In the previous chapters, we covered how many of the common techniques, used for image processing, can be implemented using MATLAB. This chapter is a little bit more advanced in terms of the complexity of the algorithms covered, but we will present them in a gentle, introductory manner. The results will be more than interesting, since you will get to learn how to mix, or combine multiple images to create new ones, either for scientific purposes, or for pure artistic results. More specifically, in this chapter, we will cover how you can work with multispectral images in MATLAB, to unveil hidden details. We will move on to panorama stitching for the creation of beautiful panoramic images and then we will show blending techniques to combine masked areas in an image with another image. Finally, we will cover the production of High Dynamic Range (HDR) images using MATLAB. All these techniques will be demonstrated in the simplest possible ways, using hands-on examples.
In this chapter, we will cover:
- What the importance of mixing images is
- What multispectral images are and how we can manipulate them in MATLAB
- How we can create composite images in MATLAB
- How we can blend selected areas of images to create interesting visual results
- What High Dynamic Range (HDR) images are and how we can create and process them in MATLAB
- How we can stitch images to create panoramas in MATLAB
So, let's get started!
An obvious question for a beginner in the field of image processing, would be about what the importance of mixing images is. The answer is that the basic idea behind mixing or combining images is to enrich them and achieve the following results:
- Enhance the information included in multimodal images, that is, images acquired from different sensors or scanners, to detect regions of interest. This is a technique often used in medical imaging applications, such as brain CT/MRI images or body PET/CT images.
- See more than the eye can see, by combining images of the same subject, taken at different frequencies of the electromagnetic spectrum. This is a way to extract information that is not normally visible to humans, such as for example, infrared wavelengths.
- Blend two or more images together to end up with an artistic result.
- Combine images of the same subject taken using different exposure levels, to end up with a resulting image with a higher dynamic range.
- Stitch together multiple images that partially overlap each other, to produce a larger, panoramic image of the photographed scene.
In the rest of this chapter, we will show how we can achieve many of the examples mentioned previously, using MATLAB.