Laplacian of Gaussian

Laplacian of Gaussian (LoG) is a computer vision and image processing technique used for edge detection and blob detection. It combines the Laplacian operator, which highlights regions of rapid intensity change, with Gaussian smoothing to reduce noise sensitivity. This results in a filter that identifies edges and blobs at various scales in an image.

Also known as: LoG, Marr-Hildreth operator, Mexican hat filter, Laplacian-Gaussian, LoG filter

Why learn Laplacian of Gaussian?

Developers should learn LoG when working on image analysis tasks requiring precise edge or blob detection, such as in medical imaging, object recognition, or feature extraction. It's particularly useful in scenarios where noise reduction is critical before edge detection, as the Gaussian smoothing step helps mitigate false positives from image artifacts.

Learning Resources

• docs: OpenCV Documentation: Laplacian
• tutorial: Image Processing: Laplacian of Gaussian (LoG)
• video: Laplacian of Gaussian (LoG) - Computer Vision Basics
• book: Digital Image Processing by Gonzalez and Woods