Detection Method of Diffusion Effect Based on Steel Liquid Image
The researchers proposed two detection indexes of particle size and uniformity,and established a detection process based on the YOLOv8-Seg model to solve the problem of lack of objectivity in detecting the diffu-sion effect of steel liquid in vacuum casting.Firstly,the researchers used CLAHE and BPDFHE combined with a processing method based on low-pass filtering and sum to preprocess the steel liquid image,and used YOLOv8-Seg to segment the steel liquid.Then,the researchers detected the granularity and evenness of the liquid steel by calcu-lating parameters such as the amount of edge pixels and the amount of binary image pixels in the processed image.The results show that,by comparing with the evaluation results of several engineers,in terms of uniformity index,the results of this method are closer to the results of engineers;in terms of the granularity index,the results are still acceptable though there are some differences.
万方数据
维普
