Image enhancement using multi scale image features extracted by top-hat transform

Bai, Xiangzhi; Zhou, Fugen; Xue, Bin

doi:10.1016/j.optlastec.2011.07.009

Cited by 157 publications

(70 citation statements)

References 35 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The algorithm proposed in this paper is a variation of the technique described by Bai, Zhou and Xue [2] for grayscale images. They propose a method of contrast enhancement that extracts image features by multi-scale top-hat transform which we call BZX algorithm.…”

Section: Proposalmentioning

confidence: 99%

See 1 more Smart Citation

Image color contrast enhancement using multiscale morphology

Román

Noguera

2017

JCIS

View full text Add to dashboard Cite

Contrast enhancement aims to improve color images in terms of human interpretation and further processing in other stages of digital image processing. The multiscale mathematical morphology technique is widely used to enhance contrast in grayscale images. The extention for mathematical morphology for color images is not an easy task, because the images can be defined by different color spaces, which are represented by n-dimensional vectors and there is no single way of ordering vectors. HSI color space is used to represent the color of pixels and a lexicographical order (I→S→H) is imposed for processing. The fundamental morphological operations will be extended from this ordering to be applied into color images. In this work, a contrast enhancement algorithm is presented using the features extracted from the color image by the transformed multiscale top-hat. To evaluate the proposed method, various tests were performed using different color images. The experimental results show the effectiveness of the method, in which contrast enhancement was evaluated by the color enhancement factor (CEF) Metric. Also the visual inspection was used.

show abstract

Section: Proposalmentioning

confidence: 99%

“…In digital image processing there are several techniques for contrast enhancement, such as histogram equalization, which improves the contrast of an image by a redistribution of the gray levels [4]; and the multiscale morphology that has shown efficiency in contrast enhancement for grayscale images [2].…”

Section: Introductionmentioning

confidence: 99%

Image color contrast enhancement using multiscale morphology

Román

Noguera

2017

JCIS

View full text Add to dashboard Cite

show abstract

“…1b. [7], Bai, Z h o u and Xue [8,9] have designed the multi-scale top-hat transform for the image enhancement with the strategy of conducting enhancement on the detailed feature above different scale levels, but there exists great redundancy when realizing the enhancement operation and they are all lacking in flexibility with the unified criterion on the detailed feature, and the cited processing expressions are (8) = + 0.5 o − 0.5 c . The calculating parameters are defined that:…”

Section: Multi-scale Morphological Filtering Theorymentioning

confidence: 99%

“…And the above theorybased process consists of ROI extraction and extracting the difference value at the adjacent scale which is able to adequately avoid the unnecessary enhancement to the detailed information. The feature extraction at different scale "i" can be fulfilled by the next equation: (9) { WTH = − ∘ , 0 ≤ ≤ , BTH = ⦁ − , 0 ≤ ≤ , WTH , BTH represent the light and dark detailed feature respectively, which are smaller than the preset structural elements.…”

Section: Multi-scale Morphological Filtering Theorymentioning

confidence: 99%

“…According to the visual effect and feature parameters evaluation result we can conclude that the global gray value of the enhanced image is relatively high, so it is necessary to adjust the global gray value to improve the enhancement performance, where the paper introduces the nonlinear function for the re-enhancement process based on analysis results in paper [9,10], the transformation is expressed as (18) e2 = ( e ( , )) = { 2( e ( , )) 2 , 0 ≤ e ≤ , 1 − 2(1 − e ( , )) 2 , < e ≤ 1. In that expression, , ∈ [0, 1], is set so as the transform the turning point with the discipline so that when e ≤ , the value is going nearer to 0 and when < e , the value is going nearer to 1.…”

Section: Re-enhancement Based On the Constructed Non-linear Functionmentioning

confidence: 99%

See 1 more Smart Citation

Hand Vein Image Enhancement Based on Multi-Scale Top-Hat Transform

Wang

et al. 2016

Cybernetics and Information Technologies

View full text Add to dashboard Cite

Enhancing microscopy images of minerals through morphological center operator‐based feature extraction

Bai

2012

Microscopy Res & Technique

View full text Add to dashboard Cite

To well enhance the mineral image and image details obtained by microscopes, an effective mineral image enhancement algorithm through feature extraction using the morphological center operator is proposed in this work. First, mineral image feature extraction based on the morphological center operator is proposed and discussed. Second, the multiscale extension of the mineral image feature extraction is given by using the multiscale structuring elements. Third, the important mineral image features at multiscales of image are extracted and used to construct the final mineral features for mineral image enhancement. Finally, the original mineral image is well enhanced through importing the extracted final mineral image features into the original mineral image. Experimental results on different types of microscopy images of minerals verified the effective performance of the proposed algorithm for microscopy mineral image enhancement.

show abstract

Image enhancement using multi scale image features extracted by top-hat transform

Cited by 157 publications

References 35 publications

Image color contrast enhancement using multiscale morphology

Image color contrast enhancement using multiscale morphology

Hand Vein Image Enhancement Based on Multi-Scale Top-Hat Transform

Enhancing microscopy images of minerals through morphological center operator‐based feature extraction

Contact Info

Product

Resources

About