China
Africa
Explore chapters and articles related to this topic
Image Edge Detection Using Fractional Conformable Derivatives in Liouville-Caputo Sense for Medical Image Processing
Published in Devendra Kumar, Jagdev Singh, Fractional Calculus in Medical and Health Science, 2020
J. E. Lavín-Delgado, J. E. Solís-Pérez, J. F. Gómez-Aguilar, R. F. Escobar-Jiménez
C1, C2 and C3 are small constants that are included to avoid instability when the denominator is very close to zero; µI1, µI2 are the means; σI1I2 is the covariance; and σI1, σI2 are the standard deviations of the ideal image (I1) and the filtered image (I2), respectively. In addition, ζ, δ, and γ are positive parameters that control the relative importance of the components. The main limitation of SSIM measure is the inability to measure highly blurred images successfully [43]. Edge-strength-similarity-based image metric (ESSIM) is an improved measure, based on SSIM, which compares the edge information between the analysed images [44]. Let be the ideal image (ground truth) as I1=[I11,…,I1i,…,I1N]∈ℝN,
Reconfigurable optimal hybrid vision enhancement system for night surveillance robot using hybrid genetic-PSO algorithm
Published in Arun Kumar Sinha, John Pradeep Darsy, Computer-Aided Developments: Electronics and Communication, 2019
L.M.I. Leo Joseph, B. Girirajan
In this analysis, we consider the visionless images for quality analysis. The quality is based on the signal to noise ratio (PSNR) and the structural similarity image index (SSIM). In general, a high value of the PSNR vary from 0 to infinity give an indication that two images might be very similar. The two images can be deemed similar when the SSIM value, which varies from 0 to 1, is close to 1. The test can be conducted in a MATLAB. The quality of proposed OHVE system is compared with existing HVE system [15]. Figure 4 gives the test visionless images obtained from night visionless images database, following images are taken as test images namely corridor, desktop, digit-max-gate, rest room, server room and station. Figure 5 illustrates the images that we have obtained from the proposed OHVE system that enhances the luminance intensities in a very high manner. The PSNR and SSIM of proposed OHVE and existing HVE system are compared in Table I. It clearly depicts the PSNR of proposed OHVE system is very high compare to existing HVE system, with the difference of 25
Utilization of Small S-Boxes for Information Hiding
Published in S. Ramakrishnan, Cryptographic and Information Security, 2018
Majid Khan, Syeda Iram Batool Naqvi
The structural similarity index matrix (SSIM) is a technique for measuring the similarity between two pictures. The SSIM record is a full reference metric; as such, the measuring of picture quality focused around an introductory uncompressed or without distortion picture as reference. The SSIM metric is figured on different windows of a picture. The measure between original and marked images of size is given below [19]: SSIM(X,Y)=(2μxμx+c1)(σxy+c2)(μx2+μy2+c1)(σx2+σy2+c2).
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
SSIM is an evaluation metric used to evaluate how well the structure of an image is retained. The similarity among two images x and y of size N × N is where µx and µy indicate the mean of the original image and enhanced image, respectively, σx, σy represent the standard deviations of each image and σxy represents the covariance of the two images. C1, C2, and C3 are introduced to avoid division by zero error. C1 = (K1L)2, C2 = (K2L)2, and C3 = C2/2, where K1 = 0.01, K2 =0.03, and L denotes dynamic range of the input. SSIM values are calculated for each channel and averaged together for colour images. If the SSIM value is in the neighbourhood of unity, more will be the preservation of edges in the pre-processed image as given in Table 3. The improved L-CLAHE algorithm also preserves the structural similarity of the enhanced images.
Phase congruency-based filtering approach combined with a convolutional network for lung CT image analysis
Published in The Imaging Science Journal, 2021
Mohamed Ben Gharsallah, Hassene Seddik
MSSIM was developed to measure the visual quality of a filtered image compared to the original image. The idea of SSIM is to measure the similarity of structure between the two images rather than a pixel-to-pixel difference. The underlying assumption is that the human eye is more sensitive to changes in the structure of the image. The SSIM metric is calculated on several windows of an image. The SSIM measurement between two images F and G is computed as follows: where are the averages of images F and G, respectively, are the variance of F, G, respectively,is the covariance of F, G. c1 and c2 two variables intended to stabilize the division when the denominator is very low chosen c1 = 0.01 and c2 = 0.03 by default. MSSIM is defined as the mean of SSIM PSNR is a distortion measure used in digital image, especially in image compression. It quantifies the performance of coders by measuring the quality of reconstruction of the compressed image compared to the original image.
A blind proposed 3D mesh watermarking technique for copyright protection
Published in The Imaging Science Journal, 2020
O. H. Khalil, Ahmed Elhadad, A. Ghareeb
There are more evaluations of the perceptual difference techniques such as in Ref. [45]. The structural similarity index measures (SSIMs) the perceptual difference between two similar images based on multiplicative combination of three visible structure characteristics of an image: luminance, contrast and structure. The difference with respect to other techniques mentioned previously such as mean square error (MSE) or peak signal-to-noise ratio (PSNR) is that these approaches estimate absolute errors; on the other hand, SSIM is a perception-based model that considers image degradation as perceived change in structural information. Therefore, Figure 8 shows the corresponding resultant structural similarity of the extracted secret image. In fact, the results show that the different values of β provide a minor impact on the accuracy of the extracted secret image.