Image Smoothing

Type: personal

Description: Image smoothing ameliorates the effect of high-frequency spatial noise from an image. It is also an important step used even in advanced critical applications like medical image processing, making operations like derivative computation numerically stable. For this beginner-level image processing project, you can implement Gaussian smoothing. To do so, you will need to create a 2-dimensional Gaussian kernel (possibly from one-dimensional kernels using the outer product) by employing the NumPy library and then convoluting it over the padded image of your choice. The above output has been obtained from the scikit-image with the Multi-dimensional Gaussian filter used for smoothing. Observe how the ‘sharpness' of the edges is lost after the smoothing operation in this image processing project. The smoothing process can also be performed on the RGB image. However, a grayscale image has been used here for simplicity. Image smoothing ameliorates the effect of high-frequency spatial noise from an image. It is also an important step used even in advanced critical applications like medical image processing, making operations like derivative computation numerically stable. For this beginner-level image processing project, you can implement Gaussian smoothing. To do so, you will need to create a 2-dimensional Gaussian kernel (possibly from one-dimensional kernels using the outer product) by employing the NumPy library and then convoluting it over the padded image of your choice. The above output has been obtained from the scikit-image with the Multi-dimensional Gaussian filter used for smoothing. Observe how the ‘sharpness' of the edges is lost after the smoothing operation in this image processing project. The smoothing process can also be performed on the RGB image. However, a grayscale image has been used here for simplicity.

GitHub: https://github.com/Lymphatus/caesium-image-compressor

Live: https://imagecompressor.com/2