This chapter described the use of PyCUDA in the development of simple computer vision applications. It described the use of PyCUDA in calculating the histogram of an array. The histogram is a very important statistical global feature of an image that can be used to find out important information about it. The concept of atomic operations and shared memory was explained in detail, using histogram calculation as an example. Images in Python are stored as numpy arrays, so manipulating images in PyCUDA is similar to modifying multidimensional numpy arrays. This chapter described the use of PyCUDA in various basic computer vision applications, such as image addition, image inversion, and color space conversion. The concepts described in this chapter can be utilized for developing complex computer vision applications using PyCUDA.
This chapter also marks an end to this book...
