An adaptive image edge detection algorithm was developed to improve edge detection accuracy. By combining global and local edge detection methods, the algorithm effectively extracts complete and rich edges without false edges. The global edge detection uses an adaptive smooth filter algorithm based on the Canny operator, while the local edge detection employs a distance weighted average method based on the k-average method. This combined approach results in powerful anti-noise ability and more effective image edge extraction compared to traditional methods like the Canny operator.