1. 3x3 spatial convolution filters

Laplace Filter

Here the brightness values of the neighboring pixels are subtracted from the central pixel. Consequently, in region of the image that is uniform in brightness, the result is to reduce the grey level to zero. When a discontinuity is present within the neighborhood in the form of a point, line or edge, the result of this filter is a non-zero value.

High Pass Filter

High pass filtering is used to enhance edges between different regions as well as to "sharpen" an image. This is accomplished using a kernel with a high central value, typically surrounded by negative weights. High pass filter remove the local mean, and output the measure of deviation of the input signal from the local mean. The High Pass filter replaces the center pixel with a value that significantly increases its contrast from its neighbors. The HiPass filter leaves only elements of high contrast.

Low Pass Filter

Low pass filtering is used for smoothing the image. For all the pixels inside the kernel, low pas filter will take the average DN of all these pixels and output to output image. Low pass filter is a linear filter used for image enhancement and sensor simulation. It preserves the local mean of the image, but it decreases the spatial resolution of the image. If image interpretation doesn't very high spatial resolution, low pass filter is usually used for filtering noise by averaging the DN value of the noisy pixel with the "good" pixels. The Low Pass filter replaces the center pixel with the mean value in its neighborhood. The Low Pass filter can also be used to remove noise..

2. Morphing filters

Morphing filters may be used for removing artifacts or enhancing, dilating or eroding edges from the image. Notice how the straight lines are removed from the above image by using a Closing filter.

Window size determines the filter strength.