Developing new antioxidants and using natural examples is of current interest. This study evaluated the antioxidant activities and the ability to inhibit soybean oil oxidation of oat oil obtained with different solvents. Oat oil extract obtained by ethanol extraction gave the highest antioxidant activity with a DPPH radical (1,1-diphenyl-2-picrylhydrazyl) scavenging activity of 88.2 % and a reducing power (A 700) of 0.83. Oat oil extracted by ethanol contained the highest polyphenol and α-tocopherol content. Significant correlation was observed between the total polyphenol contents, individual phenolic acid, α-tocopherol, and DPPH radical scavenging activity. Soybean oil with 2 % added oat oil showed low malondialdehyde content (8.35 mmol mL(-1)), suggesting that the added oat oil inhibited oxidation. Oat oil showed good antioxidant activity, especially when extracted with ethanol which could also retard the oxidation of soybean oil . DPPH radical scavenging activity was the best method to evaluate the antioxidant activity and components of oat oil.

Staircases are some of the most common building structures in urban environments. Stair detection is an important task for various applications, including the environmental perception of exoskeleton robots, humanoid robots, and rescue robots and the navigation of visually impaired people. Most existing stair detection algorithms have difficulty dealing with the diversity of stair structure materials, extreme light and serious occlusion. Inspired by human perception, we propose an end-to-end method based on deep learning. Specifically, we treat the process of stair line detection as a multitask involving coarse-grained semantic segmentation and object detection. The input images are divided into cells, and a simple neural network is used to judge whether each cell contains stair lines. For cells containing stair lines, the locations of the stair lines relative to each cell are regressed. Extensive experiments on our dataset show that our method can achieve 81.49$$\%$$ % accuracy, 81.91$$\%$$ % recall and 12.48 ms runtime, and our method has higher performance in terms of both speed and accuracy than previous methods. A lightweight version can even achieve 300+ frames per second with the same resolution.

D. Du et al. related fields. The collected dataset is formed by 3, 360 images, including 2, 460 images for training, and 900 images for testing. Specifically, we manually annotate persons with points in each video frame. There are 14 algorithms from 15 institutes submitted to the VisDrone-CC2020 Challenge. We provide a detailed analysis of the evaluation results and conclude the challenge. More information can be found at the website: http://www.aiskyeye.com/.

Lower extremity robotic exoskeletons (LEEX) can not only improve the ability of the human body but also provide healing treatment for people with lower extremity dysfunction. There are a wide range of application needs and development prospects in the military, industry, medical treatment, consumption and other fields, which has aroused widespread concern in society. This paper attempts to review LEEX technical development. First, the history of LEEX is briefly traced. Second, based on existing research, LEEX is classified according to auxiliary body parts, structural forms, functions and fields, and typical LEEX prototypes and products are introduced. Then, the latest key technologies are analyzed and summarized, and the research contents, such as bionic structure and driving characteristics, human–robot interaction (HRI) and intent-awareness, intelligent control strategy, and evaluation method of power-assisted walking efficiency, are described in detail. Finally, existing LEEX problems and challenges are analyzed, a future development trend is proposed, and a multidisciplinary development direction of the key technology is provided.

Unlike existing work in deep neural network (DNN) graphs optimization for inference performance, we explore DNN graph optimization for energy awareness and savings for power-and resource-constrained machine learning devices. We present a method that allows users to optimize energy consumption or balance between energy and inference performance for DNN graphs. This method efficiently searches through the space of equivalent graphs, and identifies a graph and the corresponding algorithms that incur the least cost in execution. We implement the method and evaluate it with multiple DNN models on a GPU-based machine. Results show that our method achieves significant energy savings, i.e., 24% with negligible performance impact.