Search

Food quality and safety oversight now urgently require the identification of similar food samples. Using infant rice cereals as an example, this study collected the Raman spectra of samples from various brands and used a statistical control chart method to reveal that samples from the same brand have controllable quality fluctuations. Additionally, samples from different brands also fall within the control limit, indicating a high degree of similarity between the samples. Under conditions of raw data, the extreme learning machine algorithm and Raman spectroscopy for sample identification demonstrated a recognition rate of 77.6%, suggesting that the machine learning algorithm has some identification effect. After optimizing the conditions, it was found that the recognition accuracy was significantly improved, reaching 99.9%, based on coif3 wavelet denoising and 0~0.1 normalization processing. During the experiment, it took 100 seconds to collect the Raman spectrum signal of a single sample and running the intelligent algorithm only took less than one second to obtain the calculation results. The optimization and identification methods proposed in this work have the advantages of efficiency and accuracy, which can provide a reference for the identification of similar samples. [ABSTRACT FROM AUTHOR]

High-resolution precipitation data is conducive to objectively describe the spatial-temporal variability of regional precipitation, and the study of downscaling techniques and spatial scale effects can provide technical and theoretical support to improve the spatial resolution and accuracy of satellite precipitation data. In this study, we used a machine learning algorithm combined with a regression algorithm RF-PLS (Random Forest-Partial Least Squares) to construct a downscaling model to obtain three types of high-resolution TRMM (Tropical Rainfall Measuring Mission) downscaled precipitation data for the years 2000–2017 at 250 m, 500 m, and 1 km. The scale effects with topographic and geomorphological features in the study area were analysed. Finally, we described the spatial and temporal variation of precipitation based on the optimal TRMM downscaled precipitation data. The results showed that: 1) The linear relationships between the TRMM downscaled precipitation data obtained by each of the three downscaled models (PLS, RF, and RF-PLS) and the precipitation at the observation stations were improved compared to the linear relationships between the original TRMM data and the precipitation at the observation stations. The accuracy of the RF-PLS model was better than the other two models. 2) Based on the RF-PLS model, the resolution of the TRMM data was increased to three different scales (250 m, 500 m, and 1 km), considering the scale effects with topographic and geomorphological features. The precipitation simulation effect with a spatial resolution of 500 m was better than the other two scales. 3) The annual precipitation was the highest in the areas with extremely high mountains, followed by the mediumhigh mountain, high mountain, medium mountain, medium-low mountain, plain, low mountain, and basin. [ABSTRACT FROM AUTHOR]

Raman spectroscopy can characterize the rich molecular vibration information of soybean milk powder samples, but difficulties arise in its direct use for sample classification and identification. Therefore, it is urgent to develop an intelligent identification technology based on Raman spectroscopy. For brand identification of soybean milk powder, this work investigates and discusses a variety of spectral processing technologies including wavelet denoising, normalization, principal component analysis, and the results show that appropriate spectral processing can improve the recognition accuracy of the random forest algorithm. Under the optimal conditions (db2 wavelet, normalization, principal component analysis, 30 decision trees), the best recognition effect of soybean milk brand identification can be achieved. The Raman spectral signal acquisition time of each sample is 40 s, and the spectra pretreatment and identification operation time only takes a few minutes. The analytical approach established in this paper has the advantages of convenient and fast Raman spectra acquisition, fast and accurate model identification. [ABSTRACT FROM AUTHOR]

The effect of integration time on Raman spectral data collected from fresh milk products was investigated and analyzed. The collected spectral data were denoised by wavelet transform to remove signal interference. Two-dimensional correlation Raman spectra of the fresh milk products were then constructed using two-dimensional correlation analysis and laser integrating time as an external perturbation. Finally, feature extraction was carried out, and the Euclidean distance was calculated. Thus, the similarities between fresh milk products of the same brand and different brands were quantitatively analyzed and calculated from Raman spectroscopy, two-dimensional correlation Raman spectroscopy, and feature extraction. The results showed that the quality of fresh milk products of the same brand fluctuated but maintained a high consistency. The differences between different brands of fresh milk products under the same detection conditions were much greater than within the same brand samples. [ABSTRACT FROM AUTHOR]

The dynamic Raman spectra of a brand of red wine (aa) as the main research object were collected over a range of laser integration times (1–5 s) to observe the changing trends of molecules in the wine under experimental conditions. The three-dimensional Raman characteristic spectrum of this wine was then constructed further by two-dimensional correlation fusion analysis. The fluctuations of the three-dimensional Raman spectra were also evaluated using a similarity algorithm. The correlation coefficients were 0.977 ± 0.011 and 0.990 ± 0.006 based on synchronous and asynchronous two-dimensional correlation Raman spectroscopy, respectively. These results suggested that the samples of wine aa were highly self-similar and could be effectively distinguished from two different brands of red wine (bb and cc) based on their different spectral responses. Therefore, this method has the potential to supplement existing methods for the classification analysis of red wine. [ABSTRACT FROM AUTHOR]

Raman spectra and ultraviolet-visible absorption spectra of four different geographic origins of Radix Astragali were collected. These data were analyzed using kernel principal component analysis combined with sparse representation classification. The results showed that the recognition rate reached 70.44% using Raman spectra for data input and 90.34% using ultraviolet-visible absorption spectra for data input. A new fusion method based on Raman combined with ultraviolet-visible data was investigated and the recognition rate was increased to 96.43%. The experimental results suggested that the proposed data fusion method effectively improved the utilization rate of the original data. [ABSTRACT FROM AUTHOR]

This study used 46 years of recent data, including glacial area, temperature, precipitation, and runoff data, to examine the glacier melting and its possible socioeconomic effects in the Manas River basin in western China. The average yearly change in the glaciated area in the Manas River basin for the entire study period was 0.41 %, and the glacier mass balance mainly keeps negative in the last 46 years. The negative glacial mass balance observed between 1986 and 2006 was 2.8 times greater than that for the period 1960-1985. Additionally, the amount of meltwater runoff was 78 % greater in 1986-2006 than in 1960-1985, with a mean depth of 478 mm year.Glacier melting and runoff in the Manas River basin during the late twentieth century were higher than at present. Annual meltwater volumes can reach 1 × 10 m, providing beneficial water resources to downstream areas. However, as the climate becomes warmer, the risk of meltwater flooding will also increase. Our calculations indicate that after the 2030s, the level of flooding risk will increase substantially. [ABSTRACT FROM AUTHOR]