Study on Detection Technology of Milk Powder Based on Support Vector Machines and Near Infrared Spectroscopy

摘要

This paper presents a novel classifier to identify standard and sub-standard milk powder, which is built by support vector machines (SVM) and near infrared spectroscopy (NIR). The training set is composed of 38 samples and the testing set is composed of 12 samples. The correct classification ratio of the training set is up to 100%, while that of the testing set is up to 100%. The result indicates that the combination of SVM and NIR can be used as a fast, convenient, and safe technology to identify standard and sub-standard milk powder.