Introduction
The research paper endeavors to explore the various steps involved in the digital image processing procedure. The first procedure of the digital image processing entails the use of a device to capture the image. Thereafter, the process of acquiring the digital image is executed. The steps of image enhancement, image restoration, image coluring, compression, segmentation and recognition, shall all be assessed. Accordingly, the research paper is an attempt to explore the process of digital image processing according to the aforementioned sequence. Question A1.1 Fundamental steps in digital processing There are some basic steps that are involved in the digital image processing procedure and they include:
Image acquisition
According to Gonzalez, image acquisition is “the process where the digital image is captured using a device. The acquisition process is different and it is determined by the kind of applications used” (Gonzalez 45). At this point, a pre processing stage comes into effect. The pre processing stage entails the sampling the quantization steps. This is an important step because it facilitates in the conversion of a continuous image into a digital one. In the second question, the research paper shall provide details of the image acquisition process.
Image enhancement
The image enhancement process entails the modification, changing, and adjusting of the image in question to facilitate in the customisation process to suit a specific process or application. In question number four, the writer has made use of histogram equalization to give an image enhancement example. At the image enhancement stage, the contrast attributed to an output image of choice is not only pleasant, but also suitable, in comparison with the input image.
Image restoration
The aim of this process is to ensure that quality image is produced. The image restoration process may be regarded as objective, compared with the image enhancement process; mainly it acts as a yardstick with which to determine the best definition of a good image. Mathematical speculations are normally used to determine the specifications needed to come up with a desirable image specification. As such, image enhancement may therefore be regarded as a subjective process.
Color Imaging
This is a term used in reference to the act of imparting colour to an image. This often occurs on coloured images. The techniques of colour imaging that are often applied in gray scale images also finds use in colour imaging or many occasions. On the other hand, one could also use specialised colour imaging techniques, on the basis of the application in question. Wavelets Wavelets may be regarded as the use of varying techniques to compress images. Wavelets have the potential to transform or change various images into smaller sizes, thereby ensuring the achievement of a computation efficacy.
Compression
In the past few decades, there has been a dramatic increase in the number of individuals using multimedia distributions and the web. Consequently, this has acted to enhance the application of compression in images. Burger defines image compression as ” the process of compressing images and therefore storing or representing them in less memory than they were initially” (Burger, 17). Through the compression of image, there is a resultant reduction in the use of space by storage devices. Morphological image processing Morphological image process is a technique that entails the extraction of the various components of an image relative to how the images have been described or represented.
Segmentation
The process of segmentation entails the subdivision of a given image into a number of parts. Segmentation enables one to easily define and distinguish the various regions attributed to a given objects in the image. One of the processes that heavily relies on segmentation is background subtraction. Here, it becomes necessary to segment the foreground as well as the background sections.
Recognition
Using this technique, one is able to recognise and identify different kinds of features, faces, and objects related to an image.
Digital image acquisition
In digital image acquisition, the co-ordinate axes and amplitude of a continuous image have to be captured. Sonka and Boyle argue contend that “In order for the extraction of a digital image from the continuous image to be done, the continuous image should be digitized in the x and y co-ordinates and also the amplitude should be digitized” (Sonka and Boyle, 90). Sampling is the name given to the digitising procedure of both the y and x coordinates. On the other hand, quantization process is the name given to a process whereby the amplitude of definite co-ordinate axis is digitised. A discrete value is often allocated to the x and y co-ordinates during the sampling step. Conversely, the image intensity often acquires a discreet value during the quantization process. In the event that the analysis is to be carried out hypothetically, we are more likely to witness a continuous number of the intensity values, as well as the y and x co-ordinates.
On the other hand, the continuous image is usually replaced by a strip of sensor once the actual analysis has been accomplished. Accordingly, the value in question the number of sensors contained in the stripe determines the co-ordinate values. The number of bits contained in a pixel usually determines the level of digitization intensity. For instance, in case whereby there are eight bits, we are likely to have a total of two hundred and fifty six intensity values.
Histogram
Histogram equalization is a procedure that involves the enhancement of an image’s contrast so that the overall look remains quite clear. In addition, Histogram equalization enables an individual to examine closely each object on the image once the processing has been accomplished. Nonetheless there are fundamental steps of the histogram equalization process that need to be followed:
Imhist() Function
The role of this particular function is to demonstrate to an individual an image’s histogram in a colour bar. By looking at the graph, one is also bale to determine the pixel value. Moreover, the Imhist function enables us to calculate the probability density of the images available. We are also in a position to decipher how the gray levels have been distributed in the individual contrast areas, as depicted by figure 2.
Histeq() Function
The image intensity value gets transformed when this function is performed. This is because we are able to represent a lot of details and as such, the sharpness of the image is enhanced once the values in the histogram have been altered, as can be seen from figure 4. From this figure, we can clearly see that the whole of the pixel shall be presented on the basis of its distributed along the colour map graph.