China
Africa
Explore chapters and articles related to this topic
Computer-aided Diagnosis (CAD) System for Determining Histological Grading of Astrocytoma Based on Ki67 Counting
Published in Varun Bajaj, G.R. Sinha, Computer-aided Design and Diagnosis Methods for Biomedical Applications, 2021
Fahmi Akmal Dzulkifli, Maryam Ahmad Sharifuddin, Mohd Yusoff Mashor, Hasnan Jaafar
The next process was to convert the color space of the image from a RGB to L*a*b* color space. This color space was selected due to the exact color representation and its device-independent color model. Then, the contrast enhancement technique was applied to the luminosity, “L” channel while the a* and b* channels remained unchanged. In this study, Contrast-Limited Adaptive Histogram Equalization was used to enhance the contrast of the astrocytoma images. The CLAHE technique is a variation of an adaptive histogram equalization which reduces noise amplification by limiting contrast amplification [17]. Instead of using the whole image, this technique was performed on small regions called “tiles.” The contrast of each tiles was enhanced, resulting in the histogram of the output region approximately matching the histogram specified by the desired histogram value [18]. Equation 11.1 demonstrates the central equation for enhancing the image: Ix,y=px,yqx,y*maxluminosity
Application of Image Processing and Data in Remote Sensing
Published in Ankur Dumka, Alaknanda Ashok, Parag Verma, Poonam Verma, Advanced Digital Image Processing and Its Applications in Big Data, 2020
Ankur Dumka, Alaknanda Ashok, Parag Verma, Poonam Verma
Histogram: Histogram equalization is a computer image processing technique that is used to improve contrast in images. It stretches out the intensity range of the image, thereby allowing areas of lower local contrast to gain a higher contrast. There are two types of Histogram Equalizations:Adaptive Histogram Equalization: Adaptive Histogram Equalization computes many histograms where each histogram corresponds to a distinct part of the image, thereby making it suitable for enhancing the definitions of the edges in each region of an image.Contrastive Limited Adaptive Equalization: Contrastive Limited Adaptive Equalization has a transformation function derived from the contrast limiting procedure applied to each neighboring pixel of an image.
Deep Learning for Retinal Analysis
Published in Ervin Sejdić, Tiago H. Falk, Signal Processing and Machine Learning for Biomedical Big Data, 2018
Henry A. Leopold, John S. Zelek, Vasudevan Lakshminarayanan
While more computationally intensive, local histogram enhancement methods are able to improve the enhanced image quality and contrast where global methods fail. Rather than sampling all pixels within an image once, histograms are generated for subsections of the image, each of which is normalized. Windows need to overlap so subsection boundaries are normalized accordingly, resulting in a the spike in computational power. One limitation for local methods is the risk of enhancing noise within the image. Contrast-limited adaptive histogram equalization (CLAHE) is one method that overcomes this limitation. CLAHE limits the maximum pixel intensity peaks within a histogram, redistributing the values across all intensities prior to histogram equalization [60]. This is the contrast enhancement method used during preprocessing in Figure 17.14. More advanced local enhancement techniques exist and often require some calibration for specific problems, such as rank enhancement where pixel values are adjusted based on their relative distance from histogram minima/maxima.
A Comprehensive Survey on the Detection of Diabetic Retinopathy
Published in IETE Journal of Research, 2022
The template-based methods draw candidate optic disc regions by adjusting the threshold value. The Hough Transformation identifies the candidates as a circle with the highest average intensity in the retinal image containing OD. The Prewitt edge detector gives OD boundary candidates [43]. The optic disc (OD) can be segmented from the retinal image after passing through the image enhancement system and low-pass FIR filter. By altering the contrast level, the adaptive histogram equalization approach promotes image quality. CLAHE is a cutting-edge technique for increasing image quality by distributing intensity levels. Green components of the colour image are passed through this process. Image preprocessing consists of a low-pass filter with an appropriate degree of order to be applied on the fundus camera retinal image to blur the blood vessels present; image enhancement is required to segment the OD. Morphological dilation and median filtering operation is useful in the segmentation of OD. OD consists of many limited pixels in a circular area [44]. Detection and segmentation of OD is a complex process. In the detection of OD, filtered retinal image, a maximum intensity pixel is collected and used as a reference point for computing the threshold. A full-intensity pixel is present approximately at the centre of OD. While calculating the threshold, the region should not be excessively huge or tiny, it should fit the OD region properly. The threshold decides the operational region and its value can be empirically obtained, as shown in Figure 6.
Retinal Blood Vessel Segmentation from Depigmented Diabetic Retinopathy Images
Published in IETE Journal of Research, 2021
T. Jemima Jebaseeli, C. Anand Deva Durai, J. Dinesh Peter
The proposed method identifies a solution to the retinal blood vessel segmentation problem. The outline of the proposed architecture is shown in Figure 2. This effort endeavors to segment the blood vessels through three phases viz. Firstly, Contrast Limited Adaptive Histogram Equalization (CLAHE) is used to remunerate the consequences of non-uniform lighting and to improve the image contrast. The blood vessels are classified using Deep Learning Based Support Vector Model (DLBSVM). The classified blood vessels are segmented by TPCNN model. The parameters of TPCNN model are fixed by PSO. Finally, the segmented results are evaluated with the ground truth image. The procedures adopted during each phase are portrayed in the following subsections.
A Novel L-CLAHE-Based Intensification Filter for Enhancement of Underwater Images and Pipeline Tracking
Published in IETE Journal of Research, 2023
Arun A. Balakrishnan, P. R. Dhanya, Syamily Anilkumar, M. H. Supriya
CLAHE is a modified version of adaptive histogram equalization in which the amplification of noises can be reduced by limiting the contrast of the pixels. In CLAHE, the slope of the transformation function gives the amount of contrast amplification. Since different images have different intensity levels, noise levels, and artefacts, instead of a predefined clip limit, it is adaptively estimated from the input image to yield better results. The entropy-based method for the automated selection of the clip limit [10] for the CLAHE seems to be ideal after testing out the different methods.