A very deep convolutional neural network (CNN) has recently achieved great success for image super-resolution (SR) and offered hierarchical features as well. However, most deep CNN based SR models do not make full use of the hierarchical features from the original low-resolution (LR) images, thereby achieving relatively-low performance. In this paper, we propose a novel residual dense network (RDN) to address this problem in image SR. We fully exploit the hierarchical features from all the convolutional layers. Specifically, we propose residual dense block (RDB) to extract abundant local features via dense connected convolutional layers. RDB further allows direct connections from the state of preceding RDB to all the layers of current RDB, leading to a contiguous memory (CM) mechanism. Local feature fusion in RDB is then used to adaptively learn more effective features from preceding and current local features and stabilizes the training of wider network. After fully obtaining dense local features, we use global feature fusion to jointly and adaptively learn global hierarchical features in a holistic way. Experiments on benchmark datasets with different degradation models show that our RDN achieves favorable performance against state-of-the-art methods.

Convolutional neural network has recently achieved great success for image restoration (IR) and also offered hierarchical features. However, most deep CNN based IR models do not make full use of the hierarchical features from the original low-quality images, thereby achieving relatively-low performance. In this paper, we propose a novel residual dense network (RDN) to address this problem in IR. We fully exploit the hierarchical features from all the convolutional layers. Specifically, we propose residual dense block (RDB) to extract abundant local features via densely connected convolutional layers. RDB further allows direct connections from the state of preceding RDB to all the layers of current RDB, leading to a contiguous memory mechanism. To adaptively learn more effective features from preceding and current local features and stabilize the training of wider network, we proposed local feature fusion in RDB. After fully obtaining dense local features, we use global feature fusion to jointly and adaptively learn global hierarchical features in a holistic way. We demonstrate the effectiveness of RDN with three representative IR applications, single image super-resolution, Gaussian image denoising, and image compression artifact reduction. Experiments on benchmark datasets show that our RDN achieves favorable performance against state-of-the-art methods for each IR task.

During physiological processes molecules undergo chemical changes involving reducing and oxidizing reactions. A molecule with an unpaired electron can combine with a molecule capable of donating an electron. The donation of an electron is termed as oxidation whereas the gaining of an electron is called reduction. Reduction and oxidation can render the reduced molecule unstable and make it free to react with other molecules to cause damage to cellular and sub-cellular components such as membranes, proteins and DNA. In this paper, we have discussed the formation of reactive oxidant species originating from a variety of sources such as nitric oxide (NO) synthase (NOS), xanthine oxidases (XO), the cyclooxygenases, nicotinamide adenine dinucleotide phosphate (NAD(P)H) oxidase isoforms and metal-catalyzed reactions. In addition, we present a treatise on the physiological defences such as specialized enzymes and antioxidants that maintain reduction-oxidation (redox) balance. We have also given an account of how enzymes and antioxidants can be exhausted by the excessive production of reactive oxidant species (ROS) resulting in oxidative stress/nitrosative stress, a process that is an important mediator of cell damage. Important aspects of redox imbalance that triggers the activity of a number of signaling pathways including transcription factors activity, a process that is ubiquitous in cardiovascular disease related to ischemia/reperfusion injury have also been presented.

Antioxidant activities of the aqueous and ethanol extracts of pigeonpea [Cajanus cajan (L.) Millsp.] leaves, as well as petroleum ether, ethyl acetate, n-butanol and water fractions and the four main compounds separated from the ethanol extract, i.e. cajaninstilbene acid (3-hydroxy-4-prenylmethoxystilbene-2-carboxylic acid), pinostrobin, vitexin and orientin, were examined by a DPPH radical-scavenging assay and a β-carotene-linoleic acid test. In the DPPH system, the antioxidant activity of the ethanol extracts was superior to that of the aqueous extracts, with IC50 values were 242.01 and 404.91 µg/mL, respectively. Among the four fractions, the ethyl acetate one showed the highest scavenging activity, with an IC50 value of 194.98 µg/mL. Cajaninstilbene acid (302.12 µg/mL) and orientin (316.21 µg/mL) showed more efficient radical-scavenging abilities than pinostrobin and vitexin. In the β-carotene-linoleic acid test, the inhibition ratio (%) of the ethyl acetate fraction (94.13%±3.41%) was found to be the highest, being almost equal to the inhibition capacity of the positive control BHT (93.89%±1.45%) at 4 mg/mL. Pinostrobin (>500 µg/mL) and vitexin (>500 µg/mL) showed insignificant antioxidant activities compared with cajaninstilbene (321.53 µg/mL) and orientin (444.61 µg/mL). In general, the ethyl acetate fraction of the ethanol extract showed greater activity than the main compounds in both systems, such results might be attributed to the synergistic effects of the components. The antioxidant activities of all the tested samples were concentration-dependent. Based on the results obtained, we can conclude that the pigeonpea leaf extracts may be valuable natural antioxidant sources and are potentially applicable in both medicine and the healthy food industry.