Madhav Institute of Technology & Science, Gwalior

(A Govt. Aided UGC Autonomous Institute Affiliated to RGPV, Bhopal)

NAAC Accredited with A++ Grade.

Lab Report
of
IMAGE PROCESSING (280602)

SUBMITTED BY
Nakshatra Trivedi
0901AM211033
3rd Year (Sixth Semester)
Artificial Intelligence and Machine Learning

SUBMITTED TO
Dr. Sunil Kumar Shukla Ms. Deepti Gupta

Assistant Professor Assistant Professor

Centre For Artificial Intelligence

Session 2023-24
 3

Experiment – 1

Aim:
The aim of this study is to comprehensively understand the various aspects of digital image processing,
including image acquisition, enhancement, color processing, and resizing, and to explore the underlying
principles and techniques involved in each of these areas.

Theory:
Image processing is a fundamental area of study in computer vision and digital signal processing,
encompassing a wide range of techniques aimed at analyzing, manipulating, and interpreting digital images.
This study delves into four key areas of image processing:
1. Image Acquisition: This involves the process of capturing digital images from different sources such
as cameras, scanners, and medical imaging devices. Understanding image acquisition techniques helps
in obtaining high-quality images suitable for further processing and analysis.
2. Image Enhancement: Image enhancement techniques aim to improve the visual quality of digital
images by emphasizing certain features, reducing noise, and enhancing contrast and sharpness. These
techniques are crucial for improving the interpretability and analysis of digital images across various
domains.
3. Color Image Processing: Color plays a significant role in digital images, and color image processing
techniques enable the manipulation and analysis of color information. This includes color space
conversion, color correction, segmentation, and filtering, which are essential for applications ranging
from photography to medical imaging.
4. Image Resizing: Image resizing is the process of changing the dimensions of a digital image while
maintaining its aspect ratio and visual content. Various interpolation techniques such as nearest-
neighbor, bilinear, and bicubic interpolation are used for resizing images, with applications in
multimedia, web development, and printing.
By studying these aspects of image processing, one can gain insights into the theoretical foundations, practical
applications, and computational techniques involved in processing digital images effectively. This
understanding forms the basis for developing advanced image processing algorithms and applications across
diverse domains.
This consolidated aim and theory provide a holistic perspective on the study of image processing, covering its
various facets and applications.
 4

Program:
5
 6

Experiment – 2

Aim:
The aim of this program is to implement image resampling, which involves resizing digital images through
both upsampling and downsampling techniques. The program will demonstrate the principles and application
of resampling methods to alter the dimensions of images while preserving their visual content and aspect ratio.

Theory:
Image Sampling:
Image sampling involves converting a continuous image into a discrete representation by selecting a finite
number of samples or pixels from the continuous image. It's crucial for digital image processing, where images
are represented as a grid of pixels, determining resolution and detailing in digital images.

Upsampling:
Upsampling, or interpolation, increases image size by adding new pixels between existing ones, aiming to
enhance resolution and maintain visual content. Common methods include nearest neighbor, bilinear, and
bicubic interpolation.

Downsampling:
Downsampling, or decimation, reduces image size by removing pixels, useful for decreasing image resolution
while retaining essential features. Techniques include average pooling, max pooling, and Gaussian filtering
with subsampling.
Understanding these concepts is fundamental in digital image processing, enabling the manipulation and
analysis of digital images across various applications.
 7

Program:
8
 9

EXPERIMENT NO. : 3

Aim :- The aim of this experiment is to detection different color from image.

Description:

Color detection in computer vision involves identifying and extracting specific colors or color ranges from an
image. Unlike traditional RGB representation, color detection often utilizes alternative color spaces like HSV
or HSL for better color representation. In these spaces, colors are described in terms of hue, saturation, and
value/lightness.

The process of color detection typically begins by converting the image to the desired color space.
Thresholding techniques are then applied to create a binary mask that isolates pixels falling within the target
color range. This involves setting criteria based on hue, saturation, and value/lightness values to determine
whether a pixel belongs to the desired color range.

After thresholding, bitwise operations are commonly used to combine multiple binary masks and filter out
regions that match specific color criteria. This helps refine the color detection results and remove any noise or
unwanted background.

Color detection finds application in various fields such as object tracking, image segmentation, gesture
recognition, and robotics. For example, in object tracking, color detection can be used to track a specific
colored object as it moves within a video frame. However, challenges such as variations in lighting conditions,
shadows, reflections, and noise in the image can affect the accuracy of color detection. Selecting appropriate
color thresholds and color spaces is crucial to overcome these challenges and achieve accurate results. Overall,
color detection is a fundamental concept in computer vision, providing valuable information for understanding
and analyzing images and videos.
 10
Program:
11