Different Techniques Used for Image Processing

Introduction

Image processing is a technique used to process images and is used for improving the images received from various sources such as cameras, satellites, sensors, etc. There are various techniques developed and also developing from the past years to the current years in order to extract image in an efficient and best manner. Thus, image processing can be termed as a technique used for analyzing and manipulating the digitized images to improve the quality of images. The techniques proposed can be applied in various applications such as diagnosis of an image analysis, object detection and matching, localization of tumors, surgical planning, background removal in videos, traffic control system, measuring the volume of tissues, location of objects such as roads, forests, etc. in satellite images, iris recognition, medical and agriculture imaging, object localization in face recognition, etc. (Rajkomar et al., 2019) The various techniques of digital image processing primarily focus on issues such as loss in the quality of images or enhancing the degraded image.

Image processing can be classified into two parts as follows:

  • 1. Analog image processing
  • 2. Digital image processing

Analog image processing is a process of processing of images by electrical means. Example of such processing is television images. As the television signal is a voltage level, hence they vary in terms of amplitude in order to represent brightness by means of image. Displayed image appearance can be altered by varying these electrical signals. The brightness and contrast controls on a TV set serve to adjust the amplitude and reference of the video signal, resulting in the brightening, darkening, and alteration of the brightness range of the displayed image.

Digital image processing on the other hand uses digital computers for processing of images. This technique uses a digital conversion method for converting the image into digital form in order to process the image in digital format. Now most of the image processing methods are based on digital image processing only and in this chapter, we focus on different techniques used in digital image processing only.

Real environment

(Discriminant analysis Neural Network etc.)

FIGURE 2.1 Operational steps of image processing.

We can broadly divide the entire digital image processing into four components as represented in Figure 2.1:

  • 1. Acquisition of images
  • 2. Image pre-processing
  • 3. Processing of images
  • 4. Decision making

Diagnostic image analysis, surgical planning, object detection and matching, background subtraction in video, localization of tumors, measuring tissue volumes, locating objects in satellite images (roads, forests, etc.), traffic control systems, locating objects in face recognition, iris recognition, agricultural imaging, and medical imaging all these application fields consist of different approaches for image processing such as feature extraction, classification, pattern matching, decision making, and validation that finally lead to the desired output.

Acquisition of an Image

The procedure of image preparation cannot start except if the image has been caught. It is perpetually the initial step of any strategy used for image processing. Image acquisition is an equipment autonomous procedure. In this step, fundamentally an optical image is caught by the camera and is changed over into a variety of numerical information as bits.

An image is obtained utilizing cameras or various types of sensors that can catch different features of the energy that was reflected by the object surface. The acquisition process of images can be categorized into three stages, where the camera focal points center that object, and when the item mirrors the whole energy, it is caught by the sensors of the camera. Various sorts of cameras are accessible to diverse applications. The images are produced by the blend of a brightening of the wellspring of light and the reflection or the assimilation of the light by the object of intrigue. The enlightenment can be produced by the diverse energy source and to detect the image, a sensor relying upon the idea of brightening will be chosen. This total procedure of image catch is known as image securing. By and large, the sensor named the charged coupled gadget or complementary metal-oxide semiconductor (CMOS) image sensors start at the same point of capturing the image and able to convert light into electrons is utilized in the cameras. The cameras utilizing the CMOS, for the most part, would get images with commotion and use a greater amount of capacity to catch an image. Anyway, the cameras with a charge coupled device (CCD) are more equipped for creating all the more great images, with less force utilization. CCD cameras are single IC gadgets that comprise a variety of photosensitive cells, where every cell delivers an electric flow when a light beam falls on the cells. CCD cameras have less mathematical twisting and give a linear visual output. Image obtaining can be done utilizing a solitary sensor; anyway, it is a moderate technique to secure images. Images can likewise be obtained orchestrating singular sensors as an array of two dimensional (2-D) form. (Zhu et al., 2019). There are five principal boundaries to be borne as a main priority while catching an image, and these boundaries are perspective on the catching gadget, separation of an article to be caught from the gaining gadget, goal, size of the sensors in the gadget, and profundity of field. When the image has been caught, it must be digitized by a gadget known as an edge store that stores the examples of the casing in its own memory, and the edge is handily moved to a record or a memory area that can be gotten to later when required.

Image Pre-Processing

Major pre-processing techniques are (Figure 2.2):

Image Enhancement

Image enhancement is enhancing the part of image which is of use in order to have a better understanding of the image so that various operations can be performed on the selected image. Enhancement programs make information more visible by redistributing the intensities of the image in the 256 gray scale levels or by carrying out the unsharp masking by emphasizing intensity changes.

Techniques used under image pre-processing

FIGURE 2.2 Techniques used under image pre-processing.

Image Analysis

This section deals with analysis of images in order to extract the required information from the image by means of various methods.

Image Compression

Image compression is used to remove the redundancy within an image and store or transmit data in its best suited format. This is done by encoding the original image with bits. Thus, image compression is used to compress the image as much as possible such that the original feature of the image will retained.

Edge Detection

The edge detection method is used to find the edges where the brightness of an image changes sharply.

Segmentation

Segmentation is segmenting the image when needed to extract the required component from the image.

 
Source
< Prev   CONTENTS   Source   Next >