Filtering is a technique for modifying or enhancing an image. For example, you can filter an image to emphasize certain features or remove other features. Image processing operations implemented with filtering include smoothing, sharpening, and edge enhancement.
|Image Region Analyzer||Browse and filter connected components in an image|
Design Image Filters
Basic Image Filtering in the Spatial Domain
|N-D filtering of multidimensional images|
|Filter region of interest (ROI) in image|
|General sliding-neighborhood operations|
|2-D Gaussian filtering of images|
|3-D Gaussian filtering of 3-D images|
|2-D adaptive noise-removal filtering|
|2-D median filtering|
|3-D median filtering|
|2-D and 3-D mode filtering|
|2-D order-statistic filtering|
|Local standard deviation of image|
|Local range of image|
|Local entropy of grayscale image|
|2-D box filtering of images|
|3-D box filtering of 3-D images|
|Enhance elongated or tubular structures in image|
|Maximum of Frobenius norm of Hessian of matrix|
|Bilateral filtering of images with Gaussian kernels|
|Estimate parameters for anisotropic diffusion filtering|
|Anisotropic diffusion filtering of images|
|Guided filtering of images|
|Non-local means filtering of image|
|Create high-resolution image from set of low-resolution burst mode images|
Filtering By Property Characteristics
Integral Image Domain Filtering
Getting Started with Image Filtering in the Spatial Domain
In a spatially filtered image, the value of each output pixel is the weighted sum of neighboring input pixels. The weights are provided by a matrix called the convolution kernel or filter.
This example shows how to filter an image with a 5-by-5 averaging filter containing equal weights.
This example shows how to create a type of special filter called an unsharp masking filter, which makes edges and detail in an image appear sharper.
When a portion of the convolution or correlation kernel extends past the edge of an image, you can extrapolate image values by zero-padding the image or by replicating boundary pixels.
Noise refers to random error in pixel values acquired during image acquisition or transmission. Removing noise can improve image quality.
This example shows how to blur an image using Gaussian smoothing filters of different strengths. The example includes isotropic and anisotropic Gaussian filtering.
This example shows how to reduce noise associated with computing image gradients.
Guided image filtering performs edge-preserving smoothing on an image. It uses the content of a second image, called a guidance image, to influence the filtering.
This example shows how to reduce noise from an image while using a guidance image to preserve the sharpness of edges.
This example shows how to segment a hot object from the background in a thermographic image.
Integral Image Domain Filtering
Integral images are a quick way to represent images for filtering. In an integral image, the value of each pixel is the summation of the pixels above and to the left of it.
This example shows how to smooth an image by different amounts by applying box filters of varying sizes to the integral image.
Frequency Domain Filtering
You can design filters that modify the frequency content of images. Filtering in the frequency domain is often faster than filtering in the spatial domain.