An Automated Dual-Threshold Wavelet Transform Edge Detection Method with Automatic Threshold Setting

This paper presents a novel edge detection method based on dual-threshold wavelet transformation. The approach incorporates automatic threshold determination capabilities, significantly improving edge detection accuracy and stability. By employing dual-threshold technology, the method effectively captures edge information from images and accurately marks boundary locations. Key implementation aspects include wavelet coefficient analysis for multi-scale edge representation and adaptive threshold calculation using statistical properties of wavelet coefficients. The algorithm first performs 2D wavelet decomposition to obtain horizontal, vertical, and diagonal detail coefficients, then applies automatically calculated high and low thresholds to distinguish strong and weak edges. Edge linking techniques connect discontinuous edges while suppressing false positives. One major advantage is its implementation simplicity, allowing researchers to easily understand and deploy the method using standard wavelet functions (e.g., Daubechies or Symlets wavelets) and basic thresholding operations. We believe this advanced edge detection methodology will significantly contribute to image processing领域 and positively impact related research developments.