Image Operations: Pyramid of Images, Graphics Image Type, Image Treatment and Image Histogram, & Image Enhancement (Especially for GATE-Geospatial 2022)

Doorsteptutor material for competitive exams is prepared by world's top subject experts: get questions, notes, tests, video lectures and more- for all subjects of your exam.

Examrace Books on Mapping, GIS, and Remote Sensing prepares you throughly for a wide range of practical applications.

Pyramid of Images

A pyramid of images is a tool used by most digital instruments because it saves time in many phases of the calculation process. The pyramid of images is not a compression technique, but a reduction method for the calculation processes and, therefore, for the information volume with which the instrument works. The pyramid is based on multiresolution. The base of the pyramid would be an image with its original resolution and, subsequently, images with lower resolutions are stored in the memory consecutively. In a pyramid-shaped image, the search processes, as shown in the figure, are carried out from lower resolutions to higher resolutions, without the need to explore the whole image.

Progressive Search in a Pyramid of Images

Progressive search in a pyramid of images

Graphics Formats of Digital Photogrammetric Images

The format is defined as the standard method used to organise and store the data of the image. Not all the standard formats of the images are used in photogrammetry, and, conversely, some photogrammetric formats are not used in other fields. The most widely used formats in photogrammetry are the following:

  • TIFF: compressed or non-compressed, it is the most used format in photogrammetry.
  • ECW: with a small compression ratio, it is used especially in the generation of orthophotos.
  • SID: compressed format like ECW.

And all the characteristic formats of the digital photogrammetric system, such as RSW (Photomod) , PIX (PCI geomatic) , IMG (Erdas imagine) etc.

Histogram of an Image

Histograms of Grey-Scale Image

The histogram of a grey-scale digital image is a discrete function that serves as a guide for the value placed on the probability that a certain grey level must appear. It plots the number of pixels for each tonal value. This function will provide, for all grey values, a global description of the image՚s appearance.

Image Og Histograms

Each image has its own histogram. Yet, as a rule, an image is considered to have good contrast if its histogram occupies almost all the tone range. Thus, the corresponding image to the histogram (c) of figure would be the image with a better distribution of values.

Histograms of Three Channels

Histograms of Three Channels

Histograms of three channels

An RGB image would be represented by its corresponding histogram in each of its channels.

Digital Treatment of Images

Before the photogrammetric process starts, images can undergo a pre-process with the aim of improving their visual quality. Some concepts need to be defined beforehand in order to comprehend what the digital treatment of the images consists of. The most widely used techniques for digital treatment to process images would change its histogram.

Management of a Histogram with a Colour Curve

Management of a histogram with a colour curve

Thus, each manipulation of the curve will influence the image:

Possible Changes in the Histogram of an Image

Possible changes in the histogram of an image

Image Enhancement by Histogram Equalisation

Histogram equalisation is an operation devoted to the uniform distribution of grey levels between the pixels of the image. This process gives a higher digital level of the output image to the most frequent digital levels in the input image. Consequently, in the enhanced image, grey levels that occupy the majority of grids in the original image are more contrasting. In general, a better-distributed histogram is obtained, with better separation between the most frequent DN of the image.

Some of the tools used to enhance the images have the capacity to work with colour curves or grey levels. Curves as the one shown in the figure below make the transformation of histograms with high precision possible, due to the fact that several checkpoints can be set from anywhere in the histogram.

Developed by: