Digital Image Processing

What is Image Processing?

Image Processing can be defined as performing operations on an image in order to extract various useful information in the image or obtain an enhanced image. Image processing is a type of signal processing that take an image as input and output an image or extracted characteristics/features associated with that image. There are three basic steps involved in image processing as follows;

1. Import the relevant image using image acquisition tool
2. Analyze and manipulate the imported image
3. output the resultant altered image or report that is based on image analysis

Types of Image Processing

Analogue Image Processing and Digital Image Processing are two main image processing types. The analog image processing is applied on analog signals and it processes only two-dimensional signals. The images are manipulated by electrical signals. In analog image processing, analog signals can be periodic or non-periodic. Hard copies such as printouts, photographs use Analogue Image Processing techniques. Digital Image Processing is applied to digital images (a matrix of small pixels and elements). For manipulating the images, there is a number of software and algorithms that are applied to perform changes.

Importance of Digital Image Processing

Image Processing can be used for following purposes;

1. Enhance the images through sharpening and restoration — Image enhancement is the process of adjusting digital images so that the results are more suitable for display or further image analysis. Using image processing techniques like filtering, masking, noise removal, contrast stretching and histogram equalization, it can remove noise, sharpen, or brighten an image, making it easier to identify the key features in the image.
2. Distinguishing different objects in the image

3. Visualization of the hidden objects in the image

4. Seek valuable information in the image

5. Measuring different patterns of objects in the image

Applications of Digital Image Processing

1. Image Processing can be used to improve the image quality through techniques such as sharpening and restoration. In some cases to obtain the desired resultant image, original images need to be altered.
2. In the field of medicine for gamma-ray imaging, PET scan, X-ray imaging, UV imaging Digital Image Processing techniques are used.
3. Image processing applications are used in Machine Learning for pattern recognition. Pattern recognition is used in computer aided diagnosis , recognition of handwriting , recognition of images.
4. In the field of Remote sensing, Image Processing applications are used to detect infrastructure damages caused by an earthquake.
5. Automatic character recognition (zip code, license plate recognition) .
6. Finger print/face/iris recognition.
7. Industrial applications (e.g., product inspection/sorting).

Steps of Digital Image Processing

Image Acquisition —This is the first and most important step of Digital Image Processing. In this step processing algorithms takes an image already in digital format and do basic preprocessing like scaling. Image Acquisition starts with the capturing of an image by the sensor and digitized. In case, the output of the camera or sensor is not in digital form then an analog-to-digital converter (ADC) digitizes it. Customized hardware is used for advanced image acquisition techniques and methods. i.e. 3D image acquisition

Image Filtering and Enhancement — Image enhancement is one of the easiest and the most important areas of digital image processing. The core idea behind image enhancement is to find out information that is obscured or to highlight specific features according to the requirements of an image. Such as changing brightness & contrast etc. Basically, it involves manipulation of an image to get the desired image than original for specific applications. Many algorithms have been designed for the purpose of image enhancement in image processing to change an image’s contrast, brightness, and various other such things. Image Enhancement aims to change the human perception of the images. Image Enhancement techniques are of two types: Spatial domain and Frequency domain.

Image Restoration —Usually Image Restoration is performed to improve the appearance of the images. Image restoration techniques are typically based on probabilistic or mathematical models of image degradation. Image restoration removes any form of a blur, noise from images to produce a clean and original image. The image information lost during blurring is restored through a reversal process. This process is different from the image enhancement method.

Color Image Processing — This includes color modeling and processing in a digital domain etc. There are various color models such as RGB, YMC, YIQ, HSI, which are used to specify a color using a 3D coordinate system. The color image processing is done as humans can perceive thousands of colors. There are two areas of color image processing full-color processing and pseudo color processing. In full-color processing, the image is processed in full colors while in pseudo color processing the grayscale images are converted to colored images. It is an interesting topic in image processing.

Wavelets and Multi Resolution Processing — Wavelets act as a base for representing images in varying degrees of resolution. Images subdivision means dividing images into smaller regions for data compression and for pyramidal representation. Wavelet is a mathematical function using which the data is cut into different components each having a different frequency. Each component is the then studied separately through a resolution matching scale. Multi-resolution processing is a pyramid method used in image processing. Use of multiresolution techniques are increasing. Information from images can be extracted using a multi-resolution framework.

Compression — Compression techniques are used for reducing storage necessary to save an image or bandwidth to transmit it. Algorithms acquire useful information from images through statistics to provide superior quality images.

Morphological Processing — Morphological processing involves extracting tools of image components which are further used in the representation and description of shape. There are certain non-linear operations in this processing that relates to the features of the image. These operations can also be applied to grayscale images. The image is probed on a small scale known as the structuring element.

Segmentation — Segmentation involves dividing an image into its constituent parts or objects. Generally, autonomous image segmentation is one of the toughest tasks in digital image processing. In simple terms, image segmentation means partitioning an image into multiple segments for simplification and changing the representation of the image. In this, a label is assigned to every pixel such two or more labels may share the same label.

Representation and Description — The behavior of representation and description depends on the output of a segmentation stage and it includes raw pixel data, constituting either all the points in the reign or only boundary of the reign. Choosing a representation is a part of the solution to transform raw data into a suitable form that allows subsequent computer processing. As description deals with extracting attributes that yield quantitative information of interest or basic to separate one class from another.

Object recognition — Object Recognition involves assigning of a label, such as, “human” to an object completely based on its descriptors. It is a method of recognizing a specific object in an image or video. There are certain techniques and models for object recognition like deep learning models, bag-of-words model etc.

Knowledge Base — Knowledge is all about detailing regions of an image to locate the information of interest that ultimately delimits the research to be conducted in seeking that information. Knowledge Base becomes complex such as an interconnected list of all major possible defects in materials assessment problems or an image database carrying high-resolution satellite images of a region in relation with change-detection applications.