Abstract:The judgment of fuming furnace's smelting terminal for long time depends on primarily the spot worker's observation on flames through their naked eyes. The accuracy of the judgment is easily and strongly influenced by subjective factors, which makes it difficult to guarantee the stability of the products. Therefore, this paper put forward a method for judging smelting terminal, which observes flames based on image processing technique through the computer and discriminates flames with a neural network. The testing result shows that this method can effectively discriminate fuming furnace's smelting terminal.