Image Variance Research Articles

Improve Person Re-Identification With Part Awareness Learning

Person re-identification (ReID) aims to predict whether two images from different cameras belong to the same person. Due to low image quality and variance in view point and body pose, it remains a difficult task. To solve the task, a model is supposed to appropriately capture features that describe body regions for identification. With the simple intuition that explicitly incorporating ReID model with part awareness could be beneficial for learning a more discriminative feature space, we propose part segmentation as an assistant body perception task during the training of a ReID model. Specifically, we add a lightweight segmentation head to the backbone of ReID model during training, which is supervised with part labels. Note that our segmentation head is only introduced during training and that it does not change network input or the way of extracting ReID feature. Experiments show that part segmentation considerably improves the performance of ReID. Through quantitative and qualitative analyses, we further reveal that body part perception helps ReID model to capture a set of more diverse features from the body, with decreased similarity between part features and increased focus on different body regions. We experiment with various representative ReID models and achieve consistent improvement on several large-scale datasets including Market1501, CUHK03, DukeMTMC-reID and MSMT17. E.g . on MSMT17, our method increases Rank-1 Accuracy of GlobalPool-ResNet-50, PCB and MGN by 2.3%, 2.9% and 3.9%, respectively. Incorporated with MGN, our model achieves state-of-the-art performance, with Rank-1 Accuracy 95.8%, 78.8%, 90.0% and 84.0% on four datasets, respectively.

Image quality assessment (IQA) using high-frequency and image variance (HFIV) for colour image

Image quality is often lost during image acquisition, transmission, and compression. Therefore, image quality assessment (IQA) is crucial in image processing. Currently, image quality can be measured from the frequency domain features, but it only applicable to blurred grayscale images. Nevertheless, noise distortion is also a common problem in digital images, and colour also affects the perception of image quality. Therefore, this paper proposes an enhanced blur and noise specific colour image quality assessment that measures high-frequency components and image variance. The number of high-frequency components is related to the edge and noise. In order to distinguish the distortion of the image, the image variance estimation is included. Experiments on public databases have shown that this method outperforms PSNR and SSIM in terms of noise and blur distortion and has low processing time of 0.0941 s/img.

Adaptive scale selection in multiscale segmentation based on the segmented object complexity of GF-2 satellite image

Image segmentation is the premise and key step of object-oriented classification, but scale selection remains a challenge in image segmentation. Over the years, scale selection methods of image segmentation have been extensively explored and developed. When the scale is chosen, all the features are generally extracted from images in these methods. In addition, in these methods, an optimal scale is generally selected based on the pre-estimation of the statistical variance of remote sensing images or determined by the post-segmentation evaluation results, rather than during the segmentation. In this study, with the central and eastern parts of Longchang City in Mid-Sichuan Hilly Region as the study area, based on the Gaofen-2 (GF-2) image, an adaptive scale selection method was proposed to determine the optimal scale of each segmentation object during the segmentation. First, the single-scale optimal image segmentation is determined by an unsupervised evaluation method which uses weighted variance and Moran’s I to measure global intra-segment homogeneity and inter-segment heterogeneity, respectively. Then, spectral, texture, and shape features of each segmented object were extracted to construct the complexity index function, and the optimal scale of the segmentation objects was determined through the iterative calculation according to the threshold value. The proposed method was applied to process GF-2 image to obtain the segmentation results and classification map, compared with the methods of determining the optimal scale through estimation local statistic variances of image and post-evaluation of segmentation results. The experimental results showed that the proposed method is of practical helpfulness and effectiveness for generating the optimal scale matching the actual ground objects.

Statistical properties of cerebral CT perfusion imaging systems. Part I. Cerebral blood volume maps generated from nondeconvolution-based systems.

The development and clinical employment of a computed tomography (CT) imaging system benefit from a thorough understanding of the statistical properties of the output images; cerebral CT perfusion (CTP) imaging system is no exception. A series of articles will present statistical properties of CTP systems and the dependence of these properties on system parameters. This Part I paper focuses on the signal and noise properties of cerebral blood volume (CBV) maps calculated using a nondeconvolution-based method. The CBV imaging chain was decomposed into a cascade of subimaging stages, which facilitated the derivation of analytical models for the probability density function, mean value, and noise variance of CBV. These models directly take CTP source image acquisition, reconstruction, and postprocessing parameters as inputs. Both numerical simulations and in vivo canine experiments were performed to validate these models. The noise variance of CBV is linearly related to the noise variance of source images and is strongly influenced by the noise variance of the baseline images. Uniformly partitioning the total radiation dose budget across all time frames was found to be suboptimal, and an optimal dose partition method was derived to minimize CBV noise. Results of the numerical simulation and animal studies validated the derived statistical properties of CBV. The statistical properties of CBV imaging systems can be accurately modeled by extending the linear CT systems theory. Based on the statistical model, several key signal and noise characteristics of CBV were identified and an optimal dose partition method was developed to improve the image quality of CBV.

The shape of the photon transfer curve of CCD sensors

The photon transfer curve (PTC) of a CCD depicts the variance of uniform images as a function of their average. It is now well established that the variance is not proportional to the average, as Poisson statistics would indicate, but rather flattens out at high flux. This “variance deficit”, related to the brighter-fatter effect, feeds correlations between nearby pixels that increase with flux, and decay with distance. We propose an analytical expression for the PTC shape, and for the dependence of correlations with intensity, and relate both to some more basic quantities related to the electrostatics of the sensor, which are commonly used to correct science images for the brighter-fatter effect. We derive electrostatic constraints from a large set of flat field images acquired with a CCD e2v 250, and eventually question the generally-admitted assumption that boundaries of CCD pixels shift by amounts proportional to the source charges. Our results show that the departure of flat field statistics from the Poisson law is entirely compatible with charge redistribution during the drift in the sensor.

Image quality enhancement in Neutron Computerized Tomography based on projection exposure time adjustment

This paper presents a proposed procedure for neutron radiography image enhancement through studying the effect of the projection exposure time on the quality of Neutron Computerized Tomography (NCT) image. A sample was exposure at the Egyptian Second Research Reactor (ETRR-2) NCT facility with projection exposure time varying from one second to six seconds at reactor power operation of 17 MWth. The NCT image was corrected against the read-out noise using flat and dark current images. Then, the Region of Interest (ROI) is selected based on an automatic threshold of the occurrence histogram of the image variance. After that, the image was normalized and filtered by median filter. The resultant images were evaluated qualitatively and compared with the images obtained from Octopus 8.5 software that is currently used in the Egyptian NCT facility. The results indicated that high-quality images had accomplished in comparison with Octopus 8.5 software without losing any information. In addition, with respect to quality measures, the image at the projection exposure time equal to 4 seconds is the best obtained image.

Recaptured Image Forensics Algorithm Based on Image Texture Feature

With the rapid development of digital phones, the digital image forensics system in current times has had a great impact. It will lead to a serious threat for us, and especially the emergence of the recaptured image makes the existing digital image forensics algorithm invalid. So, it needs an effective image detection algorithm for us to identify recaptured images. In this paper, a new detection algorithm of the recaptured image is presented based on gray level co-occurrence matrix by analyzing the differences between the real and recaptured images. In order to analyze the differences, a new image evaluation model was put forward in this paper, which is called image variance ratio. Firstly, the algorithm proposed extracted high-frequency and low-frequency information of images by wavelet transform, based on which we calculated the relative gray level co-occurrence matrices. Secondly, the features of gray level co-occurrence matrix were extracted. At last, the recaptured image was classified by the support vector machine according to the features. The experimental results showed the algorithm proposed can not only effectively identify the recaptured image obtained from different media but also have better identification rate.

Repeat tourism, destination image and behavioural intentions: implications for sustainable development in South Africa

ABSTRACTRepeat tourism represents a viable source market that provides a steady source of income for tourism managers. Specifically, only those destinations can rely on repeat tourism to survive in potentially difficult political, environmental or economic conditions who were classed as mature by Richard Butler in his concept of tourism areas cycle (TALC) in 1980. Although research confirms that tourists’ previous visitation influences their image perceptions and intention to frequent a destination, limited studies have aimed to identify these implications for sustainable development; specifically, how destinations can develop sustainably to attract and target tourists that are both financially and environmentally viable. This literature gap provides the impetus for this research note. Based on a sample of repeat tourists to South Africa (n = 337), four destination image factors were identified. The two factors, professionalism and experiential, contributed the most destination image variance for repeat South African tourists and also significantly predicted most of their behavioural intentions (i.e. repurchase intentions). Consequently, to develop a potentially sustainable market (e.g. repeat business travel) for a mature destination (South Africa), a focus on professionalism and providing an experience that encompasses the experiential factor items (i.e. fun and enjoyment) will entice positive behavioural intentions such as loyalty and positive word-of-mouth communication.

A Review of Methods to Quantify the Genomic Similarity of Topological Associating Domains.

Topologically associating domains (TADs) are the most fundamental elements and significant structures of the eukaryotic genome. Currently, algorithms have been developed to find the TADs. But few algorithms are reported to compare the similarity of TADs between genomes. In this study, mice Hi-C sequencing data of four contrasts were enrolled. Seventeen algorithms, including BPscore, Jaccard index (JI) distance, VI distance, image hash, image subtraction, image variance, and so on, were used to quantify the genomic similarity of TADs. Image subtraction, Euclidean distance, and Manhattan distance were significantly better for TAD difference detection than the others. Deferent Hash (dHash) with the best zoom size ranked the second, followed by improved Hamming distance algorithm and JI distance. Advantages and disadvantages of various algorithms for quantifying the similarity of TADs were compared. Our work could provide the fundament for TADs comparison.

Performance Comparison of Firefly and Cuckoo Search Algorithms in Optimal Thresholding of Cancer Cell Images

This research presented a performance comparison of the two methods in cancer cells image processing. Each method consisted of two stages. The first stage was image enhancement using fuzzy sets. The second stage was optimal fuzzy entropy based image thresholding. In the thresholding stage, the first method used Firefly Algorithm (FA) and the second used Cuckoo Search (CS). In both methods, four performance metrics (Mean Squared Error (MSE), Peak Signal-to-Noise Ratio (PSNR), Structured Similarity Indexing Method (SSIM), and Feature Similarity Indexing Method (FSIM)) and variance and entropy of the images were computed to validate the comparison. The image histograms of both methods show that the distribution of red, green, and blue channel is better than the histograms of original images. In terms of the four metrics, the method that uses FA shows higher performance than CS. In terms of image variance and entropy, the method using CS shows better results than FA. These results suggest that when the performance metrics used are MSE, PSNR, MSSIM, and FSIM, the method using FA is more suitable for cancer cells image enhancement and thresholding. However, when the variance and entropy of the images are used as the performance metrics, the method using CS is more suitable for cancer cells image enhancement and thresholding. Both methods will be useful to assist in the analysis of cancer cell images by the experts in the field.

A Defect Inspection for Explosive Cartridge Using an Improved Visual Attention and Image-Weighted Eigenvalue

Surface defects of explosive cartridge in the automatic sorting process are of a small area, irregular shape, and random distribution, and all problematic characteristics that hinder surface defect detection. To address these issues, a new multidefect detection method has been proposed in this paper based on a combination of an improved visual attention model and image partitioning-weighted eigenvalue (IP-WEV). First, image preprocessing is carried out by a background estimation algorithm. Then, a new fusion operator based on defects discrimination is implemented in a visual attention model to integrate intensity, orientation, and edge conspicuity into a saliency graph, in which a saliency effect of defects is considered during the fusion process. Third, a saliency map is divided into image blocks based on the image variance. This allows for the extraction of image blocks including defects, the calculation of the weighted eigenvalue, and the determination of regions containing multidefect. The IP-WEV is used to make a decision for multidefect. The experimental results show this method's detection accuracy as 98.2%, with a less computation time and quickly detection speed. Therefore, this method could be adopted for on-line detection systems for explosive cartridge surface defects.

Speckle Suppression Based on Weighted Nuclear Norm Minimization and Grey Theory

Coherent imaging systems are greatly affected by speckle noise, which makes visual analysis and features extraction a difficult task. In this paper, we propose a speckle suppression algorithm based on weighted nuclear norm minimization (WNNM) and Grey theory. First, we use logarithmic transformation to the noisy images such that the speckle noise is transformed into additive noise. Second, by matching the local blocks based on Grey theory, we will get approximate low-rank matrices grouped by the similar blocks of the reference patches. We then estimate the noise variance of the noisy images with the wavelet transform. Finally, we use WNNM method to denoise the image. The results show that our algorithm not only effectively improves the visual effect of the denoised image and preserves the local structure of the image better but also improves the objective indexes values of the denoised image.

A local mean and variance active contour model for biomedical image segmentation

To effectively segment the biomedical images with intensity inhomogeneity, a local mean and variance (LMV) active contour model is presented in this paper. In the model, the distribution of intensity belonging to each region is assumed as a Gaussian distribution with spatially varying mean and variance. At the same time, an energy function is defined and incorporated into a variational level set formulation. The curve evolution equation is then obtained from energy minimization. Because both image local mean and variance are considered, the proposed model can more effectively deal with the images with intensity inhomogeneity than the existing active contour model. Experimental results on synthetic and real images show the advantages of our method in terms of both effectiveness and robustness. Compared with the local binary fitting (LBF) model and the local hybrid image fitting (LHIF) model, our model is much most computationally efficient and much least sensitive to the initial contour.

A Gaussian random field model for de-speckling of multi-polarized Synthetic Aperture Radar data

Synthetic Aperture Radar (SAR) data have gained interest for a variety of remote sensing applications, given the capability of SAR sensors to operate independent of solar radiation and day/night conditions. However, the radiometric quality of SAR images is hindered by speckle noise, which affects further image processing and interpretation. As such, speckle reduction is a crucial pre-processing step in many remote sensing studies based on SAR imagery. This study proposes a new adaptive de-speckling method based on a Gaussian Markov Random Field (GMRF) model. The proposed method integrates both pixel-wised and contextual information using a weighted summation technique. As a by-product of the proposed method, a de-speckled pseudo-span image, which is obtained from the least-squares analysis of the de-speckled multi-polarization channels, is also produced. Experimental results from the medium resolution, fully polarimetric L-band ALOS PALSAR data demonstrate the effectiveness of the proposed algorithm compared to other well-known de-speckling approaches. The de-speckled images produced by the proposed method maintainthe mean value of the original image in homogenous areas, while preserving the edges of features in heterogeneous regions. In particular, the equivalent number of look (ENL) achieved using the proposed method improves by about 15% and 47% compared to the NL-SAR and SARBM3D de-speckling approaches, respectively. Other evaluation indices, such as the mean and variance of the ratio image also reveal the superiority of the proposed method relative to other de-speckling approaches examined in this study.

Simultaneous estimation and segmentation from projection data in dynamic PET.

Dynamic positron emission tomography (PET) is known for its ability to extract spatiotemporal information of a radio tracer in living tissue. Information of different functional regions based on an accurate reconstruction of the activity images and kinetic parametric images has been widely studied and can be useful in research and clinical setting for diagnosis and other quantitative tasks. In this paper, our purpose is to present a novel framework for estimating the kinetic parametric images directly from the raw measurement data together with a simultaneous segmentation accomplished through kinetic parameters clustering. An iterative framework is proposed to estimate the kinetic parameter image, activity map and do the segmentation simultaneously from the complete dynamic PET projection data. The clustering process is applied to the kinetic parameter variable rather than to the traditional activity distribution so as to achieve accurate discrimination between different functional areas. Prior information such as total variation regularization is incorporated to reduce the noise in the PET images and a sparseness constraint is integrated to guarantee the solution for kinetic parameters due to the over complete dictionary. Alternating direction method of multipliers (ADMM) method is used to solve the optimization problem. The proposed algorithm was validated with experiments on Monte Carlo-simulated phantoms and real patient data. Symbol error rate (SER) was defined to evaluate the performance of clustering. Bias and variance of the reconstruction activity images were calculated based on ground truth. Relative mean square error (MSE) was used to evaluate parametric results quantitatively. In brain phantom experiment, when counting rate is 1×106 , the bias (variance) of our method is 0.1270 (0.0281), which is lower than maximum likelihood expectation maximization (MLEM) 0.1637 (0.0410) and direct estimation without segmentation (DE) 0.1511 (0.0326). In the Zubal phantom experiment, our method has the lowest bias (variance) 0.1559 (0.0354) with 1×105 counting rate, compared with DE 0.1820 (0.0435) and MLEM 0.3043 (0.0644). As for classification, the SER of our method is 18.87% which is the lowest among MLEM+k-means, DE+k-means, and kinetic spectral clustering (KSC). Brain data with MR reference and real patient results also show that the proposed method can get images with clear structure by visual inspection. In this paper, we presented a joint reconstruction framework for simultaneously estimating the activity distribution, parametric images, and parameter-based segmentation of the ROIs into different functional areas. Total variation regularization is performed on the activity distribution domain to suppress noise and preserve the edges between ROIs. An over complete dictionary for time activity curve basis is constructed. SER, bias, variance, and MSE were calculated to show the effectiveness of the proposed method.

Noise power spectrum of ultrasound uniformity images

Background The Noise Power Spectrum (NPS) is an effective tool for providing information on how the total variance of uniformity images is distributed in frequency space. The NPS is a tool which can widen the understanding of noise in ultrasound as noise in this imaging modality includes both electronic noise and static structural noise. However, surprisingly, very little research is found in the literature and unlike projection X-Ray Imaging and CT, NPS is not included in ultrasound QC protocols. Objectives (1) To establish a scan protocol to measure the NPS for ultrasound scanners in the case of linear transducers; (2) To specify an IQWorks analysis tree for the calculation of All Noise, Stochastic and Static NPS for Regions of Interest (ROIs) at near field, focal region and far field; (3) To Investigate the short term reliability of the methodology. Research Design The protocol was tested on a Siemens ACUSON Sequoia using a linear transducer and two phantoms (Gammex RMI404GS LE, CIRS Model 551). An IQWorks analysis tree was developed for the calculation of NPS of ultrasound uniformity images. Three ROIs were selected at near field, focal region and far field to evaluate the mean pixel value and image variance of the ROIs, Stochastic Integral NPS, Static Integral NPS, All Noise Integral NPS and the shape of one-dimensional NPS at different depths. The short term reliability of the methodology was checked by measuring the coefficient of variation of the mean pixel value, image variance and NPS values over repeat acquisitions. Results The NPS in ultrasound is mainly of a low-frequency component. There are interesting changes of the NPS at different depths. The NPS tool can be used as part of an ultrasound QC programme using the NPS IQWorks analysis tree specified in this project.

Face segmentation based on level set and improved DBM prior shape

This paper puts forward a new method of level set image segmentation based on prior shape, which aims to provide a better solution to the challenging segmentation problems that typically occur in images with complex background, intensity inhomogeneity and partially blocked targets. First, we introduced glial cells into deep Boltzmann machine (DBM) to solve that units in the DBM layer are not connected to each other, and then the novel DBM is employed to learn prior shape. Next, we used the variational level set and the local Gaussian distribution to fit the image energy term with local mean and local variance of image. Then, the prior shape energy is integrated into the image energy term to construct the final energy segmentation model. The experimental results show that the new model has stronger robustness and higher efficiency for face images segmentation.

Artifact Suppressed Nonlinear Diffusion Filtering for Low-Dose CT Image Processing

Computed tomography (CT) images with a low-dose protocol generally have severe mottle noise and streak artifacts. In this paper, we propose a novel diffusion method named “artifact suppressed nonlinear diffusion filtering (ASNDF),” to process low-dose CT (LDCT) images. Different from other diffusion filtering methods, the proposed ASNDF not only includes image gradient as the main cue to construct a diffusion coefficient function, but also incorporates the local variances of image to be diffused and residual image between two adjacent diffusions. In detail, the classical PM diffusion is first performed to get the initial residual image, and then from the second iteration, the LDCT image is processed according to the ASNDF processing. Simulated data, clinical data and rat data are conducted to evaluate the proposed method, and the comparison experiments with other competing methods show that the proposed ASNDF method makes an improvement in artifact suppression and structure preservation, and offers a sound alternative to process LDCT images from most current CT systems.

Fusion of multi-focus images via a Gaussian curvature filter and synthetic focusing degree criterion.

The aim of multi-focus image fusion technology is to acquire an image of every scene all focused on the same visual point at different focal settings. To achieve this goal, we propose an improved multi-focus image fusion algorithm based on a Gaussian curvature filter and synthetic focusing degree criterion. First, in order to realize the salient feature extraction, a Gaussian curvature filter is applied to obtain the most sharpness regions. Then we obtain a coarse fusion map by composing a synthetic focusing degree criterion, which is a combination of the spatial frequency and the local variance of image. The coarse fusion map is further processed by morphological filters and median filters to acquire an optimized fusion map. Finally, the fusion image is obtained via a weighted fusion operation. Experimental results demonstrate that our proposed algorithm can be competitive with, or even outperform, many existing fusion methods on both qualitative and quantitative measures.

Fuzzy clustering based transition region extraction for image segmentation

Transition region based approaches are recent hybrid segmentation techniques well known for its simplicity and effectiveness. Here, the segmentation effectiveness depends on robust extraction of transition regions. So, we have proposed clustering approach based transition region extraction method for image segmentation. The proposed method initially uses the local variance of the input image to get the variance feature image. Fuzzy C-means clustering is applied to the variance feature image to separate the transitional features from the feature image. Further, Otsu thresholding is applied to the transitional feature image to extract the transition region. For extracting the exact edge image, morphological thinning operation is performed. The edge image extracted in former step is closed in nature. The morphological cleaning and region filling operation is performed on an edge image to get the object regions. Finally, objects are extracted via these object regions. The proposed method is compared with different image segmentation methods. An experimental result reveals that the proposed method outperforms other methods for segmentation of images containing single and multiple objects.