation: Inversion can be implemented using mathematical techniques such as
matrix inversion algorithms (e.g., Gaussian elimination) to find the inverse of a matrix.

4. Information Technology

IT, short for Information Technology, refers to the use of computers, software, networks, and
systems to store, retrieve, transmit, and manipulate data and information. It encompasses
technologies that enable individuals and organizations to manage and process data effectively
for various purposes, including communication, automation, and decision-making. Linear
transformations are crucial in IT for several applications. In computer graphics, they rotate,
scale, and move objects in images and animations. For data compression, they convert image
data for easier compression, like in JPEG, to reduce file sizes. Cryptography uses them in
encryption algorithms to obscure data and protect information. In machine learning, linear
transformations adjust neural network weights and biases for better pattern recognition. Finally,
in signal processing, they help filter and analyze signals by reducing noise and
extracting features.
 Types:
Linear transformations can be categorized into several types, each serving specific purposes in
applications like graphics, data analysis, and more. Here are the main types along with
examples, benefits, and sample programs:
 Rotation:
 Description:
Rotates a vector around the origin by a certain angle.
 Uses:
It is used in Rotating objects and images.
 Benefits:
Maintains the object's shape and size.
Essential for animations and 3D rendering.

Program (Python with NumPy):

import numpy as np

# Define a rotation matrix for a 45-degree rotation

theta = np.pi / 4
rotation_matrix = np.array([
[np.cos(theta), -np.sin(theta)],
[np.sin(theta), np.cos(theta)]
])

# Define a vector
vector = np.array([1, 0])

# Apply the rotation

rotated_vector = np.dot(rotation_matrix, vector)

print("Rotated vector:", rotated_vector)

 Reflection:
 Description:
Flips a vector over a specified axis.
 Uses:
Image Processing: Mirroring images.
Graphics: Creating symmetrical designs.
 Benefits:
Useful for symmetry operations.
Simple to implement.
 Program (Python with NumPy):

import numpy as np

# Define a reflection matrix over the x-axis

reflection_matrix = np.array([
[1, 0],
[0, -1]
])

# Define a vector
vector = np.array([1, 2])

# Apply the reflection

reflected_vector = np.dot(reflection_matrix, vector)

print("Reflected vector:", reflected_vector)

 Translation
 Description:
Moves a vector by adding a constant value to its coordinates.
 Uses:
Image Processing: Shifting images.
Graphics: Moving objects within a scene.
 Benefits:
Simple to implement.
Useful for positioning objects.
Program (Python with OpenCV):
import cv2
import numpy as np

# Load an image
image = cv2.imread('image.jpg')

# Define a translation matrix

tx, ty = 100, 50 # Translation offsets
translation_matrix = np.float32([[1, 0, tx], [0, 1, ty]])

# Apply the translation

translated_image = cv2.warpAffine(image, translation_matrix, (image.shape[1],
image.shape[0]))

# Save the result

cv2.imwrite('translated_image.jpg', translated_image)

 Inversion
 Description:
Inverts a matrix, reversing its transformation.
 Uses:
Linear Algebra: Solving systems of linear equations.
Computer Graphics: Reversing transformations.
 Benefits:
Restores original data after transformation.
Useful for undo operations.
Program (Python with NumPy):
import numpy as np

# Define a matrix
matrix = np.array([[4, 7], [2, 6]])

# Compute the inverse

inverse_matrix = np.linalg.inv(matrix)

print("Inverse matrix:", inverse_matrix)

These types of linear transformations are widely used in IT to manipulate and analyze data
efficiently. Each type has its specific use cases and benefits, making them essential tools in fields
such as computer graphics, data analysis, and machine learning.

5. Data analysis

Data analysis is the process of inspecting, cleansing, transforming, and modeling data with the
goal of discovering useful information, informing conclusions, and supporting decision-making.
Linear transformations are fundamental in data analysis, allowing us to manipulate data in ways
that simplify analysis, enhance visualization, and improve the performance of machine learning
models. They are mathematical operations applied to data points in a vector space, preserving
linear relationships among the points. Linear transformations include rotation, inversion,
translation, and reflection have unique properties and applications.
 Rotation
Rotation involves transforming data points around a fixed center, preserving their geometric
properties while altering their orientation in space.
 Principal Component Analysis (PCA):
PCA rotates data to align principal components along directions of maximum variance,
effectively transforming the data into a new coordinate system.
This transformation is crucial for reducing the dimensionality of high-dimensional datasets while
retaining as much variance as possible. It aids in visualizing complex data structures, identifying
dominant patterns, and extracting meaningful features for further analysis. PCA is widely used
in image processing, finance, genetics, and other fields where understanding data variability is
essential.
 Image Processing:
Rotating images corrects orientation errors, allowing for consistent interpretation and analysis.
In medical imaging, for example, correctly orienting MRI or CT scan images ensures accurate
diagnosis and treatment planning. In remote sensing, image rotation facilitates precise mapping
and monitoring of environmental changes. The ability to adjust image orientation enhances the
reliability and usability of data across various applications, from healthcare to environmental
science.
 Data Visualization:
Rotating scatter plots or 3D visualizations provides alternative perspectives on data distributions
and relationships.
Different viewpoints enable analysts to uncover hidden patterns, outliers, and correlations that
might not be apparent from a single angle. Rotating visualizations aids in exploratory data
analysis by facilitating interactive exploration and hypothesis generation. It supports decision-
making processes by providing comprehensive insights into complex datasets, enhancing
understanding and communication of data-driven findings.
 Inversion
Inversion flips data points relative to a fixed point or axis, reversing their positions while
preserving their geometric relationships.
 Signal Processing:
Inverting signals involves changing their phase or amplitude characteristics, often used in noise
reduction and signal enhancement.
In audio processing, for instance, signal inversion helps cancel out echoes and background noise
by subtracting the inverted signal from the original. This technique improves sound clarity and
fidelity in communication systems and audio recording environments. Inversion also facilitates
phase analysis, critical for studying signal coherence and synchronization in fields like
telecommunications and biomedical engineering.
 Data Transformation:
Inverting data values involves scaling or transforming them to fit within a standardized range,
such as mapping values from 1 to -1.
This transformation is essential for normalizing data distributions across different scales or
units, enabling fair comparisons and statistical analysis. In financial analysis, for example,
inverting stock price movements allows analysts to compare performance metrics consistently
over time or across different securities. In machine learning, data inversion supports
preprocessing steps that enhance model training and accuracy by reducing data variability and
bias.
 Data Augmentation:
Inverting features or labels generates synthetic data points to augment existing datasets for
machine learning models.
Augmentation through data inversion improves the robustness and generalization of machine
learning algorithms. By introducing variations in feature values or target labels, models become
more adept at handling diverse inputs and making accurate predictions on unseen data. This
technique is particularly valuable in tasks requiring large datasets or when data scarcity limits
model training effectiveness.
 Translation
Translation shifts data points along specified vectors, displacing their positions without altering
their internal structure or relationships.
 Time-Series Analysis:
Translating time-series data adjusts timestamps or chronological sequences for alignment and
comparative analysis.
In financial forecasting, translating historical stock prices or economic indicators facilitates trend
analysis and pattern recognition. It helps analysts identify seasonal fluctuations, long-term
trends, and recurring cycles that influence market behavior and investment strategies. Time-
series translation is also critical in operational analytics for optimizing resource allocation and
production scheduling based on historical performance data.
 Geographic Data Analysis:
Translating geographic coordinates involves adjusting spatial data points to align with a common
reference system or coordinate grid.
This transformation supports spatial analysis and mapping applications in geographic
information systems (GIS). By standardizing location data from diverse sources, such as GPS
devices or satellite imagery, analysts can overlay and integrate datasets for urban planning,
environmental monitoring, and disaster response. Geographic translation enhances the
accuracy and interoperability of spatial data, enabling informed decision-making and policy
development in areas ranging from infrastructure development to natural resource
management.
 Data Integration:
Translating datasets involves aligning data from different sources or formats to facilitate unified
analysis and comparison.
Integrating diverse datasets into a common framework enhances the completeness and
reliability of analytical insights. By standardizing variables or aligning data structures,
organizations can combine information from multiple systems or departments for
comprehensive business intelligence and strategic planning. Data integration supports cross-
functional collaboration and data-driven decision-making, enabling stakeholders to leverage
aggregated insights for operational efficiency, customer engagement, and competitive
advantage.
 Reflection
Reflection creates mirror images of data points across specified axes or planes, preserving their
shapes while changing their orientations.
 Pattern Recognition:
Reflecting data involves identifying symmetrical patterns and features within datasets or
images.
In computer vision and pattern recognition, reflecting image data helps detect and classify
objects based on their geometric properties and spatial relationships. By analyzing mirrored
representations, algorithms can identify recurring patterns, shapes, and structures that
contribute to accurate object detection, image segmentation, and scene understanding.
Reflection enhances the robustness and accuracy of automated systems in applications ranging
from facial recognition and object tracking to industrial inspection and autonomous navigation.
 Data Transformation:
Reflecting data standardizes orientations to ensure consistent analysis and model performance.
This transformation improves the reliability and interpretability of data by eliminating biases or
discrepancies in representation. In statistical analysis and machine learning, reflecting data
points or variables enhances model training and evaluation by aligning inputs with predefined
criteria or assumptions. It supports rigorous data preprocessing techniques that optimize
feature engineering, model calibration, and predictive accuracy across diverse domains and
application scenarios.
 Error Analysis:
Reflecting measurements involves validating data accuracy through comparative analysis
between original and mirrored results.
In scientific research, quality control, and experimental design, reflecting measurements serves
as a critical validation technique for verifying data integrity and reliability. By cross-referencing
mirrored data points or experimental outcomes, researchers can identify and correct
measurement errors, calibration discrepancies, or systematic biases that may affect research
findings or experimental conclusions. Reflection supports rigorous data validation protocols and
ensures the reproducibility and robustness of scientific investigations in fields such as physics,
chemistry, biology, and engineering.