Tool safety is an important part of machining and machine tool safety, and machine tool path image detection can effectively obtain the in-machine condition of a tool. To obtain an accurate image edge and improve image processing accuracy, a novel subpixel edge detection method is proposed in this study. The precontour is segmented by binarization, the second derivative in the neighborhood of the demand point is calculated, and the obtained value is sampled according to the specified rules for curve fitting. The point whose curve ordinate is 0 is the subpixel position. The experiment proves that an improved subpixel edge can be obtained. Results show that the proposed method can extract a satisfactory subpixel contour, which is more accurate and reliable than the edge results obtained by several current pixel-level operators, such as the Canny operator, and can be used in edge detection with high-accuracy requirements, such as the contour detection of online tools.

The current work focuses on the cutting tool condition monitoring of end milling based on direct and indirect approach in machining AISI H13 alloy steel. Indirect process parameters such as cutting force signals are measured as responses using force sensor. In order to successfully inspect the milling tool life online for direct approaches, an automated machine vision system was used for tool condition monitoring. The image processing algorithms are developed to extract different features of rotating milling tool. A detection and compensation system for tool wear based on machine vision is designed. Feedforward Back-Propagation Neural Network applied for tool wear classification developed based on many force features. Ten time-domain features extracted and the sensitive features is determined based on Pearson’s correlation coefficient. I-kaz method which integrates between kurtosis and standard deviation is added as input feature with the ten time-domain features. A strong correlation is established between most of time-domain features and tool wear with high correlation coefficient. ANN model applied for classification tool states as normal and abnormal. Experiments with vision system have shown that area of wear at bottom and flank is suitable to inspect in-process. Actual measurements of the tool wear stages are possible to identify the abnormality in cutting using vision system. ANN model showed superior results for tool states classification. The mean squared Error (MSE) for classification model was less than 6.7E-09 and R equal to 1. The model can be used to construct fault estimation mode for tool wear online classification and inspection.