PaperlayerView exact revision on arXiv

Source-linked AI summary

Enhancement of Images using Morphological Transformation

K. Sreedhar, B. Panlal

arXiv:1203.2514v1cs.CV

TL;DR

The paper addresses enhancement of poorly contrasted images and detection of their backgrounds, where global methods can preserve local detail inadequately. It proposes block-based and morphological background approximations with Weber’s-law-based contrast operators, illustrating the approaches through image examples and comparisons.

  • Problem

    Poorly lit images require context-sensitive background detection and contrast enhancement because global techniques can perform poorly in local detail preservation.

  • Method

    The paper uses block analysis, morphological operations including opening by reconstruction, and Weber’s-law-based contrast operators for grayscale and colour image enhancement.

  • Results

    The proposed operators are illustrated on several examples and compared with techniques from the literature; enhanced images reveal characteristics not initially visible.

  • Takeaways & Limitations

    Background approximation through blocks and morphological connected transformations provides the paper’s framework for enhancing contrast in poorly lit images.

Abstract

from arXiv · show

This paper deals with enhancement of images with poor contrast and detection of background. Proposes a frame work which is used to detect the background in images characterized by poor contrast. Image enhancement has been carried out by the two methods based on the Weber's law notion. The first method employs information from image background analysis by blocks, while the second transformation method utilizes the opening operation, closing operation, which is employed to define the multi-background gray scale images. The complete image processing is done using MATLAB simulation model. Finally, this paper is organized as follows as Morphological transformation and Weber's law. Image background approximation to the background by means of block analysis in conjunction with transformations that enhance images with poor lighting. The multibackground notion is introduced by means of the opening by reconstruction shows a comparison among several techniques to improve contrast in images. Finally, conclusions are presented.

MORPHOLOGICAL TRANSFORMATIONS

The supplied passages identify the paper’s authors, affiliations, and topic keywords concerning image-background analysis, morphological methods, Weber’s law, and contrast enhancement.

  • The paper is authored by K. Sreedhar and B. Panlal.
  • The authors are affiliated with electronics and communication engineering departments at VITS and VCE in Karimnagar, Andhra Pradesh.
  • The listed topics include image-background analysis by blocks, morphological methods, Weber’s law, opening and closing operations, erosion-dilation, and gray-level block analysis.

1. INTRODUCTION

The introduction frames poor-contrast enhancement and background detection as problems requiring context-sensitive methods, then presents block analysis and morphological approaches based on Weber’s law.

  • 1. INTRODUCTION: Mathematical morphology is presented alongside nonlinear and frequency-domain methods as an approach to image enhancement.
  • 1. INTRODUCTION: The paper addresses background detection in images with poor contrast using a MATLAB simulation model.
  • 1. INTRODUCTION: Two background approximations are proposed: block analysis and opening by reconstruction, with contrast operators based on logarithmic Weber’s-law processing.
  • 1.1. Existing System: Poor lighting and regional intensity differences motivate context-sensitive contrast enhancement because global techniques often preserve details poorly.

2. MORPHOLOGICAL TRANSFORMATIONS AND WEBER’S LAW

Morphology processes image structure with structuring elements and supports operations on grayscale images that can be extended to full-colour images.

  • 2.1. Morphology: Morphological processing applies a structuring element to an input image and produces an output image of the same size.
  • 2.1. Morphology: Each output pixel is determined by comparing the corresponding input pixel with neighboring pixels.
  • 2.1. Morphology: Changing the structuring element’s size and shape makes the operation sensitive to particular object shapes.

2.2. Morphological Operations

Morphological operations apply structuring elements to images, with dilation and erosion forming the basis for opening and closing transformations.

  • Morphological Operations: Morphological operations apply structuring elements to an input image and produce an output image of the same size.The selected structuring element’s size and shape determine neighborhood sensitivity.
  • Dilation: Dilation determines each output pixel from the maximum value in the neighborhood defined by the structuring element.In binary images, any neighborhood pixel set to 1 sets the output pixel to 1.
  • Erosion: Erosion determines each output pixel from the minimum value in the structuring-element neighborhood.For binary images, any neighborhood pixel set to 0 sets the output pixel to 0.
  • Opening and Closing: Opening is performed by erosion followed by dilation, while closing is performed by dilation followed by erosion.Closing is described as the reverse of opening.
  • Application: The framework applies block analysis and morphological erosion-dilation to enhance images, extending grayscale processing to color components.The color workflow separately enhances the R, G, and B components before concatenating them.

2.3. Weber’s Law

The paper adapts Weber’s logarithmic contrast relation to image gray-level intensity and uses background estimates to enhance poorly lit images.

  • Weber’s Law: The paper motivates local background analysis because abrupt luminance changes can prevent global background estimation from enhancing poorly lit images correctly.
  • Weber’s Law: Weber’s law relates image contrast logarithmically to luminance, with background and scaling parameters controlling enhancement.The paper substitutes image gray-level intensity for luminance in this relation.
  • Weber’s Law: The method adapts Weber’s law by using image intensity f and background parameter b instead of luminance L.
  • Weber’s Law: The enhancement transformation uses logarithmic terms and distinguishes clear and dark regions according to a background criterion.
  • Weber’s Law: Morphological erosion and dilation are identified as the principal operations for block-based background analysis.

3.1. Block Analysis for Gray level images

Block analysis estimates a local background from intensity extrema in each image block, then applies Weber-based enhancement; more blocks improve quality, while large structuring elements cause artifacts.

  • 3.1. Block Analysis for Gray level images: Each image block supplies minimum and maximum intensities used to compute a background criterion and select its background parameter.
  • 3.1. Block Analysis for Gray level images: The background parameter depends on the criterion value, taking the block’s maximum intensity for one specified threshold case.
  • 3.1. Block Analysis for Gray level images: The block-based enhancement adds one to the logarithmic function to avoid an undefined determination condition.
  • 3.1. Block Analysis for Gray level images: More blocks improve enhanced-image quality, and previously unclear objects become visible after enhancement.
  • 3.1. Block Analysis for Gray level images: Structuring-element size 2 (µ=2) gives the best reported results, whereas larger sizes introduce severe blurring and contouring.

4. IMAGE BACKGROUND ANALYSIS USING OPENING BY RECONSTRUCTION FOR GRAY SCALE IMAGES

Opening by reconstruction estimates image background without block partitioning, using erosion-derived markers and iterative dilation, while Weber-based enhancement preserves surviving object shapes.

  • 4. IMAGE BACKGROUND ANALYSIS USING OPENING BY RECONSTRUCTION FOR GRAY SCALE IMAGES: Opening by reconstruction performs whole-image manipulation rather than partitioning the image into blocks.
  • 4. IMAGE BACKGROUND ANALYSIS USING OPENING BY RECONSTRUCTION FOR GRAY SCALE IMAGES: Erosion-dilation provides more local background analysis than blocks by examining neighboring pixels through the structuring element, but large elements can generate undesired values.
  • 4. IMAGE BACKGROUND ANALYSIS USING OPENING BY RECONSTRUCTION FOR GRAY SCALE IMAGES: Opening by reconstruction is used because it restores the shapes of components remaining after erosion while passing through regional minima and merging components.
  • 4. IMAGE BACKGROUND ANALYSIS USING OPENING BY RECONSTRUCTION FOR GRAY SCALE IMAGES: The operation first erodes the input as a marker, then iteratively dilates that marker until stability is reached.
  • 4. IMAGE BACKGROUND ANALYSIS USING OPENING BY RECONSTRUCTION FOR GRAY SCALE IMAGES: The background is calculated with a unity-sized structuring element, after which Weber’s law is applied for contrast enhancement.

5. SIMULATION RESULTS

The simulation section presents visual pipelines for block analysis, morphological background detection, opening by reconstruction, and enhancement under multiple structuring-element sizes.

  • 5. SIMULATION RESULTS: Figures 11 and 12 compare background-detection pipelines using erosion, dilation, closing, opening, and opening by reconstruction.
  • 5. SIMULATION RESULTS: The simulation figures include original, processed, background-detection, and operation-specific images for the proposed approaches.
  • 5. SIMULATION RESULTS: Figure 13 presents block-based background detection and contrast enhancement alongside gray, infrared, erosion, dilation, opening, and closing operations.
  • 5. SIMULATION RESULTS: Figure 14 shows background images and enhanced images produced with structuring-element sizes 20, 50, 80, 110, and 180.

6. CONCLUSION

The paper introduces background estimation and contrast-enhancement methods for poorly lit images, then compares their performance with methods from the literature.

  • The proposals estimate image background through block analysis and morphological connected transformations after erosion and dilation caused difficulties.
  • Morphological contrast-enhancement transformations based on Weber’s law were introduced and illustrated through several examples.
  • The proposed operators were compared with other techniques reported in the literature.
  • The contrast-enhancement transformations work satisfactorily only for images with poor lighting, motivating future work on this limitation.

Authors

The paper lists biographical and professional information for K. Sreedhar and B. Panlal, both associated with engineering education in India.

  • K. Sreedhar holds engineering degrees from JNTUH University and works as an assistant professor in Electronics and Communication Engineering.
  • Both authors are associated with Electronics and Communication Engineering and have participated in technical conferences or professional organizations.
  • B. Panlal holds engineering degrees from JNTUH University and KU University and works at Vaageswari College of Engineering.
Loading 1203.2514v1…