Search

This paper proposes a lightweight residual network for highly accurate recognition of the numbers of wheel dials. The proposed residual network is based on an optimized ResNet-18 network with the depthwise separable convolution and Dropout function. In addition, a case study on automatic water meter readings based on captured images is presented. Compared to existing methods, the results demonstrate improved recognition accuracy as high as 99.8% and increased computation speed with reduced model parameter size.

A new natural product named longanlactone was isolated from Dimocarpus longan Lour. seeds. Its structure was determined as 3-(2-acetyl-1H--pyrrol-1-yl)-5-(prop-2-yn-1-yl)dihydrofuran-2(3H)-one by spectroscopic methods and HRESIMS.

Five previously uncharacterised polyphenols, ethyl gallate (2), 1-β-O-galloyl- d -glucopyranose (3), methyl brevifolin carboxylate (4), brevifolin (5) and 4-O-α- l -rhamnopyranosyl-ellagic acid (8), and three previously identified polyphenols, gallic acid (1), corilagin (6) and ellagic acid (7), were isolated from longan seeds. Their structures were identified by spectroscopic and chemical methods including HRESIMS and NMR. Eight polyphenols exhibited scavenging activity towards DPPH radicals with SC50 values of 0.80–5.91 μg/ml and towards superoxide radicals with SC50 values of 1.04–7.03 μg/ml. The radical-scavenging activity of the newly characterised polyphenols was comparable to that of gallic acid, corilagin and ellagic acid.