Structural Similarity Algorithm Research Articles

Surface‐enhanced Raman scattering spatial fingerprinting decodes the digestion behavior of lysosomes in live single cells

AbstractLysosome, the digestive organelle in eukaryotic cells, plays an important role in the degradation and recirculation of cellular products as well as in maintaining the stability of cellular metabolic microenvironment. Surface‐enhanced Raman scattering (SERS) is a molecular fingerprint technology with high detection sensitivity and photostability, suited for revealing various intracellular molecular information by inducing endocytosis of SERS‐active nanoparticles. However, it remains challenging to selectively extract the molecular information of specific organelles (e.g., lysosomes) from a high‐dimensional spectral set. Herein, we proposed a novel paradigm by combining label‐free SERS spectroscopy with confocal fluorescence imaging to investigate the digestion behavior of lysosomes in cells. The structural similarity algorithm was innovatively introduced and exhibited its effectiveness in screening out the wavenumbers in the SERS spectral set with high correlation with the metabolic behaviors of lysosomes. With comprehensive experiments on HeLa single cells, we captured the intracellular macromolecular digestion phenomenon and discovered the changing pattern of cellular SERS spectra after starvation‐induced autophagy, and analyzed the molecular information within the lysosomes in three‐dimensional space.

Bubble nucleation and jetting inside a millimetric droplet

In this work we present experiments and simulations on the nucleation and successive dynamics of laser-induced bubbles inside liquid droplets in free-fall motion, i.e. a case where the bubbles are subjected to the influence of a free boundary in all directions. Within this spherical millimetric droplet, we have investigated the nucleation of secondary bubbles induced by the rarefaction wave that is produced when the shock wave emitted by the laser-induced plasma reflects at the drop surface. Interestingly, three-dimensional clusters of cavitation bubbles are observed. Their shape is compared with the negative pressure distribution computed with a computational fluid dynamics model and allows us to estimate a cavitation threshold value. In particular, we observed that the focusing of the waves in the vicinity of the free surface can give rise to explosive cavitation events that end up in fast liquid ejections. High-speed recordings of the drop/bubble dynamics are complemented by the velocity and pressure fields simulated for the same initial conditions. The effect of the proximity of a curved free surface on the jetting dynamics of the bubbles was qualitatively assessed by classifying the cavitation events using a non-dimensional stand-off parameter ${\Upsilon\hskip -1,05em -\,}$ that depends on the drop size, the bubble maximum radius and the relative position of the bubble inside the drop. Additionally, we studied the role of the drop's curvature by implementing a structural similarity algorithm to compare cases with bubbles produced near a flat surface to the bubbles inside the drop. Interestingly, this quantitative comparison method indicated the existence of equivalent stand-off distances at which bubbles influenced by different boundaries behave in a very similar way. The oscillation of the laser-induced bubbles promotes the onset of Rayleigh–Taylor and Rayleigh–Plateau instabilities, observed on the drop's surface. This phenomenon was studied by varying the ratio of the maximum radii of the bubble and the drop. The specific mechanisms leading to the destabilisation of the droplet surface were identified.

Deep learning-based method for detecting anomalies in electromagnetic environment situation

The anomaly detection of electromagnetic environment situation (EMES) has essential reference value for electromagnetic equipment behavior cognition and battlefield threat assessment. In this paper, we proposed a deep learning-based method for detecting anomalies in EMES to address the problem of relatively low efficiency of electromagnetic environment situation anomaly detection (EMES-AD). Firstly, the convolutional kernel extracts the static features of different regions of the EMES. Secondly, the dynamic features of the region are obtained by using a recurrent neural network (LSTM). Thirdly, the Spatio-temporal features of the region are recovered by using a de-convolutional network and then fused to predict the EMES. The structural similarity algorithm (SSIM) is used to determine whether it is anomalous. We developed the detection framework, de-signed the network parameters, simulated the data sets containing different anomalous types of EMES, and carried out the detection experiments. The experimental results show that the proposed method is effective.

Quality evaluation method of agricultural product packaging image based on structural similarity and MTF

With the rapid development of multimedia and communication technologies, image technology has become more and more widely used in agriculture. Aiming at the advantages of structural similarity and MTF combined algorithm, this paper proposes an image quality evaluation algorithm based on structural similarity and MTF, which mainly adopts improved gamma correction method, dual homomorphic filter correction method and illumination-independent graph method. The pre-processed image was subjected to the elimination of illumination treatment, and the experimental results of the above three methods were compared, and the experimental results of the images not subjected to illumination-independent processing were compared and analyzed. The experimental results show that the evaluation indexes of SSIM algorithm are better than other evaluation algorithms in general. The distortion in JPEG2000 image is mainly caused by the ringing effect at the edge of the image caused by wavelet transform and the blurring caused by quantization. The SSIM algorithm not only considers the edge information of the image, but also develops an illumination-invariant image recognition system, so it has obtained good evaluation results. The MATLAB research platform is used to analyze the image quality evaluation of agricultural products packaging. Through the analysis of subjective and objective evaluation results, the algorithm that combines structural similarity and MTF is superior to other algorithms.

Anomaly Detection of Pantograph Based on Salient Segmentation and Generative Adversarial Networks

The pantograph is an important component of railway pantograph-catenary system, which can provide electric current for electrified railway electric locomotive. Since the pantograph is in an open roof environment, the defects of the pantograph are inevitable in the long run. In order to ensure the safe operation of trains, in this paper, we propose a new method for anomaly detection of pantograph based on salient segmentation and generative adversarial networks. First, an object location model is trained by U-Net which perform excellent properties for a small number of samples to accurately extract salient area of the pantograph. Second, a generative adversarial model is constructed to generate the reconstructed salient images of pantograph by vector mapping. Finally, the structural similarity algorithm is used to evaluate the similarity between the input salient image and the reconstructed salient image, so as to extract the image difference and realize the anomaly detection of the pantograph. Experimental results validate the effectiveness and accuracy of our approach.

Gap analysis business process model by using structural similarity

<span>Gap analysis process model is a study that can help an institution to determine differences between business process models, such as a model of Standard Operating Procedure and a model of activities in an event log. Gap analysis is used for finding incomplete processes and can be obtained by using structural similarity. Structural similarity measures the similarity of activities and relationships depicting in the models. This research introduces a graph-matching algorithm as the structural similarity algorithm and compares it with dice coefficient algorithms. Graph-matching algorithm notices parallel relationships and invisible tasks, on the contrary dice coefficient algorithms only measure closeness between activities and relationships. The evaluation shows that the graph-matching algorithm produces 76.76 percent similarity between an SOP model and a process model generating from an event log; while, dice coefficient algorithms produces 70 percent similarity. The ability in detecting parallel relationships and invisible tasks causes the graph-matching algorithm produces a higher similarity value than dice coefficient algorithms.</span>

Soil moisture assessment through the SSMMI and GSSIM algorithm based on SPOT, WorldView-2, and Sentinel-2 images in the Daliuta Coal Mining Area, China.

A set of indicators that focus only on numerical values is constructed based on remotely sensed images to assess soil moisture conditions. The quantitative evaluation of soil moisture variation in two periods is rarely referred to in the current literature. In this study, a scaled soil moisture monitoring index (SSMMI) was established to monitor the soil moisture status during 2010-2018 in the Daliuta Coal Mining Area (DCMA), China, based on SPOT-5, SPOT-6, and Sentinel-2 images. We also employed a gradient-based structural similarity (GSSIM) algorithm to quantitatively analyze the characteristics of the spatial distribution of the soil moisture in the DCMA. The optimal scale for exploring the spatial heterogeneity of the soil moisture was determined by local variance and semivariance methods. The results showed that the soil moisture decreased at a rate of 0.0213/a from 2010 to 2018. The areas with the extremely dry and dry levels, which were mainly located in the northwest, some regions of the central area, and the southeast of the DCMA, decreased from 14.48% in 2010 to 13.66% in 2018. The proportion of the no dry level was improved by 14.62%, while the area of the extremely wet and wet levels decreased by 13.79%. The mean value of the soil moisture in the unmined area was greater than that in the DCMA, which was larger than that in the mined area. The result of the GSSIM analysis indicated that the area of dramatic change, where the soil moisture changed substantially, was chiefly distributed in the north, west, some central regions, and some parts of the south and east of the DCMA. The region where the substantial change occurred was surrounded by a moderate-change area, which was encompassed by a low-change area. The area with dramatic and moderate decreases in the soil moisture accounted for 64.52% of the region, which was greater than that with incremental soil moisture changes, which accounted for 5.85% of the region. The area also showed decreased soil moisture from 2010 to 2018. Soil moisture changes are closely related to variations in land cover. For instance, vegetative cover over an open-pit mining area can cause a dramatic reduction in soil moisture. Ninety-three meters was the optimal scale used for monitoring the soil moisture in the DCMA, which indicates that we can adopt the SPOT-5, SPOT-6, and Sentinel-2 images to evaluate the soil moisture conditions in the DCMA.

Fast Implementation of Image Structural Similarity Algorithm

The traditional Structural SIMilarity (SSIM) index has higher computation complexity compared with Peak-Signal-to-Noise-Ratio (PSNR). In order to solve this problem, we propose a two-stage optimization algorithm to reduce the computation complexity of SSIM. First, floating-point operations in the calculation process of SSIM are converted to fixed-point ones. Then, the repeated calculation rule of the iteration operation in SSIM is analyzed and the repeated operations are subsequently removed via modifying SSIM calculation method. Experiment results show that, in comparison with the original SSIM algorithm, the proposed algorithm obtains 24.12% of computation time reduction with negligible accuracy decrease.

Content-Aware Retargeted Image Quality Assessment

In targeting the low correlation between existing image scaling quality assessment methods and subjective awareness, a content-aware retargeted image quality assessment algorithm is proposed, which is based on the structural similarity index. In this paper, a similarity index, that is, a local structural similarity algorithm, which can measure different sizes of the same image is proposed. The Speed Up Robust Feature (SURF) algorithm is used to extract the local structural similarity and the image content loss degree. The significant area ratio is calculated by extracting the saliency region and the retargeted image quality assessment function is obtained by linear fusion. In the CUHK image database and the MIT RetargetMe database, compared with four representative assessment algorithms and other latest four kinds of retargeted image quality assessment algorithms, the experiment proves that the proposed algorithm has a higher correlation with Mean Opinion Score (MOS) values and corresponds with the result of human subjective assessment.

Image matching using structural similarity and geometric constraint approaches on remote sensing images

Image matching has been a central issue in the field of computer vision and image processing for decades. It is normally based on the maximum similarity between two given images. We propose a similarity measure criterion based on structural similarity (SSIM) for image matching. We use the similarity measure criterion to compute the similarity of the feature points of two images to obtain the matching points. The results show that the correct matching rate of the proposed similarity measure criterion is higher than that of the normalized correlation coefficient. As for the mismatches in the initial set of matches, we use the proposed algorithm to compute the mean structural similarity (MSSIM) index of their neighborhood window image and eliminate those whose values of the MSSIM index are below the threshold. Then we use the geometric distribution of corresponding points in the image space to eliminate the mismatches that are difficult to eliminate using the SSIM algorithm. The experimental result shows that the proposed algorithm is superior to the random sample consensus algorithm in terms of mismatch detection and computational time.

Reduced respiratory motion artifacts using structural similarity in fast 2D dynamic contrast enhanced MRI of liver lesions.

The purpose of this work was to improve dynamic contrast enhanced MRI (DCE-MRI) of liver lesions by removing motion corrupted images as identified by a structural similarity (SSIM) algorithm, and to assess the effect of this correction on the pharmacokinetic parameter Ktrans using automatically determined arterial input functions (AIFs). Fifteen patients with colorectal liver metastases were measured twice with a T1 weighted multislice 2D FLASH sequence for DCE-MRI (time resolution 1.2s). AIFs were automatically derived from contrast inflow in the aorta of each patient. Thereafter, SSIM identified motion corrupted images of the liver were removed from the DCE dataset. From this corrected data set Ktrans and its reproducibility were determined. Using the SSIM algorithm a median fraction of 46% (range 37-50%) of the liver images in DCE time series was labeled as motion distorted. Rejection of these images resulted in a significantly lower median Ktrans (p<0.05) and lower coefficient of repeatability of Ktrans in liver metastases compared with an analysis without correction. SSIM correction improves the reproducibility of the DCE-MRI parameter Ktrans in liver metastasis and reduces contamination of Ktrans values of lesions by that of surrounding normal liver tissue.

A Study on Determination of an Optimized Detector for Single Photon Emission Computed Tomography.

The detector is a critical component of the single photon emission computed tomography (SPECT) imaging system for giving accurate information from the exact pattern of radionuclide distribution in the target organ. The SIMIND Monte Carlo program was utilized for the simulation of a Siemen's dual head variable angle SPECT imaging system with a low energy high resolution (LEHR) collimator. The Planar and SPECT scans for a 99mTc point source and a Jaszczak Phantom with the both experiment and simulated systems were prepared and after verification and validation of the simulated system, the similar scans of the phantoms were compared (from the point of view of the images’ quality), namely, the simulated system with the detectors including bismuth germanate (BGO), yttrium aluminum garnet (YAG:Ce), Cerium-doped yttrium aluminum garnet (YAG:Ce), yttrium aluminum perovslite (YAP:Ce), lutetium aluminum garnet (LuAG:Ce), cerium activated lanthanum bromide (LaBr3), cadmium zinc telluride (CZT), and sodium iodide activated with thallium [NaI(Tl)]. The parameters of full width at half maximum (FWHM), energy and special resolution, sensitivity, and also the comparison of images’ quality by the structural similarity (SSIM) algorithm with the Zhou Wang and Rouse/Hemami methods were analyzed. FWHMs for the crystals were calculated at 13.895, 14.321, 14.310, 14.322, 14.184, and 14.312 keV and the related energy resolutions obtained 9.854, 10.229, 10.221, 10.230, 10.131, and 10.223 %, respectively. Finally, SSIM indexes for comparison of the phantom images were calculated at 0.22172, 0.16326, 0.18135, 0.17301, 0.18412, and 0.20433 as compared to NaI(Tl). The results showed that BGO and LuAG: Ce crystals have high sensitivity and resolution, and better image quality as compared to other scintillation crystals.

A New Image Quality Assessment Algorithm based on SSIM and Multiple Regressions

Image quality assessment (IQA) is crucial in image processing algorithms. In the state-of-the-art IQA index, the structural similarity (SSIM) index has been proved to be better objective quality assessment metric. However, the accuracy of SSIM is relatively lacking when used to access blurred images. And the component weights of structural similarity (SSIM) index are fixed in some past environments. So an improved assessment algorithm incorporating multiple linear regressions and SSIM index was proposed in this paper. We use regression analysis to adjust the component weight of SSIM index. So the improved algorithm is more accuracy on different distortion types’ quality assessment. Experimental results show that the improved SSIM algorithm is better than traditional methods in nonlinear regression correlation coefficient, Spearman correlation coefficient and out ratio.

K-Neighbourhood Structural Similarity Approach for Spatial Clustering

Background/Objectives: The scope and the diversity of contexts in which graphs are used in the field of network analysis have created scope for interdisciplinary research. It has proved to be an efficient structure to understand the features, the topological structure and the dynamics of many complex systems. Methods/Statistical Analysis: This paper proposes a graph clustering method on a directed weighted graph network for detecting communities in social network, based on neighbourhood nodes and the frequency of the path traversed. The algorithm is evaluated with small, medium and large sized data. The algorithm is experimented with the data and its performance is evaluated in terms of cluster quality. Findings: Our results represent a novel method for identifying quality clusters with high intra-similarity and low intersimilarity. The significance of the algorithm is it automatically generates the parameters needed for clustering and also the number of clusters to be formed from the dataset. Experimental results obtained on the mobility and citation dataset prove the efficiency of k-Neighbourhood Structural Similarity algorithm. Application/Improvements: Clustering, a widely used data mining technique finds its application in areas of healthcare, banking, finance and decision making. This algorithm considers the spatial aspect and as a future enhancement we would analyze the influence of temporal factor in forming good clusters.

Quantitative assessment of laser-dazzling effects through wavelet-weighted multi-scale SSIM measurements

Abstract Laser active imaging systems are widespread tools used in region surveillance and threat identification. However, the photoelectric imaging detector in the imaging systems is easy to be disturbed and this leads to errors of the recognition and even the missing of the target. In this paper, a novel wavelet-weighted multi-scale structural similarity (WWMS-SSIM) algorithm is proposed. 2-D four-level wavelet decomposition is performed for the original and disturbed images. Each image can be partitioned into one low-frequency subband (LL) and a series of octave high-frequency subbands (HL, LH and HH). Luminance, contrast and structure comparison are computed in different subbands with different weighting factors. Based on the results of the above, we can construct a modified WWMS-SSIM. Cross-distorted image quality assessment experiments show that the WWMS-SSIM algorithm is more suitable for the subjective visual feeling comparing with NMSE and SSIM. In the laser-dazzling image quality assessment experiments, the WWMS-SSIM gives more reasonable evaluations to the images with different power and laser spot positions, which can be useful to give the guidance of the laser active imaging system defense and application.

Imaging through a highly scattering medium with structural similarity and genetic algorithm

A method to recover the image of an object behind a highly scattering medium with higher accuracy is presented. Instead of the Pearson correlation coefficient (PCC) used in the existing methods, structural similarity (SSIM), which is known as an excellent evaluation indicator of image quality, is employed as the cost function for the wavefront optimization. Compared to PCC, better imaging quality can be acquired with SSIM, because the latter comprehensively analyzes the luminance, the contrast, and the structure of imaging results. By comparing the performances of the three commonly used global optimization algorithms, including a genetic algorithm (GA), particle swarm optimization and differential evolution algorithm, we verify that GA has the best reliability and stability to solve this multidimensional wavefront modulation problem among these global optimization algorithms, including in strong noise environments. This work can improve the quality of imaging through a highly scattering medium with a wavefront optimization technique and can be applied to the fields of optical detection or biomedical imaging.

Structural similarity classifier for facial expression recognition

A novel Gabor filter structural similarity algorithm (GFSSIM) is proposed for facial expression recognition (FER) on noisy images. Low-resolution facial images with low SNRs are specifically dealt with FER system. The features are extracted using 40 Gabor filters, and a feature subset is selected for classification. The test image is classified based on proposed GFSSIM algorithm. The experimental results show that the recognition rate for heavily deteriorated images outperforms the conventional classifier method. In addition, the proposed method is more efficient from the computational complexity point of view.

Automated detecting and removing cloud shadows in full-disk solar images

Sky clouds affect full-disk solar observations significantly. Their shadows obscure the details of solar features in observed images. Cloud-covered images are difficult to be used for further research without pre-processing. We proposed a technique for detecting and removing cloud shadows in full-disk solar images. In the detection procedure, a two-step approach is applied: (1) identifying the deviation of a cloud-covered image from a perfect disk; (2) quantifying the cloud cover by an index that we defined in this paper. In the removal procedure, the transmittance of clouds is measured by comparing the cloud-covered image with a “Quiet Sun” that getting from a normal observation. A restored cloud-free image can be obtained after correcting the absorption by clouds. We tested the procedures using the full-disk solar Hα images of the Global Oscillation Network Group (GONG), and utilized the structural similarity (SSIM) algorithm for evaluating the performance of image restoration. The results demonstrate that both procedures are significant effective, and that the cloud-covered image is restored not only in visual effect but also in intensities of solar features.

MONTE CARLO STUDY OF THE EFFECT OF BACKSCATTER MATERAIL THICKNESS ON 99MTC SOURCE RESPONSE IN SINGLE PHOTON EMISSION COMPUTED TOMOGRAPHY

Introduction SPECT projections are contaminated by scatter radiation, resulting in reduced image contrast and quantitative errors. Backscatter constitutes a major part of the scatter contamination in lower energy windows. The current study is an evaluation of the effect of backscatter material on FWHM and image quality investigated by Monte Carlo simulation. Materials and Methods SIMIND program was used for simulation of a Siemen’s dual-head variable angle scintillation gamma camera. Planar and SPECT scanning of a 99mTc source and a Jaszczak phantom for varying thicknesses of Perspex slabs, as a backscatter media, were analyzed using the photopeak and scatter windows. Simulated planar images and reconstructed tomographic images were evaluated qualitatively, by two nuclear medicine specialists, and quantitatively, by Structural Similarity (SSIM) Index. Results In the 99mTc photopeak window, no significant change in total counts due to backscatter material was measured. In the scatter windows, scattering was overestimated compared with a simulated backscatter free SPECT system. For instance, at a thickness of 10 cm, total counts of a 99mTc source detected in the 72 keV windows eventually doubled with increasing backscatter material, compared with the situation without backscatter material. The backscatter contribution plateaued when more than 7 cm of scatter material was placed but there were optimized results for a backscatter thickness of 4.5 cm. Better image quality for the thickness was confirmed by the results of eye interpretation and also by SSIM algorithm. Conclusion Backscatter should be taken into account, particularly in model-based scatter correction methods in SPECT for an accurate simulation system optimization.

A Medical Image Segmentation Method Fusing Anisotropic Diffusion Model

The denoising principal of the anisotropic diffusion equation is studied. Adaptive filtering of image is realized by combining the improved image structural similarity algorithm and the anisotropic diffusion equation. This algorithm is applied to medical image segmentation. Experimental results show that the improved algorithm has good robustness and advantages in the application of adaptive medical image filtering and segmentation.