Framework for comprehensive enhancement of brain tumor images with single-window operation

International Journal of Electrical and Computer Engineering (IJECE)
Vol. 10, No. 1, February 2020, pp. 801~808
ISSN: 2088-8708, DOI: 10.11591/ijece.v10i1.pp801-808  801
Journal homepage: http://ijece.iaescore.com/index.php/IJECE
Framework for comprehensive enhancement of brain tumor
images with single-window operation
Deepthi Murthy T. S.1
, Sadashivappa G.2
1
School of Electronics and Communication Engineering (ECE), REVA University, India
2
Department of Telecommunication Engineering, RV College of Engineering, India
Article Info ABSTRACT
Article history:
Received Aug 10, 2018
Revised Sep 13, 2019
Accepted Sep 27, 2019
Usage of grayscale format of radiological images is proportionately more as
compared to that of colored one. This format of medical image suffers from
all the possibility of improper clinical inference which will lead to
error-prone analysis in further usage of such images in disease detection or
classification. Therefore, we present a framework that offers single-window
operation with a set of image enhancing algorithm meant for further
optimizing the visuality of medical images. The framework performs
preliminary pre-processing operation followed by implication of linear
and non-linear filter and multi-level image enhancement processes.
The significant contribution of this study is that it offers a comprehensive
mechanism to implement the various enhancement schemes in highly
discrete way that offers potential flexibility to physical in order to draw
clinical conclusion about the disease being monitored. The proposed system
takes the case study of brain tumor to implement to testify the framework.
Keywords:
Image enhancement
Medical image
Optimization
Preprocessing
Visual quality
Copyright © 2020 Institute of Advanced Engineering and Science.
All rights reserved.
Corresponding Author:
Deepthi Murthy T S,
School of Electronics and Communication Engineering (ECE),
REVA University,
Bangalore, India.
Email: deepthiresearch1983@gmail.com
1. INTRODUCTION
There has been a tremendous revolution in the area of medical image processing that is essentially
meant for accurately diagnosing the disease [1]. The existing techniques of medical imaging include
ultrasound, Magnetic Resonance Imaging (MRI), Positron Emission Tomography (PET), CT (Computed
Tomography) etc [2]. The degree of improvement in such advance imaging so high that even a micro-
structure and tissue can be seen during the visual analysis [3]. The existing MRI offers multiple slices of
the scanned region to have better look into the suspected region of human body [4]. Most importantly,
there are various advance scientific research work e.g. [5, 6], which claims that medical imaging could be
possibly used for automated diagnosis as well as making the automatic inference of the diagnosed disease
using sophisticated classification techniques. Improvement in the classification techniques potentially assists
in detecting the stage of the disease with accurate clinical information [7].
However, such research ideas are only limited to research papers and yet not make out to
the commercial markets. There are multiple reasons behind this. It has been seen that majority of the research
approaches towards disease diagnosis and classification uses datasets and very less on real-time images.
It is because data set images are already pre-processed and free from any form of errors so that it becomes
easy to check the applicability of research-based algorithms on such dataset images. However, in real-time
medical images do suffer from some realistic problems that are less addressed in existing research work
although some research papers have highlighted few facts about it [8]. The first issues is that if
the pre-processing of the medical images are not carried out effectively than it will lead the other algorithm to

 ISSN: 2088-8708
Int J Elec & Comp Eng, Vol. 10, No. 1, February 2020 : 801 - 808
802
extract error-prone information that will lead to wrong direction of research work. Second issues in medical
image processing are related to the grayscale form of the image where it is quite difficult to make out
the actual location of suspected disease. Inspite of colored radiological images being used in special cases,
usage of grayscale version of medical image is frequently utilized in real-time environment. At present, there
are various forms of pre-processing algorithms [9] as well as denoising algorithm [10], but very few of them
have been tested with complicated cases of medical images.
All the existing pre-processing techniques focuses on implementing some common methods e.g.
wavelets, Retinex, contrast enhancement, filters, histogram, etc. All these techniques are good enough for
certain set of images and don’t offer similar scale of performance if the implementation environments as well
as the datasets are altered. This will eventually mean that existing techniques doesn’t necessarily ensure
uniformity of the enhancement performance. It is quite likely that a physician might want to check one
radiological image from multiple angles within a shortest span of time. At present, apart from simple form of
contrast, brightness, sharpness, and opacity, there are not much form of enhancement that a physician can do
in order to make decision of clinical inference. This is a major research gap which calls for investigating
some reliable techniques for solving such issues. Therefore, we present a very simple and novel framework
where medical image can be subjected to multiple form of enhancement using set of algorithms that are not
only computational friendly but also offer faster response time.
This paper presents discussion of one such algorithm where series of image enhancement techniques
have been implemented. Section 1 discusses about the existing literatures where different medical image
enhancing techniques are discussed followed by discussion of research problems explored from the existing
research approaches in Section 1.1 and proposed solution to solve such research problem in 1.2. Section 2
discusses about image enhancing algorithm implementation that is meant for offering comprehensive image
enhancing operation in highly cost effective way followed by discussion of result analysis in Section 3.
Finally, the conclusive remarks are provided in Section 4.
This section discusses about the existing research approaches towards the medical image
enhancement. The most recent work carried out by Yelmanova and Romanyshyn [11] have focused on
involuntary contrast enhancement using histogram. Zhao and Zhou [12] have presented a masking principle
for the purpose embedding the multiplicative steps in it that significantly enhances the medical images.
The adoption of Verilog scheme was seen in the work of Chiuchisan [13] for prototyping the image
enhancing scheme. Usage of sliding decomposition approach for analyzing the detailed feature of medical
images in the work of Liang and Si [14]. Gong et al. [15] have introduced chaotic approach in order to
enhance the performance of immune algorithm for enhancing the brain images. Li and Kang [16] have
implemented wavelet based approach in order to enhance the resolution of medical image using both
conventional wavelets as well as inverse wavelets. Soloperto et al. [17] have introduced a clinical approach
using Halloysite nanotubes for investigating its trends during the enhancement of medical images.
Xue [18] have used contrast enhanced index for improving the medical images using Shearlet-based
approach. Usage of contourlet transform was seen in the work of Zhou et al. [19] along with a Laplace
pyramid for an effective control on multiple coefficients. Chaira [20] used a fuzzy logic for addressing
similar problems. Dhinagar and Celenk [21] have used histogram as well as homomorphic filtering for
enhancing the image generated from ultrasound. The work carried out by Hua et al. [22] have used Young-
Helmholtz based transformation scheme in order to enhance colored radiological images. Kurt et al. [23]
have used adaptive histogram as well as anisotrophic filters in order to perform enhancement of medical
images. Xiurong [24] have implemented a masking algorithm of adaptive nature in order to effectively
control the sharpening of the medical images.
Fan et al. [25] have constructed an enhanced look up table for improving the contrast in multi-scale
factor. Adoption of context and gradient smoothness towards enhancing the medical image was seen in
the work of Jing and Jie [26]. Literatures have also witnessed usage of non-linear techniques for enhancing
medical images as seen the work carried out by Hossain et al. [27]. The authors have also used histogram
matching as well as logarithmic transform scheme to further enhance it. Jiang et al. [28] have used contourlet
transform using invariance approach in order to assists in image representation and mitigate noise level
in medical images. Usage of wavelet transform was observed in the implementation carried out by
Yang et al. [29] for performing medical image enhancement. Therefore, it can be seen that there are different
techniques available in literatures with an emphasis on addressing image enhancement problems, however,
majority of the existing schemes have nearly similar forms of implementation schemes with lesser
availability of journals addressing such problems.
However, such schemes are quite capable of addressing particular problems and would succumb to
a failure if the test environment is changed. Such identified issues are outlined in next section in the form of
research problem. Bangare et al. [30] have illustarted to find out tumor ripped lesions to be make better using
fluorescence image resolution. Kahina et al. [31] have demonstarted Viterabi decoder with the window

Int J Elec & Comp Eng ISSN: 2088-8708 
Framework for comprehensive enhancement of brain tumor images with single.... (Deepthi Murthy T S)
803
technique. Verma and Sharma [32] have presented an optimal image improvement technique for color
images by protecting their strength.
The research problems identified from literature review are as follows:
a. Existing system are too much specific about the enhancement and therefore they cannot caterup
the processing requirements of multiple forms of radiological images.
b. Single algorithm or set of algorithm addressing multiple forms of enhancement issues of medical images
are very few to explore in existing system.
c. Usage of existing schemes doesn’t offer a single-handed function where there is a scope of further
enhancement or rather optimizing the enhancement process.
d. Existing approaches are also mathematically complex in implementation when it comes for computational
process leading to longer response time with external dependencies.
Therefore, the problem statement of the proposed study can be stated as “Designing a single set of algorithm
with one-window operation that assists in multiple-level of enhancement as per the clinical investigation with
retention of cost effectiveness in approach is quite challenging.
The proposed system is an extension of our prior work [33] where we have introduced segmentation
of brain tumor. This work will be considered as complimentary for the segmentation process as it offers
furnishing more pixel-based enhancement on the brain tumor images. The Figure 1 shows the schematic
diagram of the proposed system is as follows:
Gamma
Correction
Adaptive
Histogram
Histogram
Equalization
Filtering
Linear
Non-Linear
Multi-Level Enhancement
weightvariance
Image
Enhancement
Techniques
Figure 1. Schema of proposed methodology
The proposed technique offers comprehensive individual methods in one window operation that can
perform enhancement of the given radiological image. The proposed system consists of three core modules
i.e. preliminary preprocessing, applying filters, and multi-level enhancement. The preliminary preprocessing
operation consists of applying gamma correction, adaptive histogram, and histogram equalization.
The second process is about applying filter of two forms i.e. linear and non-linear forms in order to mitigate
all forms of distortions in the frequeny levels. The third process is about multiple forms of enhancement
where the emphasis is given to formulate an empirical expression in order to investigate the impact of
statistical feature with the enhancement process. The proposed system uses variance and weight in multiple
orders for investigating the impact of it towards the enhancement of an image.
The overall outcome was assessed using Peak Signal-to-Noise Ratio (PSNR). The proposed scheme
offers the physical with various alternatives of image enhancement as there is no particular enhancement
technique that can have equal performance on different forms of radiological images in existing system.
Hence, the proposed system offers a single window operation, where one radiological image of brain tumor
can be investigated more easily owing to availability of different forms of image enhancement technique.
The next section discusses about the algorithm implementation.
2. ALGORITHM IMPLEMENTATION
This algorithm is responsible for performing multiple forms of enhancement of brain tumor images
in order to assist the diagnosis process for better visualization. Hence, this algorithm implements a series of
image enhancement techniques that offers better decision process during closer investigation of radiological

 ISSN: 2088-8708
804
images of brain tumor. The first part of the study implies gamma correction method. The gamma correction
method is mathematically represented as,
𝐺(𝑥, 𝑦) = [𝑓(𝑥, 𝑦)] 𝛾
Where, 𝐺(𝑥, 𝑦) intensity of the output image, f(x,y) intensity values of the input image, T
transformation function, x, y  coordinates of the image pixels, 𝛾 the constant value defined as gamma
value. The second process deals with histogram equalization. Considering the Rayleigh distribution in AHE,
A histogram of bell shaped curve is produced, which can be defined as a function given as
𝑔 = 𝑔 𝑚𝑖𝑛 + [2 × 𝛼2
ln
1
1−𝑃(𝑓)
]
0.5
Where, 𝑔 distribution function, (in this case it is Rayleigh distribution), 𝑔 𝑚𝑖𝑛 Minimum pixel value, and
P(f)  cumulative probability distribution function which is a non-negative real scalar representing
a distribution parameter. The mathematical expression for applying linear filtering is as below:
𝐼 𝑝(𝑖, 𝑗) = √[𝑒𝑥𝑝(𝑓(𝑖, 𝑗))]
2
Where, 𝐼 𝑝 enhanced image. In non-linear filtering, the image is defined by,
𝐼 = 𝑓(𝑖, 𝑗)
Where, i number of rows and j number of columns, and L logarithmic single channel intensity image
to be processed, niter  number of iterations, Enc_img  raw enhanced output image. To obtain
the maximum colour variance of the channel, Maximum = max (L)
𝑂𝑃 = 𝑀𝑎𝑥𝑖𝑚𝑢𝑚 × ∑ 𝑎𝑖𝑗
𝑁
𝑖,𝑗
Where, a1. To perform initial shift,
𝑠ℎ𝑖𝑓𝑡 = ∑ 2(log2(min(𝑖,𝑗)))−1𝑁−1
𝑖,𝑗=1
Initializing old product, the horizontal and vertical step is given by
𝐼 𝑝 = {
𝐼 𝑝1| 𝑖 + 𝑗 > 0
𝐼 𝑝2| 𝑖 + 𝑗 < 0
𝐼 𝑝1 = (∑ ∑ 𝑂𝑃(𝑖, 𝑗)
𝑁−𝑗
1
𝑁−𝑖
1 ) + (∑ ∑ 𝑅𝑅(𝑖, 𝑗)𝑁
𝑗+1
𝑁
𝑖+1 ) − (∑ ∑ 𝑅𝑅(𝑖, 𝑗)
𝑁−𝑗
1
𝑁−𝑖
1 )
𝐼 𝑝2 = (∑ ∑ 𝑂𝑃(𝑖, 𝑗)𝑁
1−𝑗
𝑁
1−𝑖 ) + (∑ ∑ 𝑅𝑅(𝑖, 𝑗)
𝑁+𝑗
1
𝑁+𝑖
1 ) − (∑ ∑ 𝑅𝑅(𝑖, 𝑗)𝑁
1−𝑗
𝑁
1−𝑖 )
𝑠ℎ𝑖𝑓𝑡 =
𝑠ℎ𝑖𝑓𝑡
2
,
𝑁𝑃 =
(𝐼𝑃+𝑂𝑃)
2
𝑂𝑃 = 𝑁𝑃, where, 𝑁𝑃 enhanced image
In Multilevel Enhancement, the Initial parameters are a  weight initialized to a vector
(i.e. 𝑎1, 𝑎2, 𝑎3, 𝑎4, 𝑎5), var  variance initialized to a vector of finite values (i.e. 𝑤1, 𝑤2, 𝑤3, 𝑤4, 𝑤5).
Enhancement with respect to primary image enhancement is given as,
𝐺(𝑖, 𝑗) = ∑ 𝑘1 ∙ 𝑒𝑥𝑝 (−
(𝑖2+𝑗2)
𝑣𝑎𝑟2
)𝑁−1,𝑀−1
𝑖,𝑗=1 (1)
Where, G  image with respect to Gaussian distribution function, I  original image, x, y  pixel
coordinates. Performing convolution for the (1).

805
𝑐𝑜𝑛𝑣 = 𝐼⨂𝐺 (2)
Where, conv  convoluted image, computing the difference and applying log to both the variables, we get
𝐼 𝑝 = log(𝐼 + 1) − log(𝑐𝑜𝑛𝑣)
Where, 𝐼 𝑝 enhanced image
The significant steps of the algorithm are as follows:
Algorithm for Multi-Level Image Enhancement
Input: I (Input Image), χ (Correction factor)
Output: EI (Enhanced Image)
Start
1. init I, χ
2. IgcIχ
3. Iahα1(I)
4. Iheα2(I)
5. Ilf|expIFFT2(FFT2(log(I). β(I, fH, fL, c, d)))|
6. Inlfϕ(I)
7. For i=1:m
8. For j=1:n
9. ck.exp(-(i2
+j2
)/σ2
)
10. End
11. End
12. Imelog(I+1)-log(ψ(I+1), c)
13. EI={Igc, Iah, Ilf, Inlf, Ime}
End
The algorithm first performs gamma correction in order to encoding as well as decoding
the luminance factor for a given value of correction factor χ (Line-2) in order to obtain gamma corrected
image Igc. The algorithm also allows implementing adaptive histogram α1 and histogram equalization α2 in
order to obtain enhanced image Iah (Line-3) and Ihe(Line-4). The next part of the algorithm implements linear
filtering process using multiple forms of input parameters e.g. fH (high-frequency), fL (low-frequency),
c (constant that controls the slope peaks), d (threshold frequency). This process is also associated with
applying Fourier and inverse Fourier transform in order to obtain an enhanced image Ilf (Line-5). The next
part of the algorithm focuses on implementing non-linear filter ϕ (Line-6).
The input to the function ϕ is a logarithmic single-channel intensity image that is required to be
enhanced as well as iterations too whereas the output of this process is enhanced image accomplished from
non-linear filtering. This step of the algorithm uses a shift operator that initiates with initializing
the maximum color value in the image followed by design of initial shift as wel as old product towards
horizontal and vertical steps. Finally, the shift operators are updated. The next part of the algorithm
introduces multiple-level of enhancement where the input image has the scope to get itself enhance
depending on the numerical inputs offered by the user. For this purpose, the algorithm takes weight as well as
variance. Than for all the values of rows (Line-7) and column (Line-8), the algorithm implements an
empirical expression of c (Line-9) that depends on image dimensions (i-row and j-column) and variance σ
(Line-9). This step is followed by computation of finally enhanced image Ime by applying logarithmic
function and convolution operation on the image (Line-12).
The algorithm offers three different forms of enhancement in the multiple levels. The primary level
of enhance is carried out by only one values of weight and variance, whereas the secondary image
enhancement level is carried out by using dual values of weight and variance. Similarly, the teriary image
enhancement is carried out using three values of weight and variance. Incorporation of multiple forms of
weight and variable offers flexibility to assess different forms of radiological images. One interesting fact
about the proposed system is that it offers multiple forms of simple image enhancement techniques in single
window operation that offers the significant level of advantage for the physician to diagnose the different
forms of brain tumor images with more customized forms of enhancement. At the same time, majority of
the filtering process is carried out with an equal emphasis on linear and non-linear filtering techniques, which
is one of the great methods for addressing the multiplicative noise (as radiological images do have
multiplicative noises). Moreover, as the radiological images of brain tumor are normally grayscale therefore,
this algorithm offers a set of enhancement techniques (Line-13) that optimizes the appearance of

 ISSN: 2088-8708
806
the grayscale images with reduced iteration and highly in progressive forms. Therefore, the proposed system
can be claimed to offer a cost effective solution to enhance the brain tumor images within inclusion of any
complex and iterative parameters in the filter design.
3. RESULT ANALYSIS
This section discusses about the outcomes obtained from implementing the proposed system.
The analysis of image enhancement was carried out in brain dataset where certain samples of visual
outcomes were exhibited in Figure 2 where Figure 2(a) represents original sample image, Figure 2(b)
represents gamma corrected image, Figure 2(c) represents image acquired from adative histogram,
Figure 2(d) represents image obtained from histogram equalization. The outcome of linear and non-linear
filtering can be seen in Figure 2(e) and Figure 2(f). The multiple forms of enhancement for primary,
secondary, and tertiary forms with multiple weights and variance can be seen in Figure 2(g), Figure 2(h), and
Figure 2(i) respectively. Finally, Figure 2(j) shows outcome of joint forms of multiple enhancement
by maximizing the number of weights and variance level.
Figure 2. Visual outcomes of proposed system
We also closely monitor the outcome of the PSNR for all the implemented image enhancement
techniques and found that each technique have different forms of pattern based on its visual quality.
However, with the outcome shown in Figure 3, there should be any conclusion drawn about effectiveness of
any specific technique as we have perform this analysis by uniformly distributing multiple weight and
variance values. Looking into this outcome, it will not mean that multi-level enhancement offers lesser PSNR
is ineffective while gamma correction is highly effective. As for different forms of images, the outcomes
differ to some extent depending upon the visual complexity of an image.
Figure 3. Comparative performances of image enhancement techniques

807
It can only be said that proposed system offers a tool that allows the physician to perform thorough
investigation of the radiological image of a brain tumor so that the diagnosis is carried out in proper manner.
At the same time, the algorithm takes minimum of 0.052755 seconds and maximum of 1.223537 seconds in
order to provide the response of enhanced image of brain tumor. The response time is quite faster and there is
accumulation of any memory or dependency of any heuristics in order to perform the analysis. Hence, it can
be said that proposed system offers wide ranges of image enhancement techniques in one window operation
with faster response to enhanced image.
4. CONCLUSION
From the discussion presented in this paper, it is very much clear that it is feasible to construct
a framework using sub-set of algorithms for offering wide-range of enhanced image. The first contribution of
this paper is its cost effectiveness as in order to implement this scheme in reality there is no much
dependency of any external computational or hardware-based parameter and neither there is any external
dependencies. The second contribution of this paper is that it offers a physician to view the given
enhancement outcome in their possible choices which offers them better scope of decision making. The third
contribution of this approach is that statistical parameters e.g. weight and variance plays a significant role
that can offer more comprehensive extraction of information to offer more granularities in the enhanced
images for diagnosis purpose. Our future work will be completely in the direction of making classification of
diseases on the top of this framework.
REFERENCES
[1] W. Birkfellner, Applied Medical Image Processing, Second Edition: A Basic Course, Taylor & Francis, 2016.
[2] D. V. Sahani and A. E. Samir, Abdominal Imaging, E-Book: Expert Radiology Series, Elsevier Health Sciences,
2016.
[3] K. W. Lai, et al., Advances in Medical Diagnostic Technology, Springer Science & Business Media, 2014.
[4] C. H. Chen, Computer Vision in Medical Imaging, World Scientific, 2014.
[5] A. Das, et al., “Detection of abnormal regions of precancerous lesions in digitised uterine Cervix images,” 2014
International Electrical Engineering Congress (iEECON), Chonburi, pp. 1-4, 2014.
[6] S. Khobragade, et al., “Automatic detection of major lung diseases using Chest Radiographs and classification by
feed-forward artificial neural network,” 2016 IEEE 1st International Conference on Power Electronics, Intelligent
Control and Energy Systems (ICPEICES), Delhi, pp. 1-5, 2016.
[7] R. Chandra, et al., “On the Opportunities and Challenges in Microwave Medical Sensing and Imaging,” IEEE
Transactions on Biomedical Engineering, vol. 62, pp. 1667-1682, 2015.
[8] F. L. Lizzi and E. J. Feleppa, “Image Processing and Pre-Processing for Medical Ultrasound,” ACM_roceedings of
the 29th Applied Imagery Pattern Recognition Workshop, pp. 187, 2000.
[9] B. Chitradevi and P. Srimathi, “An Overview on Image Processing Techniques,” International Journal of Innovative
Research in Computer and Communication Engineering, 2014.
[10] S. Preethi and D. Narmadha, “A Survey on Image Denoising Techniques,” International Journal of Computer
Applications, vol. 58, pp. 27-30, 2012.
[11] E. S. Yelmanova and Y. M. Romanyshyn, “Medical image contrast enhancement based on histogram,” 2017 IEEE
37th International Conference on Electronics and Nanotechnology (ELNANO), Kiev, pp. 273-2, 2017.
[12] Z. Zhao and Y. Zhou, “PLIP based unsharp masking for medical image enhancement,” 2016 IEEE International
Conference on Acoustics, Speech and Signal Processing (ICASSP), Shanghai, pp. 1238-1242, 2016.
[13] I. Chiuchisan, “An approach to the Verilog-based system for medical image enhancement,” 2015 E-Health and
Bioengineering Conference (EHB), Iasi, pp. 1-4, 2015.
[14] L. Liang and Y. Si, “Medical image enhancement using sliding weighted empirical mode decomposition,” 2015
IEEE International Conference on Information and Automation, Lijiang, pp. 3145-3148, 2015.
[15] T. Gong, et al., “Improved immune algorithm for medical image enhancement,” 2015 International Workshop on
Artificial Immune Systems (AIS), Taormina, pp. 1-7, 2015.
[16] X. W. Li and Y. X. Kang, “A novel medical image enhancement method based on wavelet multi-resolution
analysis,” 2015 8th International Conference on Biomedical Engineering and Informatics (BMEI), Shenyang,
pp. 727-731, 2015.
[17] G. Soloperto, et al., “Multiparametric Evaluation of the Acoustic Behavior of Halloysite Nanotubes for
Medical Echographic Image Enhancement,” IEEE Transactions on Instrumentation and Measurement, vol. 63,
pp. 1423-1430, 2014.
[18] Q. Xue, “Enhancement of medical images in the shearlet domain,” Proceedings of 2013 3rd International
Conference on Computer Science and Network Technology, Dalian, pp. 235-238, 2013.
[19] F. Zhou, et al., “Medical image enhancement based on NSCT,” IET International Conference on Smart and
Sustainable City 2013 (ICSSC 2013), Shanghai, pp. 166-169, 2013.
[20] T. Chaira, “Medical image enhancement using intuitionistic fuzzy set,” 2012 1st International Conference on Recent
Advances in Information Technology (RAIT), Dhanbad, pp. 54-57, 2012.

 ISSN: 2088-8708
808
[21] N. J. Dhinagar and M. Celenk, “Ultrasound medical image enhancement and segmentation using adaptive
homomorphic filtering and histogram thresholding,” 2012 IEEE-EMBS Conference on Biomedical Engineering and
Sciences, Langkawi, pp. 349-353, 2012.
[22] L. Hua, et al., “A new color medical image enhancement method,” IET International Conference on Information
Science and Control Engineering 2012 (ICISCE 2012), Shenzhen, pp. 1-5, 2012.
[23] B. Kurt, et al., “Medical images enhancement by using anisotropic filter and CLAHE,” 2012 International
Symposium on Innovations in Intelligent Systems and Applications, Trabzon, pp. 1-4, 2012.
[24] T. Xiurong, “The application of adaptive unsharp mask algorithm in medical image enhancement,” Proceedings of
2011 Cross Strait Quad-Regional Radio Science and Wireless Technology Conference, Harbin, pp. 1368-1370, 2011.
[25] B. Fan, et al., “Medical image enhancement based on modified lut-mapping derivative and multi-scale layer contrast
modification,” 2011 4th International Congress on Image and Signal Processing, Shanghai, pp. 696-703, 2011.
[26] L. Jing and X. Jie, “Research and Implementation on the Digital Medical Image Enhancement Methods Based on
Gradient Smooth and Context Smooth,” 2011 Seventh International Conference on Computational Intelligence and
Security, Hainan, pp. 1193-1196, 2011.
[27] M. F. Hossain, et al., “Medical image enhancement based on nonlinear technique and logarithmic transform
coefficient histogram matching,” IEEE/ICME International Conference on Complex Medical Engineering, Gold
Coast, QLD, pp. 58-62, 2010.
[28] A. Jiang, et al., “Medical Image Enhancement Algorithm Based on Contourlet Transform,” 2010 First International
Conference on Pervasive Computing, Signal Processing and Applications, Harbin, pp. 624-627, 2010.
[29] Y. Yang, et al., “Medical image enhancement algorithm based on wavelet transform,” Electronics Letters, vol. 46,
pp. 120-121, 2010.
[30] S. L. Bangare, et al., “Neuroendoscopy Adapter Module Development for Better Brain Tumor Image Visualization,”
International Journal of Electrical and Computer Engineering (IJECE), vol. 7, no. 6, pp. 3643-3654, 2017.
[31] R. Kahina and A. Bassou, “Improving The Performance of Viterbi Decoder using Window System,” International
Journal of Electrical and Computer Engineering (IJECE), vol. 8, no. 1, pp. 611-621, 2018.
[32] Verma, et al., “Intensity Preserving Cast Removal in Color Images Using Particle Swarm Optimization,”
International Journal of Electrical & Computer Engineering (IJECE), vol. 7, no. 5.pp. 2581-2595, 2017.
[33] T. S. D. Murthy and G. Sadashivappa, “Brain tumor segmentation using thresholding, morphological operations and
extraction of features of tumor,” IEEE-International Conference on Advances in Electronics Computers and
Communications, Bangalore, pp. 1-6, 2014.
BIOGRAPHIES OF AUTHORS
Deepthi Murthy T.S, received the BE (Electronics and Communication) Degree from VTU,
India in 2005, and the M.Tech (Digital Electronics and Communication) degree from VTU,
India in 2007. She has 10 years of teaching experience. She is currently working as an
Assistant Professor in REVA University, Bangalore; her research interests include Medical
Imaging, Signal Processing and Applications.
Dr. G. Sadashivappa, has obtained his doctoral degree from VTU, Belgaum. He has 32 years
of teaching experience and 10 years of research experience. He has authored two text books.
He has chaired many international conferences. He was the Principle Investigator for
the project “On Study and development of Image Compression Algorithms for Satellite
Images” supported by ISRO. He is currently professor at Telecommunication Engineering
Department, RVCE, Bangalore. His research area of interest is Fiber Optic communication,
Medical Imaging, Signal Processing & applications. He has published many international
journals.

Framework for comprehensive enhancement of brain tumor images with single-window operation

More Related Content

What's hot (18)

Similar to Framework for comprehensive enhancement of brain tumor images with single-window operation (20)

More from IJECEIAES (20)

Recently uploaded (20)

Framework for comprehensive enhancement of brain tumor images with single-window operation