Image Quality Assessment Parameters Research Articles

Design and Implementation of a Self-Supervised Algorithm for Vein Structural Patterns Analysis Using Advanced Unsupervised Techniques

Compared to other identity verification systems applications, vein patterns have the lowest potential for being used fraudulently. The present research examines the practicability of gathering vascular data from NIR images of veins. In this study, we propose a self-supervision learning algorithm that envisions an automated process to retrieve vascular patterns computationally using unsupervised approaches. This new self-learning algorithm sorts the vascular patterns into clusters and then uses 2D image data to recuperate the extracted vascular patterns linked to NIR templates. Our work incorporates multi-scale filtering followed by multi-scale feature extraction, recognition, identification, and matching. We design the ORC, GPO, and RDM algorithms with these inclusions and finally develop the vascular pattern mining model to visualize the computational retrieval of vascular patterns from NIR imageries. As a result, the developed self-supervised learning algorithm shows a 96.7% accuracy rate utilizing appropriate image quality assessment parameters. In our work, we also contend that we provide strategies that are both theoretically sound and practically efficient for concerns such as how many clusters should be used for specific tasks, which clustering technique should be used, how to set the threshold for single linkage algorithms, and how much data should be excluded as outliers. Consequently, we aim to circumvent Kleinberg’s impossibility while attaining significant clustering to develop a self-supervised learning algorithm using unsupervised methodologies.

A method of image fusion using texture guidance based guided image filter and image statistics

A technique of image fusion by means of a guided image filter is proposed in this paper. It employs image smoothing with a guided filter by making use of texture information as guidance for the guided filter. Then, image statistics-based pixel weight computation is utilized for generating the weight maps of the source images from detail layer features. Finally, the source images are integrated based on the weighted average combining strategy. The efficacy of the proposed algorithm is tested and compared with several state-of-the-art image fusion methods in terms of several objective image quality assessment parameters. The experimental result suggests the efficacy of the proposed algorithm in image fusion.

Analysis of dual-energy mammography subtraction technique for the dose and image quality evaluation using 3D-printed breast phantom.

This study aimed to quantitatively assess the radiation dose using XR-QA2 and the image quality of the dual-energy subtraction mammography technique on an in-house phantom. The analysis was carried out to investigate the effect of targets/filters on dose value and image quality using an in-house phantom made of PLA + as an object representing compressed breasts. All irradiation parameters were performed in the craniocaudal position with manual mode. Mean glandular dose (MGD) was recorded, followed by the calculation of the signal-to-noise ratio (SNR), contrast-to-noise ratio (CNR), and modulation transfer function (MTF) for image quality assessment parameters. The results showed that the image quality was accepted at dose levels within the IAEA and BAPETEN tolerance limit for 60mm equivalent compressed breast using dual-energy mammography. Furthermore, the target/filter (W/Rh) reduced the dose by 1.03mGy compared to the Mo/Mo and Mo/Rh with an enhancement in image quality. This indicated that the target/filter (W/Rh) combination was optimal due to the image quality improvement obtained with lower MGD.

Segmentation of Microscopy images using Multi-Scale Retinex with Chromacity Preservation and Otsu Thresholding

Bacteria play a significant role in our environment by being helpful or harmful; hence, it is crucial to identify the various bacterial species. The microscopic image captured by camera with microscope is not reliable due to the poor quality of image, making bacterial counting a difficult and time-consuming task. This paper proposes improved and enhanced Multi-Scale Retinex with Chromacity Preservation and Otsu Thresholding techniques for increasing the quality of images of bacterial cells for segmentation and contrast enhancement. A combinative procedure of image enhancement and segmentation is illustrated in this paper. The parameters for Image Quality Assessment (IQA) used are Enhancement Measure Estimation and Standard Deviation of the upgraded images. The proposed approach gives better segmentation results, as proven by the incremental changes in the IQA parameters.

Segmentation of Microscopy images using Multi-Scale Retinex with Chromacity Preservation and Otsu Thresholding

Bacteria play a significant role in our environment by being helpful or harmful; hence, it is crucial to identify the various bacterial species. The microscopic image captured by camera with microscope is not reliable due to the poor quality of image, making bacterial counting a difficult and time-consuming task. This paper proposes improved and enhanced Multi-Scale Retinex with Chromacity Preservation and Otsu Thresholding techniques for increasing the quality of images of bacterial cells for segmentation and contrast enhancement. A combinative procedure of image enhancement and segmentation is illustrated in this paper. The parameters for Image Quality Assessment (IQA) used are Enhancement Measure Estimation and Standard Deviation of the upgraded images. The proposed approach gives better segmentation results, as proven by the incremental changes in the IQA parameters.

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.

Image Retrieval Using Digital Image Inpainting Techniques

Image retrieval is an inverse problem in digital image processing. In this paper, the authors deal with restoration of image using digitally image inpainting methods. In this inpainting technique, one can extract a missing an important part or can fill unwanted parts of an image, so as to recover original information without loss. They are using semi-automatic and fast image inpainting methods and compared their result in terms of image quality assessment parameters such as PSNR, MSE, as well as required timing constraints. The observational as well as statistical results show that the fast image inpainting technique is superior in all respects for any kinds of multimedia image application.

Face Anti-Spoofing System using Motion and Similarity Feature Elimination under Spoof Attacks

From border control to mobile device unlocking applications, the practical utility of biometric system can be seriously compromised due face spoofing attacks. So, face recognition systems require greater attention to combating face spoofing attacks. As, face spoofing attacks can be easily propelled through 3D masks, video replays, and printed photos so we are presented face recognition system using motion and similarity features elimination under spoof attacks against the Replay Attack and Institute of Automation, Chinese Academy of Sciences (CASIA) databases. In this paper a calculative analysis has been done by firstly segmenting the foreground and background regions from the input video using Gaussion Mixture Model and secondly by extracting features i.e., face, eye, and nose and applied 26 image quality assessment parameters on spoof face videos under different illumination lighting conditions. The results attained using Replay Attack and CASIA databases are extremely competitive in discriminating from fake traits with paralleled viz-a-viz other approaches. Different machine learning classifiers and their comparative analysis with existing approaches has been shown.

An Integrated Double Hybrid Fusion Approach for Image Smoothing

Often in practice, during the process of image acquisition, the acquired image gets degraded due to various factors like noise, motion blur, mis-focus of a camera, atmospheric turbulence, etc. resulting in the image unsuitable for further analysis or processing. To improve the quality of these degraded images, a double hybrid restoration filter is proposed on the two same sets of input images and the output images are fused to get a unified filter in combination with the concept of image fusion. First image set is processed by applying deconvolution using Wiener Filter (DWF) twice and decomposing the output image using Discrete Wavelet Transform (DWT). Similarly, second image set is also processed simultaneously by applying Deconvolution using Lucy–Richardson Filter (DLR) twice followed by the above procedure. The proposed filter gives a better performance as compared to DWF and DLR filters in case of both blurry as well as noisy images. The proposed filter is compared with some standard deconvolution algorithms and also some state-of-the-art restoration filters with the help of seven image quality assessment parameters. Simulation results prove the success of the proposed algorithm and at the same time, visual and quantitative results are very impressive.

Restoration and quality improvement of distorted tribal artworks using Particle Swarm Optimization (PSO) technique along with nonlinear filtering

Cultural artworks that are not properly cared for, degrades over time. However modern image processing techniques proves to be a boon for such artworks. Various techniques have been developed time to time for different aspects of image processing like noise removal, improving brightness and image quality, etc. One of such techniques is PSO (Particle Swarm Optimization) that can be used for image restoration. In this paper a combination of median filter and PSO has been used for the quality improvement of the distorted tribal artworks and the results are compared with few HE (Histogram Equalization) techniques like the CLAHE, BBHE and DSIHE. The comparison is done on the basis of various image quality assessment parameters, namely PSNR, AMBE, NMSE, CPP and MSE.

An efficient face anti-spoofing and detection model using image quality assessment parameters

Biometric authentication poses a significant problem as reconstructed sample or fake self-manufactured samples used by intruders for accessing the actual real legitimate traits. The other prime concern for biometrics is the increasing demand for safety in mobile devices, such as smartphones and tablets etc. So, in the present scenario security for biometrics has gained considerable attention due to various inherent qualities of biometrics. For detection of valid user in a face recognition system with photographs, videos, and 3D models, face liveness detection system is a great technique against spoofing attacks for differentiating between the fake traits from the real traits. In this paper, a novel fake biometric detection technique utilizing liveness detection is proposed for detecting deceitful access attempts in the biometric face system. The prime objective of the paper is to propose a low-complexity fake biometric detection using different image quality assessment parameters i.e. Mean Square Error, Signal to Noise Ratio,SC etc. on the extracted features of the images. The authenticity of the proposed model is confirmed by analyzing the values of MSE, which are 5.8% and 8.49% more than the threshold value of nose and eye features. The same results have also been shown for other 11 different image quality assessment parameters. The experiments were done on the database prepared using the image samples of the 500 male and female students having age between 20 to 30 years.

A Novel Coverless Information Hiding Method Based on the Most Significant Bit of the Cover Image

With the rapid development of information technology recently, information security has become the focus of public concern. Information Hiding (IH) technology is an effective method to tackle the problem of information leakage incidents. In this paper, a novel coverless information hiding method based on the Most Significant Bit (MSB) of cover image is proposed (CIHMSB). Firstly, the cover image is segmented into a number of fragments. Secondly, in order to use the MSB of cover image to represent the secret information, the average intensity of each fragment is calculated. Thirdly, a one-to-one mapping between the MSB of the image fragment and the secret information is established using the mapping sequence (denote as $Km$ ), decided by the sender and the receiver in advance. This process produces a mapping flag (denote as $Kf$ ), which is sent by the sender along with the stego image. The objective of the proposed work is to increase hiding capacity, curtail the distortion of the stego image to improve its quality and reduce the Bit Error Rate ( $BER$ ) of stego image in the case of distortion. Experimental results show that the proposed method can conceal 2601 bits secret information per carrier with peak signal-to-noise ratio ( $PSNR$ ) of $\infty $ dB. What’s more, some stego image quality assessment parameters, such as structural similarity ( $SSIM$ ) index and universal image quality index ( $Qi$ ), are slightly better than existing information hiding methods. Furthermore, the proposed method has good performance against such as AGWN, salt & pepper noise, low-pass filtering and JPEG compression attacks.

Performance Analysis of Various Filters for De-Speckling of Thyroid Ultrasound Images

Thyroid ultrasonography is the most common and extremely useful, safe, and cost effective way to image the thyroid gland and its pathology. However, an inherent characteristic of Ultrasound (US) imaging is the presence of multiplicative speckle noise. Speckle noise reduces the ability of an observer to distinguish fine details, make diagnosis more difficult. It limits the effective implementation of image analysis steps such as edge detection, segmentation and classification. The main objective of this study is to compare the performance of various spatial and frequency domain filters so as to identify efficient and optimum filter for de-speckling Thyroid US images. The performance of these filters is evaluated using the image quality assessment parameters Signal to Noise Ratio (SNR), Peak Signal to Noise Ratio (PSNR), Structural Similarity Index (SSIM), Mean Square Error (MSE) and Root Mean Square Error (RMSE) for different speckle variance. Experimental work revealed that kuan filter resulted in higher PSNR, SNR, SSIM and least MSE, RMSE values compared to other filters

Reversible Image Steganography Using Dual-Layer LSB Matching

Recently, reversible information hiding (RIH) methods have drawn substantial attention in many privacy-sensitive real-time applications, such as the Internet of Things (IoT) enabled communications, electronic health care infrastructure, and military applications. The RIH methods are proven to be competent in such hyper-sensitive infrastructures where the loss of a single bit of information is not acceptable. In this paper, dual-layered based RIH method using modified least significant bit (LSB) matching has been proposed. The objective of the proposed work is to enhance the embedding efficiency (EE) using dual-layer based embedding strategy and to curtail the distortion caused to the stego-image to improve its quality. At the first layer of embedding, each pixel conceals two bits of information using the proposed modified LSB matching method to produce the intermediate pixel pair (IPP). Further, the IPP is utilized to conceal four bits of information during the next layer of embedding. Experimental study reveals that, the proposed method can embed 1,572,864 bits of secret data with peak signal-to-noise ratio (PSNR) of 47.86 dB, 48.05 dB, 46.51 dB and 48.14 dB, for the respective images. Further, the image quality assessment parameters like structural similarity (SSIM) index and universal image quality index (Q) are as good as the existing literature. Additionally, the proposed method shows excellent anti-steganalysis ability to regular and singular (RS) and pixel difference histogram (PDH) analysis.

Performance Analysis of Quality Measurement for Biomedical Images

With the rise used of biomedical images for diagnosis of diseases of a patient at various hospitals, the quality measurement of images is necessary for better diagnosis. The need for biomedical images quality measurement due to the transmission of images from one place to another place within hospitals or remote health centers. The lots of distortions may affect the quality of biomedical images during transmission. So, in this paper, the various image quality assessment parameters discussed for biomedical images. The paper gives information on various full reference-based objective assessment (OA) image quality measurement (FR-IQM) standards. The paper also gives performance analysis of FR-IQM standards for biomedical images.

DIGITAL MAMMOGRAM ENHANCEMENT USING KAPUR MEASURE OF ENTROPY AND MATHEMATICAL MORPHOLOGY

Mammography is the most reliable, effective, low cost and highly sensitive method for early detection of breast cancer. Mammogram analysis usually refers to the processing of mammograms with the goal of finding abnormality presented in the mammogram. Mammogram enhancement is one of the most critical tasks in automatic mammogram image analysis. Main purpose of mammogram enhancement is to enhance the contrast of details and subtle features while suppressing the background heavily. In this paper, a hybrid approach is proposed to enhance the contrast of microcalcifications while suppressing the background heavily, using fuzzy logic and mathematical morphology. First, mammogram is fuzzified using Gaussian fuzzy membership function whose bandwidth is computed using Kapur measure of entropy. After this, mathematical morphology is applied on fuzzified mammogram. Mathematical morphology provides tools for the extraction of microcalcifications even if the microcalcifications are located on a nonuniform background. Main advantage of Kapur measure of entropy over Shannon entropy is that Kapur measure of entropy has α and β parameters that can be used as adjustable values. These parameters can play an important role as tuning parameters in the image processing chain for the same class of images. Experiments have been conducted on images of mini-Mammogram Image Analysis Society (MIAS) database (UK). Experiment results of the proposed approach are compared with histogram equalization (HE), contrast limited adaptive histogram equalization (CLAHE) and fuzzy histogram hyperbolization (FHH) which are well-established image enhancement techniques. In order to validate the results, several different kinds of standard test images (fatty, fatty-glandular and dense-glandular) of mini-MIAS database are considered. Objective image quality assessment parameters: Target-to-background contrast enhancement measurement based on standard deviation (TBCSD), target-to-background contrast enhancement measurement based on entropy (TBCE), contrast improvement index (CII), peak signal-to-noise ratio (PSNR) and average signal-to-noise ratio (ASNR) are used to evaluate the performance of proposed approach. The experimental results show that the proposed approach performs well. This study can be a part of developing a computer-aided diagnosis (CAD) system for early detection of breast cancer.

Optimization of Aberrated Coherent Optical Systems

The images of a straight edge in coherent illumination, produced by an optical system with circular aperture and apodised with amplitude filters have been studied. The image quality assessment parameters such as edge-ringing, edgegradient and edge-shift of the edge fringes have been studied as a function of apodisation parameter for various degrees of defocus, Coma and primary spherical aberrations. It is found that, at certain combinations of aberrations the quality of the image of straight edge objects can be improved.

Multiple distortion pooling image quality assessment

This paper presents a quality assessment method which pools some basic image quality assessment parameters and empirically combines them in such a manner to evaluate across different distortion types. The proposed quality metric (Q) is formulated by modelling an image distortion as a combined effect of structural distortion, contrast distortion and edge distortion, which may occur on account of deteriorations due to noise contamination, contrast manipulations, blurring, rotation or compression. The values of the correlation coefficient prove that this metric provides an accurate estimation for the above mentioned distortions in comparison to mean squared error (MSE) and structural similarity measure (SSIM). Results of subjective evaluation also validate the efficiency of the proposed quality assessment method and its ability to quantify the effect of different distortions under a single quality metric.

Integration of Fuzzy and Wavelet Approaches towards Mammogram Contrast Enhancement

Mammography is the most reliable, effective, low cost and high sensitive method for early detection of breast cancer. In breast cancer diagnosis, it is difficult for radiologists to detect the typical diagnostic signs because mammograms are low contrast and noisy images. In order to improve diagnosis accuracy, mammogram contrast enhancement technique is often used to enhance the contrast of mammogram and aid the radiologists. In this paper, a combined approach with fuzzy and wavelet towards improved enhancement of details and subtle features in digital mammogram images. The proposed contrast enhancement approach utilizes wavelet transform to decompose the mammogram, and then approximation coefficients are modified by fuzzy contrast enhancement approach and detail coefficients are by non-linear transformation. After that, inverse wavelet transform is applied on modified coefficients to obtain the enhanced mammogram. The proposed algorithm has been tested on a number of images in Digital Database for Screening Mammography (DDSM), comparing the results with histogram equalization which is a well-established image enhancement technique and Cheng’s enhancement algorithm based on wavelet transform. In order to accurately assess the proposed approach, enhanced mammogram images are evaluated through objective image quality assessment parameters: information entropy, contrast and peak signal-to-noise ratio (PSNR). Experimental results of the proposed approach are quite promising.

Use of a quality index in threshold contrast detail detection measurements in television fluoroscopy

The use of a single index to assist in quality control procedures of X-ray television fluoroscopy systems was investigated. A single quality index was devised incorporating a measure of threshold contrast detail detectability (TCDD) performance and taking into account image intensifier input kerma rate, field size, differences in radiation beam quality, and pulsed fluoroscopy. This was applied to a number of clinical systems to investigate changes in image quality index quantified over time. Accepted measurement protocols were used to obtain these measurements. The results show system performance for different systems and can establish the decline in performance parameters over time or assess non-optimal image quality with clinical systems in field measurements. The systems studied were assessed with a variety of performance parameters including TCDD results, low contrast sensitivity, limiting resolution, and image intensifier input kerma rate under clinical modes of operation. The TCDD quality index, and dose normalized quality index, were found to be useful image quality assessment parameters for serial testing of systems, which augment the use of graphical methods for the display of TCDD curves.