论文部分内容阅读
目的压缩感知方法通过混样过程来减少样品的检测次数,从而提高检测效率,降低检测成本,缩短检测时间。方法利用食品中污染物超标数据稀疏性的特点,将压缩感知方法应用于食品安全风险监测的样品检测。该理论的核心思想是通过混合待检测的样品,得到远少于原样品数的检测次数,然后根据相应重构算法由测量值重构原始数据。该算法可采用R统计软件实现。结果用压缩感知方法重构125份原始样品的检测值,误差平方和为3.782 652×10~(-29),其中原始样品中117份低于检出限的样品全部精准重构,高于检出限的8份样品压缩感知重构值稍稍大于真实值,但误差极小,可以忽略不计。结论压缩感知方法可以通过混合样品来减少样品的检测次数,并可由少数检测值重构每一个原始样品的食品污染物含量。
The purpose of compression sensing method through the sample mixing process to reduce the number of test samples, thereby enhancing the detection efficiency, reduce detection costs and shorten the detection time. Methods The method of compressive sensing was applied to the detection of food safety risk by taking advantage of the sparseness of data in food contaminants. The core idea of this theory is that by mixing the samples to be detected, the number of detections is far less than the number of original samples, and then the original data is reconstructed from the measured values according to the corresponding reconstruction algorithm. The algorithm can be used R statistical software. Results The compressive sensing method was used to reconstruct the original value of 125 samples. The sum of squared errors was 3.782 652 × 10 ~ (-29). All of the 117 samples below the detection limit were accurately reconstructed, Out of the eight samples, the compressed sensing reconstruction value is slightly larger than the true value, but the error is very small and can be neglected. Conclusion Compressed sensing methods reduce the number of sample detections by mixing samples and reconstruct the food contamination content of each original sample from a few detections.