An Improved Image Filtering Method for Weld Bead Inspection using Unsharp Masking Technique

There are many disturbances that occur during image capturing process. One of the common disturbances is noise. Consequently, various methods were developed to improve image quality. In this study, the proposed method consists of an enhanced Unsharp Masking Technique that is combined with common filtering methods. The method was applied in different noise situations. The image filtering methods involved were common filter which Mean, Median and Gaussian filters. The images of welding process were converted into RGB for ease of calculation. Mean Square Error (MSE) and Peak Signal to Noise Ratio (PSNR) were used to determine the quality of the image. Graph generated to reinforce the PSNR value. The results obtained proved that the proposed method could yield better filtering performance.