Texture Loss Research Articles

A Multi-Modal Edge Consistency Metric Based on Regression Robustness of Truncated SVD

In this paper, we propose a novel edge consistency metric for multi-modal correspondence. It is based on a novel observation on image truncated SVD (singular value decomposition) termed regression robustness, which describes the fact that, a good approximation from image truncated SVD can be inherited even if the eigen-images change due to expansion and channel-dependent offsets. Compared to state-of-the-arts, multi-modal edge consistency metric can simultaneously handle multiple images with complex modality changes, including local variation, gradient reverse, intensity order change, and texture loss. Its complexity is almost linear to pixel number. Remarkable accuracies have been achieved in experiments.

Multispectral Fusion of RGB and NIR Images Using Weighted Least Squares and Convolution Neural Networks

In low light condition, color (RGB) images captured by visible sensors suffer from severe noise causing loss of colors and textures. However, near infrared (NIR) images captured by NIR sensors are robust to noise even in low light condition without color. Since RGB and NIR images are complementary in low light condition, the multispectral fusion of RGB and NIR images provides a viable solution to low light imaging. In this paper, we propose multispectral fusion of RGB and NIR images using weighted least squares (WLS) and convolution neural networks (CNNs). We combine traditional WLS filtering for layer decomposition and denoising with latest deep learning for image enhancement and texture transfer into the multispectral fusion to take both advantages. We build two networks based on CNN: image enhancement network (IEN) for image enhancement and texture transfer network (TTN) for NIR texture transfer. First, we perform RGB image denoising based on WLS filtering and generate the base layer. We use both RGB and NIR images for WLS filtering as weights to filter out noise in low light RGB images. Second, we conduct IEN to enhance contrast of the base layer. Third, we perform TTN to deliver NIR details completely and naturally to the fusion result. The combination of WLS, TTN and IEN leads to noise reduction, contrast enhancement, and detail preservation in fusion. Experimental results show that the proposed method achieves good performance in both noise reduction and detail transfer as well as outperforms state-of-the-art methods in terms of visual quality and quantitative measurements.

Contemporary diagnosis and management of dental erosion.

This article is aimed at providing an overview of the topic of erosive tooth wear (ETW), highlighting the clinical signs, diagnosis, and management of dental erosion. With the increased prevalence of ETW, it is important that oral health professionals are able to recognize the early signs. Early clinical signs of dental erosion are characterized by loss of enamel texture, a silky glossy appearance, and sometimes a dulling of the surface gloss, referred to as the "whipped clay effect, cupping, and restorations 'standing proud'." The progression of ETW should be monitored by means of diagnostic models or clinical photographs. ETW can be as a result of acid attack of extrinsic or intrinsic origin. There is an increase of ETW that is being recognized by the profession. The first step in diagnosing and management is to recognize as early as possible that the process is occurring. At that point a determination of whether the primary etiology is either intrinsic or extrinsic should be made. If these findings are confirmed, appropriate prevention, and management strategies can be adopted followed by appropriate restorative therapy. The prevalence of ETW continues to increase. It is therefore important that oral health care providers have a better understanding of the etiology, pathophysiology, and management of this condition. This review aims to provide the guidelines for diagnosis and management of dental erosion.

Edge Profile Super Resolution

The purpose of single-image super resolution (SISR) is to reconstruct an accurate high-resolution image from a degraded low-resolution image. Owing to the lack of information in low-resolution images, SISR is a challenging problem. In particular, it is difficult to represent details, including high-frequency components, such as texture and structural information. We propose the edge profile super-resolution (EPSR) method to preserve structural information and restore texture. EPSR is achieved by stacking modified fractal residual network (mFRN) structures hierarchically and repeatedly. Each mFRN is composed of many residual edge profile blocks (REPBs) that extract features and preserve the edge, structure, and texture information of the image. For implementing REPB, we design three main modules: Residual Efficient Channel Attention Block(RECAB) module, Edge Profile(EP) module, and Context Network(CN) module. By repeating the procedure in the mFRN structure, the EPSR method could be used to extract high-fidelity features, thus preventing texture loss and preserving the structure with appropriate sharpness. Experimental results show that EPSR achieves competitive performance against state-of-the-art methods in terms of the peak signal-to-noise ratio(PSNR) and structural similarity index measure(SSIM) evaluation metrics, as well as visual results.

OCT-Based Angiography and Surface Topography in Burn-Damaged Skin.

There is a clinical need for an accurate, non-invasive imaging tool that can provide the objective assessment of burn wounds. The aims of this study are to demonstrate the potential of optical coherence tomography (OCT) in evaluating burn wound healing, as well as exploring the physiological basis of human wound healing. This was a retrospective study. Seven patients with severe burn wounds who were admitted to Harborview Medical Center were imaged using an in-house-built, clinical-prototype OCT system. OCT imaging was carried out at multiple scan sites on the burned skin across two time points (imaging session #1 and #2) with a field of view of ~9 × 9 mm. Due to pathological differences among burn zones, scan sites were classified into red sites (zone of hyperemia), white sites (zone of coagulation), and mixed sites. In addition to obtaining qualitative en face vascular and surface topography maps, we quantified vascular area density and surface roughness for comparative purposes. En face vascular and surface topography maps demonstrated numerous morphological changes over both imaging sessions associated with burn injury, such as altered blood flow and loss of regular texture. Quantitative analyses revealed that during imaging session #1, vascular area density was significantly increased within the red sites compared with that of a healthy control (P = 0.0130), while vascular area density was significantly decreased within the white sites compared with that of a healthy control (P < 0.0001). During imaging session #2, vascular area density was significantly reduced to a more normal range within the red sites compared with imaging session #1 (P = 0.0215); however, vascular area density was still significantly lower within the white sites compared with that of a healthy control (P < 0.0001). Furthermore, vascular area density and surface roughness were significantly increased within the white sites during imaging session #2 compared with imaging session #1 (both P < 0.0001). OCT is clinically feasible to monitor vascular changes and alterations in skin surface roughness during the process of burn wound healing. Variations in vascular area density and roughness measurements within the burn wounds revealed by OCT offer some key insights into the underlying pathophysiological mechanisms responsible for wound healing, which may become critical biological indicators in future clinical evaluation and monitoring of wound healing. Lasers Surg. Med. © 2020 Wiley Periodicals LLC.

Factors affecting performance of sprayed seals in rural Victoria

ABSTRACT Factors that significantly affect performance of in-service sprayed seals have been identified through developing comprehensive models using a multilevel analysis approach. Subjective condition data of surface distresses from five rural road networks in Victoria/Australia have been used. They include cracking, deformation, texture loss, stone loss and patching. Evaluations of individual distresses for representative segments are combined into a composite index called Surface Inspection Rating (SIR) which is used to trigger the need for resealing. Contributing factors considered include surface age, traffic volume, aggregate size, subgrade soil type, shoulder seal width, rainfall and temperature. For each network, statistically significant factors for each distress are identified and used in developing their SIR models. An overall SIR model has also been developed using condition data from all five networks. Predictions of this model have been compared with an existing model that has been developed using Markov Chains with surface age as the only predictor. The results show that both models have roughly similar deterioration rates of SIR up to 8 years of seal life, after which deterioration rate of the newly developed model becomes higher.

The effect of thermomechanical treatments on the properties of powder metallurgy Ti–5Fe alloy

In this study, the low-cost Ti–5Fe alloy was fabricated by a combination of sintering plus thermomechanical treatment (TMT). The primary aim is to improve the tensile properties of the recently widely studied powder metallurgy Ti-xFe alloys made by the press and sinter process. The effect of hot forging or extrusion and subsequent annealing treatment on the microstructure and mechanical properties of the Ti–5Fe alloy was therefore studied. TMT from temperatures in the β-phase field severely changes the microstructure of the sintered alloy. The substantial deformation of the β-grains and the formation of fine α-phase have a significant effect on the mechanical properties. The yield and ultimate tensile strengths of the TMT alloys reach high values, from 1050 MPa to over 1200 MPa. The subsequent heat treatment results in the modification of the α-phase morphology and the partial loss of the crystallographic texture, leading to the enhancement of the elongation to values above 20%, despite of the high oxygen content of the alloy.

Instant controlled pressure drop (DIC) assisted dehydrofreezing for improving freezing/thawing efficiency and apple fruit texture

Partial air-drying, instant controlled pressure drop, freezing, and thawing operations were combined to minimize the texture loss of frozen/thawed apple fruits. Fresh apple samples (700% db) and samples previously dehydrated until different water contents [200, 100, and 30% dry basis (db)] and DIC treated at constant conditions (0.2 MPa, 25 s) were completely frozen at − 30 °C and at two different freezing velocities and then thawed at 4 °C. The effects of sample water content and freezing velocity on freezing/thawing durations and apple texture were assessed. Lower sample water content implied higher freezing and thawing rates, and lower freezing and thawing times. Indeed, freezing times are 108 min and 30 min for samples with water contents of 700% and 100% db, respectively. Thawing time was also reduced from 588 to 42 min for conventionally frozen and DIC-dehydrofrozen samples (100% db). During thawing, the amount of water exudate was approximately 12% (g/100 g residual water) for conventionally frozen apples at the lower freezing velocity. Whereas, it is lowered to 1.6% for 100% db DIC-dehydrofrozen samples. Moreover, the impact of freezing velocity on water exudate was significant and very important only for conventionally frozen apples (with high water content) and remains insignificant for DIC-dehydrofrozen samples with different water content levels. Thawed apple firmness was as higher as the water content was lower, without any significant impact of freezing velocity. Dehydrofreezing assisted by instant controlled pressure drop exhibited significant effects on reducing thawing duration and enhancing frozen apple fruit texture.

Effect of ZnO nanoparticles combined radio frequency pasteurization on the protein structure and water state of chicken thigh meat

How to reduce the damage of conventional sterilization to the quality of prepared food, the primary indicator of consumer concern, remains an important challenge for food processing practitioners. The ZnO nanoparticles combined radio frequency (ZNCRF) has been developed previously by our research group. In this study, we explored the mechanism that the quality of chicken thigh meat samples of ZNCRF pasteurization was superior to that of conventional high pressure steam (HPS) sterilization from the perspective of the changes of protein and water state. The results showed that after processing by ZNCRF and HPS, the secondary structure of drumstick protein changed significantly, which the relative content of β-sheet in the samples decreased by 22.20% and 22.09%, respectively. The results of LF-NMR show that ZNCRF pasteurization (within 20 min) was helpful to improve the water retention and quality of chicken thigh meat products. The SEM showed that adopting ZNCRF processing (within 20 min) can be very good to retain the chicken thigh meat tissue microstructure. These differences were manifested in the sensory sense that the loss of color, texture, and flavor of the ZNCRF pasteurized chicken thigh meat samples was less than that of the HPS pasteurized samples. The results of this study provide a theoretical basis for the application of ZNCRF pasteurization in the chicken industry and the research of relevant sterilization equipment.

Modification of apple, beet and kiwifruit cell walls by boiling in acid conditions: Common and specific responses

Cell wall (CW) degradation causes texture loss of plant-based products after processing. However, these losses differ in intensity, which could be due to cell wall structure or plant tissue internal pH conditions. To distinguish these two factors, CWs isolated from apple, beet and kiwifruit were subjected to boiling at pH 2.0, 3.5 and 6.0. Pectin depolymerization was the least pronounced at pH 3.5, while galacturonic acid contents of all CWs decreased at pH 6.0 due to the β-elimination. Pectins were solubilized, and their size decreased with increased pH during CW treatment. At pH 6.0, degrees of methylation decreased mostly in apple and beet CWs while galactose decreased more in kiwifruit CW. At pH 2.0, arabinan was lost in apple and beet CW due to acid hydrolysis. Apple CW was the most susceptible to degradation either at pH 2.0 or 6.0, while beet CW was more degraded at pH 2.0. In contrast, kiwifruit CW was the least susceptible to degradation whichever the pH. Acid hydrolysis and β-elimination appeared to be common mechanisms that cause loss of neutral sugars, often from pectin side chains, and galacturonic acid, respectively, but their effects were of different intensities. This work has a guiding significance for improving texture in the thermally canning process.

Coupled adversarial learning for semi-supervised heterogeneous face recognition

Visible-near infrared (VIS-NIR) face matching is a challenging issue in heterogeneous face recognition due to the large spectrum domain discrepancy as well as the over-fitting on insufficient pairwise VIS and NIR images during training. This paper proposes a coupled adversarial learning (CAL) approach for the VIS-NIR face matching by performing adversarial learning on both image and feature levels. On the image level, we learn a transformation network from unpaired NIR-VIS images to transform a NIR image to VIS domain. Cycle loss, global intensity loss and local texture loss are employed to better capture the discrepancy between NIR and VIS domains. The synthesized NIR or VIS images can be further used to alleviate the over-fitting problem in a semi-supervised way. On the feature level, we seek a shared feature space in which the heterogeneous face matching problem can be approximately treated as a homogeneous face matching problem. An adversarial loss and an orthogonal constraint are employed to reduce the spectrum domain discrepancy and the over-fitting problem, respectively. Experimental results show that CAL not only synthesizes high-quality VIS or NIR images, but also obtains state-of-the-art recognition results.

Microbial and quality improvement of boiled gansi dish using carbon dots combined with radio frequency treatment

Use of carbon dots (CDs) combined with radio frequency (RF) was applied to pasteurize and reduce the microorganism population in order to improve the quality of boiled gansi dish. CDs were prepared from banana using hydrothermal method, and characterized by using TEM, XRD and FTIR. The minimal inhibitory concentration (MIC) test showed CDs can efficiently inactivate Escherichia coli (E. coli), Staphylococcus aureus (S. aureus) and Bacillus subtilis (B. subtilis). This study also evaluated the effectiveness of five treatments, including CDs alone, CDs combined RF (CDRF) heating for different time (8 min, 12 min, and 16 min), and high pressure steam (HPS) sterilization of boiled gansi dish inoculated with B. subtilis. After CDRF treated for 8 min, 12 min, and 16 min, the center temperature of samples reached to 78.92, 87.77 and 93.82 °C, and the colony forming units (CFU) of B. subtilis reduced by 2.13, 3.62, and 4.63 log, respectively. Samples with CDRF12 treatment, exhibited better product quality as evidenced by reduced loss of texture, flavor, and sensory as compared with HPS sample. The results indicated that CDRF treatment has a great potential to produce packaged boiled gansi dish with high product quality.

Performance Improvement of Fruit Ripeness Smart Label Based On Ammonium Molibdat Color Indicators

Research in fruit ripeness indicator is still experiencing especially due to major difficulties of several fruits with no color changes in its skin when it is ripen. From the previous research, there was found that ammonium molybdate [(NH4) 6Mo7O24.4H2O] embedded in the polymer matrix could be used as an indicator label to detect the ripeness of climacteric fruits base on the color change from yellow to blue and then green. However, the performance label still needs to be developed further. The surface of the label was still poor due to air or bubble trapping inside the film. It was found that mixing H2O2 and molydate agent produced air or bubble thus in this research, a pre-treatment of film solution was done to chase away the air by storing and vacuuming the solution in cold temperature and time period of storage. A variety of film drying method was also carried out to find the best temperature of the oven to produce smooth surface of the film. The sensitivity of the label to ethylene gas was improved by adding more agent solution into the film. The best form of smart labels was produced using an oven at 40°C for 18 hours. The label composition was 100 mL distilled water, 3.5 g PVOH, 2 mL glycerol, and a color indicator solution (ammonium molybdate and hydrogen peroxide ratio of 1:10) at 4 mL. The molydate solution had to be stored for 3 days in temperature of 7oC before used. The label sensitivity was improved as low as 100 ppm of pure ethylene gas. Label application in a pack of avocados showed a relationship between label color changes and fruit quality degradation. The value of hue label on days 0 to 6 changed from yellow to greenish yellow, while on the 7th to the 10th day the color of the label was still in the same color as the day of 6th. Decreasing the quality of fruit during storage can be seen from the increase in the percentage of weight loss and hardness of fruit texture.

Texture preservation and speckle reduction in poor optical coherence tomography using the convolutional neural network.

For a poor quality optical coherence tomography (OCT) image, quality enhancement is limited to speckle residue and edge blur as well as texture loss, especially at the background region near edges. To solve this problem, in this paper we propose a de-speckling method based on the convolutional neural network (CNN). In the proposed method, we use a deep nonlinear CNN mapping model in the serial architecture, here named as OCTNet. Our OCTNet in the proposed method can fully utilize the deep information on speckles and edges as well as fine textures of an original OCT image. And also we construct an available pertinent dataset by combining three existing methods to train the model. With the proposed method, we can accurately get the speckle noise from an original OCT image. We test our method on four experimental human retinal OCT images and also compare it with three state-of-the-art methods, including the adaptive complex diffusion (ACD) method and the curvelet shrinkage (Curvelet) method as well as the shearlet-based total variation (STV) method. The performance of these methods is quantitatively evaluated in terms of image distinguishability, contrast, smoothness and edge sharpness, and also qualitatively analyzed at aspects of speckle reduction, texture protection and edge preservation. The experimental results show that our OCTNet can reduce the speckle noise and protect the structural information as well as preserve the edge features effectively and simultaneously, even where the background region near edges. And also our OCTNet has full advantages on excellent generalization, adaptiveness, robust and batch performance. These advantages make our method be suitable to process a great mass of different images rapidly without any parameter fine-turning under a time-constrained real-time situation.

Directional fractional-order total variation hybrid regularization for image deblurring

To overcome the staircase effect and texture loss of total variation (TV), a nonblind image deblurring method that combines the multidirectional fractional-order TV and the traditional TV is proposed. First, a four-directional fractional-order TV (FOTV4) model is put forward to explore more directional information of an image. Then the FOTV4 model and the traditional TV model are related in a unified regularization framework. Finally, the split Bregman algorithm and fast Fourier transform theory are utilized to optimize the proposed hybrid regularization model. Numerical experiments demonstrate that the proposed method can protect the texture and edge while eliminating the staircase effect effectively.

Features of Vanadium-Surface Damage Induced by Pulsed Laser Radiation

The features of the damage of the surface layer of vanadium under the action of pulsed laser radiation are studied. Laser irradiation is carried out in air using a GOS 1001 setup in the Q-switching mode with the following parameters: flux power density q = 1.2 × 108 W/cm2, pulse duration τ0 = 50 ns, and number of pulses N = 1–6. The typical surface damages induced by the laser pulses are found to include the melting of material, a microcrack network, a wavy relief, and drop-like particles. The central region characterized by the greatest degree of damage contains also individual drops of metal, which crystallized like a spiral. The heat-affected zone (HAZ) adjacent to the central one is damaged to a significantly weaker extent. Surface degradation increases as the number of pulses increases. Laser irradiation is revealed to change the X-ray diffraction (XRD) patterns: loss of texture, presence of vanadium-oxide signal, peak broadening, and lattice parameter increase (from 3.022(2) to 3.027(3) A). It is shown that preliminary irradiation with argon ions (dose of 1022 m–2, E = 20 keV) affects no surface damage of the central region, while in the adjacent heat-affected zone, there is a spallation of local surface regions.

Evaluation of Beef Meat Quality at Slaughterhouse in Kendari City

. This study aims to find out whether DFD (Dark, Firm, and Dry) and PSE (Pale, Soft, and Exsudative) happen/ occur to Bali Cattle Carcas slayed in Anggoeya Animal Slaugtherhouse in Kendari City. This study is useful to give information about the meats’ DFD and PSE cases and carcas quality at Anggoeya Slaugtherhouse in Kendari City. The sample of this study are 6 bulls and 6 cows taken from slaugtherhouse in Kendari City. The obtained data was, then, analyzed descriptively based on the mean and deviation standard (χ ± SD).The variables observed in this studyincludes PH measurement for 1 – 6 hours after slashing, color, loss of mass, texture, hardness, and humidity. The meats were found to be around normal PH 5.55 – 5.73. Based on the color score AUS_MEAT Colour, the color of the meats were around the score of 4 – 5, loss of mass were around 30,64 – 43,28%. Based on panelist judgement, most of the meats texture were scored 3 or were in moderate level. Most of the meats’ elasticity were scored 3 or were rather soft, while most of the meats’ humidity were scored 3 or moderate (62.50%). On the meats produced by the Animals Slaugtherhouse in Kendari City, no changes to DFD (Dark, Firm, and Dry) or PSE (Pale, Soft and Exudative) was found.

Avaliação pós-colheita de tangerinas ‘Dekopon’ submetidas a aplicação de cloreto de cálcio em pré-colheita

As tangerinas têm apresentado grande aumento na produção devido a adoção de diversos tratos culturais, potencializando a redução de perdas e a comercialização. No entanto, faz-se necessários estudos que possam manter a qualidade pós-colheita e incremento no campo. Portanto, objetivou-se com o presente trabalho avaliar física e quimicamente os frutos de tangerina variedade ‘Dekopon’, analisando a influência da aplicação de cloreto de cálcio em 100, 200, 300, 400% da dosagem recomendada, além do controle. As amostras dos frutos foram colhidas em pomar experimental no município de Anápolis-GO e as análises foram realizadas no laboratório da EMATER-GO. O experimento foi realizado em Delineamento Inteiramente Casualizado (DIC) composto por cinco doses de cloreto de cálcio, três repetições durante quatro semanas de avaliação (0, 7, 14 e 21 dias), respectivamente, sendo as variáveis avaliadas submetidas a análise de regressão polinomial em função das semanas. Foram avaliados: massa fresca dos frutos, coloração, sólidos solúveis, acidez titulável e textura. Os dados obtidos foram submetidos a análise de regressão por meio de um teste F no nível de probabilidade de 5%, para medir o significado do modelo proposto. Conclui-se que o uso do cloreto de cálcio, não atuou de forma positiva na manutenção pós-colheita da qualidade dos frutos, no entanto, pode-se destacar que foram observadas características desejáveis com frutos maiores e manutenção da cor para aplicações de 100 e 200% da dosagem recomendada durante 21 dias de avaliação.

Generative Adversarial Network for Image Super-Resolution Combining Texture Loss

Objective: Super-resolution reconstruction is an increasingly important area in computer vision. To alleviate the problems that super-resolution reconstruction models based on generative adversarial networks are difficult to train and contain artifacts in reconstruction results, we propose a novel and improved algorithm. Methods: This paper presented TSRGAN (Super-Resolution Generative Adversarial Networks Combining Texture Loss) model which was also based on generative adversarial networks. We redefined the generator network and discriminator network. Firstly, on the network structure, residual dense blocks without excess batch normalization layers were used to form generator network. Visual Geometry Group (VGG)19 network was adopted as the basic framework of discriminator network. Secondly, in the loss function, the weighting of the four loss functions of texture loss, perceptual loss, adversarial loss and content loss was used as the objective function of generator. Texture loss was proposed to encourage local information matching. Perceptual loss was enhanced by employing the features before activation layer to calculate. Adversarial loss was optimized based on WGAN-GP (Wasserstein GAN with Gradient Penalty) theory. Content loss was used to ensure the accuracy of low-frequency information. During the optimization process, the target image information was reconstructed from different angles of high and low frequencies. Results: The experimental results showed that our method made the average Peak Signal to Noise Ratio of reconstructed images reach 27.99 dB and the average Structural Similarity Index reach 0.778 without losing too much speed, which was superior to other comparison algorithms in objective evaluation index. What is more, TSRGAN significantly improved subjective visual evaluations such as brightness information and texture details. We found that it could generate images with more realistic textures and more accurate brightness, which were more in line with human visual evaluation. Conclusions: Our improvements to the network structure could reduce the model’s calculation amount and stabilize the training direction. In addition, the loss function we present for generator could provide stronger supervision for restoring realistic textures and achieving brightness consistency. Experimental results prove the effectiveness and superiority of TSRGAN algorithm.

Recent advances in mushrooms preservation

Mushroom is considered as a functional food rich in both nutrients and medicinal properties. Over the years due to increased recognition of its medicinal and nutritional values and income generating potential the demand for mushrooms has been on the rise. However, mushroom have shorter shelf-life because of their high respiration rate and lack barriers to prevent water loss and microbial attack. Various physico-chemical changes takes place in mushroom during postharvest period like moisture loss, discoloration, texture loss, nutrient loss, microbial spoilage, etc. In order to improve self-life and preserve quality of mushroom, various preservation methods are used which involve thermal, physical and chemical processes. This article highlights recent methods and techniques of preservation of postharvest mushrooms.