Types of images and their main characteristics

Department of Data Communication Networks and Systems

Lecturer Shukhrat Palvanov

What is a Image?

• A image = 2D representation of a scene/object.
• Composed of pixels (picture elements), each storing brightness or color.

Image Representation in Computers

• Stored as a grid (matrix) of rows × columns.
• Pixel values depend on image type:
• Binary → 0 or 1
• Grayscale → 0–255 (8-bit)
• Color → R, G, B channels

Why Image Processing Matters

• Enables analysis, enhancement, and compression.
• Applications in medicine, security, remote sensing, and daily technology.

Introduction to Digital Image

It is a type of image which is formed from pixels. Each pixel has some finite size and is represented by some finite intensity to show the image. The pixels are set of a properly arranged rectangular array. The image size can be determined by the pixel array dimensions. It has finite coordinates i.e. x-coordinate and y-coordinate. It is of two types i.e. raster type and vector type. The raster image is basically used for reference to a digital image.

What is Digital Image?

It is the type of image which is made of pixels. It includes printed texts, photographs, and artwork. It is captured from the digital camera, any image machine. As it is made up of pixels it is used to define the quality of the image. The pixel value is used to measure the quality of the image. The binary code is used to represent the value of a pixel. The resolution can be measured by the value of pixel per inch value. For storing it in the computer there are different file formats used which are TIFF, GIF, JPEG, BMP.

Formats: JPEG�

The full form of JPEG is the Joint Photographic Experts Group. The JPEG format is used to store the information in a small size of the file. The digital camera produces the digital image in the format of JPEG because the file size is smaller for JPEG format. JPEG is not used when the image is used as logos or drawings as the image is in compressed form and when it is zoomed the quality of the image gets decreased.

TIFF

The full form of TIFF is the Tagged Image File Format. The file size of TIFF digital image format is comparatively large from the JPEG format. The file size is larger because the images are present in uncompressed form. The photo software produces the TIFF format image in Photoshop.

PNG

The full form of PNG is Portable Network Graphics. Most web images present on the internet is present in the format of PNG. The PNG format is not used as the file size is larger for PNG images compare to JPEG images. But the image of text is of small size with great quality of image resolution. When any user takes the screenshot of the screen of a computer the image is stored in the format of PNG. As the image is a mixed image that contains text and pictures the image is stored in the format or PNG format.

Types of Digital Images

• Binary (black & white)
• Grayscale
• Color (RGB, indexed, etc.)
• Multispectral / Hyperspectral

Binary Images

A binary image is a digital image where each pixel can take only

two possible intensity values: black or white.

• Key Features:
• Pixel values: 0 = black, 1 = white
• Requires 1 bit per pixel (smallest file size)
• Simplest image type in digital image processing
• Common Uses:
• Text recognition (OCR)
• Shape/edge detection
• Document scanning, barcodes, QR codes

​

Binary Images Characteristics

• Pixel Values:
• Only two values → 0 (black) and 1 (white)
• File Size Efficiency:
• Requires minimal storage (1 bit per pixel)
• Much smaller than grayscale or color images
• Processing Simplicity:
• Easy to analyze and process computationally
• Ideal for pattern recognition and segmentation
• Limitations:
• Cannot represent shades or colors
• Not suitable for detailed images

​

Grayscale Images

• A grayscale image is a digital image where each

pixel represents an intensity level between black and white.

• Pixel Values:
• Typically 8-bit → values range from 0 (black) to 255 (white)
• Higher bit-depths (10-bit, 12-bit, 16-bit) allow more intensity levels
• Characteristics:
• Captures details through different shades of gray
• Does not include color information
• Common in scientific, medical, and technical imaging

​

Grayscale Images (Applications)

• Medical Imaging
• X-rays, CT scans, and MRI scans use grayscale to highlight tissue differences
• Document Processing
• Scanning books, newspapers, and handwritten texts
• Easier compression compared to color images
• Texture & Pattern Analysis
• Used in industrial quality inspection (e.g., detecting surface defects)
• Remote Sensing & Astronomy
• Satellite images and telescope captures often stored in grayscale
• Computer Vision
• Object detection and recognition models often use grayscale preprocessing

​

Color Images

• A color image is a digital image where each pixel is represented

by a combination of values from multiple color channels.

• RGB Model (most common):
• Each pixel is a triplet (R, G, B)
• Example: (255, 0, 0) = Red
• 24-bit depth → over 16 million possible colors
• Other Representations:
• Indexed color images (palette-based)
• Advanced models like CMYK, HSV, HSI
• Key Advantage:
• Provides realistic representation of objects and scenes

​

Color Models

• Color Models
• RGB (Red, Green, Blue)
• Additive model: colors created by mixing light
• Used in monitors, cameras, digital displays
• CMYK (Cyan, Magenta, Yellow, Black)
• Subtractive model: colors created by mixing pigments/inks
• Standard in printing industry
• HSV / HSI (Hue, Saturation, Value/Intensity)
• Perceptual models closer to human vision
• Useful for image analysis, segmentation, and enhancement
• Other Models (optional mention):
• YUV, Lab → used in video and advanced imaging

​

Image Properties Overview

Resolution

• Number of pixels in the image (width × height)
• Determines the level of detail

Brightness

• Average intensity of pixels
• Controls lightness/darkness

Contrast

• Difference between light and dark regions
• Enhances visibility of features

Pixel Depth (Bit Depth)

• Number of bits per pixel
• Defines the range of intensity levels or colors

​

Resolution

• Number of pixels in image (width × height)
• Higher resolution = more detail

​

Resolution is the number of pixels used to represent an image, usually expressed as width × height.

• Key Points:
• Higher resolution → more detail, sharper image
• Lower resolution → less detail, pixelated appearance
• Resolution directly affects image quality and file size
• Example:
• 1920 × 1080 = 2,073,600 pixels (~2 Megapixels)

​

​

Resolution (Examples)

• Low Resolution
• 640 × 480 (VGA) → Basic detail, often pixelated
• Standard / High Definition
• 1280 × 720 (HD) → Clearer detail
• 1920 × 1080 (Full HD) → Widely used in media
• Ultra High Resolution
• 3840 × 2160 (4K UHD) → Very sharp and detailed
• 7680 × 4320 (8K UHD) → Extreme clarity, professional use
• Key Insight:
• Higher resolution = better quality, but larger storage and processing requirements

​

Brightness (Definition)

Brightness is the perceived intensity of light

in an image, determined by the average pixel values.

Key Points:

• Low brightness → darker image
• High brightness → lighter image

Can be adjusted by modifying pixel intensity values

Importance:

• Affects visibility of details
• Critical in applications like night vision, photography, and medical imaging

​

Brightness (Visualization)

Dark Image (Low Brightness):

• Most pixels have low intensity values
• Image looks dim and lacks visibility

Bright Image (High Brightness):

• Most pixels have high intensity values
• Image looks clearer but may lose detail in bright areas

Adjustment Method:

• Performed by adding or subtracting a constant to pixel values
• Example: useful for night vision enhancement

​

Contrast (Definition)

Contrast is the difference in intensity

between the darkest and brightest areas of an image.

Key Points:

• Low contrast → image looks flat or washed out
• High contrast → sharper details, clearer distinction between objects
• Optimal contrast is essential for image analysis and visual perception

Applications:

• Medical imaging (detecting tissue differences)
• Satellite and remote sensing images
• Photography and computer vision tasks

​

Contrast (Visualization)

Low Contrast Image:

• Small difference between light and dark areas
• Appears dull, flat, and lacks detail

High Contrast Image:

• Large difference between light and dark regions
• Objects and edges are more distinct and clear

Enhancement Method:

• Contrast Stretching / Histogram Equalization
• Widely used in preprocessing for computer vision and medical imaging

​

Pixel Depth (Definition)

Pixel depth (or bit depth) is the number of bits

used to represent each pixel in an image.

Key Points:

• Determines how many intensity levels or colors an image can display
• More bits → more shades/colors → higher image quality
• Directly affects file size and storage requirements

Formula:

• Number of possible values = 2^n (where n = number of bits per pixel)

​

Pixel Depth Examples

1-bit Image:

• 2 possible values → Black & White (Binary)

8-bit Image:

• 256 possible intensity levels (0–255) → Grayscale

24-bit Image:

• 16.7 million possible colors (2^24) → True Color (RGB)

Higher Bit Depths (30-bit, 48-bit, etc.):

• Used in professional imaging, HDR, medical and scientific applications
• Provide more color accuracy and smoother gradients

​

File Size and Pixel Depth

Relationship:

• Higher pixel depth → more bits per pixel → larger file size

Example (512 × 512 image):

• 8-bit (Grayscale): ~262 KB
• 24-bit (Color): ~786 KB
• 32-bit (Color + Alpha channel): ~1 MB

Key Points:

• More depth improves image quality
• But requires more storage and processing power
• Important trade-off in image compression and transmission

​

Image Representation in Matrix Form

Binary Images:

• Stored as a matrix of 0 (black) and 1 (white)

Grayscale Images:

• Stored as a matrix of intensity values (0–255 for 8-bit)

Color Images (RGB):

• Represented by three separate matrices:
• R (Red channel)
• G (Green channel)
• B (Blue channel)
• Each pixel = (R, G, B) triplet

Importance:

• Matrix form makes images suitable for mathematical operations,

filtering, and transformations

​

Sampling

Sampling is the process of selecting discrete

pixel locations to represent a continuous image.

Key Points:

• Determines the spatial resolution of an image
• Higher sampling rate → more pixels → better detail
• Lower sampling rate → fewer pixels → loss of detail, aliasing

Example:

• High sampling → clear, sharp photo
• Low sampling → blocky, pixelated image

Applications:

• Image scanning, digitization, remote sensing

​

Quantization

Quantization is the process of mapping continuous

intensity values into a finite set of discrete levels.

Key Points:

• Determines the number of intensity levels an image can display
• Fewer levels → reduced image quality (banding effects)
• More levels → smoother transitions and higher detail

Example:

• 2 levels → Binary image (Black & White)
• 256 levels → Standard Grayscale image
• Millions of levels → Color images (24-bit and above)

Applications:

• Image compression
• Digital signal and image processing

​

Compression

Compression reduces the amount of data required

to represent an image while maintaining acceptable quality.

Types of Compression:

1. Lossless Compression
• No loss of information
• Original image can be perfectly reconstructed
• Examples: PNG, GIF, TIFF
2. Lossy Compression
• Some information is lost, but file size is much smaller
• Suitable for applications where perfect accuracy is not required
• Examples: JPEG, WebP

Benefits:

• Saves storage space
• Faster transmission over networks
• Essential for multimedia and web applications

​

Lossless Compression�

�A compression method where the original image can be fully reconstructed without any loss of data.

Key Techniques:

• Run-Length Encoding (RLE):
• Stores consecutive repeated values as a single value with a count.
• Useful for simple images with large uniform areas (e.g., icons, text images).
• Huffman Coding:
• Assigns shorter codes to frequent pixel values and longer codes to less frequent ones.
• LZW (Lempel-Ziv-Welch):
• Builds a dictionary of repeated patterns in the image data.

Applications:

• PNG (Portable Network Graphics)
• GIF (Graphics Interchange Format)
• TIFF (Tagged Image File Format, lossless option)

​

Lossy Compression

• A compression method where some data is lost permanently

to achieve higher compression ratios.

The reconstructed image is an approximation of the original.

Key Techniques:

1. Transform Coding (e.g., DCT – Discrete Cosine Transform):
• Converts spatial data into frequency components.
• Removes less noticeable frequencies to reduce file size.
2. Quantization:
• Approximates pixel values by grouping them into ranges.
• Causes irreversible data loss but reduces storage needs.
3. JPEG Compression:
• Widely used; combines DCT and quantization.

Advantages:

• High compression ratio → smaller file sizes.

Disadvantages:

• Quality loss (blurring, blockiness, artifacts).
• Not suitable for medical or scientific imaging where accuracy is critical.

Applications:

• JPEG, WebP, HEIF (used in smartphones), MP4 video frames.

​

Lossless vs Lossy Compression

Summary:

• Lossless → Preserves accuracy, larger size.
• Lossy → Saves space, but sacrifices some detail.

Image Quality Factors

Image quality refers to the perceived visual fidelity

and how well an image represents the original scene or data.

Key Factors Affecting Image Quality:

1. Resolution – Higher resolution gives more detail.
2. Brightness & Contrast – Proper balance improves visibility.
3. Noise – Random variations that degrade clarity.
4. Compression – Lossy compression may reduce quality.
5. Sharpness – Determines edge clarity and fine detail.
6. Color Accuracy – Correct color reproduction enhances realism.
7. Artifacts – Unwanted effects (blurring, blockiness) from processing.

Measurement Methods:

• Objective: PSNR (Peak Signal-to-Noise Ratio), SSIM (Structural Similarity Index).
• Subjective: Human visual perception and comparison.

​

Comparison of Image Types

Summary:

• Binary → Simplest, minimal storage.
• Grayscale → Good balance of detail & storage.
• Color → Most realistic but requires more memory.

Future Directions

High Dynamic Range (HDR) Imaging

• Captures a wider range of light intensities
• Produces more realistic and detailed images
• Widely used in modern cameras, smartphones, and displays

Multispectral and Hyperspectral Imaging

• Capture information beyond the visible spectrum (infrared, ultraviolet, etc.)
• Applications in agriculture, remote sensing, environmental monitoring, defense, and medicine

3D Imaging & Holography

• Provides depth information in addition to 2D image data
• Used in robotics, AR/VR, and medical visualization

AI-Powered Image Processing

• Deep learning for image recognition, enhancement, and generation
• Automatic image restoration and super-resolution

​

Conclusion

Summary of Image Types

• Binary: Simplest, used for shape/edge detection.
• Grayscale: Provides detail with less storage.
• Color: Realistic representation, higher storage requirement.

Key Image Properties

• Resolution, brightness, contrast, and pixel depth determine image quality.
• Compression (lossless & lossy) balances quality and file size.

Importance in Image Processing

• Foundation for computer vision, medical imaging, multimedia, and AI.
• Critical for applications in robotics, surveillance, and digital media.

Future Outlook

• Advances in HDR, multispectral/hyperspectral, and 3D imaging.
• AI will continue to revolutionize image analysis and enhancement.

📌 Takeaway:�Digital images are the core building blocks of modern visual technologies, bridging human vision and machine understanding.

​