Tool wear monitoring needs high accuracy that can be done with electron microscope which needs long period of time. Instead, this research is to simplified the tool wear monitoring with image processing-based computer vision using Dino-Lite attached to robotic arm manipulator. The development uses OpenCV on Python with the following steps: (1) gathering the new and the broken tool images using Dino-Lite; (2) importing the image to Python and convert to HSV; (3) giving a noise reduction using Gaussian Blur; (4) giving a color detection to obtain masking of the HSV thresholding variable adjustment; (5) uses image Canny to detect contour area from the thresholding; (6) the new and the broken tool face area will be displayed; (7) these two values will be compared and generate the wear percentage. The image processing calculates the tool face area and the experiment uses the variation of Gaussian Blur for noise reduction, with the given values of 0, 1, 3, 5, 7, 9 11, 13, 15, 17. Few data cannot be obtained due to the unsupported image condition. The results show that the tool area on the images is more potential to be detected due to the increasing number of Gaussian Blur value.