0% found this document useful (0 votes)
2 views26 pages

prepcvip

The document provides an overview of computer vision and image processing, detailing the definitions, types, applications, and workflows involved in human and computer vision. It highlights the advantages and limitations of both human and computer vision, as well as the historical development of these technologies. Additionally, it covers image acquisition methods, image representation, and the various applications of computer vision across different fields.

Uploaded by

srinurandom
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
2 views26 pages

prepcvip

The document provides an overview of computer vision and image processing, detailing the definitions, types, applications, and workflows involved in human and computer vision. It highlights the advantages and limitations of both human and computer vision, as well as the historical development of these technologies. Additionally, it covers image acquisition methods, image representation, and the various applications of computer vision across different fields.

Uploaded by

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

UNIT–1

Overview of Computer Vision and Image


Processing: Definitions and Scope

Vision
Vision is the ability to perceive, identify, and understand objects and the surrounding
environment through light entering the eyes.

Types
1. Human Vision
2. Computer Vision

Real-World Examples
 A person recognizing a friend's face.
 A driver seeing traffic signals.
 A student reading a book.

Diagram
Light

Human Eye

Brain

Understand Scene

Applications
 Reading books
 Driving vehicles
 Identifying objects
 Recognizing faces

Human Vision
Human Vision is the natural ability of humans to observe, recognize, and interpret the
surrounding environment using the eyes and brain.

Components
 Eye
 Retina
 Optic Nerve
 Brain

Work flow
Object

Light Reflection

Eye

Retina

Optic Nerve

Brain

Recognized Object

Real-World Example
You see a mango on a table and immediately recognize it because your brain has learned its
shape and color.

Advantages
 Learns from experience
 Adapts to different environments
 Excellent object recognition

Limitations
 Gets tired
 Affected by lighting conditions
 Subject to human error

Computer Vision
Computer Vision is a branch of Artificial Intelligence (AI) that enables computers to acquire,
process, analyze, and understand images or videos to make decisions automatically.

Types (Major Tasks)


 Image Classification
 Object Detection
 Image Segmentation
 Face Recognition
 Image Tracking

Computer Vision Workflow


Real World Object


Image Acquisition
(Camera / Sensor)


Image Preprocessing
(Resize, Filter, Threshold)


Feature Extraction


Machine Learning / AI Model


Object Recognition / Classification


Decision / Output

Real-World Examples
 Face Unlock on smartphones
 Self-driving cars detecting pedestrians
 CCTV detecting intruders
 Google Lens identifying objects
 Medical diagnosis from X-rays

Advantages
 High speed
 High accuracy
 Works continuously without fatigue
 Processes large volumes of images

Limitations
 Requires quality data
 Computationally intensive
 Sensitive to poor lighting and image quality

Image Processing
Definition
Image Processing is the process of performing operations on a digital image to improve its
quality or extract useful information.

Types
Based on Processing

1. Analog Image Processing


2. Digital Image Processing

Common Operations

 Brightness Adjustment
 Contrast Adjustment
 Filtering
 Thresholding
 Histogram Processing

Image Processing Workflow


Input Image


Image Enhancement


Noise Removal


Feature Extraction


Output Image

Scope of Computer Vision


Definition
The scope of Computer Vision refers to the wide range of fields where image analysis and visual
understanding techniques are applied to solve real-world problems.

Applications of Computer Vision


Applications of Computer Vision are the real-world areas where computers analyze images or
videos to perform useful tasks automatically.

Major Application Area


1. Healthcare

 MRI Scan Analysis


 CT Scan Analysis
 Tumor Detection
 Disease Diagnosis

Example:
AI detecting lung cancer from X-ray images.

2. Security and Surveillance

 Face Recognition
 CCTV Monitoring
 Intruder Detection

Example:
Airport security systems.

3. Autonomous Vehicles

 Lane Detection
 Traffic Sign Recognition
 Pedestrian Detection
 Obstacle Detection
Example:
Self-driving cars.

4. Agriculture

 Crop Disease Detection


 Fruit Quality Inspection
 Weed Detection

Example:
Drone monitoring crop health.

5. Manufacturing

 Defect Detection
 Product Quality Inspection
 Robot Guidance

Diagram
Computer Vision

┌──────────┬──────────┬──────────┬──────────┐
│ │ │ │ │
Healthcare Security Robotics Agriculture Transport

Real-World Examples
Healthcare

Detecting tumors from MRI and CT scan images.

Security

Face recognition at airports.

Agriculture

Identifying plant diseases from leaf images.

Transportation

Automatic number plate recognition.


Industry

Detecting defective products on manufacturing lines.

Advantages of Computer Vision


 Reduces human effort
 Fast image analysis
 Improves accuracy
 Works 24×7
 Supports automation

Limitations
 Needs high-quality images
 High computational cost
 Performance decreases with poor lighting
 Requires training data for AI-based systems

Historical Development
The historical development of Computer Vision and Image Processing refers to the evolution
of technologies, techniques, and algorithms that enable computers to acquire, process, analyze,
and understand images from the real world.

Timeline of Historical Development


Year Development Description
Digital Image
1950s Scientists began processing digital images using computers.
Processing
NASA used image processing to enhance images from satellites
1960s Space Image Processing
and spacecraft.
Medical Image
1970s CT scans and MRI imaging became popular in healthcare.
Processing
1980s Pattern Recognition Computers started recognizing shapes, numbers, and characters.
Early face detection and biometric security systems were
1990s Face Recognition
developed.
Computer vision systems learned from training data to improve
2000s Machine Learning
recognition.
Year Development Description
CNNs greatly improved image classification and object detection
2010s Deep Learning
accuracy.
AI-Based Vision Used in autonomous vehicles, smart surveillance, robotics, and
2020s
Systems healthcare.

Timeline Diagram
1950s → Digital Image Processing

1960s → Satellite Image Processing

1970s → Medical Imaging

1980s → Pattern Recognition

1990s → Face Recognition

2000s → Machine Learning

2010s → Deep Learning

2020s → AI-Based Computer Vision

Evolution of Computer Vision


Image Capture

Image Processing

Pattern Recognition

Machine Learning

Deep Learning

Artificial Intelligence

Smart Computer Vision Systems

Advantages
 Fast image analysis
 High accuracy
 Reduces manual effort
 24×7 operation
 Supports automation
 Improves decision making

Limitations
 Requires high-quality images
 High computational requirements
 Performance affected by poor lighting
 Requires trained AI models and datasets

Image Formation
Image Formation is the process by which light reflected from an object passes through an optical
system (camera lens or human eye) and forms an image on a sensor or retina.

Types
1. Binary Image

 Pixel values: 0 or 1
 Black and White only

Example:
QR Code

2. Grayscale Image

 Pixel values: 0–255


 Black to White

Example:
X-ray image

3. Color Image

 Three color channels


o Red
o Green
o Blue

Example:
Mobile phone photograph
Image Formation Process (Workflow)
Real World Object


Light Reflection


Camera Lens / Human Eye


Image Sensor / Retina


Digital Image


Computer Storage

Image Representation
Image Representation is the process of storing an image digitally as a collection of pixels, where
each pixel contains intensity or color information.

Inside a computer, an image is stored as rows and columns of pixels.

Example:

Image

┌──┬──┬──┬──┐
│25│80│60│40│
├──┼──┼──┼──┤
│55│90│75│20│
├──┼──┼──┼──┤
│15│45│65│95│
└──┴──┴──┴──┘

Each number represents the intensity value of one pixel.

Real-World Examples
Mobile Camera

The mobile camera captures your selfie and stores it as millions of pixels.

CCTV Camera
Captures continuous video frames.

Each frame is an image.

Satellite Imaging

Earth images captured by satellites are represented as digital pixels.

Medical Imaging

MRI and CT scans are stored as grayscale images.

Applications
 Digital Photography
 Medical Imaging

Advantages
 Easy to store
 Easy to process
 High accuracy
 Supports AI applications
 Enables image enhancement

Limitations
 Large images require more storage.
 High-resolution images require more processing time.
 Image quality depends on camera resolution and lighting conditions.

Image Acquisition
Image Acquisition is the process of capturing an image from the real world using an imaging
device (camera or sensor) and converting it into a digital image that can be processed by a
computer.

Why is Image Acquisition Important?


Without image acquisition, a computer has no visual data to process.
It is the starting point for all Computer Vision tasks such as:

 Face Recognition
 Medical Imaging
 Object Detection
 OCR
 Satellite Image Analysis

Image Acquisition Workflow


Real World Object


Light Falls on Object


Camera / Sensor


Analog Signal


Analog-to-Digital Converter (ADC)


Digital Image


Computer Vision Processing

Image Acquisition Methods


There are several methods to acquire images.

Method 1: Camera-Based Acquisition


Definition

A digital camera captures light reflected from an object and converts it into a digital image.

Examples

 Mobile Camera
 DSLR Camera
 Webcam
 CCTV Camera

Applications

 Photography
 Face Recognition
 Video Conferencing
 Surveillance

Method 2: Scanner-Based Acquisition


Definition

A scanner captures printed documents or photographs and converts them into digital images.

Examples

 Flatbed Scanner
 Document Scanner
 Barcode Scanner

Applications

 Document Digitization
 OCR
 Banking
 Libraries

Method 3: Medical Imaging Acquisition


6

Definition

Medical devices capture images of internal body structures for diagnosis.

Examples

 MRI
 CT Scan
 X-ray
 Ultrasound

Applications

 Tumor Detection
 Bone Fracture Analysis
 Organ Examination
Method 4: Satellite and Remote Sensing
6

Definition

Satellites capture images of the Earth's surface from space.

Applications

 Weather Forecasting
 Agriculture
 Disaster Management
 Urban Planning
Method 5: Industrial Image Acquisition
6

Definition

Industrial cameras capture images during manufacturing for inspection and automation.

Applications

 Quality Inspection
 Defect Detection
 Robotics
 Automated Assembly

5. Comparison of Image Acquisition Methods


Method Device Applications
Camera-Based Mobile, DSLR, Webcam Photography, Face Recognition
Scanner-Based Flatbed Scanner OCR, Document Scanning
Medical Imaging MRI, CT, X-ray Healthcare
Satellite Imaging Satellites Agriculture, Mapping, Weather
Industrial Imaging Machine Vision Cameras Manufacturing, Inspection

6. Real-World Examples
Mobile Phone

Take a selfie → Camera captures the image → Stored digitally.

ATM

Cheque scanner reads a cheque and converts it into a digital image.

Hospital

CT Scanner captures internal body images.

Google Maps

Satellites capture Earth's surface images.

Factory

Machine vision cameras inspect products for defects.

7. Applications
 Computer Vision
 Medical Diagnosis
 OCR
 Autonomous Vehicles
 Robotics
 Smart Agriculture
 Traffic Monitoring
 Security Surveillance
 Industrial Automation

8. Advantages
 Captures real-world information digitally.
 Enables automated image analysis.
 High-speed acquisition.
 Supports AI and Machine Learning applications.

9. Limitations
 Poor lighting affects image quality.
 Expensive sensors may be required.
 Noise can reduce image quality.
 Weather conditions can affect outdoor image capture.

You might also like