Peng Shan

To examine a potential correlation between food waste and lunar phases, we have devised a randomized controlled trial. The experiment spanned from 31 March to 10 July 2022, during which we employed the direct weighing method to collect 1903 valid data points on food waste. Utilizing propensity score matching, we meticulously controlled for various factors, including dining dates, the number of diners, dining times, spending levels, and store activities. The study revealed a close relationship between lunar phases and food waste. During the new moon phase, there was an increase in both orders and waste generated by consumers. Specifically, individuals, on average, squandered an additional 6.27% of animal protein (0.79 g), 24.5% of plant protein (1.26 g), 60.95% of starchy foods (3.86 g), and 61.09% of vegetables (5.12 g), resulting in an aggregate food waste of 32.14% (10.79 g). Conversely, during the full moon phase, consumers decreased their orders and subsequently decreased food waste. On average, individuals wasted 44.65% less animal protein (5.76 g), 43.36% less plant protein (2.5 g), 85.39% less seafood (0.73 g), and 8.43% less vegetables (0.93 g), resulting in a 20.52% (7.81 g) reduction in food waste. Furthermore, we validated our conclusions through various validation methods, including model replacement, to ensure robustness and reliability.

Recently, many spectral-special classification models have emerged one after another in the remote sensing community. These models aim to introduce the spatial information of the pixel to improve the accuracy of the class attribute of the pixel. However, for the spectral-spatial classification algorithms, not all pixels need to introduce the corresponding spatial information since the use of a large amount of spatial information has a costly time. To solve this problem, this paper proposes a robust dual-stage spatial embedding (RDSSE) model for spectral-spatial classification of hyperspectral image, which is composed of the following several main steps: First, an over-segmentation algorithm is employed to cluster original hyperspectral image into many superpixel blocks with shape adaptive characteristics. Then, we design a k-peak criterion to fuse the spectral feature of pixels within and between superpixels. Next, a low time-consumption spectral classifier is introduced to conduct primary classification for a testing pixel to achieve the corresponding probability distribution. Specifically, the difference between the probability of the largest class and that of the second-largest class is served as class confidence. Finally, the predicted label of the low-confidence testing pixels is reclassified based on a high-accuracy spectral-spatial classification method. Experimental results on several real images illustrated that the proposed RDSSE can obtain superior performance with respect to several competitive methods.