What is image processing?

Image processing is a way to manipulate different operations on any image data using software or hardware to get specific output. Image processing has a vast coverage area, including blurring the background, restoring the image, increasing the image's brightness, detecting subjects, sharpening the images, and much more.

Why do we need image processing?

The demand for image processing is increasing daily in this era of technology. We need image processing for:

• Storage

• Transmission

• Representation for autonomous machine perception

• To improve image quality for human and robotic interpretation and many others.

How it works

Every software uses different algorithms to perform the desired operation on an image. The given image is considered as input on which algorithm of DIP works. After completing image processing, we get an enhanced image as the output. See the visual to understand it elaborately.

In the figure above, the image has been sent as an input to the image processing system to focus on the woman's face and remove all the other details so that the quality of the image remains the same. Moreover, the image shape has been changed from a rectangle to a circle to make it entirely focused.

Types of images

A combination of a finite number of pixels $f(x,y$) is an image where$x$ and $y$ are spatial coordinates and $f$ is the gray level of the image at $x$. In computer language, an image is a two-dimensional array with $x$ and $y$ coordinates arranged as rows and columns. There are two types of images:

• Digital images: The discrete quantities of $x,y,$ and $f$ in images.

• Analog images: The continuous range of values of $x,y,$ and $f$ in images.

Phases of image processing

• Image acquisition: This involves preprocessing and fetching the image from the source.

• Image enhancement: This includes changing certain features of images, such as increasing the brightness.

• Image restoration: In this phase, probabilistic models are used to increase the quality of the image.

• Color image processing: This includes pseudo colors and a full-colored model for image processing.

• Wavelets and multi-resolution processing: The image is divided into waveletsWavelets are used to represent images in various degrees of resolution. for data compression and to represent it in various degrees of resolution.

• Compression: Image size and resolution are changed to transmit or save it.

• Morphological processing: This changes the shapes of images by performing morphing operations on the image for better representation.

• Segmentation: This involves dividing the image into its components.

• Representation and description: Representation includes characters and regions of the image, while the description consists of quantitative information about objects.

• Recognition: This adds labels to the object of the image depending upon its description.