SoC FPGA Accelerated Sub-Optimized Binary Fully Convolutional Neural Network for Robotic Floor Region Segmentation.

In this article, a new Binary Fully Convolutional Neural Network (B-FCN) based on Taguchi method sub-optimization for the segmentation of robotic floor regions, which can precisely distinguish floor regions in complex indoor environments is proposed. This methodology is quite suitable for robot vision in an embedded platform and the segmentation accuracy is up to 84.80% on average. A total of 6000 training datasets were used to improve the accuracy and reach convergence. On the other hand, to reach real-time computation, a PYNQ FPGA platform with heterogeneous computing acceleration was used to accelerate the proposed B-FCN architecture. Overall, robots would benefit from better navigation and route planning in our approach. The FPGA synthesis of our binarization method indicates an efficient reduction in the BRAM size to 0.5-1% and also GOPS/W is sufficiently high. Notably, the proposed faster architecture is ideal for low power embedded devices that need to solve the shortest path problem, path searching, and motion planning.

Antioxidant activity and hepatoprotective effect of 10 medicinal herbs on CCl4-induced liver injury in mice.

BACKGROUND: Many natural products confer health benefits against diverse diseases through their antioxidant activities. Carbon tetrachloride (CCl4) is often used in animal experiments to study the effects of substances on liver injury and the related mechanisms of action, among which oxidative stress is a major pathogenic factor. AIM: To compare antioxidant and hepatoprotective activities of ten herbs and identify and quantify phytochemicals for the one with strongest hepatoprotection. METHODS: The antioxidant activity of ten medicinal herbs was determined by both ferric-reducing antioxidant power and Trolox equivalent antioxidant capacity assays. The total phenolic and flavonoid contents were determined by Folin-Ciocalteu method and aluminum chloride colorimetry, respectively. Their effects on CCl4-induced oxidative liver injury were evaluated and compared in a mouse model by administrating each water extract (0.15 g/mL, 10 mL/kg) once per day for seven consecutive days and a dose of CCl4 solution in olive oil (8%, v/v, 10 mL/kg). The herb with the strongest hepatoprotective performance was analyzed for the detailed bioactive components by using high-performance liquid chromatography-electrospray ionization source-ion trap tandem mass spectrometry. RESULTS: The results revealed that all tested herbs attenuated CCl4-induced oxidative liver injury; each resulted in significant decreases in levels of serum alanine transaminase, aspartate transaminase, alkaline phosphatase, and triacylglycerols. In addition, most herbs restored hepatic superoxide dismutase and catalase activities, glutathione levels, and reduced malondialdehyde levels. Sanguisorba officinalis (S. officinalis) L., Coptis chinensis Franch., and Pueraria lobata (Willd.) Ohwi root were the three most effective herbs, and S. officinalis L. exhibited the strongest hepatoprotective effect. Nine active components were identified in S. officinalis L. Gallic acid and (+)-catechin were quantified (7.86 ± 0.45 mg/g and 8.19 ± 0.57 mg/g dried weight, respectively). Furthermore, the tested herbs displayed a range of in vitro antioxidant activities proportional to their phenolic content; the strongest activities were also found for S. officinalis L. CONCLUSION: This study is of value to assist the selection of more effective natural products for direct consumption and the development of nutraceuticals or therapeutics to manage oxidative stress-related diseases.

Influence of Grape Berry Maturity on Juice and Base Wine Composition and Foaming Properties of Sparkling Wines from the Champagne Region.

In sparkling wine cool-climate regions like Champagne, it is sometimes necessary to pick the healthy grape clusters that have a relatively low maturity level to avoid the deleterious effects of Botrytis cinerea. In such conditions, we know that classical oenological parameters (sugars, pH, total acidity) may change but there is little information concerning the impact of grape berry maturity on wine proteins and foaming properties. Therefore, healthy grapes (Chardonnay and Pinot meunier) in 2015 and 2016 were picked at different maturity levels within the range of common industrial maturity for potential alcohol content 8⁻11% v/v in the Champagne region. Base wine protein content and foamability, and oenological parameters in grape juice and their corresponding base wines, were investigated. The results showed that base wine protein contents (analyzed by the Bradford method and by electrophoresis) and foamability were higher when the grapes were riper. The Pearson's correlation test found significant positive correlations (r = 0.890⁻0.997, p < 0.05) between Chardonnay grape berry maturity degree (MD) and base wine foamability in both vintages. Strong correlations between MD and most of the oenological parameters in grape juice and base wine were also found for the two cultivars. Under the premise of guaranteed grape health, delaying harvest date is an oenological decision capable of improving base wine protein content and foamability.