Structural Similarity Index Metric Research Articles

An optimized EBRSA-Bi LSTM model for highly undersampled rapid CT image reconstruction

COVID-19 has spread all over the world, causing serious panic around the globe. Chest computed tomography (CT) images are integral in confirming COVID positive patients. Several investigations were conducted to improve or maintain the image reconstruction quality for the sample image reconstruction. Deep learning (DL) methods have recently been proposed to achieve fast reconstruction, but many have focused on a single domain, such as the image domain of k-space. In this research, the highly under-sampled enhanced battle royale self-attention based bi-directional long short-term (EBRSA-bi LSTM) CT image reconstruction model is proposed to reconstruct the image from the under-sampled data. The research is adapted with two phases, namely, pre-processing and reconstruction. The extended cascaded filter (ECF) is proposed for image pre-processing and tends to suppress the noise and enhance the reconstruction accuracy. In the reconstruction model, the battle royale optimization (BrO) is intended to diminish the loss function of the reconstruction network model and weight updation. The proposed model is tested with two datasets, COVID-CT- and SARS-CoV-2 CT. The reconstruction accuracy of the proposed model with two datasets is 93.5 % and 97.7 %, respectively. Also, the image quality assessment parameters such as Peak-Signal to Noise Ratio (PSNR), Root Mean Square Error (RMSE) and Structural Similarity Index metric (SSIM) are evaluated, and it yields an outcome of (45 and 46 dB), (0.0026 and 0.0022) and (0.992, 0.996) with two datasets.

Technical note: Extended longitudinal and lateral 3D imaging with a continuous dual-isocenter CBCT scan.

The clinical benefits of intraoperative cone beam CT (CBCT) during orthopedic procedures include (1) improved accuracy for procedures involving the placement of hardware and (2) providing immediate surgical verification. Orthopedic interventions often involve long and wide anatomical sites (e.g., lower extremities). Therefore, in order to ensure that the clinical benefits are available to all orthopedic procedures, we investigate the feasibility of a novel imaging trajectory to simultaneously expand the CBCT field-of-view longitudinally and laterally. A continuous dual-isocenter imaging trajectory was implemented on a clinical robotic CBCT system using additional real-time control hardware. The trajectory consisted of 200° circular arcs separated by alternating lateral and longitudinal table translations. Due to hardware constraints, the direction of rotation (clockwise/anticlockwise) and lateral table translation (left/right) was reversed every 400°. X-ray projections were continuously acquired at 15frames/s throughout all movements. A whole-body phantom was used to verify the trajectory. As comparator, a series of conventional large volume acquisitions were stitched together. Image quality was quantified using Root Mean Square Deviation (RMSD), Mean Absolute Percentage Deviation (MAPD), Structural Similarity Index Metric (SSIM) and Contrast-to-Noise Ratio (CNR). The imaging volume produced by the continuous dual-isocenter trajectory had dimensions of L=95cm×W=45cm×H=45cm. This enabled the hips to the feet of the whole-body phantom to be captured in approximately half the imaging dose and acquisition time of the 11 stitched conventional acquisitions required to match the longitudinal and lateral imaging dimensions. Compared to the stitched conventional images, the continuous dual-isocenter acquisition had RMSD of 4.84, MAPD of 6.58% and SSIM of 0.99. The CNR of the continuous dual-isocenter and stitched conventional acquisitions were 1.998 and 1.999, respectively. Extended longitudinal and lateral intraoperative volumetric imaging is feasible on clinical robotic CBCT systems.

A high‐efficient imprecise discrete cosine transform block based on a novel full adder and Wallace multiplier for bioimages compression

SummarySophisticated systems take advantage of approximate circuits for energy management. A gate‐level structure is introduced to make a novel imprecise full adder (FA) cell by the gate diffusion input (GDI) strategy. The carbon nanotube field‐effect transistor (CNTFET) technology lowers the FA power, and the swing issue is resolved by the dynamic threshold (DT) technique. The FA specifications are 10 transistors, two internal nodes, 0.191‐μm2 area, three errors, and an error rate (ER) of 37.5%. Some inputs are directly passed to the final stage of the FA to reduce the delay and dynamic power. The FA is used in three 4:2 compressors with 12, 16, and 20 transistors, respectively. The compressors are based on the modified stacking concept to attain low power, small area, and high speed. The compressors are used up in the partial product reduction tree (PPRT) of a new 8‐bit inexact multiplier. Regarding the multiplier, the power–delay product (PDP) is decreased by 25.21%, and the normalized mean error distance (NMED) shows high accuracy. An imprecise discrete cosine transform (DCT) is implemented by the multiplier for image compression, and compared with the literature, the peak signal‐to‐noise ratio (PSNR) and structural similarity index metric (SSIM) are amended by 8.26% and 7.29%, respectively.

Encipher GAN: An End-to-End Color Image Encryption System Using a Deep Generative Model

Chaos-based image encryption schemes are applied widely for their cryptographic properties. However, chaos and cryptographic relations remain a challenge. The chaotic systems are defined on the set of real numbers and then normalized to a small group of integers in the range 0–255, which affects the security of such cryptosystems. This paper proposes an image encryption system developed using deep learning to realize the secure and efficient transmission of medical images over an insecure network. The non-linearity introduced with deep learning makes the encryption system secure against plaintext attacks. Another limiting factor for applying deep learning in this area is the quality of the recovered image. The application of an appropriate loss function further improves the quality of the recovered image. The loss function employs the structure similarity index metric (SSIM) to train the encryption/decryption network to achieve the desired output. This loss function helped to generate cipher images similar to the target cipher images and recovered images similar to the originals concerning structure, luminance and contrast. The images recovered through the proposed decryption scheme were high-quality, which was further justified by their PSNR values. Security analysis and its results explain that the proposed model provides security against statistical and differential attacks. Comparative analysis justified the robustness of the proposed encryption system.

Resampling Estimation Based RPC Metadata Verification in Satellite Imagery

Recent advances in machine learning and computer vision have made it simple to manipulate a variety of media, including satellite images. Most of the commercially available satellite images go through the process of orthorectification to remove potential distortions due to terrain variations. This orthorectification process typically involves the use of rational polynomial coefficients (RPC) that geometrically remap the pixels in the original image to the rectified image. This paper proposes the first method to verify the authenticity of RPC metadata in an orthorectified satellite image. The steps include calculating the Residual Discrete Fourier Transform (DFT) pattern from the image using a linear predictor based residual spectral analysis and comparing with Expected Residual DFT pattern that is obtained using the RPC metadata associated with the image. If the metadata associated with orthorectified image is correct, then the Residual-DFT pattern (which represents image data) and the Expected-Residual-DFT pattern (which represents metadata) should be similar. We use SSIM (Structural Similarity Index Metric) to quantify the similarity and thereby verify if the data has been tampered or not. Detailed experimental results demonstrate that our method achieves over 97% accuracy in the majority of binary tampering detection tests.

Proposed Privacy Preservation Technique for Color Medical Images

Nowadays, the security of images or information is very important. This paper introduces a proposed hybrid watermarking and encryption technique for increasing medical image security. First, the secret medical image is encrypted using Advanced Encryption Standard (AES) algorithm. Then, the secret report of the patient is embedded into the encrypted secret medical image with the Least Significant Bit (LSB) watermarking algorithm. After that, the encrypted secret medical image with the secret report is concealed in a cover medical image, using Kekre’s Median Codebook Generation (KMCG) algorithm. Afterwards, the stego-image obtained is split into 16 parts. Finally, it is sent to the receiver. We adopt this strategy to send the secret medical image and report over a network securely. The proposed technique is assessed with different encryption quality metrics including Peak Signal-to-Noise Ratio (PSNR), Correlation Coefficient (Cr), Feature Similarity Index Metric (FSIM), and Structural Similarity Index Metric (SSIM). Histogram estimation is used to confirm the matching between the secret medical image before and after transmission. Simulation results demonstrate that the proposed technique achieves good performance with high quality of the received medical image and clear image details in a very short processing time.

Image Denoising using Wavelet Cycle Spinning and Non-local Means Filter

Removing as much noise as possible in an image while preserving its fine details is a complex and challenging task. We propose a wavelet-based and non-local means (NLM) denoising method to overcome the problem. Two well-known wavelets: dual-tree complex wavelet transform (DT-CWT) and discrete wavelet transform (DWT), have been used to change the noise image into several wavelet coefficients sequentially. NLM filtering and universal hard thresholding with cycle spinning have been used for thresholding on its approximation and detail coefficients, respectively. The inverse two-dimensional DWT was applied to the modified wavelet coefficients to obtain the denoised image. We conducted experiments with twelve test images on the set12 data set, adding the additive Gaussian white noise with variances of 10 to 90 in increments of 10. Three evaluation metrics, such as peak signal noise to rate (PSNR), structural similarity index metric (SSIM), and mean square error (MSE), have been used to evaluate the effectiveness of the proposed denoising method. From these measurement results, the proposed denoising method outperforms DT-CWT, DWT, and NLM almost in all noise levels except for the noise level of 10. At that noise level, the proposed denoising method is lower than NLM but better than DT-CWT and DWT.

Optimal Video Compression Parameter Tuning for Digital Video Broadcasting (DVB) using Deep Reinforcement Learning

DVB (digital video broadcasting) has undergone an enormous paradigm shift, especially through internet streaming that utilizes multiple channels (i.e., secured hypertext transfer protocols). However, due to the limitations of the current communication network infrastructure, video signals need to be compressed before transmission. Whereas most recent research has concentrated and focused on assessing video quality, little to no study has worked on improving the compression processes of digital video signals in lightweight DVB setups. This study provides a video compression strategy (DRL-VC) that employs deep reinforcement learning for learning the suitable parameters used in digital video signal compression. The problem is formulated as a multi-objective one, considering the structural similarity index metric (SSIM), the delay time, and the peak signal-to-noise ratio (PSNR). Based on the findings of the experiments, our proposed scheme increases bitrate savings while at a constant PSNR. Results also show that our scheme performs better than the benchmarked compression schemes. Finally, the root means square error values show a consistent rate across different video streams, indicating the validity of our proposed compression scheme.

Comparing Different Algorithms for the Pseudo-Coloring of Myocardial Perfusion Single-Photon Emission Computed Tomography Images.

Single-photon emission computed tomography (SPECT) images can significantly help physicians in diagnosing patients with coronary artery or suspected coronary artery diseases. However, these images are grayscale with qualities that are not readily visible. The objective of this study was to evaluate the effectiveness of different pseudo-coloring algorithms of myocardial perfusion SPECT images. Data were collected using a Siemens Symbia T2 dual-head SPECT/computed tomography (CT) scanner. After pseudo-coloring, the images were assessed both qualitatively and quantitatively. The qualities of different pseudo-color images were examined by three experts, while the images were evaluated quantitatively by obtaining indices such as mean squared error (MSE), peak signal-to-noise ratio (PSNR), normalized color difference (NCD), and structure similarity index metric (SSIM). The qualitative evaluation demonstrated that the warm color map (WCM), followed by the jet color map, outperformed the remaining algorithms in terms of revealing the non-visible qualities of the images. Furthermore, the quantitative evaluation results demonstrated that the WCM had the highest PSNR and SSIM but the lowest MSE. Overall, the WCM could outperform the other color maps both qualitatively and quantitatively. The novelty of this study includes comparing different pseudo-coloring methods to improve the quality of myocardial perfusion SPECT images and utilizing our collected datasets.

Deep-Learning Architecture-Based Approach for 2-D-Simulation of Microwave Plasma Interaction

This article presents a convolutional neural network (CNN)-based deep-learning (DL) model, inspired from UNet with a series of encoder and decoder units with skip connections, for the simulation of microwave–plasma interaction. The microwave propagation characteristics in complex plasma medium pertaining to transmission, absorption, and reflection primarily depend on the ratio of electromagnetic (EM) wave frequency and electron plasma frequency, and the plasma density profile. The scattering of a plane EM wave with fixed frequency (1 GHz) and amplitude incident on a plasma medium with different Gaussian density profiles (in the range of <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$1 \times10^{17}$ </tex-math></inline-formula> <inline-formula xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink"> <tex-math notation="LaTeX">$- 1\times 10^{22} {\text {m}^{-3}}$ </tex-math></inline-formula> ) have been considered. The training data associated with microwave–plasma interaction has been generated using 2-D finite-difference time-domain (FDTD)-based simulations. The trained DL model is then used to reproduce the scattered electric field values for the 1-GHz incident microwave on different plasma profiles with an error margin of less than 2%. We propose a complete DL-based pipeline to train, validate, and evaluate the model. We compare the results of the network, using various metrics like structural similarity index metric (SSIM) index, average percent error, and mean square error, with the physical data obtained from well-established FDTD-based EM solvers. To the best of our knowledge, this is the first effort toward exploring a DL-based approach for the simulation of complex microwave–plasma interaction. The DL technique proposed in this work is significantly fast when compared to the existing computational techniques and can be used as a new, prospective, and alternative computational approach for investigating microwave–plasma interaction in a real-time scenario.

A Robust technique for Pre-Processing Liver CT images using Double Stage Guassian Filtering with Texture and Contrast Enhancement

Liver is the greatest and largest glandular organ in human body by weight. It is absolutely fundamental to human life as it performs some vital functions such as digestion, metabolism and filtering of possibly harmful bio-chemicals from blood. Due to changing lifestyle and excess consumption of packaged food, liver diseases such as liver cancer are becoming very common. Liver cancer is one of the cancers having very low survival rates. Early detection and proper medication is the key to survival from liver cancer. Many imaging techniques are used to detect liver cancer, with Computed Tomography (CT) images being the most widely used technique. Computer Aided Diagnosis (CAD) involves studying CT scan images to detect liver cancer. Due to noises present in CT images, sometimes the effectiveness of algorithm gets affected. It is therefore important to pre-process CT images before being used as input for cancer detection algorithms. This Paper presents a robust technique for pre-processing CT images to remove any possible noise present in CT images. The evaluation metrics used for this technique are Mean Square Error (MSE), Peak Signal to Noise Ratio (PSNR) and Structural Similarity Index Metrics(SSIM). Results obtained are SSIM is .85, PSNR is 57.56 dB, and MSE is 0.13.

Registration of photoacoustic tomography vascular images: Comparison and analysis of automatic registration approaches

Image registration is crucial in the clinical application of photoacoustic tomography (PAT) for vascular growth monitoring. Aiming to find an optimized registration scheme for PAT vascular images acquired at different times and with varying imaging conditions, we compared and analyzed different commonly used intensity-based and feature-based automatic registration schemes. To further improve the registration performance, we proposed a new scheme that combines phase correlation with these commonly used intensity-based registration methods and compared their performances. The objective evaluation measures: peak signal-to-noise ratio (PSNR), structural similarity index metric (SSIM), root mean square error (RMSE), and quantitative visual perception (jump percentage P), as well as subjective evaluation using mean opinion score (MOS), were combined to evaluate the registration performance. Results show that the feature-based approaches in this study were not suitable for PAT image registration. And by adding phase correlation as rough registration, the overall registration performance was improved significantly. Among these methods, the proposed scheme of phase correlation combined with mean square error (MSE) similarity measure and regular-step-gradient-descent optimizer provides the best visual effect, accuracy, and efficiency in PAT vascular image registration.

Securing grayscale image using improved Arnold transform and ElGamal encryption

The security of sensitive data is critical, and it opens up a wide area of research to find efficient and effective methods to prevent unauthorized access. Our study provides a secure framework for sending visual information over an untrusted channel, such as a social networking site. The proposed framework is a combination of scrambling and encryption techniques. Initially, a hybrid block-wise and pixel-wise scrambling approach is administered to the grayscale image, followed by the Arnold transform, which causes all pixel points to move within the image. Finally, to improve the efficiency of the diffusion process, the asymmetric encryption ElGamal algorithm has been mastered. Peak-signal-to-noise ratio (PSNR), structural similarity index metric (SSIM), number of pixels change rate (NPCR), and unified average changing intensity (UACI) are the metrics utilized to evaluate the efficiency of the proposed scheme. The efficiency of the suggested scheme validates the low-average PSNR value of <9 dB and the SSIM average value of <0.01 for the encrypted images. The NPCR and UACI values achieved in our study are above the threshold values of 99.6% and 33.33%, respectively, exhibiting the strength of the proposed framework.

Impulse noise recuperation from grayscale and medical images using supervised curve fitting linear regression and mean filter

Acquisition of images from electronic devices or Transmission of the image through any medium will cause an additional commotion. This study aims to investigate a framework for eliminating impulse noise from grayscale and medical images by utilizing linear regression and a mean filter. Linear regression is a supervised machine learning algorithm that computes the value of a dependent variable based on an independent variable. The value of the recuperating pixel is measured using a curve-fitting, direction-based linear regression approach or applying a mean filter to the noise-free pixels. The efficiency of the proposed technique experiments with benchmark test images and the images of the USC-SIPI and TESTIMAGES data sets. Peak signal-to-noise ratio and structural similarity index metrics are determined to prove the performance of the proposed method. The evaluated results, when compared with the seven recent state-of-the-art techniques, show the superiority of the proposed method in terms of visual quality and accuracy. The proposed model achieves an average PSNR value of 65.21dB and an SSIM value of 0.999 for the reconstruction of medical images, proving its accuracy and efficiency. The impulse noise restoration process helps the radiologist get a clear visual clarity of the DICOM image for diagnosis purposes.

Closed-loop experimental optimization of tunable lenses.

Tunable lenses (TLs) are optical devices that can change their optical power in response to an electrical signal. In many applications, they are often pushed to or beyond their temporal limits. Fast periodic and/or abrupt variations of the optical power induce undesired distortions in their transient response and produce a decrease in their performance. A low-cost focimetry system, along with a custom closed-loop iterative optimization algorithm, was developed to (1) characterize a TL's response at high speed and (2) optimize their performance in realistic TL working conditions. A significant lens performance improvement was found in about 23 iterations with a decrease in the area under the error curve and an improved effective time. Applying the closed-loop optimization algorithm in a depth scanning experiment enhanced the image quality. Quantitatively, the image quality was evaluated using the structural similarity index metric that improves in individual frames, on average, from 0.345 to 0.895.

Cross-Sensor Fingerprint Enhancement Using Adversarial Learning and Edge Loss.

A fingerprint sensor interoperability problem, or a cross-sensor matching problem, occurs when one type of sensor is used for enrolment and a different type for matching. Fingerprints captured for the same person using various sensor technologies have various types of noises and artifacts. This problem motivated us to develop an algorithm that can enhance fingerprints captured using different types of sensors and touch technologies. Inspired by the success of deep learning in various computer vision tasks, we formulate this problem as an image-to-image transformation designed using a deep encoder–decoder model. It is trained using two learning frameworks, i.e., conventional learning and adversarial learning based on a conditional Generative Adversarial Network (cGAN) framework. Since different types of edges form the ridge patterns in fingerprints, we employed edge loss to train the model for effective fingerprint enhancement. The designed method was evaluated on fingerprints from two benchmark cross-sensor fingerprint datasets, i.e., MOLF and FingerPass. To assess the quality of enhanced fingerprints, we employed two standard metrics commonly used: NBIS Fingerprint Image Quality (NFIQ) and Structural Similarity Index Metric (SSIM). In addition, we proposed a metric named Fingerprint Quality Enhancement Index (FQEI) for comprehensive evaluation of fingerprint enhancement algorithms. Effective fingerprint quality enhancement results were achieved regardless of the sensor type used, where this issue was not investigated in the related literature before. The results indicate that the proposed method outperforms the state-of-the-art methods.

Deep-TOF-PET: Deep learning-guided generation of time-of-flight from non-TOF brain PET images in the image and projection domains.

We aim to synthesize brain time‐of‐flight (TOF) PET images/sinograms from their corresponding non‐TOF information in the image space (IS) and sinogram space (SS) to increase the signal‐to‐noise ratio (SNR) and contrast of abnormalities, and decrease the bias in tracer uptake quantification. One hundred forty clinical brain 18F‐FDG PET/CT scans were collected to generate TOF and non‐TOF sinograms. The TOF sinograms were split into seven time bins (0, ±1, ±2, ±3). The predicted TOF sinogram was reconstructed and the performance of both models (IS and SS) compared with reference TOF and non‐TOF. Wide‐ranging quantitative and statistical analysis metrics, including structural similarity index metric (SSIM), root mean square error (RMSE), as well as 28 radiomic features for 83 brain regions were extracted to evaluate the performance of the CycleGAN model. SSIM and RMSE of 0.99 ± 0.03, 0.98 ± 0.02 and 0.12 ± 0.09, 0.16 ± 0.04 were achieved for the generated TOF‐PET images in IS and SS, respectively. They were 0.97 ± 0.03 and 0.22 ± 0.12, respectively, for non‐TOF‐PET images. The Bland & Altman analysis revealed that the lowest tracer uptake value bias (−0.02%) and minimum variance (95% CI: −0.17%, +0.21%) were achieved for TOF‐PET images generated in IS. For malignant lesions, the contrast in the test dataset was enhanced from 3.22 ± 2.51 for non‐TOF to 3.34 ± 0.41 and 3.65 ± 3.10 for TOF PET in SS and IS, respectively. The implemented CycleGAN is capable of generating TOF from non‐TOF PET images to achieve better image quality.

Secure communication over wireless sensor network using image steganography with generative adversarial networks

Over the last several years, various researchers have been driven to the challenge of addressing security issues in Wireless Sensor Networks (WSN). Steganography is one of the most secure encryption techniques used for network security around the globe. Steganography is a secret communication technique used in wireless sensor networks where the aggregated data can be hidden as a message behind a cover image that appears normal across an untrusted network. Steganography using Generative Adversarial Networks (GAN) transforms the secret information to noise vector and feeds the noise vector to GAN to generate the image like containers or stegno image. Secret information is extracted from the stegno by updating noise vector with gradient descent operation. Though this method is secure against steganalysis attacks, the error introduced in transmission of images over lossy network introduces fluctuations in the image as result the data extracted has noises. Also anyone with the GAN decoder can access the secret information. There is no protection mechanism to ensure only a valid user is able to extract the secret information. This work proposed an improvised GAN Steganography addressing the two problems of security and reducing loss to secret information. The proposed approach is evaluated experimentally using Peak Signal to Noise Ratio (PSNR), Mean Square Error (MSE) and Structural Similarity Index Metric (SSIM), resulting in improved image quality and security that exhibiting the method's significance.

GAN-Based Encoding Model for Reversible Image Steganography

In carrying out reversible image steganography, the Generative Adversarial Networks (GANs-based) models have proven to be the most suitable deep learning models for image steganography. Image steganography is a steganography system that hides secret data in an image cover medium without arousing suspicion, and it is defined by the ability to reconstruct the cover medium with no visible distortion after the steganography system has been decoded by extracting the hidden data. In this study, we try achieve the encoding phase in image steganography, where two GAN-base models (CycleGAN and DCGAN) were proposed. Empirical analysis was done to determine a better model for the encoding of image steganography. The Peak Signal-to-Noise Ratio (PSNR), the Structural Similarity Index Metric (SSIM), and bit per pixel (bpp) were used as the metrics for the analysis. The outcome of DCGAN yielded (SSIM=0.48; PSNR=19.86; bpp=24.79) and the outcome of using CycleGAN yielded (SSIM=0.97; PSNR=41.45; bpp=24.97). These values concluded that the CycleGAN was preferable over the DCGAN. Hence, the CycleGAN was adopted as the encoding model.

Multi-layer noise reshaping and perceptual optimization for effective adversarial attack of images

Adversarial attack aims to fail the deep neural network by adding a small amount of perturbation to the input image, in which the attack success rate and resulting image quality are maximized under the lp norm perturbation constraint. However, the lp norm is not accurately correlated to human perception of image quality. Attack methods based on l0 norm constraint usually suffer from the high computational cost due to the iterative search for candidate pixels to modify. In this work, we explore how perceptual quality optimization can be incorporated into the adversarial attack design and propose a two-stage attack method to reshape the adversarial noise by an initial attack and optimize the visual quality of the attacked images without sacrificing the attack success rate. Specifically, we construct a visual attention network to generate a perceptual attention map to modulate the adversarial noise generated by a base attack method. The network is trained to maximize the visual quality in Structural Similarity Index Metric (SSIM) while achieving the same attack success rate. To improve the image perceptual quality further, we propose a fast search algorithm to perform an iterative block-wise pruning of the adversarial noise. We evaluate our method on the mini-ImageNet dataset against three different defense schemes. The results have demonstrated that our method can achieve better attack performance in image quality, attack success rate, and efficiency than the state-of-the-art attack methods.