WaveSegNet: Wavelet Transform and Multi-scale Focusing Network for Scrap Steel Segmentation

Scrap steel is a green renewable resource that can be infinitely recycled, and its recycling is of great significance in reducing carbon emissions and promoting the green transformation of the steel industry. However, the current scrap steel recycling faces a series of challenges, such as high labor intensity and occupational risks for inspectors, complex and diverse sources of scrap steel, varying types of materials, and difficulties in quantifying and standardizing manual visual inspection and rating. To overcome these challenges, we propose WaveSegNet, which is based on wavelet transform and multi-scale focusing structure for scrap steel segmentation. Firstly, we utilize wavelet transform to process images and extract features at different frequencies to capture details and structural information in the images. Secondly, we introduce a mechanism of multi-scale focusing to further enhance the accuracy of segmentation by extracting and perceiving features at different scales. Through experiments conducted on public dataset Cityscapes and scrap steel dataset, we have found that WaveSegNet demonstrates outstanding performance and efficiency in the field of semantic segmentation, outperforming other advanced models. These experimental results attest to the immense potential of WaveSegNet in intelligent rating and provide a new solution for the scrap steel recycling industry.