A Detection System of Impurity in Transparent Liquid

In the fields of transparent liquid impurity detection based on machine vision technology, how to effectively detect impurities in the liquid is a difficult problem which has not yet been solved, mainly in the low recognition rate, the slow recognition speed, and the phenomenon of error detection and undetected. Therefore, this paper presents a new impurity detection method. Firstly, the hardware structure of the system is introduced in this paper. Then the flow diagram of impurity detection is presented. Finally, the algorithm of impurity detection is studied. Experiments show that the system introduced in this paper can identify impurities in liquid well on condition of ensuring the detection speed and detection accuracy.

Impurity Detection Using Machine Vision

In the fields of transparent liquid impurity detection based on machine vision technology, how to detect and identify a variety of impurities effectively in the liquid is a difficult problem which has not yet been solved, mainly in the low recognition rate, the slow recognition speed, and the phenomenon of error detection and undetected. Therefore, a new impurity detection method based on machine vision is presented in this paper. The experiments show that this method can decrease the recognition speed to 100ms for impurities’ diameter greater than 1mm and achieved a 99% recognition rate.

Structure and Magnetic Properties of Sm-Fe-N Prepared by Nitriding High Purity Sm2Fe17 Grown from Sm-Rich Melt

We report on a novel method for preparation of high purity Sm2Fe17phase by removing the liquid impurity phase from the Sm-rich precursor at high temperatures. With decreasing Sm addition, the percentage of liquid phase left as impurity is decreasing. Nitrogen absorption to the Sm2Fe17powders results in a complete formation of Sm2Fe17N3, which shows a weak disproportionation reaction at 773 K. The magnetic performance of the Sm2Fe17N3from precursors with different Sm content is improving with decreasing Sm content.