Image is one of the most important forms of information expression in multimedia. It is the key factor to determine the visual effect of multimedia software. As an image restoration task, image deraining can effectively restore the original information of the image, which is conducive to the downstream task. In recent years, with the development of deep learning technology, CNN and Transformer structures have shone brightly in computer vision. In this paper, we summarize the key to success of these structures in the past, and on this basis, we introduce the concept of a layer aggregation mechanism to describe how to reuse the information of the previous layer to better extract the features of the current layer. Based on this layer aggregation mechanism, we build the rain removal network called DenseformerNet. Our network strengthens feature promotion and encourages feature reuse, allowing better information and gradient flow. Through a large number of experiments, we prove that our model is efficient and effective, and expect to bring some illumination to the future rain removal network.
2
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
Nondestructive methods to estimate leaf area (LA) by leaf length (L) and/or width (W) are useful in plant physiology and ecology studies. However, both environmental and ontogenic factors may influence leaf size and/or shape, which may alter the coefficient of LA models. We carried an investigation along an altitudinal gradient in the Sergyemla Mountains, southeast Tibet. In August 2009, we selected nine sites at about every 50 m in altitude from 4,250 m to 4,640 m a.s.l. A total of 4,245 different leaf-aged Rhododendron aganniphum var. schizopeplum (a dominant overstory species) leaves were measured. Compared with the single dimensional models, the two-dimensional model encompassing both L and W (model 5) reflected higher R2 (0.98–0.99) and lower MSE (1.19–3.21) across different leaf age groups for each site, implying that such model could provide the best fit for LA estimation. Analysis of covariance further illustrated that two leaf dimensions model was irrespective of leaf age effects in eight out of the nine sites. Leaf shape (L:W ratio) varied between sites and tended to decrease at higher altitudes (4500–4640 m a.s.l.), leading to significant differences in coefficients of the two-dimension model between every two adjacent sites. For overstory species in alpine habitats, altitude rather than leaf age may affect leaf shape which alters the coefficients of LA estimation models. Since leaf shape of different species (overstory species versus understory ones) may show different responses to a certain environmental gradient, researchers must pay attention to the variation of leaf shape when estimating species-specific LA by measuring L and W, especially when leaves of the top overstory species were collected at different sites.
3
Dostęp do pełnego tekstu na zewnętrznej witrynie WWW
In this study, a simple and rapid liquid chromatography-mass spectrometry method was developed to simultaneously determinate five 16-deoxybarringtogenol C triterpenoid saponins with the potential of neuroprotection in rat plasma following the oral administration of the Xanthoceras sorbifolia Bunge husks extract. With digoxin as the internal standard, the plasma samples were pre-treated by ethyl acetateisopropanol (1:1, v/v). The chromatographic separation of the five analytes was performed using a Phenomenex C18 column (250 mm 34.6 mm, 5.0 mm) with a mobile phase of 0.05% formic acid (A)acetonitrile (B). The mass spectrometric detection was carried out in the selected ion mode in positive ionization. The extraction recoveries of the five analytes were all over 71.28%. The established method was fully validated in line with the ICH and Food and Drug Administration (FDA) guidelines and successfully applied to the pharmacokinetic study on the five analytes in rat plasma. The terminal half-life (t1/2) of the five analytes was 2.92 ± 0.57, 5.52 ± 1.75, 2.48 ± 0.62, 2.95 ± 0.94, and 2.34 ± 0.81, respectively. This study was purposed to investigate the oral pharmacokinetic parameters and gain an in-depth insight into the reasonable preclinical use of the husks extract derived from X. sorbifolia Bunge.
The simultaneous scouring and bleaching of cotton/linen blends was performed in a nearneutral activated peroxide system (pH = 7.2). A response surface quadratic model (RSQM) based on the central composite design (CCD) was established to investigate and optimise the bleaching performance. Research results showed that hydrophobic impurities in the blends had little impact on the bleaching. Temperature was the most significant factor affecting the fabric’s whiteness index (WI), followed by the concentration and duration of the activator tetraacetylethylenediamine (TAED). The optimised process was performed at 70°C for 40 min, incorporating 20 mmol/l of TAED and 42 mmol/l of H2O2. Compared with the conventional process carried out at 95 °C for 60 min and adding 80 mmol/l H2O2, the activation process provided the fabric with comparable WI and wettability, a smoother and cleaner microcosmic surface, as well as the advantages of saving energy and preserving fabric. This study confirmed the feasibility of an efficient one-step process for neutral scouring and bleaching.
PL
Równoczesne pranie i bielenie mieszanek bawełny i lnu przeprowadzono w prawie neutralnym układzie nadtlenkowym (pH = 7,2). W celu zbadania i zoptymalizowania wydajności wybielania został utworzony model kwadratowy powierzchni odpowiedzi (RSQM) oparty na centralnej konstrukcji kompozytowej (CCD). Wyniki badań wykazały, że zanieczyszczenia hydrofobowe w mieszankach miały niewielki wpływ na bielenie. Temperatura była najważniejszym czynnikiem wpływającym na wskaźnik bieli tkaniny (WI), a następnie stężenie i czas trwania aktywatora tetraacetyloetylenodiaminy (TAED). Zoptymalizowany proces prowadzono w temperaturze 70 °C przez 40 min., stosując 20 mmol/litr TAED i 42 mmol/litr H2O2. W porównaniu z konwencjonalnym procesem prowadzonym w 95 °C przez 60 min. i dodawaniu 80 mmol/litr H2O2, proces aktywacji zapewnił tkaninie porównywalny wskaźnik WI i zwilżalność oraz gładszą i czystszą powierzchnię, a także korzyści wynikające z oszczędności energii. Przeprowadzone badanie potwierdziło mozliwość wykonania wydajnego jednoetapowego procesu prania wstępnego i bielenia.
JavaScript jest wyłączony w Twojej przeglądarce internetowej. Włącz go, a następnie odśwież stronę, aby móc w pełni z niej korzystać.