Personal tools

Digital Image Processing

University of Texas at Austin_091921A
[ The University of Texas at Austin]

- Overview 

For more details on Digital Image Processing, see Wikipedia:Digital Image Processing.

- Digital Image Processing

Digital Image Processing (DIP) is a type of software for image manipulation. For example: computer graphics, signals, photography, camera mechanisms, pixels, etc. DIP provides a platform to perform various operations such as image enhancement, analog and digital signal processing, image signals, speech signals, etc. DIP provides images in different formats. 

DIP is the use of digital computers to process digital images through algorithms. As a subclass or field of digital signal processing, DIP has many advantages over analog image processing. It allows a wider range of algorithms to be applied to the input data and avoids problems such as the accumulation of noise and distortion during processing. 

Since images are defined in two dimensions (perhaps more), DIP can be modeled in the form of multidimensional systems. 

The emergence and development of DIP are mainly affected by three factors: one is the development of computers; the other is the development of mathematics (especially the creation and improvement of discrete mathematics theory); Increased demand for a wide range of applications in fields such as medicine.


- Digital Imaging

Digital imaging (or digital image acquisition) is the creation of a digitally encoded representation of the visual characteristics of an object, such as a physical scene or the interior structure of an object. The term is often assumed to imply or include the processing, compression, storage, printing, and display of such images. A key advantage of a digital image, versus an analog image such as a film photograph, is the ability make copies and copies of copies digitally indefinitely without any loss of image quality.

Digital imaging can be classified by the type of electromagnetic radiation or other waves whose variable attenuation, as they pass through or reflect off objects, conveys the information that constitutes the image. In all classes of digital imaging, the information is converted by image sensors into digital signals that are processed by a computer and made output as a visible-light image. For example, the medium of visible light allows digital photography (including digital videography) with various kinds of digital cameras (including digital video cameras). X-rays allow digital X-ray imaging (digital radiography, fluoroscopy, and CT), and gamma rays allow digital gamma ray imaging (digital scintigraphy, SPECT, and PET). Sound allows ultrasonography (such as medical ultrasonography) and sonar, and radio waves allow radar. Digital imaging lends itself well to image analysis by software, as well as to image editing (including image manipulation).   

[Bergen, Norway - eirikbjo]

- The Tasks of Digital Image Processing

Digital image processing allows the use of more complex algorithms and thus can provide more complex performance in simple tasks and achieve methods that cannot be achieved by analog methods. 

In particular, digital image processing is based on specific applications and practical techniques for:

  • Classification
  • Feature extraction
  • Multi-scale signal analysis
  • Pattern recognition
  • Projection

Some techniques which are used in digital image processing include:

  • Anisotropic diffusion
  • Hidden Markov models
  • Image editing
  • Image restoration
  • Independent component analysis
  • Linear filtering
  • Neural networks
  • Partial differential equations
  • Pixelation
  • Point feature matching
  • Principal components analysis
  • Self-organizing maps
  • Wavelets

- Steps of Digital Image Processing

Digital image processing is the use of computer algorithms to perform image processing on digital images. The fundamental steps in any typical Digital Image Processing pipeline are as follows:
  • Image Acquisition: Typically, image acquisition involves capturing an image through a sensor such as a camera. If there is a non-digital output, convert it to digital using an analog-to-digital converter. This process also includes preprocessing such as image scaling.
  • Image Enhancement: The process related to image processing to obtain relevant results for a specified task to be performed is called image enhancement. Ideally, this process is associated with image filtering, improving the quality of the originally captured image by performing tasks such as noise removal, contrast adjustment, brightness, and sharpening of the image. 
  • Image Restoration: Image restoration involves improving the appearance of images that may have been degraded by mathematical and probabilistic models. An ideal example is reducing blur in images. 
  • Color Image Processing: Extract features from images using color-based methods. 
  • Wavelet and multi-resolution processing: It involves representing images in terms of the various available resolutions commonly used for image compression. This is also useful for image data compression. 
  • Compression: Reducing the storage space required to save an image or the bandwidth required to display it is done with the help of compression. Techniques that involve image size reduction and resizing to minimize quality degradation fall under the image compression process. 
  • Morphological Processing: Extraction of fundamental components in an image describes the shape of a specific object in the image. Some typical morphological operations are erosion and dilation for generating image properties. 
  • Split: Image segmentation is one of the necessary steps in image processing, which involves dividing an image into parts. This process allows locating objects in an image and identifying the boundaries of objects. An important thing to note is that the accuracy of the segmentation will lead to better recognition and classification accuracy. 
  • Representation and Description: This representation is associated with displaying the image output in the form of boundaries or regions. It can involve features or region representations of corner shapes, such as textures or bone shapes. On the other hand, description is most often called feature selection and is responsible for extracting meaningful information from images. The extracted information helps to accurately distinguish objects of different categories. 
  • Object recognition/image annotation: The process of assigning labels to objects based on descriptions for classification purposes. This is a very important step in computer vision. In order to train a model, a large enough corpus of images needs to be processed and labeled so that the computer vision model can be used to detect similar objects in other images.



[More to come ...] 

Document Actions