Imaging Workshop

Below are our experiments and finding for the Imaging Workshop, which pertains to image and video analysis, using both hardware and software techniques.

The original assignment can be found here.

Convolution 1

A convolution is the process of adding each element of the image to its local neighbors, weighted by the kernel. In this case, the kernel is the identity, so...

Convolution 2

This is a convolution using the edge detection kernel:

Convolution 3

This is a convolution using the sharpen kernel:

Convolution 4

This is a convolution using the box blur kernel:

Convolution 5 (Hardware)

Now, here’s an example similar to the last one, but using a video sample. By clicking you can switch the kernel here, too:

Convolution 5 (Hardware)

The previous convolutions were made by software, this example uses hardware, and by clicking on the image you can switch the kernel using the examples mentio...

Grayscale Average & Luma

Here we have an image, to which we apply a gray scale filter using the both the raw average of the RGB channels and the Luma weighted average for each pixel.

Grayscale Video

Here we have a grayscale filter:

Grayscale Histogram

This sketch shows the histogram of a sample grayscale image.

Grayscale Luma

Here we have a gray scale filter using the Luma weighted average of the RAW values for each pixel:

Grayscale RAW Average

Here we have a gray scale filter using the average of the RAW values for each pixel:

Threshold Video

Here we have a threshold filter using threshold:

Grayscale Threshold

Here we have a gray scale filter using thresshold: